Identifying transdiagnostic predictors and mechanisms of ...

Identifying transdiagnostic predictors and mechanisms of treatment response to repetitive transcranial magnetic

stimulation over the dorsomedial prefrontal cortex

by

Katharine Dunlop

A thesis submitted in conformity with the requirements for the degree of Doctor of Philosophy

Institute of Medical Science University of Toronto

© Copyright by Katharine Dunlop, 2018

ii

Identifying transdiagnostic predictors and mechanisms of treatment response to repetitive transcranial magnetic stimulation

over the dorsomedial prefrontal cortex Katharine Dunlop

Doctor of Philosophy Institute of Medical Science

University of Toronto 2018

Abstract

There are new treatment options, such as repetitive transcranial magnetic stimulation

(rTMS), for patients with longstanding, treatment-resistant mental illness. Identifying brain

markers predictive of, and changes associated with, rTMS treatment response could improve

patient selection and thus outcomes to this intervention. Transdiagnostic abnormalities in brain

networks for attention (e.g., salience network, SN) and cognitive control have been described,

and the dorsomedial prefrontal cortex (dmPFC) as a potential therapeutic target. The aim of this

thesis was to identify pre-treatment abnormalities of the SN and dmPFC that predicted and

changed with response to dmPFC-rTMS, across three psychiatric disorders: eating disorders

(ED), obsessive-compulsive disorder (OCD), and treatment-resistant depression (TRD). All three

patient groups showed significant clinical benefit following dmPFC-rTMS, although TRD

response was not superior to placebo. Baseline dmPFC resting state functional MRI functional

connectivity (FC) in the SN correlated with the clinical response to dmPFC-rTMS across all

three disorders. FC change in the SN also accompanied symptomatic improvement in OCD and

ED. In TRD patients who received either active or placebo dmPFC-rTMS, clinical response was

associated with normalized between-network functional connectivity between the SN and the

ventromedial network (VMN), implicated in emotion processing. These findings emphasize the

role of the SN as a transdiagnostic substrate of psychiatric disorders, while at the same time

iii

underscoring the significance of inter-individual variability in SN FC for predicting response and

characterizing mechanisms of improvement on rTMS. These findings also highlight the

importance of FC changes between networks responsible for emotion processing and cognitive

control associated with treatment-non-specific symptom improvement. Looking forward, these

results have important implications for understanding common elements of psychopathology

across many psychiatric disorders, and the neurobiological mechanisms of rTMS and other focal

non-invasive brain stimulation techniques.

iv

Acknowledgments I would like to extend my deepest gratitude to my supervisor, Jonathan Downar, for his patience,

generous support and mentorship. I feel privileged to have worked alongside you, and you have

given me every opportunity to learn, collaborate with others, and grow into my own as an

academic. I would also like to acknowledge my committee members, Drs. Karen Davis and

Blake Woodside, for their guidance and unwavering support throughout my degree and in

preparation of this thesis. Thank you for everything!

Translational and clinical research takes a village, and so I am forever indebted to the

outstanding clinical team at the TWH MRI-Guided rTMS Clinic. Huge thank-yous to Sunny for

tirelessly booking and rebooking patients; to Jack, Mike and Iggy for treating all of the patients

in the RCT; and to Terri, Eileen, and Vanathy for managing the administrative aspects of my

research.

I am grateful for the support of the Canadian Institutes of Health Research in awarding me with a

Vanier Canada Graduate Scholarship, and the support of the University of Toronto Collaborative

Program in Neuroscience in awarding me the Jonathan Dostrovsky Award in Neuroscience.

I must also thank my neuroimaging mentors, Drs. Tim Salomons, Adrian Crawley, and Massieh

Moayedi for their support. It was such an honour to have worked alongside you.

To my “Zappies” – thank you and love you guys! It was such a pleasure to have worked with

each of you.

To the patients and healthy controls who dedicated their time in these studies, I thank you for

your kindness, your time, and your support.

Finally, I must acknowledge my cheerleaders: my husband (Sam), parents (Hamish and Martha),

and brother (Graeme) for their unfaltering love and support throughout my degree. Sam – thank

you for your patience, for challenging me through lively discussion and debate, and for your

support in editing my writing.

v

Contributions This project was possible thanks to the contribution of the following individuals:

Dr. Jonathan Downar (supervisor), who provided mentorship and guidance well before my

acceptance into the Institute of Medical Science; oversaw and guided my academic progress;

directed the UHN MRI-Guided rTMS Clinic; recruited and assessed patients involved in this

research; assisted me with the successful completion of this thesis.

Drs. Karen Davis and Blake Woodside (committee members), who offered their support and

mentorship as committee members and collaborators; oversaw the progress of my thesis and

development as an independent scientist.

Drs. Peter Giacobbe, Jeff Daskalakis, Daniel Blumberger, Fidel Vila-Rodriguez, Sidney

Kennedy, Ray Lam, Alastair Flint, Marion Olmsted, and Patricia Colton (PI’s and co-I’s),

who provided guidance as principal investigators and co-investigators for the studies in this

project; referred patients to the UHN MRI-Guided rTMS Clinic.

Drs. Tim Salomons, Massieh Moayedi, Adrian Crawley, and Aaron Kucyi (neuroimaging

mentors), who generously shared their expertise on neuroimaging.

Ms. Terri Cairo and Ms. Eileen Lam (clinical research managers), who helped develop the

clinical framework for Study 1; managed the research lab’s administrative duties.

Ms. Sunny Hong (clinic administrative assistant), who painstakingly scheduled all patients

involved in this research.

Ms. Vanathy Niranjan, Ms. Aisha Dar, Mr. Bernard Ma, Mr. Jack Shen, Mr. Michael

Aiello, Mr. Iggy Uwadiae, and Dr. Umar Dar (rTMS treatments), who provided rTMS

treatments for patients involved in these studies.

Mr. Keith Ta and Mr. Eugen Hlasny (MRI technicians), who acquired neuroimaging data

from all patients and healthy controls.

vi

Dr. Nathan Bakker, Ms. Sarah Peters, Ms. Laura Schulze, Mr. Farrokh Mansouri, Mr.

Peter Fettes, and Mr. Arsalan Mir-Moghtadaei (current and former graduate students), who

assisted with data collection, particularly with all healthy control recruitment and as back-up

assessors for Study 1.

My own contributions included: Study design and data collection for patients and healthy

controls in study 1; healthy control data collection in study 2; neuroimaging, psychometric, and

behavioural data analysis design for all studies 1-3; quality control and inspection for all

anatomical and functional scans, quality control of all clinical and psychometric data;

preprocessing all neuroimaging data using FSL, SPM and MATLAB; performing statistical tests

for all clinical, behavioural, psychometric, and neuroimaging data from all three studies;

visualizing and interpreting all results; and writing this thesis.

This work was funded in part by the Canadian Institutes of Health Research Vanier Scholarship

(2015 - 2018), a School of Graduate Studies Conference Grant, the Collaborative Program in

Neuroscience Jonathan Dostrovsky Award in Neuroscience, a UHN Office of Research Trainees

Conference Travel Award, a Society for Neuroscience Trainee Professional Development

Award, and a WCBR Travel Fellowship.

vii

Table of Contents ACKNOWLEDGMENTS........................................................................................................................................IV

CONTRIBUTIONS.................................................................................................................................................V

TABLEOFCONTENTS.........................................................................................................................................VII

LISTOFTABLES..................................................................................................................................................XII

LISTOFFIGURES...............................................................................................................................................XIII

LISTOFABBREVIATIONS...................................................................................................................................XVI

LITERATUREREVIEW...................................................................................................................................1

INTRODUCTION&GENERALAIMS......................................................................................................................1 INTRINSICFUNCTIONALNETWORKSOFTHEBRAIN.................................................................................................3

HistoricalandPhilosophicalPerspectivesofBrainConnectivity............................................................3

WhatareIntrinsicBrainNetworks?.......................................................................................................5 RelationshipofFunctionalNetworkstoBrainStructure........................................................................5 RelationshipofFunctionalNetworkstoHumanBehaviour...................................................................6 RelationshipofFunctionalNetworkstoElectrophysiologicalStudies...................................................7

FunctionalNetworksPertinenttotheNeurobiologyofPsychiatricDisorders.......................................9 HowdoIBNsIntegrateInformationAcrossNetworks?.......................................................................23

PSYCHIATRICDISORDERSINTHECONTEXTOFBRAINNETWORKCONNECTIVITY.........................................................30

DefiningAbnormalBrainConnectivity.................................................................................................30 TransdiagnosticAlterationsofBrainNetworks...................................................................................31

PSYCHOPATHOLOGY1:MAJORDEPRESSIVEDISORDER.........................................................................................34

MDDDiagnosis....................................................................................................................................35 MDDPrevalence..................................................................................................................................36 MDDEtiology.......................................................................................................................................36 MDDComorbidity................................................................................................................................37

AssessmentofTreatmentSeverity......................................................................................................37 StructuralandFunctionalDisruptionsinMDD....................................................................................39

PSYCHOPATHOLOGY2:OBSESSIVE-COMPULSIVEDISORDER...................................................................................42

OCDDiagnosis.....................................................................................................................................43 OCDPrevalence...................................................................................................................................43 OCDEtiology........................................................................................................................................44 OCDComorbidity.................................................................................................................................44

AssessmentofSymptomSeverity........................................................................................................45

viii

StructuralandFunctionalAbnormalitiesofOCD.................................................................................45 PSYCHOPATHOLOGY3:ANOREXIANERVOSAANDBULIMIANERVOSA......................................................................47

DiagnosisofAnorexiaandBulimiaNervosa........................................................................................48 ANandBNPrevalence.........................................................................................................................49 EtiologyofANandBN.........................................................................................................................50 ANandBNComorbidity.......................................................................................................................50

AssessmentofSymptomSeverity........................................................................................................51 StructuralandFunctionalAbnormalitiesinANandBN.......................................................................52

TRANSCRANIALMAGNETICSTIMULATIONASAPROBEOFIBNS..............................................................................55

WhatisTranscranialMagneticStimulation?......................................................................................55 ApplicationsofrTMSasaNetwork-Probe...........................................................................................60 TherapeuticApplicationsofrTMS........................................................................................................63

CLINICALAPPLICATIONSOFRTMS....................................................................................................................64 TreatmentResistantMDD...................................................................................................................64 Obsessive-CompulsiveDisorder...........................................................................................................71 AnorexiaandBulimiaNervosa............................................................................................................75

IMPROVINGRTMSRESPONSE.........................................................................................................................79 NumberofTotalTreatmentSessions..................................................................................................80 NumberofDailyTreatmentSessions...................................................................................................80

NeuronavigationDuringrTMS.............................................................................................................81 NovelStimulatoryProtocols................................................................................................................82 NovelrTMSStimulationTargetsinPsychiatricIllness.........................................................................84 Summary:PsychiatricIllness,IBNs,andrTMSTreatments..................................................................85

PREDICTINGTREATMENTRESPONSE..................................................................................................................85 WhyPredictTreatmentResponse?.................................................................................................85 ClinicalPredictorsofResponsetoConventionalInterventions.......................................................86

IBNPredictorsofResponsetoConventionalInterventions.............................................................89 PredictorsofResponsetorTMS......................................................................................................91

CHARACTERIZINGMECHANISMSOFTREATMENTRESPONSE...................................................................................94

WhyCharacterizeMechanismsofTreatmentResponse?...............................................................94 IBNChangesinResponsetoConventionalInterventions................................................................95 MechanismsofClinicalResponsetoDLPFC-rTMS..........................................................................97

MRIANDRESTING-STATEFUNCTIONALMRIMETHODS.....................................................................................100

Blood-OxygenLevelDependentfMRIPhysiology.........................................................................100 Resting-StateFunctionalMRI.......................................................................................................104 rs-fMRIPreprocessing...................................................................................................................106

ix

StatisticalMethodstoCharacterizeIBNs......................................................................................111 ConsiderationsforSelectingtheOptimalrsFCMethod................................................................120

AssessingGroup-LevelComparisonsAcrossSubjects...................................................................121 SUMMARY.................................................................................................................................................123

RATIONALE,SPECIFICAIMS&HYPOTHESES.............................................................................................128

GENERALAIMS&APPROACH........................................................................................................................128

RATIONALEFORASEED-TO-VOXEL-BASEDRSFCAPPROACH................................................................................129 STUDYI.BASELINEPREDICTORSANDMECHANISMSOFDMPFC-RTMSRESPONSEINANOREXIAANDBULIMIANERVOSA134

RationaleofStudyI............................................................................................................................134

SpecificAimsofStudyI......................................................................................................................135 HypothesesofStudyI........................................................................................................................136

STUDYII.BASELINEPREDICTORSANDMECHANISMSOFDMPFC-RTMSRESPONSEINOBSESSIVE-COMPULSIVEDISORDER....

136 StudyRationale..................................................................................................................................136 SpecificAimsofStudyII.....................................................................................................................137 HypothesesofStudyII.......................................................................................................................137

STUDYIII:BASELINEPREDICTORSANDMECHANISMSOFHIGH-ANDLOW-FREQUENCYDMPFC-RTMSINTRDUNDER

TRIPLE-BLINDSHAMCONTROLLEDSETTINGS....................................................................................................................138 RationaleofStudyIII..........................................................................................................................138

SpecificAimsofStudyIII....................................................................................................................139 HypothesesofStudyIII......................................................................................................................139

GENERALMETHODS................................................................................................................................140

PROJECTOVERVIEW....................................................................................................................................140

SUBJECTRECRUITMENT................................................................................................................................141 MRIANDRS-FMRIACQUISITION...................................................................................................................142 MOTORTHRESHOLDASSESSMENT..................................................................................................................143

DMPFC-RTMSTREATMENT..........................................................................................................................143

STUDYI:BASELINEPREDICTORSANDMECHANISMSOFDMPFC-RTMSRESPONSEINANOREXIAAND

BULIMIANERVOSA...........................................................................................................................................147

INTRODUCTION...........................................................................................................................................147 AIMS&HYPOTHESES...................................................................................................................................149

SpecificAimsofStudyI......................................................................................................................149 HypothesesofStudyI........................................................................................................................149

METHODS..................................................................................................................................................150

x

Subjects..............................................................................................................................................150 ClinicalOutcomes..............................................................................................................................151

NeuroimagingAcquisition.................................................................................................................151 NeuronavigationandrTMSTreatment..............................................................................................152 MRIPreprocessing,SeedSelection&StatisticalAnalysis..................................................................152

RESULTS....................................................................................................................................................153

PrimaryClinicalOutcomes.................................................................................................................153 SecondaryClinicalOutcomes.............................................................................................................156 FunctionalConnectivityPredictorsofResponse................................................................................157

FunctionalConnectivityChangesAssociatedwithResponse.............................................................159 DISCUSSION&CONCLUSION.........................................................................................................................162

STUDYII:BASELINEPREDICTORSANDMECHANISMSOFDMPFC-RTMSRESPONSEINOCD......................168

INTRODUCTION...........................................................................................................................................168 AIMS&HYPOTHESES...................................................................................................................................170

SpecificAimsofStudyII.....................................................................................................................170 HypothesesofStudyII.......................................................................................................................170

METHODS..................................................................................................................................................171 Subjects..............................................................................................................................................171 ClinicalMeasures...............................................................................................................................172

Intervention.......................................................................................................................................172 NeuroimagingAcquisitionandAnalysis............................................................................................173

RESULTS....................................................................................................................................................177 ClinicalOutcomes..............................................................................................................................177

Resting-StatefMRIPredictorsofTreatmentResponse......................................................................180 Resting-StatefMRICorrelatesofTreatmentResponse.....................................................................186

DISCUSSION&CONCLUSION.........................................................................................................................191

STUDYIII:BASELINEPREDICTORSANDMECHANISMSOFHIGH-ANDLOW-FREQUENCYDMPFC-RTMSIN

TRDUNDERTRIPLE-BLINDSHAMCONTROLLEDSETTINGS................................................................................196

INTRODUCTION...........................................................................................................................................196

AIMS&HYPOTHESES...................................................................................................................................198 SpecificAimsofStudyIII....................................................................................................................198 HypothesesofStudyIII......................................................................................................................199

METHODS..................................................................................................................................................199

PatientRecruitment...........................................................................................................................200 PrimaryClinicalMeasures.................................................................................................................201

xi

RandomizationandrTMSTreatment................................................................................................201 MRIAcquisitionandAnalysis.............................................................................................................203

SupplementaryClinicalMeasures......................................................................................................206 HealthyControlRecruitment&StudyVisits......................................................................................207

RESULTS....................................................................................................................................................208 DemographicandPrimaryClinicalResults........................................................................................208

SupplementaryClinicalResults..........................................................................................................216 BaselineDifferencesinrs-fMRIfunctionalconnectivity.....................................................................218 Changesinrs-fMRI............................................................................................................................227

DISCUSSION&CONCLUSION.........................................................................................................................230

GENERALDISCUSSION.............................................................................................................................236

SUMMARYOFRESULTSANDCOMPARISONTOHYPOTHESES................................................................................236

IMPLICATIONSFORTHETREATMENTOFPSYCHIATRICDISORDERS..........................................................................243 COMPARINGTHECLINICALEFFICACYOFDMPFC-RTMSTODLPFC-RTMS............................................................249 DMPFC-RTMSASATRANSDIAGNOSTICINTERVENTIONINPSYCHIATRICILLNESS......................................................251 TRANSDIAGNOSTICEFFICACYOFDMPFC-RTMS:ABIOLOGICALPERSPECTIVE.........................................................252

FRONTOSTRIATALFUNCTIONALCONNECTIVITYASAPOTENTIALTRANSDIAGNOSTICBIOMARKEROFDMPFC-RTMS

RESPONSE................................................................................................................................................................255 DMPFC-RTMSRESPONSECORRELATESWITHCSTCCONNECTIVITYCHANGE.........................................................258

TREATMENTNON-SPECIFICCLINICALRESPONSEINTRDISRELATEDTOEMOTIONREGULATIONANDIMPULSIVITY.........260 LIMITATIONSANDCHALLENGES......................................................................................................................262

CONCLUSION...........................................................................................................................................268

FUTUREDIRECTIONS...............................................................................................................................269

REFERENCES.....................................................................................................................................................274

APPENDICES.....................................................................................................................................................390

COPYRIGHTACKNOWLEDGEMENTS..................................................................................................................470

xii

List of Tables Table 4-1: Descriptive Statistics for all patients, rTMS responders and non-responders. .......... 154

Table 4-2: Brain regions where pre-treatment functional connectivity to dmPFC and dACC seeds

differed significantly between rTMS responders and non-responders. ...................................... 159

Table 4-3: Brain regions where the change in functional connectivity to dACC and dmPFC seed

from pre- to post-treatment differed significantly between rTMS responders and non-responders.

..................................................................................................................................................... 162

Table 5-1: Centre of gravity coordinates for regions of interest created from parcellation atlases,

and MNI coordinates for sphere-based regions of interest. ........................................................ 175

Table 5-2: Summary of Primary and Secondary Clinical Measures in All Patients, and in Y-

BOCS Responders and Non-Responders. ................................................................................... 179

Table 5-3: Brain regions where baseline resting-state functional connectivity to exploratory seed

regions differed significantly between responders and non-responders. .................................... 184

Table 5-4: Brain regions where the pre-to-post treatment change in functional connectivity to the

dmPFC that differed significantly between rTMS responders and non-responders. .................. 187

Table 5-5: Brain regions where the pre-to-post treatment change in functional connectivity to

exploratory ROIs that differed significantly between rTMS responders and non-responders ... 189

Table 6-1: Summary of demographic and primary clinical outcomes for healthy controls and

TRD treatment arms. ................................................................................................................... 211

Table 6-2: Summary of concomitant medications by treatment intervention. ............................ 212

Table 6-3: Summary of psychiatric comorbidities. ..................................................................... 213

Table 6-4: Summary of Baseline Secondary Clinical Outcomes ................................................ 215

Table 6-6: Summary of Post-Treatment Secondary Clinical Outcomes. .................................... 220

xiii

List of Figures

Figure 1-1: The Default Mode Network. ...................................................................................... 10

Figure 1-2: The Salience Network. ............................................................................................... 14

Figure 1-3: The Central Executive Network. ................................................................................ 19

Figure 1-4: The Affective Ventromedial Network. ...................................................................... 21

Figure 1-5: Overview of the general structural and pathways of cortico-striato-thalamo-cortical

circuits. .......................................................................................................................................... 25

Figure 1-6: Regions of frontal cortex that have cortico-striatal projections. ................................ 26

Figure 1-7: Rostrocaudal organization of frontostriatal inputs. .................................................... 27

Figure 1-8: Schematic of the Triple Network Theory of Psychopathology. ................................. 34

Figure 1-9: Example of a TMS system. ........................................................................................ 56

Figure 1-10: Configuration of an angled figure-of-eight rTMS head coil. ................................... 57

Figure 1-11: Areas of striatal dopamine release following active dmPFC-rTMS.. ...................... 62

Figure 1-12: Decreases in cerebral blood flow following active dmPFC-rTMS. ......................... 62

Figure 1-13: Targeting the dmPFC and dACC with rTMS. ......................................................... 71

Figure 4-1: Probability distribution function of binge and purge percent improvement across all

patients following dmPFC-rTMS. .............................................................................................. 156

Figure 4-2: Baseline functional connectivity differences between dmPFC-rTMS Responders and

Non-Responders. ......................................................................................................................... 159

Figure 4-3: Changes in functional connectivity in dmPFC-rTMS Responders and Non-

Responders. ................................................................................................................................. 162

xiv

Figure 5-1: Regions of interest created from parcellation atlases and coordinates. ................... 177

Figure 5-2: Probability distribution function (a) and ranked individual-patient plot (b) of

treatment outcomes for dmPFC-rTMS in OCD.. ........................................................................ 179

Figure 5-3: High baseline frontal-striatal-thalamic-subthalamic connectivity predicts response to

dmPFC-rTMS in OCD. ............................................................................................................... 183

Figure 5-4: Reductions in cortical-striatal-thalamic connectivity correlate to improvements in

OCD symptoms following dmPFC-rTMS. ................................................................................. 189

Figure 6-1: The custom MagVenture active/placebo coil. .......................................................... 204

Figure 6-2: Regions of interest used for seed-to-voxel based analyses. ..................................... 206

Figure 6-3: CONSORT diagram of patients enrolled in the study. ............................................ 211

Figure 6-4: Mean HAMD scores across the three treatment interventions. ............................... 214

Figure 6-5: Mean BDI-II scores across the three treatment interventions. ................................. 215

Figure 6-6: Mean BAI scores across the three treatment intervention. ...................................... 215

Figure 6-7: DERS improvement is significantly correlated with antidepressant response on the

BDI-II across all subjects. ........................................................................................................... 218

Figure 6-8: DERS improvement is significantly correlated with antidepressant response on the

BDI-II in the active 20 Hz arm. .................................................................................................. 219

Figure 6-9: Regions showing significantly lower resting-state functional connectivity to the

inferior ventral striatal seed in controls versus all MDD patients. ............................................. 222


VMPFC seed in controls versus all MDD patients. .................................................................... 223

Figure 6-11: Regions showing significantly higher resting-state functional connectivity to the

dorsal caudate seed in the 20 Hz versus other MDD treatment arms. ........................................ 224

xv

Figure 6-12: Regions where baseline rsFC to ventral striatum showing significantly different

correlations to BDI-II improvement across the three treatment groups. .................................... 225

Figure 6-13: Regions where baseline rsFC to VMPFC showed significant correlations to BDI-II

improvement in all three treatment groups. ................................................................................ 226

Figure 6-14: Regions where baseline rsFC to Superior VS showed significant correlations to

HAMD improvement to 20 Hz dmPFC-rTMS. .......................................................................... 227

Figure 6-15: Regions where baseline rsFC to Inferior VS showed significant correlations to BDI-

II improvement to 20 Hz dmPFC-rTMS. .................................................................................... 228

Figure 6-16: Regions of dACC rsFC change showing significantly different correlations to

HAMD improvement across the three treatment groups. ........................................................... 230

Figure 6-17: Regions of VMPFC rsFC change showing a significant positive correlation to BDI-

II improvement in all three treatment groups. ............................................................................ 231

Figure 7-1: Summary of differences in pre-treatment resting-state functional connectivity

differences between dmPFC-rTMS responders and non-responders for AN-BP/BN, OCD and

TRD............................................................................................................................................. 240

Figure 7-2: Summary of pre- to post-treatment changes in resting-state functional connectivity

that accompany dmPFC-rTMS response in AN-BP/BN, OCD and TRD. ................................. 243

xvi

List of Abbreviations ACC Anterior Cingulate Cortex AI Anterior Insula ALFF Amplitude of Low Frequency Fluctuations AN Anorexia Nervosa AN-BP Anorexia Nervosa, Binge/Purge Subtype AN-R Anorexia Nervosa, Restricting Subtype BA Brodmann Area BAI Beck Anxiety Inventory BDI-II Beck Depression Inventory II BDNF Brain-Derived Neurotrophic Factor BIS-11 Barratt Impulsiveness Scale BIS/BAS Behavioural Inhibition Scale/Behavioural Approach Scale BMI Body Mass Index BN Bulimia Nervosa BOLD Blood Oxygen Level Dependent CBT Cognitive Behavioural Therapy CEN Central Executive Network CSTC Cortico-Striato-Thalamo-Cortical cTBS Continuous Theta Burst Stimulation dACC Dorsal Anterior Cingulate Cortex DAN Dorsal Attention Network DBS Deep Brain Stimulation DERS Difficulties in Emotion Regulation Scale DLPFC Dorsolateral Prefrontal Cortex DMN Default Mode Network dmPFC Dorsomedial Prefrontal Cortex DSM-IV Diagnostic and Statistical Manual, fourth edition DSM-5 Diagnostic and Statistical Manual, fifth edition dTMS Deep Transcranial Magnetic Stimulation ECoG Electrocorticography ECT Electroconvulsive Therapy ED Eating Disorders EDE Eating Disorder Examination EDE-Q Eating Disorder Example - Questionnaire EEG Electroencephalography ERP Exposure and Response Prevention fALFF Fractional Amplitude of Low Frequency Fluctuations FC Functional Connectivity FDG-PET Fludeoxyglucose Positron Emission Tomography FDR False Discovery Rate fMRI Functional Magnetic Resonance Imaging FOV Field of View FSL FMRIB Software Library FWE Family-Wise Error FWHM Full Width Half Maximum GABA γ-Aminobutyric Acid

xvii

GPe Globus Pallidus, External Segment GPi Globus Pallidus, Internal Segment GSR Global Signal Regression HAMD Hamilton Rating Scale for Depression IBN Intrinsic Brain Network ICA Independent Component Analysis ICD-10 International Statistical Classification of Behavioral Disorders, tenth edition IDS Inventory of Depressive Symptoms IFG Inferior Frontal Gyrus IPL Inferior Parietal Lobule IPT Interpersonal Therapy lOFC Lateral Orbitofrontal Cortex ISI Inter-Stimulus Interval iTBS Intermittent Theta Burst Stimulation LFP Local Field Potential LTD Long-Term Depression LTP Long-Term Potentiation M1 Primary Motor Cortex MAO Monoamine Oxidase Inhibitors MBCT Mindfulness-Based Cognitive Therapy MCQ Monetary-Choice Questionnaire MDD Major Depressive Disorder MDT Medial Dorsal Thalamus MEG Magnetoencephalography MEP Motor-Evoked Potential MINI Mini International Neuropsychiatric Interview MNI Montreal Neurological Institute mOFC Medial Orbitofrontal Cortex mPFC Medial Prefrontal Cortex MPS Multidimensional Perfectionism Scale MRI Magnetic Resonance Imaging MSN Medium Spiny Neuron MST Magnetic Seizure Therapy NAc Nucleus Accumbens NDRI Norepinephrine-Dopamine Reuptake Inhibitors NEO Revised NEO Questionnaire OCD Obsessive-Compulsive Disorder OFC Orbitofrontal Cortex PCC Posterior Cingulate Cortex PET Positron Emission Tomography PHQ-9 Patient Health Questionnaire-9 QIDS Quick Inventory of Depressive Symptoms QPS Quadripulse Stimulation rACC Rostral Anterior Cingulate Cortex rCBF Regional Cerebral Blood Flow RDoC Research Domain Criteria ReHo Regional Homogeneity rs-FC Resting-State Functional Connectivity

xviii

rs-fMRI Resting-State Functional Magnetic Resonance Imaging ROI Region of Interest RRS Rumination Responses Scale rTMS Repetitive Transcranial Magnetic Stimulation RTPJ Right Temporoparietal Junction SCID-I Structured Clinical Interview for DSM-IV-Axis-I Disorders SD Standard Deviation sgACC Subgenual Anterior Cingulate Cortex SMA Supplementary Motor Area SN Salience Network SNc Substantia Nigra, Pars Compacta SNr Substantia Nigra, Pars Reticulata SNRI Serotonin-Norepinephrine Reuptake Inhibitors SSRI Selective Serotonin Reuptake Inhibitors STAR*D Sequenced Treatment Alternatives to Relieve Depression STN Subthalamic Nucleus TBS Theta Burst Stimulation TCA Tricyclic Antidepressants tDCS Transcranial Direct Current Stimulation TE Echo Time TI Inversion Time TMS Transcranial Magnetic Stimulation TR Repetition Time TRD Treatment-Resistant Depression UPPS-P UPPS Impulsive Behaviour Scale VAN Ventral Attention Network VLPFC Ventrolateral Prefrontal Cortex VMN Ventromedial Affective Network VMPFC Ventromedial Prefrontal Cortex VNS Vagal Nerve Stimulation VRP Ventral Rostral Putamen VS Ventral Striatum VSi Inferior Ventral Striatum VSs Superior Ventral Striatum VTA Ventral Tegmental Area Y-BOCS Yale-Brown Obsessive-Compulsive Scale II

Literature Review

Introduction & General Aims In 2013, Statistics Canada reported that approximately 1 in 10 Canadians (2.8 million

people) met criteria for at least one mental illness or substance use disorder in the past 12

months, while approximately 1 in 3 Canadians (9.1 million people) met the lifetime criteria for a

mental illness or substance use disorder (Pearson et al, 2013). Given this high prevalence, the

burden of these illnesses is likewise remarkably high. In Canada, the total economic burden of all

mental illness was estimated at $51 billion in 2003 (Lim et al, 2008), and in Ontario, the

economic burden for major depressive disorder alone was greater than the burden of breast,

colorectal, lung, and prostate cancer combined (Ratnasingham et al, 2013). Despite the

substantial prevalence of these disorders and their associated personal and economic costs, there

has been limited progress in developing new treatment options, and even more limited progress

in improving long-term clinical outcomes, over the past 50 years. New treatment approaches, and

methods to select the optimal treatment for any given patient, are urgently needed – especially

for those patients who are unresponsive to conventional pharmacotherapy and psychotherapy.

In the 1990s, non-invasive neuromodulation emerged as a potential novel treatment for

psychiatric patients who do not respond to conventional medications or psychotherapies. One

such intervention, repetitive transcranial magnetic stimulation (rTMS), uses focused, powerful

magnetic field pulses to induce changes in the activity of targeted and downstream brain regions.

A course of several weeks of daily stimulation can achieve therapeutic effects in patients

unresponsive to conventional psychotherapy and pharmacotherapy. Following the first

demonstrations of the efficacy of rTMS for major depression in the mid-1990s (George et al,

1995; Pascual-Leone et al, 1996b), subsequent studies suggested that other disorders including

mood, anxiety, and eating disorders, may be improved by rTMS (Lefaucheur et al, 2014). Given

the favorable safety and tolerability profile of rTMS, the hope was that this intervention might at

last deliver a meaningful improvement in outcomes over our longstanding, conventional

approaches of pharmacotherapy and psychotherapy.

However, despite 20 years of research into therapeutic rTMS, response rates have

remained low compared to more intensive neuromodulation treatments such electroconvulsive

2

therapy (Micallef-Trigona, 2014; Xie et al, 2013), and only a subset of patients who undergo

rTMS achieve full remission. This limited efficacy may arise in part from the substantial

heterogeneity of clinical features and neurobiological abnormalities seen among patients who

ostensibly meet the formal diagnostic criteria for any given psychiatric disorder (Drysdale et al,

2017). This heterogeneity could impose a ceiling of efficacy on any ‘one-size-fits-all’ rTMS

treatment protocol for a given disorder. As such, a better understanding is needed of the

neurobiological mechanisms of rTMS, and the neurobiological features of those patients for

whom these mechanisms are most effective.

To improve the overall remission rates for rTMS in the clinical setting, it is key to

identify neurobiological predictors and correlates of rTMS treatment response. Neurobiological

markers that can reliably predict which patients will respond to treatment could lead to a

predictive test for clinical treatment planning, reduce the incidence of futile treatment, and

thereby improve remission rates among those patients who undergo rTMS. Similarly,

neurobiological markers that show reliably different changes over the course of treatment in

responders versus non-responders can reveal the neural mechanisms of successful treatment.

Thus, the problem of identifying neural predictors and correlates of rTMS response in psychiatric

illness has important and direct translational significance.

In the early 1990s, functional magnetic resonance imaging (fMRI) and later resting-state

functional MRI (rs-fMRI) arose as a novel, safe, non-invasive, and readily accessible tool to

study and localize task-evoked and spontaneous patterns of activity in the human brain in vivo

(Ogawa et al, 1990; Snyder and Raichle, 2012). To date, over 10,000 studies involving fMRI

have been published, with many of these studies focused on individuals with psychiatric

disorders (Raichle, 2009a). fMRI has provided a groundbreaking tool for characterizing the

abnormalities of brain activity, as well as the abnormalities of connectivity within or between

functional brain networks, that underlie the many varieties of psychiatric illness (for a review,

see (Menon, 2011)). From a translational perspective, the hope is that neuroimaging techniques

such as fMRI might lead to a better understanding of the pathophysiology of psychiatric

illnesses, and eventually to new treatments that might better target these underlying disease

mechanisms.

3

Although there are many tools to study the neurobiological predictors and correlates of

treatment response, rs-fMRI offers several advantages that are specific to the intervention of

rTMS in the psychiatrically ill population. First, fMRI is widely accessible, safe, and non-

invasive. Second, fMRI has relatively good spatial resolution– a key advantage for anatomically

focal interventions. Third, rs-fMRI can be used to track the coordinated activity of networks of

regions across the whole brain. This is important because network activity plays a central role

both in psychopathology (Greicius et al, 2007; Menon, 2011) and in the neurobiological effects

of rTMS (Eldaief et al, 2011; Fox et al, 2013a). As such, rs-fMRI is uniquely well-positioned to

predict rTMS response, and to characterize neural mechanisms of successful versus

unsuccessful rTMS treatment.

Therefore, the goal of this thesis is to identify rs-fMRI predictors and correlates of

treatment response to rTMS across three different categories of psychiatric disorder: eating

disorders, obsessive-compulsive disorder, and major depressive disorder. The potential

translational significance of the work, is to inform the development of predictive tests to reduce

the rates of non-response among patients who undergo treatment, and toward novel treatment

strategies to treat non-responders.

Intrinsic Functional Networks of the Brain “The fact that the body is lying down is no reason for supposing that the mind is

at peace. Rest is… far from restful” – Seneca, ~60 A. D.

Historical and Philosophical Perspectives of Brain Connectivity

Major neuroscientific paradigm shifts are preceded by methodological and technological

advances. A classical example is in Cajal and Sherrington’s reconstruction of neuron histology,

and their proposal that the neuron is the structural and functional unit of the nervous system.

Arguably, this neuron doctrine is a consequence of the development of the light microscope and

Golgi stain (as reviewed by (Yuste, 2015)). Yet in spite of the impact this doctrine had on

4

modern neuroscience, the study of individual neurons alone is necessary but not sufficient for a

holistic understanding of brain function.

Theories surrounding the functional importance of neural networks arose as early as the

1940s, positing that brain function arises as a result of the activation of populations of neurons,

(Hebb, 1949; McCulloch and Pitts, 1943). At the time, models of neuron networks were

strengthened by evidence for structural and and extensive and complex neuronal connectivity,

and the discovery of local cortical processing ‘units’ called cortical columns (Hubel, 1988;

Mountcastle, 1957).

Two opposing views of brain function were conceived alongside the development of the

neuron doctrine by Cajal and Sherrington (as reviewed by (Raichle, 2010)). On one hand,

Sherrington theorized that brain function is reflexively driven by momentary demands in the

environment (bottom-up). Arguably, much of the field of cognitive neuroscience, including its

offshoots within functional neuroimaging, is implicitly underpinned by this theory. For example,

task-based functional neuroimaging involves measuring brain responses to carefully designed

and controlled stimuli. On the other hand, Brown posited that brain activity and function are

intrinsic, meaning that brain function involves the acquisition of information from the

environment and the integration of this information with endogenous ongoing brain activity, with

the purpose of interpreting it, predicting future environmental demands, and responding to

current demands in a top-down fashion.

The discovery of electroencephalography (EEG) and the Berger Block (Berger, 1929)

gave empirical support to the idea that populations of neurons might be associated with

spontaneous, intrinsic oscillations involved with the aforementioned functions. With the further

development of EEG techniques and the advent of fMRI and rs-fMRI in the 1990s and 2000s (as

reviewed by (Raichle, 2009b, 2009a)), spontaneous fluctuations in intrinsic brain networks have

become a mainstream topic of study, aligning well with Brown’s alternative, intrinsic-activity

account of brain function (Buckner, 2012).

5

What are Intrinsic Brain Networks?

The ongoing activity of the human brain as a whole may be organized into a set of brain

networks consisting of regions that are highly functionally connected, meaning that they activate

and deactivate at roughly the same time (Beckmann et al, 2005; Sporns, 2011). In other words,

the two regions possess synchronous activity. While these networks are often labelled as

‘resting-state’ networks (Beckmann et al, 2005), this term is somewhat misleading, as these

networks can be discerned within ongoing brain activity not only while the participant is not

actively engaged in a particular task (from spontaneous fluctuations in brain activity, or at ‘rest’)

(Beckmann et al, 2005; Damoiseaux et al, 2006; Greicius et al, 2003), but also while the

participant is engaged actively in a cognitive or behavioural task (Cole et al, 2014; Krienen et al,

2014). Consequently, this thesis will refer to these networks as intrinsic brain networks (IBNs):

distinct sets of functionally (and potentially structurally) coupled regions whose activity is

correlated over time, either at rest or on task (Yeo et al, 2011). The topological patterns of IBNs

are highly replicable, with general topology congruent both across subjects (Beckmann et al,

2005; Cole et al, 2010a; Damoiseaux et al, 2006) and within a single subject (Laumann et al,

2015).

Although the precise number of IBNs is still a matter of debate, 7-20 IBNs can be

distinguished consistently across large samples of. For example, one large scale study of

neuroimaging data from the Human Connectome Project in 000 individuals identified seven

reproducible cortical IBNs that could be further subdivided into 17 sub-IBNs (Yeo et al, 2011).

Studies from the same group also reported IBN-specific subdivisions in the striatum (Choi et al,

2012) and cerebellum (Buckner et al, 2011). This set of 7 or 17 IBNs provides a preliminary

reference ‘atlas’ of the major IBNs that can be used to understand the psychopathology and/or

the therapeutic mechanisms of rTMS.

Relationship of Functional Networks to Brain Structure

IBNs identified from the functional coupling of activity across distant brain regions are

hypothesized to reflect the underlying structural (white matter) architecture of the brain between

these regions. Multiple studies have found close coherence in terms of the underlying white

6

matter architecture of the brain and the functional coupling of cortical regions during various

task paradigms and during spontaneous fluctuations of brain activity (Greicius et al, 2009; van

den Heuvel et al, 2009). Moreover, one study reported that the relative strength of white matter

connections between two regions are indicative of how strong their functional coupling is at rest

(Hermundstad et al, 2013). However, other studies have reported that some strong IBN

connections exist despite no direct structural association; instead, these strong functional

relationships were related to indirect structural connections, potentially through intermediate

structures such as the thalamus, cerebellum, or striatum (Honey et al, 2009). One recent study

found that spontaneous interhemispheric electrophysiological activity displayed a temporal lag of

up to 50ms, likely suggesting that such brain-wide spontaneous fluctuations are constructed via

polysynaptic connections (Baek et al, 2016). Whether defined from white matter connections or

correlations of activity, the particular contributions of a given network to overall brain function

(sensory, motor, cognitive, or otherwise) may be considered to arise from the position of that

network within the overall ‘topology’ of connections across the brain as a whole (Bullmore and

Sporns, 2009).

Relationship of Functional Networks to Human Behaviour

While the architecture and brain-wide hierarchy of IBNs is not yet fully understood, IBN

functions have been linked to specific components of human perception, cognition, and

behaviour. Inter-individual variability in the functional coupling of resting-state IBNs is

associated with areas of association cortex (particularly those of recent evolutionary expansion),

and ongoing fluctuations in the coupling of these regions appears to be related to fluctuations in

cognitive demands, task performance, and even states of consciousness. First, one recent study

reported ‘common core’ spatial patterns of IBN activity across 14 different tasks, concluding that

distinct tasks harness reconfigurable IBN ‘modes’ to allow for flexible, rapidly adaptable modes

of behaviour (Krienen et al, 2014). Second, inter-individual variability in the functional coupling

of IBNs has been related to behavioural variability, as measured by response time variability

during a Flanker task (Kelly et al, 2008). Such inter-individual variability has also been related

to behavioural variability in other tasks; for example, stronger within-network IBN coupling in

cortical regions responsible for motor and language function were associated with higher

7

performance on a reading task (Koyama et al, 2011), and higher rates of behavioural impulsivity

were related to the coupling of attentional and executive control regions to premotor regions

(Shannon et al, 2011). Finally, IBNs persist during sleep, and their relative strength is more

variable during light sleep than during wakefulness (Larson-Prior et al, 2009). IBN strength is

correlated with levels of consciousness, such that less network functional coupling is observed in

patients with brain damage at different levels of consciousness (for example, locked-in syndrome

versus coma) (Vanhaudenhuyse et al, 2010). Consequently, the overall consensus is that IBNs

act as functional ‘modules’ that operate in concert to facilitate complex cognitive and

behavioural processes, and that variability in IBNs reflect the overall integrity of brain-wide

functional networks.

Relationship of Functional Networks to Electrophysiological Studies

IBNs extracted from spontaneous fluctuations in neuronal signal have been associated

with electrophysiological sources; however, the relationship between IBNs and

electrophysiological properties remains elusive. Numerous studies have reported that spatial

patterns of frequency-specific power correlate with patterns of IBNs generated from rs-fMRI.

For example, the spatial network organization of electrocorticography (ECoG) in epilepsy

patients correlates with the spatial organization of IBNs generated using rs-fMRI (He et al, 2008;

Nir et al, 2008), and reflect opposing (anti-correlated) relationships between IBNs (Keller et al,

2013). Studies of simultaneous EEG and fMRI have demonstrated that fMRI-generated IBNs

have distinct patterns of frequency-specific power and phase synchrony (Mantini et al, 2007;

Sadaghiani et al, 2012). Similarly, rs-fMRI IBNs are also associated with distinct frequency-

specific power patterns on magnetoencephalography (MEG) (Brookes et al, 2011; Hipp et al,

2012; de Pasquale et al, 2010, 2012; Wens et al, 2014).

Co-activity between IBN brain regions is associated with covariation in gamma power

(50-100 Hz) (Fox and Raichle, 2007; Schölvinck et al, 2010, 2013). For example, spontaneous

gamma fluctuations in monkey visual cortex, as measured by local field potentials (LFP),

correlate with spontaneous fluctuations of rs-fMRI signal (Shmuel and Leopold, 2008). In

humans, spontaneous gamma oscillations correlate between brain regions that are highly

functionally connected on rs-fMRI (Nir et al, 2008). Keller et al. also showed that regional

8

ECoG gamma power correlates with distinct IBNs (Keller et al, 2013). Studies of intracranial

EEG IBNs rs-fMRI connectivity have also supported the idea that IBNs are generated by

synchronous increases in gamma power that reflect increased local spiking in nodes of IBNs

(Logothetis et al, 2001; Manning et al, 2009; Mukamel et al, 2005; Nir et al, 2007; Ray and

Maunsell, 2011).

IBNs are also characterized by activity in other frequency bands. Recent research has

suggested that IBNs are also related to infraslow cortical potentials (<0.1 Hz) (He et al, 2008)

and have been found to be more tightly correlated to spontaneous fluctuations in brain activity

relative to gamma (Lu et al, 2016). Spontaneous fluctuations in infraslow LFPs using EEG and

MEG have been shown to be related to spontaneous fluctuations in rs-fMRI (Hiltunen et al,

2014). In rodents, infraslow LFPs are state-dependent, as infraslow activity and phase coupling

differ between wakefulness and anesthesia (Mitra et al, 2018), and can differ depending on the

kind of anesthesia used (Thompson et al, 2014a). Interhemispheric infraslow activity also

directly associated with neuronal activity as this activity is attenuated by voltage-gated sodium

channel or glutamate receptor antagonists (Chan et al, 2015). Other frequency-limited power

bands, including alpha, theta and beta, have also been shown to resemble functional connections

of brain regions (Brookes et al, 2011; Hipp et al, 2012; de Pasquale et al, 2010, 2012; Wens et

al, 2014). Consequently, it has been proposed that IBNs and their functional connections are

related to a complex interaction of multiple frequency bands (Buzsáki and Draguhn, 2004; Fox

and Raichle, 2007) (Buzsaki & Draguhn, 2004; Fox & Raichle, 2007).

The relationship between infraslow LFPs, gamma power and other frequency bands

might be related to attentional processes and network architecture, as fluctuations in infraslow

LFPs correlate with gamma (Vanhatalo et al, 2004). He and colleagues demonstrated that IBNs

generated using ECoG have a similar spatial structure to IBNs in the gamma band, but only

during wakefulness and during rapid-eye movement sleep. Infraslow potentials, however,

exhibited consistent spatial correlation structure to that of IBNs during wakefulness, rapid-eye

movement sleep, and during slow-wave sleep (He et al, 2008). However, a recent study using

intracranial EEG reported that both infraslow and high frequency oscillations representing IBNs

are reproducible both during wakeful rest and sleep (Kucyi et al, 2018).

9

Functional Networks Pertinent to the Neurobiology of Psychiatric Disorders

1.2.6.1 Default Mode Network

Early human neuroimaging studies reported a consistent set of regions that were active

during stimulus-free control (‘rest’) conditions used as a contrast to study task-induced

activations. These regions were active during the control condition and deactivated in response

to a certain cognitive task (as reviewed in (Raichle et al, 2001)). At the time, however, this

phenomenon was largely ignored as an experimental confound (as reviewed in (Buckner, 2012)).

Shulman and colleagues were the first to report a common pattern of increased cerebral blood

flow during stimulus-free control conditions of task-based neuroimaging, localized to medial

parietal structures, including the posterior cingulate cortex (PCC), precuneus and retrosplenial

cortex, the medial prefrontal cortex (mPFC), the bilateral inferior parietal lobules (IPL), and the

medial and lateral temporal cortices (Buckner et al, 2008; Shulman et al, 1997). From 2001

onwards, functional relationship of these regions were delineated, ultimately known collectively

as the ‘default mode’ or ‘task-negative’ network (DMN; Figure 1-1) (Fox et al, 2005; Greicius et

al, 2003; Gusnard et al, 2001; Raichle et al, 2001). The DMN has also been observed in

homologous brain regions in other mammals, including the anesthetized macaque (Vincent et al,

2007), chimpanzee (Rilling et al, 2007), cat (Popa et al, 2009), rat (Lu et al, 2012), and mouse

(Stafford et al, 2014).

10

Figure 1-1: The Default Mode Network. Adapted from (Dunlop et al, 2017b; Yeo et al, 2011).

Medial and Lateral Parietal Regions of the DMN

Medial parietal structures associated with the DMN include the PCC, precuneus and

retrosplenial cortex. The PCC is comprised of Brodmann Areas (BA) 23 and 31 (Morris et al,

2000) and receives afferents from higher order sensory and associative cortices in the parietal

and frontal lobes (Adey and Meyer, 1952; Pandya et al, 1971; Pandya and Kuypers, 1969). The

PCC, particularly from BA 23, has reciprocal connections to other brain regions of the DMN,

including the medial temporal gyrus, mPFC, and IPL (BA 7a) (Kobayashi and Amaral, 2003,

2007). In the macaque, studies using retrograde tracers to study the anatomical of PCC white

matter tracts have also reported reciprocal white matter connections between the PCC and IPL,

especially in BA 39 and BA 40 (Kobayashi and Amaral, 2003; Vincent et al, 2007). The PCC is

a key node of the DMN function, even after controlling for the connectivity strength of other

DMN regions (Fransson and Marrelec, 2008). In non-human primates, switching between two

tasks, a measure of cognitive control, is associated with decreased basal neuronal activity of

single neuron recordings in the PCC (Hayden et al, 2010).

The retrosplenial cortex, comprised of BAs 29 and 30, lies posterior to the corpus

callosum, and ventral to the PCC (Morris et al, 2000; Vogt et al, 2001). The retrosplenial cortex

receives 40% of its input from medial temporal regions, including the hippocampal formation

and parahippocampus (Kobayashi and Amaral, 2003; Morris et al, 1999; Suzuki and Amaral,

11

1994), and has efferent white matter connections to both the medial temporal lobe and mPFC

(Kobayashi and Amaral, 2007). The precuneus is comprised of BA 7m and lies in the

dorsomedial portion of the parietal lobe (Cavanna and Trimble, 2006; Parvizi et al, 2006).

Precuneus connectivity is unlike other posteromedial structures of the DMN, in that BA 7m

predominantly has white matter connections with occipital and parietal regions responsible for

visual processing and with frontal motor planning regions (Cavada and Goldman-Rakic, 1989;

Leichnetz, 2001). Consequently, the membership of the precuneus in the DMN proper has been

debated extensively (see (Buckner et al, 2008; Scheperjans et al, 2008; Utevsky et al, 2014;

Vogeley et al, 2004)). The IPL node of the DMN consists of inferior regions of the lateral

parietal cortex, including BA 7a, BA 39 (the angular gyrus) and BA 40 (the supramarginal

gyrus), and has reciprocal white matter connections to the PCC (Kobayashi and Amaral, 2003,

2007; Vincent et al, 2007), and to medial temporal lobe via BA 7a (Clower et al, 2001; Lavenex

et al, 2002; Suzuki and Amaral, 1994).

Medial Prefrontal Cortex and the DMN.

Many regions of the mPFC are thought to be of relatively recent evolutionary origin, and

as a result are significantly larger in humans relative to nonhuman primates. This is especially

the case for prelimbic regions (BA 32pl), the dorsal anterior cingulate cortex (BA 32ac; dACC),

and frontopolar cortex (BA 10) (Ongür et al, 2003; Semendeferi et al, 2001). mPFC regions of

the DMN include the frontopolar cortex (BA 10m, 10r and 10p), rostral ACC (rACC) (BA 24,

32a and 32c) and medial superior frontal gyrus (BA 9) (Buckner et al, 2008). The nonhuman

primate homologues of these structures have nearly no afferent or efferent white matter

connections to low-level sensory regions; instead, the mPFC has reciprocal connections with the

PCC, retrosplenial cortex, superior temporal gyrus, hippocampal formation, and

perirhinal/parahippocampal regions (Barbas et al, 1999; Price, 2007). Of note in the psychiatric

context, electrical stimulation of the dACC induces a will to persevere in the face of a

psychological or physical challenge (Parvizi et al, 2013).

12

Temporal and Hippocampal Regions of the DMN.

As reviewed in detail in Section 2.5.1.2, the hippocampal formation and surrounding

limbic medial temporal lobe brain regions are connected to key posterior nodes of the DMN,

including retrosplenial cortex (Kobayashi and Amaral, 2003, 2007; Morris et al, 1999; Suzuki

and Amaral, 1994), PCC (Kobayashi and Amaral, 2003, 2007) and IPL (Clower et al, 2001;

Lavenex et al, 2002; Suzuki and Amaral, 1994). In humans, functional coherence between these

parietal and hippocampal structures has been associated with successful episodic memory

recollection (Vincent et al, 2006).

As a functional network, early neuroimaging studies reported that the DMN deactivation

was associated with increasing task difficulty or task switching, possibly suggesting a

reallocation of resources from ‘passive, resting-state’ processing to ‘active’ attentional or

cognitive control processes (McKiernan et al, 2003; Raichle et al, 2001; Shulman et al, 1997).

However, more recent studies have reported that DMN activation is related to a number of

behaviours related to internally-generated cognition and self-referential processing (Crittenden et

al, 2015; Dixon et al, 2014), including mind wandering (Mason et al, 2007), hippocampal-

dependent autobiographical memory retrieval (Addis et al, 2007; Svoboda et al, 2006), spatial

navigation, and thinking about the thoughts of others (Spreng et al, 2009).

The strength of DMN functional coupling is also related to levels of consciousness. In a

study of individuals with differing levels of consciousness due to brain damage (i.e., vegetative

and coma patients), DMN coupling was negatively correlated with the level of consciousness

(Vanhaudenhuyse et al, 2010). Some neuroimaging studies have also reported that the DMN

consists of at least three sub-IBNs: (1) a midline ‘core’ network comprised of the mPFC and

PCC that is consistently activated for all DMN-relevant functions; (2) a dmPFC subnetwork

comprised of the dmPFC, angular gyrus and temporal pole that is active for self-referential or

affective processes; and (3) a temporoparietal lobe network comprised of the bilateral IPL,

medial temporal and lateral temporal cortices for memory retrieval and scene reconstruction

from memory (Andrews-Hanna et al, 2010; Kim, 2012; Maillet and Rajah, 2014; Shapira-Lichter

et al, 2013; Yeo et al, 2011). Consequently, more recent theories of DMN function posit that

DMN activity has a role in self-referential, internally-generated cognition opposed to simply a

‘default mode’ of the brain.

13

1.2.6.2 Salience Network

Prior to 2007, a key limitation to the study of brain networks and task-related brain

activity was that many distinct IBNs co-activate across a wide variety of tasks. These areas,

collected termed the ‘task-activation ensemble’ or ‘task-positive network,’ include the dACC,

lateral areas of the prefrontal cortex, lateral parietal brain regions, the inferior frontal gyrus

(IFG), and the anterior insula (Fox et al, 2005; Seeley et al, 2007). However, the results of

multiple neuroimaging studies at that time suggested the dissociation of this ensemble into at

least two networks, one responsible for attention and working memory, and the other responsive

to uncertainty, response selection, and salience detection (for example, see (Ridderinkhof et al,

2004a)). In 2007, Seeley and colleagues were the first to identify two IBNs that had dissociable

topology and behavioural correlates: first, a ‘salience’ network (SN; Figure-1-2) comprised of

the dACC and anterior insula (AI) that correlated with anxiety ratings; and second, a ‘central

executive’ network (CEN; Figure 1-3; described below) comprised of dorsolateral prefrontal and

parietal cortices that correlated with task performance on The Trail Making Test, a measure of

executive functioning (Seeley et al, 2007). A subsequent review defined the SN by four basic

mechanisms: first, the detection of salient stimuli; second, brain network switching for the

refocus of attention to salient stimuli; third, autonomic reactions to salient stimuli; and fourth,

projections to the motor cortex to facilitate an action (Menon and Uddin, 2010).

14

Figure 1-2: The Salience Network. Adapted from (Dunlop et al, 2017b; Yeo et al, 2011).

Dorsal Anterior Cingulate Cortex and the SN.

The dACC node of the SN is comprised of BAs 23 and 24 (Smith, 1945; Walker, 1940).

Both BA 23 and BA 24 are heterogeneous in terms of their cytoarchitecture, with poorly-

differentiated laminae nearest to the corpus callosum to well-defined cortical laminae nearest the

cingulate sulcus (Smith, 1945; Walker, 1940). The dACC has reciprocal white matter

connections to the PCC (Pandya et al, 1981; Vogt et al, 1979), a major node of the DMN. BAs

23 and 24 also receive white matter projections from the AI (Morecraft et al, 2012), and medial

dorsal nucleus of the thalamus (Vogt et al, 1987). Only BA 24 receives projections from the

dorsolateral prefrontal cortex (BA 9), orbitofrontal cortex (BA 12), the subgenual subregion of

the anterior cingulate cortex (BA 25), and middle frontal gyrus (BA 46) (Vogt and Pandya,

1987). The dACC projects directly to motor and premotor cortices (Muakkassa and Strick,

1979), and to striatal nuclei (Royce, 1982; Yeterian and Van Hoesen, 1978).

dACC activity has been linked to goal-directed action selection and initiation via direct

white matter projections to motor and premotor cortices (Picard and Strick, 1996). For example,

patients with dACC lesions exhibit difficulty initiating complex movements (Rushworth et al,

2004), and also have difficulty switching from one complex behaviour to another (Williams et al,

2004). Another proposed function of the dACC is the ability to resolve conflicting stimuli during

action selection (and before top-down modulation of motor regions), as temporarily disrupting

15

dACC activity using non-invasive brain stimulation increased errors rates to ambiguous stimuli

(e.g., incongruent trials on a flanker task) (Taylor et al, 2007).

Insular Regions of the SN.

The insular cortex, BA 13, is located within the lateral fissure and hidden below the

frontal, parietal, and temporal opercula. Broadly speaking, the insula is associated with wide-

ranging functions including homeostatisis, interoception and sensorimotor processing, and the

integration of interoceptive, autonomic, affective and cognitive processing (Craig, 2009a). The

insula is subdivided into three subregions based on cytoarchitecture: an agranular region located

on the anterior ventral field of the insular cortex; a granular region located in the dorsal and

posterior insular cortex; and a transitional dysgranular field between these two subregions (Jones

and Burton, 1976; Mesulam and Mufson, 1982; Morel et al, 2013). Some cytoarchitectural

studies report that four insular subregions exist, as the anterior agranular insula can be further

subdivided into dorsal and ventral anterior insula subregions (Preuss and Goldman-Rakic, 1989).

Furthermore, the insula has been subdivided based on its white matter connectivity, revealing

anterior, dorsomedial and posterior subregions (Cloutman et al, 2012; Jakab et al, 2012), while

subdivisions based on insular function have yielded two to four subregions (Cauda et al, 2012;

Chang et al, 2013; Kurth et al, 2010; Mutschler et al, 2009). The function of the insula and the

inter-relationships of its subregions are not fully understood, although multiple studies have

described the role of the posterior insula in interoception, urges related to interoception such as

hunger and thirst, and sensorimotor processing (Craig, 2002; Del Parigi et al, 2002; Tataranni et

al, 1999; Wager et al, 2004).

The AI is the key insular node of the SN (see (Craig, 2009b, 2010a, 2010b)). The AI has

widespread afferent connections from and efferent connections to frontal and parietal cortices,

and the posterior insula (Mufson and Mesulam, 1982). The AI also has efferent projections to the

dACC (van den Heuvel et al, 2009; Morecraft et al, 2012). Functionally, the AI has a role in

decision-making and stimulus identification independent from the sensory modality of the

stimulus (Ho et al, 2009), and is active during auditory object identification (Binder et al, 2004),

and audio-visual timing discrimination (Kosillo and Smith, 2010). AI activity is also associated

with the integration of perceptual information from multiple sensory modalities, such that higher

16

AI activation is associated with greater difficulty discriminating items associated with

ambiguous stimuli from two sensory modalities (for example, with asynchronous auditory and

visual stimuli) (Bushara et al, 2001; Calvert, 2001; Kosillo and Smith, 2010; Lewis et al, 2000).

The AI is also involved in the anticipatory response to potential rewards or losses (Grosenick et

al, 2008; Knutson et al, 2007) that is correlated to decision ambiguity and risk prediction rather

than the value of expected rewards or losses (Huettel et al, 2006; Preuschoff et al, 2006, 2008).

The AI is also active during anticipatory responses to rewards that evoke positive or negative

arousal (Knutson and Greer, 2008), during error commission, and in task paradigms where

performance monitoring is required (Klein et al, 2007; Preuschoff et al, 2006, 2008). In line with

these studies, patients with insular lesions report intense positive or negative images as less

arousing (Berntson et al, 2011), and are insensitive to changes in the expected value of rewards

when making a decision under uncertainty (Weller et al, 2009).

Lateral Parietal and Frontal Regions of the SN.

The right temporoparietal junction (RTPJ, BA 40 and 22), a region of the lateral parietal

cortex, and the bilateral dorsolateral prefrontal cortex (DLPFC), are coactive with other regions

of the SN (Jakobs et al, 2012). Before the existence of the SN was proposed by Seeley et al.

(Seeley et al, 2007), research by Downar and colleagues demonstrated that the RTPJ activates

together with the dACC and AI during the detection of external, salient stimuli irrespective of its

sensory modality (Downar et al, 2000, 2001, 2002). Patients with right parietal lesions that

include the RTPJ exhibit impaired ability to attend to the sensory stimuli (Halligan et al, 2003;

Halligan and Marshall, 1998; Vallar, 1998). Lateral areas of the parietal cortex are also active

during other forms of SN-dependent perceptual decision-making, including during the

assessment of the relatedness of words (Kuperberg et al, 2008) and value-based decisions, such

as during the evaluation of costs and benefits (Kahnt and Tobler, 2013). Other studies

demonstrate that RTPJ and DLPFC activity are related to the integration of multimodal sensory

information, including temporal order judgment (Adhikari et al, 2013). RTPJ activity has also

been associated with social cognition, including during Theory-of-Mind paradigms (Young et al,

2010). However, a recent publication by Krall et al. demonstrated that the RTPJ can be divided

into two sub-regions: an anterior RTPJ active during sensory integration, attentional/social

17

processes, and decision-making; and a posterior RTPJ connected exclusively to regions related to

social-cognitive processing and Theory-of-Mind tasks, likely unrelated to the SN proper (Krall et

al, 2015). The RTPJ and other nodes of the SN also co-activate with the ventrolateral prefrontal

cortex (VLPFC) and IFG (BA 44, 55) as a region of a particular frontoparietal network, coined

the ‘ventral attention network’ (VAN), that reorients attention based on the detection of salient or

novel environmental stimuli (for a review see (Corbetta et al, 2008; Corbetta and Shulman,

2002)). At this time, the functional or anatomical distinction between the VAN and SN is not

well understood.

SN activity is related to goal-directed behaviour, meaning that the AI, DLPFC, and RTPJ

nodes of the SN operate in concert to identify, compare, and filter relevant external sensory

information to support one action choice over another (Pleger et al, 2006; Ploran et al, 2007)

before projecting to motor and premotor regions (often via the dACC) for action initiation

(Lamichhane and Dhamala, 2015; Landmann et al, 2007). Activity of the dACC and AI also

tracks decision uncertainty, as the activation of these two regions before perceptual decision-

making is modulated by the ambiguity of sensory stimuli (Grinband et al, 2006) or by the

possible outcomes of choices (Woolgar et al, 2011). The AI and dACC are also active during

affective face discrimination (Thielscher and Pessoa, 2007), sustained attention to a task, and

error-signal monitoring (Dosenbach et al, 2006; Landmann et al, 2007). For example, these two

regions are active following the commission of an error and changes in dACC activity following

an error predicted behavioural adaptations in the subsequent trial (Ham et al, 2013). SN function

has also been related to risky decision-making and impulsivity in the presence of ambiguous

conditions. Individuals who exhibit strong AI and dACC connectivity show less risky behavior

during ambiguous or affectively-arousing financial decision-making (Jung et al, 2014). AI and

dACC activity is also positively correlated with the magnitude of expected rewards and its

associated arousal during reward anticipation (Wu et al, 2012).

1.2.6.3 Central Executive Network

Higher-order executive functioning encompasses a diverse array of behaviours,

including: planning an action; sustained attention during a task (vigilance); the initiation of

complex behaviours; the ability to keep information in mind during a task (working memory);

18

behavioural inhibition; and the ability to shift between goals or tasks (cognitive flexibility)

(Niendam et al, 2012). As previously discussed, the ‘task-activation ensemble’ or ‘task-positive

network,’ appears to be active for a similarly wide range of functions. However, the results of

multiple neuroimaging studies have suggested that this ensemble of regions may be dissociated

into at least two networks: the SN for uncertainty, response selection, and salience detection, and

the CEN (Figure 1-3) for attention and working memory (Ridderinkhof et al, 2004a). As reported

by Seeley et al., the CEN and SN were dissociable based on executive functioning performance

and anxiety ratings (Seeley et al, 2007). Classical and current models of neurobiological basis of

cognitive control and attention suggest that the CEN and SN operate cooperatively. Specifically,

sustained task-related SN activity serves to identify and integrate salient internal and external

stimuli to select and initiate an action, while transient CEN activity serves to identify changes in

the environment that necessitate the inhibition of an action or a behavioural adjustment to a

different strategy (Dosenbach et al, 2008; Posner and Petersen, 1990).

The frontoparietal brain regions of the CEN display remarkable overlap with other

reported networks related to top-down control and attention, including the cognitive control

network and the dorsal attention network (DAN) (Corbetta and Shulman, 2002; Fox et al, 2006).

The emerging consensus is that the network known as the CEN is comprised of the DLPFC,

frontal eye fields, superior parietal lobule, and intraparietal sulcus (Cole and Schneider, 2007;

Corbetta and Shulman, 2002; Niendam et al, 2012; Vincent et al, 2008; Yeo et al, 2011).

However, the functional relationship between the DAN and the CEN is not yet fully understood.

19

Figure 1-3: The Central Executive Network. Adapted from (Dunlop et al, 2017b; Yeo et al,

2011).

Lateral Prefrontal Regions of the CEN.

The DLPFC, comprised of the middle frontal gyrus (BA 9) and SFG (BA 46), contains

the core lateral prefrontal regions of the CEN. Both BAs 9 and 46 receive multimodal input from

the frontal, parietal and temporal lobes, including the anterior cingulate cortex (ACC), PCC,

retrosplenial cortex, superior temporal sulcus, and rostral superior temporal gyrus (Petrides,

2005). The DLPFC has widespread efferents, including projections to the retrosplenial cortex,

and motor and premotor areas (Cole et al, 2010b, 2013; Morris et al, 1999; Power et al, 2011).

The DLPFC is active during tasks that require sustained retention of information (working

memory) (for a review see (Curtis and D’Esposito, 2003)) and during adjustments of cognitive

control (task-switching) (Koechlin et al, 2003; MacDonald et al, 2000). Patients with lateral

prefrontal lobe lesions exhibit impaired mental planning and working memory on classic

executive functioning paradigms (Milner, 1982), including the Tower of London task (Owen et

al, 1990). Some studies also report that VLPFC and IFG activity contribute to the CEN;

however, these studies suggest that the VLPFC and IFG function to inhibit prepotent responses,

rather than to support working memory or task-switching (Aron et al, 2003; Brass et al, 2005;

Ridderinkhof et al, 2004b).

20

Lateral Parietal Regions of the CEN.

Regions of the superior parietal lobule and intraparietal sulcus are also associated with

the CEN. The superior parietal lobule is comprised of lateral portions of BAs 5 and 7, and

receives and integrates input from higher-order somatosensory and visuomotor cortices

(Scheperjans et al, 2008). The superior parietal lobule is thought to monitor shifts in the spatial

coordinates of salient external objects (Molenberghs et al, 2007). The intraparietal sulcus serves

as the anatomical border between the superior and inferior parietal lobules, and is activated by

visuomotor and visuospatial information and projects to ventral and dorsal prefrontal and

premotor cortex, including the frontal eye fields (Grefkes and Fink, 2005). The medial and

lateral intraparietal sulcus subregions are considered part of the CEN and are preferentially

active during behavioural planning, action initiation, and during task monitoring (Bisley and

Goldberg, 2003; Desmurget and Grafton, 2000; Grefkes and Fink, 2005; Kalaska et al, 2003).

1.2.6.4 Ventromedial Affective Network

The ventromedial affective network (VMN; Figure 1-4) is comprised of the rostroventral

portion of the ACC and the medial orbitofrontal cortex (for a review, see (Dunlop et al, 2017b;

Fettes et al, 2017b)). The ventral aspect of the prefrontal cortex is thought to have a role in three

major functions (Hiser and Koenigs, 2018). First, the VMN contributes to reward processing,

prediction, and reward-based reversal learning (Boorman et al, 2013; Fellows, 2007;

Kringelbach, 2005; Montague and Berns, 2002); second, the VMN has a role in generating

negative affective states and during emotional appraisal (Johnstone et al, 2007; Wager et al,

2008); and third, the VMN is active during social cognition (Rushworth et al, 2007).

21

Figure 1-4: The Affective Ventromedial Network. Adapted from (Dunlop et al, 2017b; Yeo et

al, 2011).

Ventromedial Prefrontal and Orbitofrontal Regions of the VMN.

The rostroventral aspect of the ACC is comprised of two cytoarchitectonically distinct

sub-regions: the pregenual cingulate cortex (BAs 24 and 32) and the subgenual cingulate cortex

(BAs 25 and 33; sgACC) (Vogt et al, 1992). BAs 24, 25 and 33 have extensive reciprocal white

matter connections with subcortical, limbic and autonomic structures, including the ventral

striatum, amygdala, orbitofrontal cortex (OFC), AI, hippocampus, periaqueductal gray, and

autonomic nuclei in the brainstem (Devinsky et al, 1995). The OFC is broadly divided into

medial (mOFC; BA 12 and 25) and lateral (lOFC; BA 10, 11, and 47) sub-regions, based on

distinct cytoarchitecture and afferent/efferent innervations (Cavada, 2000; Chavis and Pandya,

1976; Henssen et al, 2016; Uylings et al, 2010). The mOFC has reciprocal white matter

connections with the amygdala, hippocampal formation and parahippocampal structures, the

retrosplenial cortex, anterior nucleus of the thalamus, and the hypothalamus, while the lOFC has

widespread connections to the DLPFC, AI, mediodorsal thalamus, IPL, amygdala, and temporal

pole (as reviewed by (Kringelbach and Rolls, 2004)). Patients with lesions in the ventromedial

prefrontal cortex (VMPFC) exhibit deficits in reward-based decision-making on gambling tasks

(Bechara et al, 2000), have poor prosocial cognition (Boes et al, 2011), and have blunted affect

and arousal (Schneider and Koenigs, 2017).

22

VMN Function and Reward Processing.

The activity of the VMPFC and mOFC is thought to track the relative value of expected

rewards. For example, one meta-analysis reviewing 142 human neuroimaging studies on reward-

related decision-making concluded that the mOFC and VMPFC preferentially activate during the

presentation of positive stimuli and receipt of rewards (Liu et al, 2011). VMPFC-dependent

reward value encoding is also conserved in monkeys (Tremblay and Schultz, 1999), and mice

(Lopatina et al, 2016). The VMPFC also has strong reciprocal connections with the ventral

striatum, such that coactivation of these regions is related to dopamine-dependent bottom-up

reward evaluation and reward anticipation, and glutaminergic top-down processing over the

striatum for reward learning and adaptive decision-making (Cauda et al, 2011; Christakou et al,

2004; Richard and Berridge, 2013; Sesack et al, 1989; Smith and Graybiel, 2013). In humans,

non-human primates, and mice, the central nucleus of the amygdala also projects to VMPFC to

mediate reward value encoding (Baxter et al, 2000; Fiuzat et al, 2017; Hampton et al, 2007;

Rudebeck et al, 2013; Seo et al, 2016a).

Some studies suggest that medial and lateral OFC sub-regions differentially encode

appetitive (rewarding) and aversive stimuli. For example, Small et al. (2001) fed healthy

individuals pieces of chocolate in a sequential fashion, and demonstrated that the mOFC is active

while individuals eat pleasurable foods prior until satiety. However, once satiety is reached, the

same chocolate stimuli that were once pleasurable then switched to preferentially activating the

lOFC over the mOFC, signifying a reappraisal of the chocolate from a pleasant to an aversive

stimulus (Small et al, 2001). Similar distinctions between mOFC/lOFC and appetitive/aversive

neural representations have been reported in some (Cheng et al, 2016; Hornak et al, 2004;

O’Doherty et al, 2001), but not all (Small et al, 2003) studies. The lOFC also plays a role in

adapting choice to favour a previously unrewarded stimulus (reversal learning) or to the most

rewarding outcome (as reviewed by (Fettes et al, 2017b; Kringelbach and Rolls, 2004)).

23

VMN Function During Emotion Regulation and Negative Affect.

Activity in the mOFC and pregenual cingulate activity has been associated with,

amongst other functions, inhibition of negative affect. For example, a number of studies have

reported that mOFC stimulation reduces conditioned fear responses via direct inhibitory

connections to the amygdala (Ghashghaei and Barbas, 2002; LeDoux, 2003; Likhtik et al, 2005;

Mcdonald et al, 1996; Quirk et al, 2003; Rosenkranz et al, 2003). Increased mOFC and

decreased amygdala activity has also been demonstrated during the voluntary suppression of

negative emotions (Delgado et al, 2008; Johnstone et al, 2007; Urry et al, 2006). Top-down

VMPFC activity to regions related to visceromotor function, including the hypothalamus and

periaqueductal gray, is also thought to reflect inhibitory regulation of autonomic responses to

negative stimuli and affect (Atlas and Wager, 2014; Barbas et al, 2003; Eippert et al, 2009).

Unlike the mOFC and other areas of the VMPFC, increased sgACC (BA 25) activity is

associated with the generation of negative mood and sadness (Mayberg et al, 1999; Ramirez-

Mahaluf et al, 2018), while increased DLPFC, dmPFC, and VLPFC/lOFC activity is correlated

with successful negative emotion regulation (Wager et al, 2008). Taken together, it appears that

individual nodes of VMPFC and VLPFC are involved with the regulation of negative affect

(mOFC, pregenual ACC, and VLPFC/lOFC in concert with dorsal cognitive regions), and the

generation of negative affect (sgACC).

How do IBNs Integrate Information Across Networks?

1.2.7.1 Corticocortical Connections

At present, the functional architecture and hierarchical arrangement of IBNs necessary to

execute complex and diverse behaviour is not fully understood. Fox, Raichle, and colleagues

originally proposed the existence of two broad functional networks: a task-positive network that

was activated, and an anticorrelated task-negative (default mode) network that was deactivated

(i.e, suppressed, inactive) during a wide range of tasks, stimuli, and overt behaviours (Fox et al,

2005). This interpretation of Fox et al.’s results is likely a somewhat oversimplified model of

brain network architecture, as it is now established that regions associated with a ‘task-negative’

network are active during a variety of internally-initiated tasks such as episodic memory retrieval

and spatial navigation (Spreng and Grady, 2010). More recently, a seminal publication by Power

24

and colleagues demonstrated that the brain regions of many IBNs cluster together based on

similar function and connectivity to form relatively self-contained and self-integrated ‘processing

units.’ IBNs that clustered in this way included the visual, sensorimotor and default mode

networks, while the frontoparietal CEN instead exhibited widespread inter-network connectivity

indicative of a higher-level control network linking lower-level, self-contained IBN ‘processing

units’ (Power et al, 2011). Building on these results, Cole and colleagues (2014) found that the

IBN structure described by Power et al. appears consistent across many task states (Cole et al,

2014). Of note, at least one study reported a method to generate single-participant level maps of

brain architecture across multiple tasks by taking into account dynamic shifts of brain

connectivity (Saggar et al, 2018).

Irrespective of the whole-brain functional architecture and interactions of all IBNs, a

number of human neuroimaging studies signify the importance of IBN cross-communication for

complex cognition and behaviour. For example, a study by Menon et al. demonstrated that the

SN plays a causal role in ‘switching’ between DMN- and CEN-specific states during an auditory

attention task, a visual attention task, and a task-free condition (Sridharan et al, 2008). Cognitive

load on a working memory task was also found to be correlated with the strength of within- and

between-network connectivity of the DMN and CEN (Newton et al, 2011). Communication

within and between the DMN and CEN is also correlated with successful emotion regulation

(Pan et al, 2018) and individual differences in creative thinking (Beaty et al, 2015).

1.2.7.2 Corticostriatal Connections

Loop-like circuits from regions of the cortex to the basal ganglia operate with the

thalamus and brainstem to prepare and implement planned, goal-directed behaviour (Figure 1-5;

as reviewed by (Haber, 2016)). Historically, studies of corticostriatal circuitry originally focused

on the cortico-striatal-thalamic loop serving the primary motor cortex, and its role in the control

of movement (Künzle, 1975), as well its association with neurodegenerative movement

disorders, including Huntington’s and Parkinson’s Disease. However, subsequent studies over

the last 40 years have found that basal ganglia loop circuits of similar architecture also serve

most of the cortical regions beyond primary motor cortex, and these do not have solely motor

functions, but also play a key role in motivational, affective, and cognitive functions. These

25

include corticostriatal projections from limbic and associative cortices that lead to goal-directed

action (Alexander et al, 1986). The prefrontal regions of the DMN, SN, CEN, and VMN all

possess their own corticostriatal projections (Figure 1-6), and it is theorized that IBNs facilitate

such behaviours via both parallel and integrative cortico-striato-thalamo-cortical (CSTC)

circuitry.

Figure 1-5: Overview of the general structural and pathways of cortico-striato-thalamo-

cortical circuits. Dark blue = direct pathway; light blue = indirect pathway. Amy = amygdala;

DS = dorsal striatum; GPe = globus pallidus, external segment; GPi = globus pallidus, internal

segment; Hipp = hippocampus; SN = substantial nigra; STN = subthalamic nucleus; Thal =

thalamus; VP = ventral pallidum; VS = ventral striatum; VTA = ventral tegmental area.

Reproduced with permission from (Haber, 2016).

26

Figure 1-6: Regions of frontal cortex that have distinct CSTC circuits. 10 = Brodmann area

10 (frontopolar cortex); dACC = dorsal anterior cingulate cortex; dlPFC = dorsolateral prefrontal

cortex; dmPFC = dorsomedial prefrontal cortex; Motor = primary motor cortex; OFC =

orbitofrontal cortex; PrM = premotor cortex; vlPFC = ventrolateral prefrontal cortex; vmPFC =

ventromedial prefrontal cortex. Reproduced with permission from (Haber, 2016).

There is now a large body of evidence indicating that a complex arrangement of CSTC

circuits, each serving a different prefrontal region via a corresponding striatal-thalamic circuit, is

involved in coordinating a diverse array of frontal lobe functions, including of reward, choice,

planning, and cognition, as well as movement. For example, Kim and Hikosaka recently posited

that the caudate nucleus has a rostrocaudal organization, such that more rostral striatal neurons

that have DLPFC, dmPFC and VMPFC afferents respond during reward-based learning and

complex actions that require conscious attention, while more caudal striatal neurons that have

sensorimotor afferents respond during simpler actions that do not require conscious attention

(Figure 1-7, as review by (Kim and Hikosaka, 2015)). Two recent neuroimaging studies have

also demonstrated that different aspects of reward function can be anatomically dissociated in the

striatum. During a reinforcement learning task, Smittenaar et al. reported that caudate nucleus

activity correlated to the expected value of rewards, and putamen activity instead correlated with

the type of motor response elicited during the task (e.g., hand vs. foot movement) (Smittenaar et

27

al, 2017). Likewise, Smith et al. demonstrated that the affective and informative (for example

spatial representation) properties of a reward stimulus are represented by distinct anatomical

circuits in the striatum (Smith et al, 2016). While it is clear from these recent studies that basal

ganglia circuits are involved in complex goal-directed behaviour and cognition, the actual role of

these circuits in planning, executing, and modifying behaviours based on internal or external

stimuli is still not fully understood.

Figure 1-7: Rostrocaudal organization of frontostriatal inputs. dACC = dorsal anterior

cingulate; dPFC = dorsal prefrontal cortex (dmPFC and DLPFC); OFC = orbitofrontal cortex;

Pre/Motor = premotor and primary motor cortex; vmPFC = ventromedial prefrontal cortex.

Reproduced with permission from (Haber, 2016).

The striatum itself is a large subcortical structure that can be subdivided functionally into

the dorsal striatum (caudate and dorsal putamen) and the ventral striatum (ventral putamen and

the core and shell of the nucleus accumbens [NAc]) (DiFiglia and Carey, 1986; Záborszky et al,

1985). Medium spiny neurons (MSNs) make up the majority of the neurons in the striatum,

28

while the rest consist of interneurons (Graveland and DiFiglia, 1985), forming clusters of cells

called cell islands (Goldman-Rakic, 1982). Striatal MSNs receive excitatory (glutamatergic)

inputs from pyramidal neurons in layers V and III of the prefrontal cortex (McFarland and

Haber, 2000), the thalamus, and dopaminergic inputs from brainstem nuclei (Haber et al, 2006;

Selemon and Goldman-Rakic, 1985; Somogyi et al, 1981) that each play distinct roles to regulate

striatal circuitry (Ding et al, 2008; Moss and Bolam, 2008; Smith et al, 2004). MSNs are also

phasically active, meaning that their firing rate is highest during execution of behavioural tasks

and their spontaneous firing rate is low (Crutcher and DeLong, 1984; Kimura, 1990; Kimura et

al, 1996).

Prefrontal projections to the striatum generally also follow a rostrocaudal topology

(Figure 1-7), although there is considerable overlap among the frontostriatal projections of

neighbouring cortical regions (Selemon and Goldman-Rakic, 1985). The primary sensorimotor

cortices innervate the caudal regions of the striatum, including the dorsolateral and central

putamen (Haber, 2016), and follow a somatotopic organization linked to specific patterns of

movements (Flaherty and Graybiel, 1994; Nambu, 2011; Yin et al, 2009). Prefrontal regions,

such as the ACC and OFC, project to the rostral striatum, including the medial caudate, ventral

rostral putamen, and the NAc (Chikama et al, 1997; Haber et al, 1995; Kunishio and Haber,

1994; Pandya et al, 1981). Some prefrontal regions project to distinct striatal structures. For

example, the NAc shell primarily receives input from the VMPFC (BA 25 and 32) (Haber et al,

1995). Other regions, such as those from the dACC and DLPFC, have extensive frontostriatal

connections, likely indicating crosstalk or integration with the CSTC circuitry of nearby nodes of

IBNs (Arikuni and Kubota, 1986; Calzavara et al, 2007). Medial and inferior temporal cortex

and limbic structures also innervate the striatum, and are thought to be responsible for integrating

memories and behavioural goals with sensorimotor and affective frontostriatal inputs (Friedman

et al, 2002).

Inhibitory striatal MSNs (Rav-Acha et al, 2005) project topographically to the pallidal

structures (internal and external globus pallidus [GPi and GPe, respectively], and ventral

pallidum) (Cummings, 1993) and substantial nigra (both the pars reticulata [SNr] and pars

compacta [SNc]) (Carpenter et al, 1976; Haber et al, 1990; Lynd-Balta and Haber, 1994; Parent

and Hazrati, 1995; Smith et al, 1998). Generally, the GPe and GPi receive input from the

putamen and caudate, while the ventral pallidum receives its input from the ventral striatum and

29

NAc (Haber et al, 1993; Lyons et al, 1996). Outputs from the pallidal structures form two

parallel yet functionally distinct pathways (as reviewed by (Haber, 2016)): first, the GPi and SNr

output to nuclei of the thalamus that, in turn, project back to the cortex (the ‘direct’ pathway,

considered to serve an excitatory role in regulating cortical activity); and second, the GPe and

ventral pallidum output first to the subthalamic nucleus (STN). The STN then projects to the

thalamus via the GPi (the ‘indirect’ pathway, considered to serve an inhibitory role in regulating

cortical activity). Alongside these two classical striatal pathways, the STN also receives

widespread cortical input from the frontal cortex via the more recently described ‘hyperdirect’

pathway (Haynes and Haber, 2013). Brainstem dopaminergic neurons from the SNc and ventral

tegmental area (VTA) also project to the dorsal and ventral striatum, respectively, modulating

the activity of corticostriatal projections (for a review, see (Haber, 2014)). The modulatory

effects of dopamine on cognitive and affective frontostriatal circuitry are thought to be involved

in reward-based learning and directing attention to relevant stimuli (Schultz, 1998; Wise and

Rompre, 1989).

Early studies reported that motor and cognitive/affective CSTC circuitry formed distinct,

functionally segregated loops. For example, Alexander et al. described five separable CSTC

loops stemming from the motor cortex, oculomotor cortex, DLPFC, lOFC and ACC (Alexander

et al, 1986). While there is a general rostrocaudal separation between cognitive/affective

(prefrontal) and motor (sensorimotor) projections in the striatum (Hanakawa et al, 2017;

Lehéricy et al, 2004; Di Martino et al, 2008; Parent and Hazrati, 1995), more recent evidence

from white matter tracing in monkeys (Averbeck et al, 2014; Haber et al, 2006) and mice

(Hintiryan et al, 2016), and neuroimaging in humans (Pauli et al, 2016; Tziortzi et al, 2014)

supports the existence of both parallel and integrative CSTC loops (Haber, 2003; Voorn et al,

2004). Early theories on integrative CSTC loops posited that the basal ganglia acts as a ‘funnel,’

integrating widespread inputs of diverse function and outputting to the motor cortex via the

ventrolateral thalamus to update or initiation behaviour (Allen and Tsukahara, 1974; Evarts and

Wise, 1984; Kemp and Powell, 1971). More recently, the thalamus has been proposed as an

integrative hub of IBNs (for a review see (Hwang et al, 2017)). However, the notion of a single

site acting as a universal ‘integrative’ hub is likely an overly simplistic model of CSTC function,

as there is considerable crosstalk within and between basal ganglia circuits.

30

Psychiatric Disorders in the Context of Brain Network Connectivity

Defining Abnormal Brain Connectivity

Neurobiological disturbances are typically identified by comparing individuals with a

particular psychiatric disorder relative to sex- and age-matched healthy individuals. There have

been two proposed ways to describe disturbances in IBNs related to psychiatric disorders: either

as abnormalities associated within a brain region itself, or abnormalities in how one brain region

connects to or interacts with other regions (as reviewed by (Menon, 2011)). Considering regional

abnormalities, these can include structural alterations (such as in cortical thickness or volume)

and/or functional disruptions to a brain region within an IBN, which could give rise to abnormal

behavior; lesion studies, examining the impact of focal lesions on behavior (for example,

(Robinson et al, 1984)), are a common example of this approach. Considering inter-regional

abnormalities, the study of the behavior of large-scale networks of regions and their interactions

within and amongst one another, can also contribute to our understanding of abnormal

behaviour. For example, psychiatric disorders can be characterized by abnormal white matter or

functional connections of a brain region to other regions (either to regions within the same IBN

or between different IBNs). Regional structural alterations or functional disruptions also appear

to have functional implications across widely distributed regions connected to the disrupted

region; thus, functional or behavioural deficits ‘downstream’ from structural abnormalities can

arise. For example, focal lesions following stroke cause contralateral structural and functional

abnormalities (Crofts et al, 2011) and inter-individual differences in the gray matter volume have

been correlated with behavioural performance on tasks (McTeague et al, 2016).

Another major consideration in defining abnormal brain connectivity is whether to search

for abnormalities associated with traditional, categorical psychiatric diagnoses (MDD, ED, OCD,

etc.), or instead to search for abnormalities that are associated trans-diagnostically with broad

categories of neurocognitive deficits that are apparent across diverse psychiatric disorders. As an

example of the latter approach, the National Institute of Mental Health’s Research Domain

Criteria (RDoC) offers a research framework to characterize broad dimensions of normal

behaviour, such as cognitive control or positive valence systems (Insel, 2014). The RDoC

approach works on the assumption that psychopathology is best understood not via categorical

disorders, but as the result of neurobiological abnormalities that lead to broader dimensions of

31

cognitive and behavioural abnormalties, which in turn interact to give rise to diverse patterns of

psychopathology associated transdiagnostically with a wide range of psychiatric disorders.

Transdiagnostic Alterations of Brain Networks

Consistent with the principles of RDoC, a number of studies have identified common

dimensions of abnormal behaviour, brain structure, and brain function across a wide range of

psychiatric disorders. This raises the possibility that IBN dysfunction contributes

transdiagnostically to psychiatric illness. In particular, abnormalities in the domain of cognitive

control may be a common element across many psychiatric disorders (as reviewed by

(McTeague et al, 2016)), because deficits in tasks related to cognitive control are observed

irrespective of psychiatric disorder class (Snyder et al, 2015). Additionally, large phenotypic

epidemiological studies across disorders have demonstrated that specific patterns of clinical

symptoms are strongly related to an underlying ‘general psychopathology’ factor (Carragher et

al, 2016; Krueger, 1999; Lahey et al, 2012). Notably, recent twin studies have reported that this

general psychopathology dimension has a strong genetic component, and consequently acute

disorder-specific clinical phenotypes are product of environmental influences (Kendler et al,

2003, 2011; Rhee et al, 2015). It also appears that heritable cognitive control deficits are distinct

from psychopathology, as these behavioural deficits often precede clinical disorder diagnosis,

worsen with disorder chronicity, and persist after remission. This phenomenon has been

observed across many different disorders, including major depression (Shilyansky et al, 2016),

psychosis (Bora and Murray, 2014), and substance abuse (Boelema et al, 2016; Peeters et al,

2014). Taken together, trait-like (genetic) deficits in cognitive control in combination with

environmental influences may be a risk factor for acute, state-like impairments specific to a

particular disorder class. The relationship between these broad cognitive control deficits and the

specific forms of psychopathology encountered in conventional psychiatric disorders such as

major depression, OCD, or eating disorders is not well understood.

It is likely that common structural and functional neural abnormalities underlie these

common broad behavioural deficits across disorders. Regarding structural abnormalities, three

large neuroimaging studies have reported that abnormalities of SN regions are common to many

diverse psychiatric disorders. Goodkind and colleagues (2015) were the first to demonstrate, in a

32

meta-analysis of 193 publications and 15,892 participants, that smaller gray matter volume

relative to healthy individuals in dACC and bilateral AI (regions of the SN) was a common

element across six disorders (schizophrenia, bipolar disorder, depression, addiction, obsessive-

compulsive disorder, and anxiety), and that gray matter loss in these regions correlated with poor

cognitive control (Goodkind et al, 2015). This result was replicated in a meta-analysis of 86

datasets and 6,508 individuals by Wise et al., who showed that gray matter loss in the bilateral

AI and dACC was common to both unipolar and bipolar depression (Wise et al, 2017). Finally

and most recently, Chang and colleagues have replicated this finding in a sample of 329

individuals with either schizophrenia, bipolar or unipolar depression and 156 healthy controls

(Chang et al, 2018). Structural abnormalities of the SN across diverse forms of psychopathology

thus appears to be a robust finding in current literature.

Regarding functional abnormalities, a number of studies have reported abnormal SN

activity during task-based and task-free fMRI, albeit there is considerably more heterogeneity in

their findings. Shanmugan and colleagues were the first to show that SN hypoactivity during a

working memory task was associated with overall psychopathology in 1,129 youth (including

individuals with phobias, depression and anxiety, behavioural disturbances such as attention-

deficit hyperactivity disorder, and psychosis) (Shanmugan et al, 2016). This group also reported

that youth with these same psychopathologies exhibited abnormally elevated cerebral blood flow

and abnormally low frontostriatal connectivity at rest in the dACC (Kaczkurkin et al, 2017).

Similarly, McTeague and colleagues recently demonstrated in a meta-analysis that SN hypo- or

hyperactivity was present across a variety of disorders during various cognitive control tasks

relative to controls (McTeague et al, 2017). Two other publications have examined

transdiagnostic abnormalities related to SN connectivity. First, patients with psychotic disorders

(either in bipolar disorder or schizophrenia-spectrum disorders) were found to exhibit SN

impairments in integrating information from the rest of the brain (global efficiency) (Sheffield et

al, 2017). Second, impaired reward processing was associated with hyperconnectivity between

the SN and VMN and hypoconnectivity between the SN and DMN across 5 different disorders

(Sharma et al, 2017). Like structural SN deficits, functional SN abnormalities appear to be

associated transdiagnostically with psychopathology, although there is substantial heterogeneity

in terms of whether this abnormality is related high versus low activity or connectivity.

33

Three extant theories attempt to integrate these transdiagnostic deficits in cognitive

control and SN dysfunction in terms of abnormal crosstalk to other IBNs. McTeague and

colleagues have proposed that transdiagnostic abnormalities in SN structure, function, and in

cognitive control relate to this network’s close relationship with the CEN (as discussed in

(McTeague et al, 2016)). As previously discussed, the SN and CEN co-activate during a wide

variety of tasks and are sometimes collectively grouped into a ‘task-positive network’ (although

McTeague and colleagues refer to this network as the ‘multiple demand network’). Studies of

causal interactions between these networks during cognitive tasks demonstrate that causal

interactions from the AI to the dACC during tasks requiring salience detection are proportionate

to task cognitive demand, and, in turn, lead to activations in CEN regions in the lateral frontal

and parietal cortex (Cai et al, 2016; Chen et al, 2015; Jiang et al, 2015). Consequently,

McTeague et al. posit that the deficits in cognitive control across diverse forms of

psychopathology represent disruptions in the relationship between the SN and CEN (McTeague

et al, 2016). Alternatively, Menon’s ‘unifying triple network’ hypothesis relates to the SN’s

possible role as a functional ‘switch’ between the CEN during externally-driven cognition, and

DMN during internally-driven cognition (Figure 1-8; as discussed in (Menon, 2011; Menon and

Uddin, 2010)). Consequently, SN dysfunction may lead to aberrant activity in these other

networks, resulting in difficulties integrating salient external or physiological events, as well as

difficulties accessing networks for attention and working memory or self-

referential/autobiographical thought. Finally, a theory proposed by Marsh and colleagues reviews

the possible transdiagnostic role of frontostriatal circuits and cognitive control (as discussed in

(Marsh et al, 2009a)). Briefly, they theorize that deficits in self-regulatory control across diverse

forms of psychopathology (including obsessive-compulsive disorder and eating disorders) are

related to either dysregulated parallel (distinct) or integrated frontostriatal circuits in medial and

lateral prefrontal cortex and orbitofrontal cortex.

To summarize, structural or functional abnormalities, either of specific brain regions or of

the brain-wide connections of IBNs, appear to underlie a common broad set of deficits in

cognition and behaviour evident transdiagnostically across a diverse range of psychiatric

disorders. The next three sections will discuss three psychiatric disorders of direct relevance to

this thesis, reviewing currently available evidence on structural and functional neuroimaging

abnormalities from a disorder-specific perspective.

34

Figure 1-8: Schematic of the Triple Network Theory of Psychopathology. ACC = dorsal

anterior cingulate cortex; DLPFC = dorsolateral prefrontal cortex; Ins = insula; mPFC = medial

prefrontal cortex; PCC = posterior cingulate cortex; PPC = posterior parietal cortex. Reproduced

with permission from (Nekovarova et al, 2014).

Psychopathology 1: Major Depressive Disorder Major depressive disorder (MDD) is a psychiatric condition characterized by depressed

mood and/or a marked reduction in one’s interest in, or one’s pleasure of derived from, daily

activities, often with associated disruptions of sleep, appetite, concentration, motivation, and

movement, as well as guilty, hopeless, or suicidal thoughts (American Psychiatric Association,

2013). MDD is also characterized by a profound impairment of day-to-day function, and

therefore the personal and economic burden of MDD is extremely high: MDD is the leading

cause of years lived with a disability (Ferrari et al, 2013), and is the fourth largest contributor to

35

the global burden of disease (Murray and Lopez, 1997). Quality of life is significantly impaired

in individuals with MDD, and this impairment increases with MDD severity (Ishak et al, 2013).

In the United States, MDD is associated with US$2 billion per month in direct workplace

productivity losses (Birnbaum et al, 2010), and the broader economic burden of MDD in the

USA accounted for US$210.5 billion in 2010 (Greenberg et al, 2015). MDD is also related to

increased short-term disability and more days out of work (Alonso et al, 2011; Kessler et al,

1999). In Canada, the total economic burden of all mental illness was $51 billion in 2003 (Lim et

al, 2008), and in Ontario, MDD-attributed burden was greater than the burden of breast,

colorectal, lung, and prostate cancer combined (Ratnasingham et al, 2013).

MDD Diagnosis

Current diagnostic criteria for MDD are specified in the fifth edition of the American

Psychiatric Association’s Diagnostic and Statistical Manual (DSM-5) (American Psychiatric

Association, 2013), and the tenth edition of the World Health Organization’s International

Statistical Classification of Behavioral Disorders (ICD-10) (World Health Organization, 1992).

There are a number of studies comparing the diagnostic performance of the DSM-IV, the fourth

edition of the DSM, versus the ICD-10; these studies reported subtle differences in assessing

disorder severity but overall concordance in those with moderate-to-severe MDD (Saito et al,

2010). Since the DSM-5 is predominantly used by clinicians in North America, this thesis will

focus on the DSM-5 diagnostic criteria for MDD.

According to the DSM-5, MDD is characterized by the near daily presence of a number

of symptoms during the same 2-week period that trigger a decline from the patient’s normal

functioning. These symptoms must not be attributable to another medical condition, and at least

one of the symptoms must be either depressed mood or a reduction in interest or pleasure. Other

symptoms can include: significant weight and appetite change; insomnia or hypersomnia;

psychomotor increases or decreases; fatigue; excessive worthlessness or guilt; impaired

concentration or decision-making; and recurrent suicidal ideation or thoughts of death (American

Psychiatric Association, 2013). Multiple subtypes of MDD are considered to exist because of the

diverse presentation of these symptoms (Insel and Cuthbert, 2015). Small, but meaningful,

differences exist between the diagnostic criteria of MDD between the fourth and fifth editions of

36

the DSM; for example, a definition of hopelessness has been introduced to the core depressed

mood criterion (for a review, see (Uher et al, 2014)).

A number of instruments can be used to support a MDD diagnosis, including clinician-

administered semi-structured interviews such as Structured Clinical Interview for DSM-IV-Axis-

I Disorders (SCID-I) (First et al, 2002), the Mini International Neuropsychiatric Interview

(MINI) (Sheehan et al, 1998), as well as self-administered questionnaires such as the Patient

Health Questionnaire-9 (PHQ-9) (Kroenke et al, 2001). In a recent meta-analysis, all three

batteries demonstrated good sensitivity and specificity (>80%) to diagnose and assess the

severity of MDD (Pettersson et al, 2015).

MDD Prevalence

In the United States, the lifetime prevalence of MDD is 13-16.2%, while the 12-month

prevalence is approximately 5-6.6% (Hasin et al, 2005; Kessler et al, 2003). Of these MDD

cases, 10.4% are mild, 38.6% are moderate, 38.0% are severe, and 12.9% are very severe using

standard reference ranges (Kessler et al, 2003). The mean age of onset of MDD is 30.4 years

(Hasin et al, 2005). The mean duration of a MDD episode is approximately 16 weeks (Kessler et

al, 2003); however 26.5% experience episodes lasting at least two years (Rubio et al, 2011). In

Canada, the 12-month prevalence of MDD is between 4.9-8.2% (Patten et al, 2015; Vasiliadis et

al, 2007), and the lifetime prevalence is 11.3% (Patten et al, 2015). Women are at approximately

double the risk of developing MDD compared to men, and, as is also true in the USA, most

Canadian cases of MDD are episodic in course; 50% of episodes will resolve within 3 months

(Patten et al, 2015). Furthermore, about one third of remitted participants will experience a

subsequent episode within 3 years of remission (Hasin and Grant, 2015), and the mean number

of lifetime episodes is 4.7 (Hasin et al, 2005).

MDD Etiology

MDD is a complex and heterogeneous disorder thought to arise from an interaction of

genetic, biological, environmental, and psychosocial influences. MDD is a somewhat familial

37

disorder, being 2.84 times more common among first-degree relatives of an individual with

MDD (Sullivan et al, 2000). Furthermore, twin and sibling studies have found that

approximately one-third of the risk for developing MDD is heritable, while the remaining two-

thirds are attributable to environmental sources (Sullivan et al, 2000). The remaining variability

in MDD risk is likely attributable to individual-specific environmental factors, including early

life trauma and other stressors (Mullins et al, 2016). Notably, predominant gene-environment

interaction models highlight the role of genotypes related to monoamine function, including

serotonin, norepinephrine, and dopamine systems, and their interaction with environmental

stressors and early life trauma (for a review, see (Saveanu and Nemeroff, 2012) and (Lopizzo et

al, 2015)).

MDD Comorbidity

72.1% of lifetime and 78.5% of 12-month cases of MDD have a comorbid DSM-IV

diagnosis (Kessler et al, 2003). In the 2003 National Comorbidity Survey Replication Study,

59.2% of MDD patients had a comorbid anxiety disorder and 24% had a comorbid substance-use

disorder (Kessler et al, 2003). Personality disorders (most commonly, obsessive-compulsive

personality disorder) are also frequently comorbid with MDD (Hasin et al, 2005) (37.9% of

cases). Furthermore, the presence of somatic abnormalities such as chronic fatigue and pain are

often comorbid with MDD (for a review, see (Bair et al, 2003)). MDD patients with

comorbidities show poorer treatment outcomes relative to MDD patients unaffected by such

comorbidities. For example, MDD patients with comorbid anxiety tend to have higher suicide

rates and disorder chronicity, and poorer response to first-line treatments relative to MDD cases

without comorbid anxiety (Seo et al, 2011).

Assessment of Treatment Severity

The symptom severity of MDD can be assessed through standardize clinician- or self-

administered instruments. In clinician-administered batteries such as the Hamilton Rating Scale

for Depression (HAMD) (Hamilton, 1960) or Inventory of Depressive Symptoms (IDS) (Rush et

al, 2003), trained assessors or clinicians assess the severity of MDD using a semi-structured

38

interview that asks directed questions on the severity or frequency of a given symptom,

supplanted by direct observations of the patient’s appearance and behaviour. Self-administered

questionnaires such as the Beck Depression Inventory II (BDI-II) (Beck et al, 1961, 1996) and

Quick Inventory of Depressive Symptoms (QIDS) (Rush et al, 2003) are an alternative to

clinician-administered scales that quicker to administer and positively correlate to clinician-

administered metrics (Enns et al, 2000; Schneibel et al, 2012). However, multiple studies have

noted factors that bias scoring on both self-reported and clinical-administered scales. For

example, more severe scores on self-reported measures are significantly correlated with

personality traits, such that individuals who report high neuroticism and low extraversion also

report higher severity on the BDI-II (Enns et al, 2000; Schneibel et al, 2012). Clinician-

administered assessments such as the HAMD also show a greater reduction of symptoms scores

during treatment relative to patient-reported questionnaires (Schneibel et al, 2012).

The HAMD is a clinician-administered assessment that is one of the most commonly

used metrics assessing MDD severity over the period of one week (Hamilton, 1960). The 17-

item version of the scale assesses the following items: 1) Depressed mood; 2) Feelings of guilt

and self-criticism; 3) Suicidal thoughts and ideation; 4-6) Disturbances in sleep (early, middle,

late insomnia); 7) Interest and pleasure of daily activities, including social activities and self-

care; 8) psychomotor slowing; 9) Agitation; 10) Anxious thoughts and irritability; 11) Somatic

symptoms of nervousness; 12) Loss of appetite; 13) Fatigue and loss of energy; 14) Interest in

sex; 15) Hypochondriasis; 16) Weight loss; and 17) Acknowledgment or insight of depression.

The scale contains additional items to assess MDD subtypes, but they are not as widely used

(Hamilton, 1960). Each item is rated on a scale of 0-2 or 0-4 and item ratings are summed to

generate a score. The total HAMD score has reference ranges from subclinical/minimal

depression (0-7 HAMD score), mild depression (8-16 HAMD score), moderate depression (17-

23 HAMD score), and severe depression (24-52 HAMD score) (Zimmerman et al, 2013).

Response to a given intervention is conventionally defined as a >50% reduction in symptom

score, and remission is typically defined at a 17-item HAMD score <8 (O’Donovan, 2004).

The BDI-II is a 21-item patient-administered questionnaire that assesses MDD severity

(Beck et al, 1961, 1996). Patients rate the following symptoms on a 0-3 scale: 1) Sadness; 2)

Pessimism; 3) Past failures; 4) Loss of pleasure; 5) Guilty feelings; 6) Feelings of punishment; 7)

Self-dislike; 8) Self-Criticalness; 9) Suicidal thoughts or wishes; 10) Crying; 11) Agitation; 12)

39

Loss of interest; 13) Indecisiveness; 14) Worthlessness; 15) Loss of energy; 16) Changes in

sleeping pattern; 17) Irritability; 18) Changes in appetite; 19) Concentration difficulty; 20)

Tiredness or fatigue; and 21) Loss of interest in sex. Item ratings are summed to indicate overall

depression severity. A BDI-II score of 0-13 indicates no or minimal depression; 14-19 indicates

mild depression; 20-28 indicates moderate depression; 29-63 reflects severe depression (Beck et

al, 1996). Response to a given treatment on the BDI-II is defined as a >50% reduction, and

remission is typically defined at a BDI-II score <12 (Riedel et al, 2010).

Structural and Functional Disruptions in MDD

The neuroimaging literature on MDD is extensive, beginning in the late 1980s with early

PET studies and growing steadily to hundreds of articles a year as of the present day. This

section will focus on structural and functional neuroimaging findings in MDD that are of

particular relevance to the experimental work of the present thesis.

Abnormal gray matter volume of both cortical and subcortical regions has been linked to

MDD. Studies have consistently found that, relative to healthy controls, MDD patients exhibit

reduced gray matter volumes in cortical regions of the VMN and SN, including the dACC,

dmPFC, mOFC, sgACC, and in striatal regions such as the caudate and putamen (Bora et al,

2012c; Webb et al, 2014; Zhao et al, 2014). Other recent studies have reported lower gray matter

volume in paralimbic regions, including the inferior hippocampus (Zou et al, 2010), and in

dorsolateral prefrontal regions, including the DLPFC and IFG (Du et al, 2012). In terms of the

effects of gender, one meta-analysis found that the proportion of women in each study included

in the meta-analysis negatively correlated with gray matter volume in the right hippocampus

(Zhao et al, 2014). Such structural abnormalities highlight the prominent role of prefrontal CSTC

and paralimbic gray matter abnormalities in MDD pathophysiology (Bora et al, 2012c;

Shepherd, 2013).

Functional abnormalities have also been reported since the earliest days of neuroimaging

studies in MDD, with classic and widely-cited examples being observations of hyperactivity in

the ventromedial prefrontal cortex and specifically in the subgenual cingulate cortex (reviewed in

(Drevets et al, 2008)). Subsequent work highlighted the important role of altered functional

40

network connections from this region to other cortical networks in the pathophysiology of MDD

(Mayberg, 2003). In perhaps the most widely-cited example, Grecius and colleagues (2007)

reported increased functional connectivity between the subgenual cingulate cortex and the

DMN’s CSTC in patients with MDD (Greicius et al, 2007). More recently, abnormalities in

terms of local fMRI signal synchrony and oscillatory power have also been observed in MDD

patients relative to healthy controls. For example, MDD is associated with less local oscillatory

synchrony in the SN left insula (Li et al, 2014b) and ACC (Xue et al, 2016), and higher local

synchrony in the DMN mPFC region (Iwabuchi et al, 2015; Xue et al, 2016). MDD is also

characterized by low oscillatory power in the superior and middle frontal gyri (Zhang et al,

2017), left DLPFC (Wang et al, 2012a), sgACC (Lai and Wu, 2015), bilateral mOFC (Wang et

al, 2012a), inferior and middle temporal cortices (Shen et al, 2014) and the parahippocampal

gyrus (Liu et al, 2013). The relative amplitude of oscillations of different frequencies also

significantly differs in MDD relative to controls in the VMPFC, IFG and PCC (Wang et al,

2016). From a network perspective, there is abnormal oscillatory power in low frequency

fluctuations in the brain regions of the DMN, VMN, and corticostriatal circuits (Guo et al, 2013;

Tadayonnejad et al, 2015).

Maladaptive, self-referential rumination, a hallmark clinical feature of MDD, is

associated with a functional imbalance between IBNs. Rumination in adolescent, adult, and late-

life MDD is associated with hyperactivity localized to regions affiliated with the VMN and

DMN, and hypoactivity localized to regions affiliated with the SN and CEN (Alexopoulos et al,

2012; Bartova et al, 2015; Belleau et al, 2015; Hamilton et al, 2011b; Lemogne et al, 2012;

Rzepa and McCabe, 2016; Sheline et al, 2009; Zhu et al, 2017). For example, one resting-state

fMRI study reported that sgACC (VMN) and hippocampal (DMN) hyperactivity was inversely

correlated to hypoactivity localized to the dmPFC/dACC (SN) and DLPFC (CEN) (Hamilton et

al, 2011a). In another resting-state fMRI study from the same group, DMN hyperactivity was

positively correlated with levels of maladaptive rumination in MDD, while the onset of SN and

CEN activity was correlated with reflective, adaptive rumination (Hamilton et al, 2011b). The

authors interpret this maladaptive relationship to poorer top-down SN and CEN processes over

the DMN.

Similarly, MDD patients, relative to healthy controls, exhibit abnormal DMN and VMN

hyperactivity to aversive stimuli and negative affect. For example, the ACC and ventral striatum

41

are hyperactive during negative self-referential rumination (Wagner et al, 2013, 2015). Striatal

nodes of the VMN are similarly hyperactive while viewing aversive images (McCabe et al,

2009). Hyperactivity from regions of the DMN, the sgACC, and the lOFC have also been related

to heightened levels of negative self-focus and negative affect (Cooney et al, 2010; Mayberg et

al, 1999; Philippi et al, 2015). For example, MDD patients exhibit hyperconnectivity between

the PCC and the lOFC region of the VMN relative to healthy controls; the authors interpret this

finding that negative affect and rumination is associated with abnormally high connectivity

between regions associated with non-reward and autobiographical memory (Cheng et al, 2018).

Furthermore, the sgACC appears to play a clear role in generating negative affect in healthy

controls. One study recently reported that, in healthy controls, the sgACC regulates interactions

dorsal cognitive and ventral affective networks during sadness provocation (Ramirez-Mahaluf et

al, 2018).

Conversely, mesocorticolimbic hypoactivity of the SN and VMN have also been linked

to poor incentive salience and anhedonia, another hallmark clinical feature of MDD (Nestler and

Carlezon, 2006; Tremblay et al, 2005). During task-based fMRI, MDD patients display ACC and

nucleus accumbens hypoactivity to reward feedback relative to healthy controls (Knutson et al,

2008; Pizzagalli et al, 2009; Robinson et al, 2012), and also display an absence of VMN activity

for pleasant sights and tastes (McCabe et al, 2009). Relative to controls, MDD patients also

exhibit hypoactivity in the OFC, ACC and ventral striatum during the receipt of a reward, and

during the reward anticipation phase of monetary incentive delay tasks (Admon et al, 2015;

Chantiluke et al, 2012; Manelis et al, 2016; Schiller et al, 2013; Smoski et al, 2011).

MDD patients also display deficits in cognitive control compared to their non-depressed

counterparts, and this deficiency has been attributed to abnormal functioning of the SN and CEN.

For example, MDD patients exhibit dACC hyperactivity to tasks that assess working memory

and cognitive control, including the n-back (Harvey et al, 2005) and Stroop tasks (Holmes and

Pizzagalli, 2008; Wagner et al, 2006). Volumetric reductions in SN nodes have also been related

to behavioural deficits on such cognitive tasks (Li et al, 2010a). Furthermore, MDD patients

display higher connectivity at rest between the striatum and DLPFC relative to controls;

frontostriatal hyperconnectivity also positively correlated with disease severity (Kerestes et al,

2015). Abnormal CEN and SN frontostriatal connectivity has also been attributed to

42

inappropriate top-down processing of negative affect and aversive stimuli (Furman et al, 2011;

Kerestes et al, 2015).

It is important to note that a substantial number of MDD neuroimaging studies report

considerable inter-individual variability in findings, perhaps reflecting the substantial clinical

heterogeneity of the disorder. To examine this inter-individual variability, two recent

neuroimaging publications have reported on MDD subtypes, or on MDD dimensions of illness,

characterized by distinct patterns of clinical presentation and/or functional connectivity. In one

study, patients with MDD were positioned along 3 RDoC-inspired dimensions of illness

encompassing reward sensitivity (positive valence systems), neuroticism (negative valence

systems), and cognitive control; each dimension was associated with a distinctive EEG signature,

with deficits in cognitive control being associated with low gamma power in Sn regions

including the DLPFC (Webb et al, 2016). In another notable study, a novel clustering method

was used to define robust MDD subtypes by shared patterns of abnormal resting-state

connectivity; the authors showed in a large multisite sample that MDD patients could be

partitioned into four functional connectivity-based subtypes. The biotypes showed that the

variability in patients’ functional signatures was associated with two important dimensions of

clinical symptoms: anhedonia and anxiety, each with a distinctive pattern of whole-brain

functional connectivity. Of particular clinical relevance, the authors also showed that biotype

diagnoses were longitudinally stable over a one-month period in actively depressed patients, and

that the biotypes could be used to predict response rates to rTMS treatment (Drysdale et al,

2017).

Psychopathology 2: Obsessive-Compulsive Disorder Obsessive-compulsive disorder (OCD) is characterized by intrusive, distressing thoughts,

and repetitive behaviours aimed at reducing the aforementioned distress (American Psychiatric

Association, 2013). OCD tends to begin in childhood or early adulthood, and is a chronic and

often treatment-refractory condition (Ruscio et al, 2010). OCD is a severely disabling disorder

for both the sufferer and their family in terms of social and work functioning (for a review, see

(Steketee, 1997)). Even mild-to-moderate symptoms of OCD can lead to severe reductions in

43

quality of life (Moritz et al, 2005), and OCD is related to reductions in virtually all dimensions

of quality of life measures (Fontenelle et al, 2010).

OCD Diagnosis

As with MDD, the diagnosis of OCD in North America is typically made using DSM-5

criteria. OCD falls under the obsessive-compulsive disorders section of the DSM-5, which also

encompasses related disorders such as body dysmorphic disorder, trichotillomania and

excoriation disorder (American Psychiatric Association, 2013). According to the DSM-5

guidelines, OCD is defined by two criteria: first, the presence of obsessions, compulsions, or

both; and second, that the obsessions or compulsions are present for more than 1 hour daily or

cause clinically-significant distress in some important area of functioning. The DSM-5 defines

obsessions using two criteria: first, that the recurrent thoughts or urges are intrusive and

unwanted, and cause marked anxiety or distress; and second, that the patient tries to suppress

these thoughts or dispel them with some action, referred to as a compulsion. The DSM-5 defines

two criteria for a behaviour to be considered a compulsion: first, compulsions must be repetitive

behaviours or thoughts, including washing, checking, or counting, that the patient is driven to

perform in response to an obsession; and second, these acts are aimed at reducing or preventing

or reducing the distress associated with an obsession notwithstanding that they are not

realistically connected (American Psychiatric Association, 2013). According to one study, most

people with OCD tend to have both obsessions and compulsions (Foa et al, 1995).

A number of instruments can be employed by a clinician to support a OCD diagnosis,

including clinical-administered semi-structured interviews like the SCID-I (First et al, 2002), the

MINI (Sheehan et al, 1998), and patient-rated questionnaires like the PHQ-9 (Kroenke et al,

2001).

OCD Prevalence The lifetime prevalence of OCD is approximately 1-3% (Abramowitz et al, 2009), while

the 12-month prevalence of OCD in the United States and internationally is 1.2% and 1.1-1.8%,

44

respectively (Kessler et al, 2005; Ruscio et al, 2010). The lifetime and annual prevalence of

OCD is comparable in Canada (Weissman et al, 1994). The mean age of onset for OCD is 19.5

years, and a quarter of patients’ OCD onset begins by 14 (Kessler et al, 2005; Ruscio et al,

2010). Males tend to have an earlier onset of OCD relative to females, as nearly a quarter of

males have an onset of OCD before the age of 10 (Weissman et al, 1994). Disease prevalence is

approximately equal between the two sexes (Ruscio et al, 2010).

OCD Etiology Akin to MDD, OCD is a fairly heritable disorder, and its etiology likely involves an

interaction effect of genetic and biopsychosocial factors (Taylor and Jang, 2011). In one review,

the prevalence of OCD among individuals with adult relatives affected by OCD was two times

higher relative to controls, and the OCD prevalence rate among individuals with adolescent or

child relatives with OCD was nearly 10 times higher than that of controls (Pauls, 2010). Genetic

factors are often linked to a number of polymorphisms in genes associated with serotonin and

catecholamine function (Taylor, 2016). Additionally, environmental insults like perinatal insults

and physical abuse are associated with an increased risk of OCD onset (Grisham et al, 2011).

OCD Comorbidity OCD is often comorbid with other psychiatric disorders. The most common

psychopathologies include an anxiety disorder in 76% of OCD cases, MDD or bipolar disorder

in 41% of cases (Ruscio et al, 2010), and obsessive-compulsive personality disorder in 23-32%

of cases (Eisen et al, 2010). Another disorder commonly comorbid with OCD is tic disorder (in

30% of OCD cases); patients with this comorbidity are predominately male, and tend to differ in

symptom prevalence and prognosis compared to those without tic disorder (Leckman et al,

2010). This difference is suggestive of OCD subtypes reflected potentially by an etiological

difference.

45

Assessment of Symptom Severity The Yale-Brown Obsessive-Compulsive Scale II (Y-BOCS) is considered the gold

standard measure of OCD symptom severity (Storch et al, 2010). The Y-BOCS is a clinician-

rated symptom checklist assessing past or current obsessions and compulsions. The symptom

checklist contains 14 items: 1) Aggressive obsessions, such as a fear of harming oneself or

others; 2) Contamination obsessions, such as a concern with environmental contaminants or

animals; 3) Sexual obsessions; 4) Religious obsessions; 5) Obsession with need for symmetry or

exactness; 6) Somatic obsessions like a concern with illness or disease; 6) Hoarding or saving

obsessions; 7) Miscellaneous obsessions like a fear of saying certain things or other

superstitions; 8) Cleaning or washing compulsions like excessive handwashing; 9) Checking

compulsions like repeatedly checking locks; 10) Repeated rituals; 11) Counting compulsions;

12) Ordering or arranging compulsions; 13) Hoarding/collecting compulsions; and 14)

Miscellaneous compulsions like excessive list making or other mental rituals. Y-BOCS scores

range between 0-40, with scores over 7 indicating clinically significant OCD symptoms.

Multiple studies report high reliability and construct validity for this measure (Goodman et al,

1989; Goodman and Price, 1992; Woody et al, 1995). Signal detection analyses suggest that a

reduction of at least 25-35% on the Y-BOCS is required to demonstrate OCD treatment

response, while treatment remission is often defined by a score of less than 13 on the Y-BOCS

(Farris et al, 2013).

Structural and Functional Abnormalities of OCD

As with MDD, neuroimaging studies in OCD report structural and functional

abnormalities in the DMN, SN and VMN and to affiliated frontostriatal circuitry related to the

self-regulation of thoughts, emotions, and behaviour (van den Heuvel et al, 2010; Lipsman et al,

2013a; Marsh et al, 2009a). Structurally, OCD patients have smaller gray matter volume in SN

and VMN cortical nodes like the ACC and OFC (Gilbert et al, 2008; Kühn et al, 2013; Reess et

al, 2016) and fronto-parietal regions (Rotge et al, 2010), and poorer white matter integrity

between these networks (Hoexter et al, 2012; Rotge et al, 2010). OCD patients also have

increased basal ganglia grey matter volume in regions like the thalamus and ventral striatum

(Atmaca et al, 2016; Hou et al, 2013).

46

Hyperactive VMN and orbitofrontal frontostriatal circuits are often associated with

aversive, obsessive thoughts and compulsive behaviours. OCD patients display VMN-striatal

hyperconnectivity to punishments during a monetary incentive delay task (Beucke et al, 2012),

during symptom provocation, and at rest (Anticevic et al, 2014; Cocchi et al, 2012; Figee et al,

2013; Harrison et al, 2009). While VMN-CSTC hyperconnectivity has been replicated in many

studies, one study found VMN-CSTC hypoconnectivity during the absence of a rest task

condition in unmedicated OCD patients relative to controls (Posner et al, 2014). Abnormal

frontostriatal hyperactivity from the VMPFC and OFC is thought to underlie dysfunctional

negative or aversive obsessive thoughts and ritualistic behaviours (Graybiel and Rauch, 2000;

Nakamae et al, 2014). Most strikingly, one study reported that OCD patients have DMN and

VMN hyperconnectivity related to abnormal reward processing and the inability to inhibit

compulsive behaviours (Koch et al, 2018). The role of these circuits in the etiology of OCD is

further supported by one recent animal study: optogenetically-induced chronic medial PFC

CSTC hyperactivity has produced repetitive grooming and other OCD behaviours in mice

(Ahmari et al, 2013).

The SN and CEN, and their frontostriatal connections are also linked to deficits in

cognitive control in OCD. For example, relative to controls, in OCD patients, the dACC is

abnormally hyperactive during symptom provocation and while committing errors in cognitive

control tasks (Bourne et al, 2012). Frontostriatal abnormalities related to the SN and CEN also

appear to be hyper-connected or hyperactive at rest; hyperconnectivity relative to controls has

been observed in OCD between the medial dorsal nucleus of the thalamus, dACC (Bourne et al,

2012; Rauch et al, 2006), caudate (Graybiel and Rauch, 2000) and DLPFC (Figee et al, 2013).

Furthermore, DLPFC frontostriatal connectivity and insulo-striatal connectivity is positively

correlated to OCD symptom severity (Bernstein et al, 2016; Vaghi et al, 2017).

The local function of brain regions in terms of oscillation synchrony and power is also

abnormal in OCD patients relative to controls. Local temporal synchrony is abnormally high in

the right ACC, OFC, bilateral middle frontal gyrus and precuneus, and is abnormally low in the

bilateral caudate and PCC (Chen et al, 2016; Yang et al, 2010, 2015). The amplitude of low-

frequency fluctuations of fMRI activity is also abnormal in OCD relative to controls. OCD

patients exhibit decreased amplitude in the mOFC and dmPFC (Giménez et al, 2017), and in

CSTC circuits (Qiu et al, 2017) relative to controls. Furthermore, the amplitude of low frequency

47

fluctuations in the DMN is also disrupted, with higher amplitude found in the ACC and

decreased in the PCC (Cheng et al, 2013).

Many OCD neuroimaging studies report substantial inter-individual variability, and this

variability is thought to represent the diverse clinical presentations and symptom severity of the

disorder (Piras et al, 2015). However, only a handful of studies attempt to parse this clinical

heterogeneity into underlying neurobiological substrates. Two MRI studies compared OCD

patients with intact versus poor insight. One study found that OCD patients with poor insight had

a significantly higher frequency of morphological abnormalities (Aigner et al, 2005). Another

more recent resting-state fMRI study reported a positive correlation in the activity between the

right anterior insula (SN) and right OFC (VMN), and the level of insight (Fan et al, 2017). Other

studies have attempted to differentiate functional or structural differences by clusters of OCD

symptoms. For example, two recent publications have reported volumetric and functional

connectivity differences between patients with reactive (i.e., contamination or asymmetry) versus

autogenous (aggressive or religious) obsessions (Harrison et al, 2013; Subirà et al, 2013).

Another structural MRI study found differences in gray matter volume of the cerebellum and

insula in patients with aggression/checking obsessions versus those with contamination

obsessions (Okada et al, 2015). Further research is necessary to further disentangle the complex

cognitive and behavioural underpinnings of OCD symptomatology.

Psychopathology 3: Anorexia Nervosa and Bulimia Nervosa Eating disorders such as anorexia nervosa (AN) and bulimia nervosa (BN) carry a

disproportionately high burden of illness, both socially and individually (reviewed in further

detail below). Eating disorders are also associated with a high mortality rate. For example, one

study reported that approximately 10% of AN sufferers die within 10 years of disease onset

(Sullivan, 1995); more recent reports suggest that the current crude mortality rate is

approximately 5% and 2% per decade for AN and BN respectively; death is most often

associated with disorder-associated complications or suicide (Arcelus et al, 2011; Hoek, 2006).

According to a recent meta-analysis, the overall standard mortality ratio for AN is 5.86,

substantially higher than schizophrenia (2.8), MDD (1.6) and bipolar disorder (2.1) (Arcelus et

al, 2011). More broadly, high eating disorder symptom severity and low weight are indicators of

48

low quality of life (Bamford and Sly, 2010). Specialized eating disorder treatment programs

currently lack adequate capacity to meet demand (Hart et al, 2011). Patients who can access

these programs are often met with high out-of-pocket costs and economic hardship that impedes

treatment adherence, highlighting the need for more accessible and effective treatment options

(Gatt et al, 2014).

Diagnosis of Anorexia and Bulimia Nervosa AN and BN are categorized in the Feeding and Eating Disorders section of the DSM-5,

amongst pica, rumination disorder, and binge-eating disorder (American Psychiatric Association,

2013). AN is diagnosed using three criteria: first, the patient restricts energy intake to less than

what they require, leading to a significantly low body weight for the individual’s age, sex,

developmental trajectory, and physical health; second, the patient has an intense fear of gaining

weight, despite their significantly low weight or continued weight loss; and third, the patient

experiences some disturbance in the way they perceive their body weight or shape, or do not

recognize the seriousness of their current low body weight (American Psychiatric Association,

2013). The DSM-5 also distinguishes between two subtypes of AN: restricting (AN-R) and

binge-eating/purging (AN-BP) (American Psychiatric Association, 2013). These two subtypes

differ depending on whether the patient has engaged in recurrent binge eating or purging

behaviour, including self-induced purging or laxative misuse, in the last three months. AN can be

in partial remission if the patient no longer meets the low body weight criteria for a sustained

period of time, but meets one of the two other criteria for AN.

The DSM-5 provides the following five diagnostic criteria for BN: first, recurrent

episodes of binge eating; second, recurrent compensatory purging behaviours, like self-induced

purging or laxative misuse, to prevent weight gain; third, both binge-eating and compensatory

purging behaviours occur, on average, at least once a week for three months; fourth, the patient’s

self-evaluation is excessively influenced by body shape or weight; and fifth, these disturbances

in binge-eating and purging do not occur during episodes of AN. The DSM-5 defines an episode

of binge-eating as a discrete period of time in which the patient eats an amount of food larger

than what most other individuals would eat in a similar period of time. This rapid eating is

49

accompanied by a feeling that they cannot stop or control what or how much they were eating

(American Psychiatric Association, 2013).

A number of clinical measures are in use to assess AN or BN symptom severity. One of

the primary measures for AN is the body mass index (BMI) as a measure of expected body

weight for height (American Psychiatric Association, 2013). However, weight assessment in AN

is challenging, because healthy weights vary across individuals substantially and cut-offs for

underweight and normal weight statuses are debated (Thomas et al, 2009). Nevertheless, the

World Health Organization and United States Centers for Disease Control defines a low weight

as a BMI lower than 18.5 kg/m2. As with MDD and OCD, the semi-structured, clinician-

administered MINI can be used to establish a diagnosis for both AN and BN (Golden et al, 2003;

Sheehan et al, 1998). The Eating Disorder Examination and Questionnaire, discussed below, can

also be used to determine an eating disorder diagnosis and assess elements of symptom severity,

such as the frequency of food binge episodes, or compensatory behaviours such as purging

episodes.

AN and BN Prevalence The lifetime prevalence of all eating disorders has been reported at 5.7% for females and

1.2% in males (Hudson et al, 2007). By diagnosis, the lifetime prevalence of AN is 0.9% for

females and 0.3% for males, and the lifetime prevalence of BN is 1.5% among females, and

0.5% among males (Hudson et al, 2007). In a more recent study amongst female adolescents, the

reported lifetime prevalence was 1.7% for AN, and 0.8% for BN (Smink et al, 2014). The 12-

month prevalence of AN among adolescent females is 0.4%, and 1% for BN (Hoek, 2006). As

indicated by disparities in prevalence, AN and BN predominantly affect females; studies report

an estimated 10:1 female-to-male ratio in AN clinical populations (Götestam et al, 1998; Lucas

et al, 1991). A 10:1 female-to-male ratio is also reported in BN (Keski-Rahkonen et al, 2009;

Swanson et al, 2011).

50

Etiology of AN and BN Like other psychiatric disorders, the etiologies of both AN and BN are theorized to

involve a complex interaction of genetic and environmental factors (for a review, see (Pinheiro et

al, 2009)). First-degree relatives of those with AN have a 10 times higher lifetime risk of AN

than relatives of unaffected individuals (Lilenfeld et al, 1998; Strober et al, 2000, 2001). In a

large twin study, it was reported that the heritability of AN was 0.56; the remaining variance was

attributed to shared (0.05) and unique environmental factors (0.38) (Bulik et al, 2006; Klump et

al, 2001; Kortegaard et al, 2001; Wade et al, 2000). Genotype linkage analyses of AN suggest a

number of susceptible loci, particularly so among serotonergic and opioid receptor genes

associated with AN psychopathology or feeding behaviour (Devlin et al, 2002; Grice et al,

2002). Likewise, trait neuroticism is associated with an increased risk of AN (Bulik et al, 2006).

It is also worthwhile to note that the prevalence and presentation of AN and BN varies by

socioeconomic status and culture, further emphasizing the role of environmental factors in

disorder etiology (Hoek et al, 2005; Keel and Klump, 2003). Similarly, BN is also proposed to

have a strong familial component (for a review, see (Striegel-Moore and Bulik, 2007)), with twin

studies in BN reporting heritability estimates between 28-83% (Bulik et al, 1998b; Lilenfeld et

al, 1998; Thornton et al, 2011). In one large family study, first-degree female relatives of those

with AN or BN had at least a fourfold increased risk of developing BN (Strober et al, 2000).

AN and BN Comorbidity Comorbidities are common in AN and BN. One survey found that 56.2% of patients with

AN and 94.5% of patients with BN met criteria for another psychiatric disorder (Hudson et al,

2007). The lifetime prevalence of at least one comorbid anxiety disorder in AN and BN is

approximately 71% (Godart et al, 2003). Personality disorders such as obsessive-compulsive

personality disorder and borderline personality disorder are also frequently comorbid in AN and

BN (for a review, see ((Bruce and Steiger, 2005; Crane et al, 2007)). Mood disorders such as

MDD and dysthymia are present in 42.1% of AN and 70.7% of BN patients (Hudson et al,

2007). Impulse control disorders and substance use disorder are also present in 30.8% and 27%

of AN patients, respectively, and in 63.8% and 36.8% of BN patients, respectively (Hudson et al,

2007). The presence of any comorbidity is frequently associated with poorer treatment outcomes

51

and high relapse rates (up to 90% of recovered patients) in AN and BN (Keel and Brown, 2010;

Mischoulon et al, 2011).

Assessment of Symptom Severity The DSM-5 assesses the degree of symptom severity with the aid of the BMI for AN, and

by considering the frequency of compensatory/purging episodes for BN (American Psychiatric

Association, 2013). BMI ranges are defined according to World Health Organization norms for

body mass by height; however, the level of AN severity on this metric can also reflect other

clinical symptoms or functional disabilities. Typically, mild, moderate, severe and extreme AN

are defined by at BMI ranges of >16.99, 16-16.99, 15-15.99, and <15, respectively. Mild,

moderate, severe and extremely severe BN are typically defined at ranges of 1-3, 4-7, 8-13, and

14 or more episodes of compensatory behaviour per week, respectively.

The Eating Disorder Examination (EDE) is a semi-structured clinician-assessed interview

widely considered to be the leading technique for assessing eating disorder symptom severity

(Fairburn and Cooper, 1993), as it is extensively validated in adults (Berg et al, 2012; Cooper et

al, 1989; Rizvi et al, 2000; Rosen et al, 1990). The Eating Disorders Examination –

Questionnaire (EDE-Q) is a supplementary self-administered version of the EDE that is also

well-validated and widely used (Berg et al, 2012; Fairburn and Cooper, 1993). Both the EDE and

EDE-Q assess the frequency and severity of eating disorders using the following subscales:

restraint over eating, eating concern, shape concern, and weight concern. Restraint over eating

refers to avoidance of food and eating, dietary rules and the desire to have an empty stomach.

Eating concern refers to a preoccupation with food, eating or calories, a fear of losing control

over eating, eating in secret, and guilt after eating. Shape concern refers to a preoccupation with

shape or weight, a fear of gaining weight, dissatisfaction and discomfort with shape or body,

avoidance of body exposure or seeing one’s body, and a feeling of fatness. Weight concern refers

to the importance, dissatisfaction or preoccupation of weight, reaction to a prescribed weighing,

and a desire to lose weight. Scoring on individual EDE items vary – questions assessing

symptom frequency are reported on a 6-point scale based on a 28-day month, whereas severity-

based questions score the item on a 7-point severity scale (0-6). To obtain an overall score for a

particular subscale, the scores from the relevant items are summed and divided by the total

52

number of items forming the subscale. To obtain an overall severity rating on the EDE or EDE-

Q, the four subscale scores are summed and divided by the number of subscales (four). In a

community-based sample of 243 young women, the mean global EDE score was 0.93±0.81, and

the mean global EDE-Q score was 1.40±1.13 (Fairburn and Beglin, 1994).

Response and remission are ill-defined endpoints for AN and BN. The most conservative

definition of BN treatment remission is an absence of binge-eating and compensatory purging

behaviours for a minimum of 8 weeks (Herzog et al, 1999; Keller et al, 1992; Mitchell et al,

1985), or, less stringently, for a minimum of 4 weeks (Agras et al, 2000; Fairburn et al, 1993;

Field et al, 1997; Halmi et al, 2002). Another definition of BN remission is 1 weekly symptom

episode or less over a 4-week period (Maddocks et al, 1992; Olmsted et al, 1994) or two week

period (Pyle et al, 1990). Generally, BN treatment response is characterized by a >50% reduction

in weekly binge-eating and compensatory purging behaviours (Agüera et al, 2013).

Definitions of AN treatment response and remission take into account the following four

items: first, the patient’s weight should be restored to a normal BMI given their age and sex;

second, restriction, binge-eating or purging should significantly improve (by >50% for response

or to relapse levels defined above); third, cognitive distortions such as body/shape dysmorphia

should meaningfully improve; and finally, medical stabilization of cardiac or other medical

problems should normalize due to weight restoration (Pike, 1998).

Structural and Functional Abnormalities in AN and BN

The neuroimaging literature on structural and functional abnormalities in AN and BN is

substantially less extensive than that for MDD, bipolar disorder, or other common Axis I

disorders. In addition, standard neuroimaging measures in AN and BN can be affected by

confounding factors, such as global volumetric changes in brain size associated with starvation

or caloric insufficiency. Notwithstanding these issues, an emerging neuroimaging literature does

exist regarding the substrates of illness for both AN and BN, and its key findings are summarized

here.

53

Structurally, both AN and BN display abnormally small gray matter volumes in brain

regions affiliated with the SN and VMN. Generally, volumetric reductions are reported in the

striatum, ACC and insula (Coutinho et al, 2015; Frank et al, 2013; Friederich et al, 2012;

Schäfer et al, 2010; Titova et al, 2013). AN patients also display altered thalamic connectivity

and white matter integrity to the DLPFC and mPFC (Biezonski et al, 2016; Frieling et al, 2012;

Hayes et al, 2015). Similarly, SN structural connectivity alterations between the caudate and the

insula have also been reported in this population (Frank et al, 2016; Shott et al, 2016).

Relative to controls, AN patients display increased activation in SN and CEN nodes for

disease-relevant stimuli and aversive (punishment) stimuli, perhaps reflecting heightened

salience of these stimuli in AN patients. Studies also report hyperactivity in the AI and DLPFC

during pain anticipation, and hyperactivity to monetary losses and pain in the DLPFC and ACC

(Bailer et al, 2017; Bär et al, 2015; Bischoff-Grethe et al, 2013; Strigo et al, 2013). AN patients

also display exaggerated responses to food stimuli and aversive tastes in the insula, striatum and

ACC (Cowdrey et al, 2011). AN patients exhibit hyperactivity for food stimuli in limbic regions

that activate for fear-responses, such the amygdala, instead of VMN reward regions such as the

OFC and ventral striatum (Vocks et al, 2010). Similarly, symptom provocation in AN elicits

hyperactivity in limbic regions like the amygdala and hippocampus, and in SN nodes like the

dACC, insula and medial PFC (Ellison et al, 1998; Frank et al, 2002, 2012b; Friederich et al,

2010; Seeger et al, 2002; Uher et al, 2004; Vocks et al, 2010). One study by Friederich and

colleagues in AN reported insular hyperactivation and dACC hypoactivation while viewing

idealized female bodies, perhaps associated with altered goal-directed cognition or interoceptive

awareness related to abnormal SN function (Friederich et al, 2010). Finally, increased harm

avoidance in AN is associated with altered striatal dopamine and cingulate serotonergic binding

(Bailer et al, 2004, 2007; Frank et al, 2005), suggesting that there may be important

neurochemical alterations underlying the observed abnormalities in functional activation in this

population.

Relative to healthy controls, BN patients show a dual pattern of increased VMN activity

related to increased reward valuation and sensitivity, alongside altered SN activity related to

deficits in cognitive control and impulsivity. BN patients will work harder for food rewards

(Schebendach et al, 2013), and display higher rates of overall impulsivity on a number of clinical

scales and behavioural paradigms (Chan et al, 2014; Manwaring et al, 2011; Mole et al, 2015).

54

On fMRI, BN patients have exaggerated VMN and SN activity to rewards in the insula, medial

OFC and NAc (Frank et al, 2011, 2012a; Oberndorfer et al, 2013b; Radeloff et al, 2014;

Schienle et al, 2009; Weygandt et al, 2012). SN hypoactivity is also associated with food

impulsivity; for example, lower dACC activity in response to food reward anticipation predicts

how much a BN patient will overeat (Bohon and Stice, 2011; Frank et al, 2006). Also, SN

hypoactivity during a cognitive control (Stroop) task is associated with poor dietary restraint in

binge-eating patients (Balodis et al, 2013). Neurochemically, serotonergic and dopaminergic

binding is altered in SN and VMN nodes like the ventral striatum, PFC and insula relative to

healthy controls (Broft et al, 2012; Galusca et al, 2014).

In contrast to BN patients, AN patients display SN and CEN alterations suggestive of

enhanced cognitive control relative to healthy controls. For example, DLPFC hyperactivity to the

anticipation of rewards has been reported in restrictive AN, suggesting enhanced cognitive

control over food reward stimuli (Ehrlich et al, 2015; Sanders et al, 2015). AN-R patients also

exhibit a broader pattern of enhanced temporal delay discounting relative to healthy controls,

suggesting that the enhanced cognitive control reported for AN patients may represent a broader

trait that applies beyond food stimuli alone (Steinglass et al, 2012). Some studies, however, have

reported cognitive control deficits in AN for specific disease-relevant stimuli: for example, to

distorted images of the patient’s body (Lee et al, 2014), food stimuli (Oberndorfer et al, 2013a;

Sanders et al, 2015), social stimuli (Tapajóz P de Sampaio et al, 2015), and during cognitive

flexibility paradigms (Zastrow et al, 2009).

There is also mounting evidence that frontostriatal circuitry is abnormal in both AN and

BN, and these abnormalities contribute to deficits in cognitive control, the self-regulation of

emotional and motor responses, and goal-directed behaviour (for a review see (Berner and

Marsh, 2014)). BN-associated deficits in impulse control are related to abnormal functioning in

SN and VMN corticostriatal circuitry. For example, BN patients show hypoactivity in

frontostriatal circuitry during cognitive control tasks like the Simon Spatial Incompatibility task

(Celone et al, 2011; Marsh et al, 2009b, 2011) and the go/no-go task (Skunde et al, 2016).

Behavioural deficits and frontostriatal hypoconnectivity are also reported in AN patients during

cognitive control tasks, including during response inhibition (Oberndorfer et al, 2011; Wierenga

et al, 2014), the Wisconsin Card Sorting Task (Lao-Kaim et al, 2015), and delay discounting

(Decker et al, 2015; Wierenga et al, 2014).

55

Some research has also reported altered incentive salience for specific stimuli. For

example, AN-R patients report physical exercise as significantly more pleasant than primary

rewards, including food (Giel et al, 2013). While AN patients typically exhibit deficits in

cognitive control and frontostriatal hypoconnectivity to primary rewards (food and money), this

effect is reversed for physical exercise. For example, one study demonstrated that AN patients

exhibited frontostriatal hyperconnectivity and enhanced cognitive control during a response

inhibition task when exercise-related stimuli were used as cues. This result signifies the

importance of reward valuation and incentive salience in cognitive control processes (Kullmann

et al, 2014).

Finally, AN patients also exhibit abnormal SN and DMN connectivity in the absence of

task-evoked activations (at rest). For example, resting-state SN connectivity of the dACC and

insula is altered in BN relative to AN (Amianto et al, 2013). Cowdrey and colleagues recently

reported that AN patients have abnormal hyperconnectivity within the DMN and between the

DMN and DLPFC (Cowdrey et al, 2014). Insular resting-state connectivity is also abnormal in

AN; one study reported abnormal low functional connectivity between the insula and thalamus

(Geisler et al, 2016). In summary, AN and BN patients’ distinctive profiles of abnormality

within the RDoC domains of cognitive control and positive/negative valience systems appear to

be reflected in distinctive patterns of abnormality in the nodes of the SN, VMN, and their

respective CSTC loop circuits (Dunlop et al, 2016b).

Transcranial Magnetic Stimulation as a Probe of IBNs

What is Transcranial Magnetic Stimulation?

1.7.1.1 Transcranial Magnetic Stimulation Apparatus Overview

A transcranial magnetic stimulation (TMS) device has several key components. At

minimum, the device requires an electromagnetic inductor consisting of a wire coil inside a

protective housing (Figure 1-9A), placed on the participant’s head during stimulation. This

inductor is fed by a power system used to generate electrical pulses, which are converted into

brief (<1 ms) but powerful (1-2 Tesla) focal magnetic pulses by the inductor itself (Figure 1-9B).

Some hardware configurations include an air or liquid coolant (Figure 1-9C) intended to

56

maintain coil temperatures over long stimulation periods. Some systems also incorporate

neuronavigation via a frameless stereotaxic system. MRI-guided neuronavigation systems use

high resolution magnetic resonance images to create a 3-dimensional representation of the

participant’s head and brain in order to target the desired stimulation site with greater precision.

TMS coils can vary in shape (for a review, see Hallett, 2007). The shape of TMS coils

impacts how the magnetic field emitted from the coil interacts with the target brain region to

elicit neural activity. For example, round coils produce a fairly consistent ring-shaped field

around the coil windings, while figure-of-eight coils (Figure 1-10) produce a more focal field

relative to round coils; the field is strongest and deepest at the intersection of the two round

windings of the figure-of-eight coil (Cohen et al, 1990; Hallett, 2007). Figure-of-eight coils can

also be made where the two round components intersect at an angle more oblique than 180º (e.g.,

120º, 135º 150º) – this modification increases the depth and focality of the field and thereby

increases the depth at which a current can be induced in the cortex. Another common design is

the helmet-shaped H-coil, also referred to as deep TMS (dTMS), where large coil windings are

arranged in a complex array over the scalp in order to stimulate deeper brain targets relative to

figure-of-eight coils, albeit in a less focal manner (Hallett, 2007).

Figure 1-9: Example of a TMS system. A. Magnetic inductor coil. B. Power and control

system to generate electrical pulse sequences. C. Coil-cooling device.

57

Figure 1-10: Interior construction of an angled figure-of-eight TMS head coil. This

computed tomography image of a standard figure-8 TMS coil in clinical use (Cool B70,

MagVenture, Farum, Denmark) illustrates the arrangement of the windings of the two round

components of the figure-of-eight, at whose intersection is produced the strongest and most focal

region of the magnetic field, yielding the highest induced current in the target brain region.

1.7.1.2 Biophysical Mechanisms of TMS

During a single pulse of TMS, a brief (<1 ms), high-current electrical pulse is produced

and directed through the inductor coil placed over target brain region. Consequently, a short-

lived but very powerful magnetic field is produced orthogonally to the plane of the coil, and it

passes largely undistorted through the participant’s skin and skull. This magnetic field in turn

induces a current within the underlying neural tissue that is parallel to the coil. Above a certain

threshold, this current is strong enough to induce action potentials in local populations of

neurons. Repeated trains of stimulation, over time, can induce durable changes in brain activity

via the classical mechanisms of synaptic long-term potentiation and depression (Hallett, 2000,

2007). The exact biophysical mechanism by which TMS pulses is not yet well understood, but it

is believed that the rapidly-changing TMS pulse is sufficiently strong to depolarize the neuronal

58

membrane, thereby generating an action potential in neurons that are near the TMS coil and

oriented optimally with respect to the field (as reviewed by (Cirillo et al, 2017).

1.7.1.3 Neurophysiological Effects of TMS and rTMS

Because a TMS pulse induces a brief disruption or excitation of population neural

activity in the target cortical region, it can be used to spatially or temporally localize brain

function. A common target of TMS is the primary motor cortex (M1) as stimulating M1 using

single pulses of TMS induces observable motor evoked potentials (MEPs) in the corresponding

muscles. For example, TMS-MEP studies have been used to generate complex maps of

overlapping muscle innervations (such as leg and arm muscles) in M1 (Wassermann et al, 1992).

Additionally, TMS studies targeting sensory and cognitive brain areas have been used localize

other functions. For example, TMS over the occipital cortex can induce temporary partial

functional scotomas. One early study of visual cortex-TMS found that TMS delivered at a delay

interval of 80-100ms from a visual stimulus was sufficient to interfere with perception of the

stimulus (Amassian et al, 1989).

The effects of repeated pulses of TMS at a specified frequency or pattern (repetitive

TMS; rTMS) can also be used as a tool to explore fundamental mechanisms of brain function.

For example, rTMS can be used to temporarily disrupt normal brain activity to localize the

cognitive or behavioural function of a target cortical region; this approach is referred to as a

“virtual lesion” technique. One such study investigated the role of the RTPJ on moral decision-

making and found that 10 Hz rTMS over the RTPJ during moral decision-making altered the

ratings participants made during certain moral scenarios (Young et al, 2010). rTMS can also be

used to study how repetitive high or low frequency stimulation produces long lasting

neuroplastic changes on cognition or behaviour. Classically, high frequency (>5 Hz) rTMS is

considered to induce long-term plasticity (LTP), or increased regional excitability and MEP

amplitude, while low frequency (1 Hz) rTMS induces long-term depression (LTD), or decreased

regional excitability and MEP amplitude (Chen et al, 1997; Pascual-Leone et al, 1994).

As with single TMS pulses, the short- and long-term neuronal effects of rTMS trains in

humans are similarly not yet fully characterized. In vitro studies of the effects of repetitive

59

magnetic stimulation on neuronal plasticity (LTP and LTD) suggest that rTMS affects the

balance of cortical excitation and inhibition (as reviewed by (Lenz and Vlachos, 2016)) and the

modulation of both excitatory (glutamatergic) and inhibitory (γ-aminobutyric acid [GABA];

GABAergic) neurotransmission via calcium-dependent signaling pathways (Huang et al, 2007;

Labedi et al, 2014; Lenz et al, 2015, 2016; Vlachos et al, 2012). Many of these studies support

the idea that rTMS increases or decreases the activity of glutamatergic cortical neurons that

possess long-range axons (principal neurons) (Gersner et al, 2011; Ghiglieri et al, 2012; Lenz et

al, 2015; Levkovitz et al, 1999; Ma et al, 2013; Sykes et al, 2013; Tang et al, 2017; Tokay et al,

2009; Vlachos et al, 2012; Volz et al, 2013). Repetitive magnetic stimulation also appears to

alter dendritic growth and sprouting related to LTP and LTD in in vitro hippocampal cell

cultures (Ma et al, 2013; Vlachos et al, 2012). Human in vivo studies suggest a neuroplastic

mechanism inovlving brain-derived neurotrophic factor (BDNF), a growth factor involved in

LTP; however, some studies report increases or decreases in serum BDNF levels with rTMS,

while other report no significant change (Gedge et al, 2012; Müller, 2000; Wang et al, 2011;

Yukimasa et al, 2006; Zanardini et al, 2006).

More recent theories on the mechanisms of rTMS posit that stimulation also modifies the

activity of GABAergic cortical interneurons that act to inhibit principal neurons. Briefly,

excitatory rTMS appears to reduce the inhibitory effect of GABAergic interneurons on principal

cells (called rTMS-induced disinhibition) (Lenz and Vlachos, 2016). Excitatory and inhibitory

rTMS protocols differentially alter populations of interneurons and consequently alter different

aspects of network activity or function (Funke and Benali, 2011; Mix et al, 2010; Volz et al,

2013). Excitatory magnetic stimulation protocols have been recently shown to include LTD-like

effects on inhibitory interneurons for somatic inhibition, while inhibitory protocols alter the

activity of interneurons for dendritic inhibition (Benali et al, 2011; Funke and Benali, 2011;

Labedi et al, 2014; Mix et al, 2015; Trippe et al, 2009; Volz et al, 2013).

Another proposal draws upon studies of rTMS in animal models, and posits that rTMS

has neuroprotective effects, enhancing neurogenesis, and inhibiting neuronal cell death

(apoptosis). For example, two studies reported that high frequency magnetic stimulation

increased neurogenesis in in vitro mouse hippocampal slices (May, 2011; May et al, 2007).

Other studies studying the effects of rTMS in ischemic stroke mouse models have also

demonstrated that high frequency stimulation aids in the recovery of limb function post-stroke

60

(Feng et al, 2005), and inhibits apoptosis in the infarct zone (Fujiki et al, 2003; Gao et al, 2010;

Ogiue-Ikeda et al, 2005; Yoon et al, 2011).

The neuroplastic effects following nominally ‘excitatory’ or ‘inhibitory’ rTMS pulse

sequences can vary considerably in magnitude or in the direction of effect (increased or

decreased excitability) across individuals. One study reported substantial inter-individual

variability of rTMS on motor evoked potentials at 1, 10, 15, and 20 Hz, with the greatest

variability in the direction of effect at 1 Hz and 10 Hz, and the most consistent excitatory effects

at 20 Hz (Maeda et al, 2000). Similar findings have emerged with functional MRI, where both

‘excitatory’ 20 Hz and ‘inhibitory’ 1 Hz rTMS induced changes in connectivity to the target site,

but the direction of effect widely varied across subjects, particularly for 1 Hz rTMS (Eldaief et

al, 2011). Furthermore, the two rTMS protocols modified different kinds of connections to the

target site: 20 Hz rTMS engaged corticocortical circuits, while 1 Hz rTMS altered corticolimbic

circuits (Eldaief et al, 2011). This difference in network effects of inhibitory and excitatory

rTMS could potentially be related to the different populations of inhibitory interneurons through

which rTMS exerts its effects, thereby differentially altering the excitatory/inhibitory balance of

principal cells.

Applications of rTMS as a Network-Probe

One of the advantages of TMS is that it can focally alter the activity of cortical brain

regions to causally test brain structure-function relationships. For example, publications using

single-pulse TMS have elucidated the specific role of motor and premotor structures in

behaviour (for a review, see (Chouinard and Paus, 2010)). TMS and rTMS have also been shown

to modulate monosynaptic downstream targets, allowing for the study of not only the stimulated

brain region, but also the network that region interconnects with (for a review, see (Ruff et al,

2009)). For example, different rTMS protocols have been reported to modulate different kinds of

functional connectivity. Eldaief et al. demonstrated that excitatory and inhibitory rTMS over the

left IPL elicited different patterns of inter-regional change in network connectivity, and not

simply changes in the strength of rsFC: excitatory rTMS decreased corticocortical IPL rsFC

between nodes of the DMN, while inhibitory rTMS increased corticolimbic IPL rsFC with the

hippocampal formation (Eldaief et al, 2011).

61

Effects on CTSC circuits are also a recurring theme in the available TMS-fMRI literature

to date. For example, using interleaved TMS-fMRI, both DLPFC- and OFC-rTMS has been

demonstrated to activate CSTC circuitry, including the ventral/dorsal striatum (Dowdle et al,

2018; Hanlon et al, 2013). These effects on CSTC circuits may be of direct relevance to the

behavioural effects of rTMS. For example, rTMS has been shown to modulate cognitive control,

frontostriatal function and dACC activity when targeted at the dorsomedial prefrontal cortex

(dmPFC), a key region of the SN that is implicated in many psychiatric disorders (Section 1.3.2).

Cho et al. demonstrated that, relative to sham, active medial PFC excitatory rTMS released

striatal dopamine and improved cognitive control on delay discounting, with the behavioral

effects showing a relationship to the amount of dopamine release (Cho et al, 2015) (Figure 1-11).

Similarly, disrupting dACC function using single-pulse TMS or inhibitory rTMS impairs conflict

monitoring and increases errors during a cognitive control task (Duque et al, 2012; Taylor et al,

2007). Hayward and colleagues reported that excitatory mPFC-rTMS reduced dACC cerebral

blood flow (Hayward et al, 2007). Stimulating the left DLPFC, a region of the CEN, has also

been shown to increase glutamate in the contralateral DLPFC and dACC (Michael et al, 2003).

62

Figure 1-11: (A) Areas of striatal dopamine release, as measured by positron emission

tomography with a [11C]-(+)-PHNO dopamine receptor 2 and 3 radioligand, following

active dmPFC-rTMS. (B) Bars representing left and right striatal dopamine binding

potential following control and dmPFC rTMS. Reproduced with permission from (Cho et al,

2015).

Figure 1-12: Decreases in cerebral blood flow following active dmPFC-rTMS. (A)

Stimulation sites for control stimulation (a) and active dmPFC-rTMS (b). (B) Reduced cerebral

blood flow in the dmPFC and dACC (indicated by the white arrows). Reproduced with

permission from (Hayward et al, 2007).

63

Therapeutic Applications of rTMS

Psychiatric disorders are often associated with IBN dysfunction, and so it is reasonable to

posit that rTMS might induce its therapeutic effects via focal synaptic plasticity and neuronal

excitability at the cellular scale, and via modifications to downstream cortico-cortical, cortico-

limbic or cortico-striatal connections at the macroscopic, whole-brain scale of IBNs. In support

of this perspective, recent studies reporting on resting-state IBNs suggest that psychiatric and

neurological disorders can be characterized as ‘network disorders’ (as reviewed in Sections 1.4,

1.5, and 1.6). If so, this implies that by perturbing the function of one node of an IBN, rTMS

may be capable of achieving therapeutic effects via anatomical and functional connections with

other nodes of the same IBN, or connections with related IBNs.

In a well-cited study supporting this proposal, Fox and colleagues recently demonstrated

that clinically effective targets for invasive brain stimulation (i.e., deep brain stimulation, DBS)

belong to the same IBN as the known effective targets for non-invasive brain stimulation (i.e.,

rTMS), across a range of 19 neurological and psychiatric disorders including MDD, OCD,

Tourette’s Disorder, Parkinson’s Disease, chronic pain, and Alzheimer’s Disease (Fox et al,

2014). Although beyond the scope of this thesis, rTMS has been been explored as a therapeutic

intervention for a variety of neurological conditions, including post-stroke aphasia (Naeser et al,

2010), tinnitus (Lehner et al, 2016), and Alzheimer’s Disease (Rutherford et al, 2015). However,

as of this writing, the majority of therapeutic rTMS sessions worldwide are still used for

psychiatric disorders: most commonly major depression, alongside less common and usually off-

label treatment of a limited number of other psychiatric illnesses.

In order to provide background for further discussion of potential IBN mechanisms for

therapeutic rTMS, the following three sections will first review current treatments, and novel

stimulatory treatments for three psychiatric disorders, MDD, OCD and AN/BN, as well as the

available literature from published clinical trials of invasive and non-invasive brain stimulation

for all three of these disorders.

64

Clinical Applications of rTMS

Treatment Resistant MDD

1.8.1.1 First Line Treatments for MDD Pharmacological and psychotherapeutic interventions have been the mainstays of

conventional treatment strategies for MDD over the last six decades. Medications for MDD that

have proven efficacy in placebo-controlled studies include selective serotonin reuptake inhibitors

(SSRI) (Hieronymus et al, 2016), serotonin-norepinephrine reuptake inhibitors (SNRI)

(Montgomery et al, 2015), tricyclic antidepressants (TCA), monoamine oxidase inhibitors

(MAO) and norepinephrine-dopamine reuptake inhibitors (NDRI) (Rosenblat et al, 2015).

Psychotherapies including cognitive behavioural therapy (CBT) (Parikh et al, 2016),

interpersonal therapy (IPT) (van Hees et al, 2013; Jakobsen et al, 2012) and mindfulness-based

cognitive therapy (MBCT) (Kuyken et al, 2015; van der Velden et al, 2015) have also shown

superior efficacy rates in acute MDD and relapse prevention. Recent studies have found that

conventional psychotherapies can be as effective as medication in appropriate populations of

MDD patients (Cuijpers, 2015), but a combination of both treatment modalities is superior to

either monotherapy (Cuijpers et al, 2013; de Maat et al, 2007; Wiles et al, 2016).

MDD patients often do not respond to the first treatment they receive. One landmark

study, the Sequenced Treatment Alternatives to Relieve Depression (STAR*D) trial

(ClinicalTrials.gov ID: NCT00021528), assessed the efficacy of a sequence of up to four

successive antidepressant trials in achieving remission, in a sample of 4041 adults with MDD

(Rush et al, 2006). The results of the trial showed that approximately one third of patients

achieved remission with their first trial of antidepressant medication, and that the probability of

remission declined markedly with successive medication trials (Rush et al, 2006). The STAR*D

remission rates, defined as a score of >6 on the QIDS, for four successive pharmacotherapy

attempts were 36.8%, 30.6%, 13.7% and 13%, and the overall remission rate was 67%.

Medication non-adherence is common in MDD, and may be an important factor

underlying treatment non-response. In one study, almost 20% of participants did not properly

adhere to their medication, as indicated by a blood sample (Roberson et al, 2016). Another study

showed that up to 20% of patients missed up to four consecutive days of treatment

65

(Demyttenaere et al, 2001). Medication adherence is strongly associated with better response

rates (Akerblad et al, 2006; Demyttenaere et al, 2001).

1.8.1.2 Treatment-Resistant Depression

Those who do not respond to two or more trials (of adequate dose and duration) of

conventional pharmaco- and psychotherapies are considered to have treatment-resistant

depression (TRD) (Berlim and Turecki, 2007; Nemeroff, 2007). However, the exact definition of

TRD is contentious, because the TRD population is heterogeneous; ‘pseudo’-TRD may arise

from misdiagnosis, treatment non-adherence, or inadequate dose and duration of treatment

attempts (Souery et al, 2006); other ostensibly treatment-resistant patients may have experienced

a partial response to a past treatment (Kennedy et al, 2016). These inconsistencies make

assessing the prevalence of TRD challenging; however it is estimated that TRD, as defined

above, has a prevalence of approximately 2% of the general population (Nemeroff, 2007) and

approximately 35% of MDD patients (Trivedi et al, 2006). TRD may be addressed using

combination/augmentation therapies (for a review, see (Papakostas and Ionescu, 2015)), despite

relatively poor outcomes. However, an emerging and increasingly popular alternative is

therapeutic brain stimulation.

1.8.1.3 Brain Stimulation Treatments for TRD

One of the most potent available interventions in TRD is electroconvulsive therapy

(ECT). ECT involves the induction of a therapeutic generalized seizure under general anesthesia,

by applying a pulsed, square-wave electrical current of ~ 800 mA via electrodes that are placed

on the scalp in either a bilateral temporal, bilateral frontal or right unilateral montage (Milev et

al, 2016). Patients usually undergo 2-3 treatments per week for a total of 6-15 sessions of ECT to

achieve response or remission (Charlson et al, 2012; Daly et al, 2001). ECT is one of the most

effective treatments for TRD; recent randomized controlled trials report remission rates from

65% to as high as 90% in specific subpopulations, such as those with psychotic depression or

catatonia (Kellner et al, 2010). In one meta-analysis, higher response rates are seen in individuals

who are older, and are psychotic (Haq et al, 2015). A number of side effects and barriers to

66

treatment exist for ECT, however, including cognitive impairment post-ECT (for a review, see

(Fraser et al, 2008)), the use of general anesthesia, high associated stigma and limited resources

(predominately a lack of anesthesiologists and space to perform the procedure) (Andrade and

Thyagarajan, 2007; Delva et al, 2011). As a result of this stigma and the invasiveness of the

procedure, most TRD patients do not consent to ECT; fewer than 1% of TRD patients undergo

treatment in any given year (Delva et al, 2011).

Given the limitations and issues related to ECT, alternative brain stimulation treatments

for TRD are currently under investigation. For example, magnetic seizure therapy (MST) is a

treatment that, like ECT, induces a generalized seizure, albeit via large rTMS-type magnetic

inductors rather than electrodes. MST may have comparable efficacy rates (69% response, 46%

remission) (Kayser et al, 2011, 2015) and minimal cognitive adverse effects relative to ECT

(McClintock et al, 2011); however, this treatment is still considered investigational, and very few

devices are currently available worldwide.

More invasive therapies requiring surgical intervention are also currently under

investigation for TRD: vagus nerve stimulation (VNS) and deep brain stimulation (DBS). VNS is

delivered by a subdermal pulse generator that controls an electrode implanted around the vagus

nerve in the neck region. Stimulation of the vagus nerve is thought to activate areas of the

brainstem and brain via the nucleus tractus solitarius, which modulates activity in a variety of

monoaminergic brainstem regions to achieve symptom improvement (Nemeroff et al, 2006).

VNS is well-tolerated (Aaronson et al, 2013), and is associated with superior efficacy rates

compared to sham stimulation (Rush et al, 2005), albeit with modest response rates, estimated at

~31.8% in a recent review (Martin and Martín-Sánchez, 2012). Due to the modest results to date,

and the need for neurosurgical expertise for the implantation, uptake of VNS is substantially

lower than even that of ECT (<1%) in the TRD population.

DBS is delivered from an electrode implanted via MRI and stereotactic guidance into the

target brain region; the technique is uniquely capable of stimulating deeper brain regions and

smaller (<1 cm) brain regions in a focal and ongoing manner. DBS is most commonly used for

movement disorders such as Parkinson’s Disease, but is drawing increasing interest as an

intervention in psychiatric disorders resistant to other kinds of treatment (Lozano and Lipsman,

2013). Among the best-known examples is DBS for TRD, where an early report of successful

67

treatment in some patients following DBS of the sgACC (Mayberg et al, 2005) led to widespread

interest in DBS for refractory MDD cases. Long-term treatment outcomes of DBS for TRD

targeting the sgACC have been encouraging in early studies, with reported remission rates of

approximately 40-50% at 2 years post-implantation (Holtzheimer et al, 2012; Kennedy et al,

2011). Although some individual patients appear to show robust and reproducible benefit, a

recent study found that sgACC-DBS effects were not superior to sham for MDD patients overall

(Holtzheimer et al, 2017). Although promising, both VNS and DBS are limited by the

requirement for invasive surgical implantation, and have significant geographical barriers to

treatment access (including geographical barriers and resource limitations).

One other emerging non-invasive alternative to VNS, ECT, and DBS is transcranial

direct current stimulation (tDCS). tDCS delivers a low-amplitude (2 mA) electrical current over

a particular scalp target to modulate neuronal excitability in the underlying neural tissue, and

repeated use is thought to induce neuronal plasticity (Stagg and Nitsche, 2011). tDCS is also

portable, safe, well-tolerated, relatively easy to use, inexpensive, and suitable for home use.

Randomized controlled trials and meta-analyses have supported the efficacy of tDCS in MDD

relative to placebo tDCS (Brunoni et al, 2013), but evidence to date suggests that tDCS may

have limited efficacy in TRD populations (Brunoni et al, 2016).

1.8.1.4 rTMS as a Treatment for TRD

Aside from ECT, the most widely-used therapeutic brain stimulation technique at present

is rTMS. rTMS uses powerful, focused magnetic field pulses, delivered by an electromagnetic

induction coil applied to the scalp, to manipulate the excitability of target brain regions and their

associated IBNs (For a review, see (Hallett, 2007)). To date, medication-resistant MDD is the

best-studied therapeutic indication for rTMS. rTMS for depression was initially proposed in the

early 1990s after it was observed that stimulation of the motor cortex in healthy controls induced

an elevated mood in some individuals. Preclinical studies reported that prefrontal rTMS could

induce self-reported negative or positive mood in healthy individuals, depending on the

stimulation target (George et al, 1996; Pascual-Leone et al, 1996a). By 1994, the left prefrontal

cortex had been proposed as a stimulatory site for therapeutic rTMS in MDD (George and

Wassermann, 1994), based on early neuroimaging studies highlighting the role of left

68

dorsolateral prefrontal cortex (DLPFC) dysfunction in MDD. For example, MDD patients

displayed reduced glucose metabolism in the left DLPFC relative to controls (Baxter et al, 1989),

and left prefrontal lesions were reported to be associated with increased MDD incidence

(Robinson et al, 1984). The first rTMS studies in MDD reported only modest improvements in

depression severity; this is perhaps because these studies had small sample sizes, as well as

suboptimal stimulation targets and very short treatment courses compared to modern practice

(Höflich et al, 1993; Kolbinger et al, 1995).

The first randomized controlled trials of DLPFC-rTMS for TRD were conducted in the

mid-1990s using left high frequency rTMS (George et al, 1995, 1997; Pascual-Leone et al,

1996b). Low-frequency rTMS applied over the right DLPFC was introduced in the late 1990s,

and was found to be an approach of similar treatment efficacy relative to left DLPFC rTMS

(Klein et al, 1999). Since then, dozens of randomized controlled trials have supported the

efficacy DLPFC-rTMS for TRD, as summarized by recent meta-analyses (Berlim et al, 2014;

Kedzior et al, 2014). In one meta-analysis assessing the efficacy of 54 sham-controlled trials of

DLPFC-rTMS for TRD, Kedzior and colleagues found significant reductions in TRD severity

after left-high frequency, right-low frequency or bilateral DLPFC-rTMS (Kedzior et al, 2014).

Furthermore, data from 1371 subjects with TRD receiving high-frequency rTMS showed that

29.3% and 18.6% of patients were classified as responders and remitters, respectively (Berlim et

al, 2014). However, this statistic likely underestimates response and remission rates of DLPFC-

rTMS for TRD, as this meta-analysis included early sub-optimal trials discussed above, which

employed an average of only 10-15 sessions, versus the 20-36 sessions used in modern practice.

In the most recent generation of large trials, response and remission rates for TRD are between

50-55% and 30-35%, respectively (Brunelin et al, 2014; Fitzgerald et al, 2011).

Real-world efficacy rates for rTMS in naturalistic community practice are comparable to

add-on pharmacotherapies or a change in monotherapy (Berlim et al, 2014; Kedzior et al, 2014).

Response rates to rTMS may be further improved by adequate course length; at least 26-28

sessions of rTMS are required for maximum benefit (Carpenter et al, 2012). The most widely

used rTMS protocols involve 10 Hz left DLPFC-rTMS and/or 1 Hz right DLPFC-rTMS.

Unfortunately, these conventional protocols limit treatment capacities due to the long duration of

the treatment sessions; for example, a single session of left DLPFC-rTMS using the standard

69

FDA-approved 10 Hz rTMS protocol requires approximately 37.5 minutes (O’Reardon et al,

2007), rendering conventional rTMS treatment costly and low-volume.

One promising advance is the development of shorter theta-burst stimulation (TBS)

protocols to treat TRD (Huang et al, 2005). Intermittent theta burst stimulation (iTBS) delivers

600 pulses in 50 Hz triplet bursts, 5 bursts per second, 2 seconds on and 8 seconds off, for 20

trains in 3 minutes; this protocol has excitatory effects, as evidenced by increases in MEP

amplitude, matching or exceeding much longer conventional high-frequency protocols (Di

Lazzaro et al, 2011). Continuous theta burst (cTBS) uses the same theta burst pattern delivered

continuously over 600 pulses in 40 seconds; this protocol has inhibitory effects, as evidenced by

decreases in MEP amplitude, that may surpass those of longer conventional low-frequency

stimulation protocols (Di Lazzaro et al, 2011). In pilot studies in TRD, iTBS has been used for

left DLPFC rTMS, while cTBS has been employed for right DLPFC stimulation. Relative to

conventional rTMS protocols, both iTBS over the left DLPFC and cTBS over the right DLPFC

have shown comparable clinical efficacy in TRD (Blumberger et al, 2018; Chistyakov et al,

2010; Plewnia et al, 2014).

Limited efficacy rates and high relapse rates are among the limitations of rTMS for TRD.

For comparison’s sake, current rTMS protocols are not superior to ECT for response or

remission. A number of meta-analyses have found that the clinical efficacy of rTMS for TRD is

inferior to ECT (Micallef-Trigona, 2014; Xie et al, 2013). rTMS response is also poor in those

who have previously failed to respond to ECT (Downar et al, 2014), but it is not yet known

whether this relationship suggests that ECT and rTMS alter brain activity in similar ways.

Conversely, it is not known whether ECT response is poor in patients who have previously failed

rTMS. The median relapse time following rTMS in one naturalistic study was 120 days, with

relapse rates of 25%, 40%, 57%, and 77% at 2, 3, 4, and 6 months, respectively (Cohen et al,

2009). Notwithstanding these limitations, rTMS is attractive to patients and providers for the

absence of cognitive adverse effects, the rarity of serious side effects, and the lack of a need for

anesthesia during treatment. In current clinical practice, rTMS is beginning to occupy a

therapeutic niche in TRD treatment strategies, after >1-2 failed pharmacotherapy trials, but

before attempting ECT.

70

More recently, attention has turned to the potential of other target regions beyond the

conventional right or left DLPFC to improve response and remission rates in TRD (Downar and

Daskalakis, 2013). In particular, the dmPFC and dACC have been proposed as novel therapeutic

targets for rTMS in TRD, based on convergent evidence from neuroimaging, lesion, and

connectivity studies (Downar and Daskalakis, 2013).

Studies of the efficacy of dmPFC-rTMS are relatively few, however. One recent sham-

controlled trial of rTMS found a significant group-by-time interaction on depression severity in a

modest sample of TRD patients between active left DLPFC 10 Hz, active dmPFC 10 Hz, and

sham stimulation. Neither active rTMS treatment arms were superior to placebo, but there was a

significant clinical superiority of dmPFC-rTMS relative to left DLPFC-rTMS (Kreuzer et al,

2015). Furthermore, our group recently published the results of two studies involving 20 daily

sessions of open-label 10 Hz dmPFC-rTMS in treatment-refractory depression (Figure 1-11). In

one, HAMD scores decreased by 45%(± 31%) over the course of treatment (Salomons et al,

2014). Our group also recently published a chart review of 185 patients who received excitatory

(10 Hz or iTBS) dmPFC-rTMS in TRD, finding bimodal or trimodal patterns of response.

Response rates, as measured by the HAMD, for 10 Hz and iTBS over the dmPFC were 50.6%

and 48.5%, respectively, and remission rates were 38.5% and 27.9%, respectively, for 10 Hz and

iTBS protocols (Bakker et al, 2015).

These reported outcomes for dmPFC-rTMS overall did not appear to be markedly

superior to those previously reported for DLPFC-rTMS. However, a notable finding in another

second series of 47 subjects, was that the response to dmPFC-rTMS followed a bimodal

distribution, with individual patients showing either minimal or marked improvement over

treatment (Downar et al, 2014). The non-unimodal distribution suggests a heterogeneity among

the MDD population undergoing treatment, and leaves open the possibility that remission rates

could be improved by identifying and targeting a specific subpopulation of MDD patients. The

exploration of this possibility is central to the aims of the thesis, as outlined in detail in Section 2.

71

Figure 1-13: Targeting the dmPFC and dACC with rTMS. A. An angled figure-of-eight

rTMS head coil placed over the scalp, targeting the dmPFC. B. Structural MRI visualizing the

Broadmann areas targeted by dmPFC-rTMS. The white dot on the scalp indicates the placement

of the vertex of the coil during stimulation.

Obsessive-Compulsive Disorder

1.8.2.1 First-Line Treatments for OCD

Pharmacological and behavioural interventions are the first-line treatments for OCD.

Numerous randomized controlled trials have shown that a variety of SSRIs are superior to

placebo for OCD, including escitalopram (Stein et al, 2007), fluvoxamine (Nakatani et al, 2005),

fluoxetine (Tollefson et al, 1994), paroxetine (Zohar and Judge, 1996), and sertraline (Greist et

al, 1995). Additionally, studies report that tricyclic antidepressants including clomipramine have

superior efficacy relative to placebo, with Y-BOCS percent improvement ranging between 38-

44% for active groups, and 3-5% improvement for placebo (“Clomipramine in the treatment of

patients with obsessive-compulsive disorder. The Clomipramine Collaborative Study,” 1991).

Some studies have reported that pharmacotherapies for OCD require longer-term treatment

72

relative to other psychiatric disorders; OCD response rates may increase up to 70% after 24

weeks of medication treatment (Stein et al, 2007).

CBT and Exposure and Response Prevention (ERP) therapy are commonly-prescribed

psychotherapies with efficacy in OCD. For all anxiety disorders (among which OCD was

formerly classified), meta-analyses of randomized-controlled trials for CBT show that the pooled

odds ratio for response rates relative to placebo is 4.06, and that the strongest effect size was

seen in OCD. This suggests that across multiple studies CBT shows superior efficacy in treating

OCD relative to placebo (Hofmann and Smits, 2008). Studies of ERP therapy have also reported

superior efficacy relative to placebo, with response rates between 62-86% and 8-10% for ERP

therapy and placebo, respectively (Foa et al, 2005). The general consensus is that ERP therapy

and CBT are equally effective in terms of their efficacy in OCD (McLean et al, 2001; Whittal et

al, 2005), although one study found that CBT was superior to ERP therapy (van Oppen et al,

1995).

1.8.2.2 Treatment-Resistance in OCD Unfortunately, a substantial percentage of OCD sufferers receive no benefit from

conventional treatments. It is estimated that between 30-60% of OCD patients do not respond to

these therapies (Pallanti et al, 2002; Simpson et al, 2006). Alternative strategies like secondary

treatments or augmenting pharmacotherapies also do not usually achieve long-term improvement

(Matsunaga et al, 2009), and carry significant side effects. Even those whose symptoms initially

respond to conventional pharmaco- and behavioural therapies often show persistent functional

impairment (Steketee, 1997), and substantial future relapse rates (Simpson et al, 2005). This

problem is exacerbated by the fact that very severe and chronic OCD cases are the most

functionally disabling, and also the most difficult to treat (Bloch et al, 2010).

1.8.2.3 Brain Stimulation Treatments for OCD DBS targeting a number of different sites has been proposed for treatment-refractory

OCD, and many of these target regions were identified because of the clinical efficacy of

surgical ablations such as bilateral capsulotomy (Lippitz et al, 1999). The first DBS study for

73

treatment-refractory OCD targeted the same site used for ablative bilateral capsulotomy and

reported that 3 of 4 patients experienced significant improvements in their OCD severity and

quality of life (Nuttin et al, 1999). Subsequent randomized studies of DBS targeting the anterior

limb of the internal capsule reported that Y-BOCS severity significantly improved during the

postoperative stimulation-on condition and at 21-months relative to the stimulation-off condition

(Nuttin et al, 2003). Symptom improvements due to DBS for OCD are accompanied by long-

lasting, broad-spectrum improvements in quality of life (Ooms et al, 2014).

Other DBS targets have shown efficacy for treatment-resistant OCD, including the

ventral striatum (VS) and the inferior thalamic peduncle. One staggered-onset study using DBS

of the internal capsule and adjacent ventral striatum reported that two-thirds of patients achieved

response after 12-months of real stimulation compared to sham (Goodman et al, 2010).

Treatment response appears to persist years after electrode implantation. One study showed that

50% of participants maintained improvements over 36 months (Greenberg et al, 2006). Another

follow-up study of VS-DBS reported that the 67% of patients who achieved treatment response

over the first 12-months exhibited the same treatment response over 6-9 years of follow-up

(Fayad et al, 2016). One open-label study reported that DBS to the inferior thalamic peduncle

decreased Y-BOCS scores by 51% (Jiménez et al, 2013). More recent targets for OCD have been

proposed, notably the STN (for a review, see (Mulders et al, 2016)), and the medial

dorsal/ventral anterior thalamus, although the latter region was found ineffective for OCD

(Maarouf et al, 2016).

A number of critical limitations exist for DBS as a treatment for OCD. Most obviously,

DBS is an invasive procedure. One recent survey of adults with self-reported OCD showed that

DBS is the least accepted and least preferred novel treatment compared to other novel

behavioural therapies, possibly reflecting a lack of public awareness and evidence for this

intervention and indication (Patel et al, 2017). Only 2 of the aforementioned DBS studies were

blinded (crossover or staggered onset design) (Goodman et al, 2010; Nuttin et al, 2003), and all

of the aforementioned studies had small sample sizes (N ≤ 10), limiting the generalizability of

clinical results. Even assuming greater patient acceptance, however, the number of OCD patients

who might benefit from treatment likely exceeds by orders of magnitude the treatment capacity

of the specialized neurosurgical centres capable of performing DBS implantations.

74

To a lesser extent, tDCS and VNS have been successfully used to treat OCD, although

there are to date no randomized sham-controlled trials. The first case of tDCS for treatment-

resistant OCD is a case study in published in 2015 – the patient experienced a 40% reduction on

the Y-BOCS following 20 sessions of anodal supplemental motor area (SMA), cathodal supra-

orbital (brow bone) tDCS (Narayanaswamy et al, 2015). In the same year, another case report

showed that cathodal left OFC stimulation also induced 26% symptom severity reduction in

treatment-resistant OCD (Mondino et al, 2015). VNS has also shown some efficacy as a

treatment for OCD in one open-label study (George et al, 2008).

1.8.2.4 rTMS as a Treatment for OCD rTMS offers a non-invasive neurostimulatory alternative that could be offered to OCD

sufferers where conventional medication and pharmacotherapy fail. The first rTMS studies for

OCD targeted the DLPFC with limited success (Sachdev et al, 2001); earlier studies showed

some benefit to 20 Hz left DLPFC-rTMS in an open-label setting (Greenberg et al, 1997), while

other studies using 1 Hz right DLPFC-rTMS showed no differences relative to sham (Alonso et

al, 2001). High frequency right DLPFC-rTMS also does not appear to be effective for treatment-

resistant OCD (Mansur et al, 2011). Subsequent randomized sham-controlled trials of 1 Hz right

DLPFC-rTMS (Prasko et al, 2006) and 20 Hz left DLPFC-rTMS (Sachdev et al, 2007) have

confirmed the limited clinical efficacy of DLPFC-rTMS for OCD. However, two recent

randomized controlled trials have reported superior response rates to 1 Hz right DLPFC rTMS

over sham (Elbeh et al, 2016; Seo et al, 2016b).

Inhibitory bilateral SMA stimulation currently appears to be one of the few effective

rTMS targets available for treatment-resistant OCD (Berlim et al, 2013; Ruffini et al, 2009). In

2006, Mantovani and colleagues were the first to target the SMA with 1 Hz rTMS in an open-

label setting. In 4 OCD subjects, they found a >40% reduction in symptom severity, and this

improvement persisted at 3-months follow-up (Mantovani et al, 2006). A follow-up randomized

sham-controlled trial from the same group replicated these findings – patients in the active arm

experienced a nearly 50% reduction of symptom severity after 8 weeks of active rTMS

(Mantovani et al, 2010). Another randomized sham-controlled trial from a separate group saw

similar efficacy rates that persisted at 3-months follow-up (Gomes et al, 2012). Furthermore,

75

recent studies of low-frequency stimulation of the medial prefrontal cortex for OCD using deep

TMS acheived a mean reduction of 39% on the Y-BOCS in an open-label setting (Modirrousta et

al, 2015). One recent study saw limited efficacy for inhibitory 1 Hz rTMS over the SMA (Hegde

et al, 2016), although several others have replicated the long-term efficacy of this treatment

protocol (Hawken et al, 2016). The therapeutic mechanisms of rTMS in OCD remain poorly

characterized in the literature to date, however.

Anorexia and Bulimia Nervosa

1.8.3.1 First-Line Treatments for AN and BN To date, there is no single well-established treatment for eating disorders in general.

Treatments, or treatment-combinations, vary according to eating disorder type, severity, and

clinician experience (for a review, see (Halmi, 2005)). Treatments also take place in various

settings, including community programs, outpatient day hospital programs, hospitalization, and

group-based programs. Of note, much of the research on the efficacy of ED treatment strategies

is hampered by a lack of evidence from randomized-controlled trials.

For AN, a typical treatment course could incorporate medical management, behavioural

therapy, and cognitive therapy. Medical management refers to nutritional strategies to normalize

weight, either orally or via nasal gastric feeding tube. Recent research supports the relationship

between response and relapse rates to a patient’s BMI at discharge, indicating that the

achievement of a normal BMI throughout treatment may be a necessary component for long term

success (Kaplan et al, 2009; Rigaud et al, 2011). Cognitive and behavioural therapies aim to

reduce and prevent binge-eating and purging episodes, and also address distortions in body

image or weight and self-worth. Pharmacotherapies are more often used as adjunct therapies to

improve response rates or delay relapse. For example, adjunct fluoxetine after inpatient weight

gain was found to significantly reduce relapse rates one year after discharge in a double-blind

placebo-controlled study (Kaye et al, 2001).

Cognitive-behavioural therapy is also a conventional treatment for BN. In one

randomized controlled trial comparing CBT to IPT, it was found that individuals in the CBT arm

experienced faster and greater improvement by 6 weeks relative to individuals randomized to

76

IPT (Wilson et al, 2002). Group-based CBT also appears to be an effective treatment for BN,

although it is not yet clear whether it is as effective as, or superior to, individual CBT (Polnay et

al, 2014). Likewise, the methods of CBT trials vary widely, and studies often have modest

sample sizes (Hay et al, 2009). Pharmacotherapies are also associated with superior outcomes

relative to placebo for BN; for example, the World Federation of Societies of Biological

Psychiatry recommends tricyclic antidepressants and fluoxetine as effective treatments for BN

(Aigner et al, 2011). One other randomized-controlled trial of pharmacotherapies and

psychotherapies reported that concurrent pharmacotherapy and CBT has improved efficacy in

terms of binge reduction relative to monotherapies (Claudino et al, 2007), although these

findings have not been observed consistently (Walsh et al, 1997).

Response and remission rates for AN vary widely across studies and treatment

interventions. On average, studies that assess short-term treatment outcomes report the poorest

remission rate; in one study AN remission rates were 29% (Clausen, 2008). Other naturalistic

follow-up studies report that remission rates increase over time, with remission increasing from

68 to 84% between 8 and 16 year follow-up visits (Nilsson and Hägglöf, 2005). Another long-

term 2-year follow-up study of AN reported a recovery rate of 60.3%, a partial response rate of

25.8%, and a non-response rate of 12.8% (Rigaud et al, 2011).

1.8.3.2 Treatment-Resistance in AN and BN As indicated by the above studies with up to 20-years of follow-up data, both AN and BN

are chronic disorders. Only a third of AN patients will recover within 4 years of disease onset,

almost half will recover by 10-years after onset, and up to 75% after 10 years (Berkman et al,

2007; Steinhausen, 2002). Approximately 25% of AN patients will continuously relapse or

become chronic in their course of illness (Berkman et al, 2007; Steinhausen, 2002). While the

general consensus is that patients respond well to inpatient-based programs (Olmsted et al,

2010), many will relapse following discharge. The overall relapse rate following inpatient weight

restoration programs is 35%, and the median survival time of successful weight restoration is 18

months (Carter et al, 2004). Other studies report a range of rates of successful weight restoration

between 9 and 65% (Carter et al, 2004; Keel et al, 2005; Strober et al, 1997; Walsh et al, 2006).

As with AN, studies of BN report relapse rates between 25-63%, depending on the definition of

77

relapse and the duration of follow-up (Grilo et al, 2012; Halmi et al, 2002; Herzog et al, 1999;

Keel et al, 2005; Olmsted et al, 2005).

1.8.3.3 Brain Stimulation Treatments for AN and BN Given the limitations of current treatment strategies in AN and BN, novel treatment

options are needed, and brain stimulation treatments figure prominently in current research.

Particularly in the case of AN, where mortality rates are high, invasive options involving

neurosurgery may be considered worth pursuing if effective. Novel neurosurgical interventions

for intractable AN are well-documented, but these studies report mixed clinical results (for a

review, see (Lipsman et al, 2013c)). DBS is a promising invasive therapy more recently tested in

AN. In 2010, the first case study of successful sgACC-DBS in a TRD patient with comorbid AN

reported that the individual experienced weight restoration and eating disorder remission at 3-

years follow-up (Israël et al, 2010). Another target under study is the VS, with one case series of

VS-DBS reporting that 65% of patients achieved increased weight at 38-months follow-up (Wu

et al, 2013). The first phase 1 trial of DBS for AN was published in 2013 – 6 patients with

intractable AN underwent sgACC-DBS and were observed for 9 months in follow-up. Lipsman

and colleagues reported that 3/6 of these patients achieved and maintained a higher BMI than

baseline, and experienced significant improvements in mood, AN-associated obsessions and

compulsions, with benefits maintained at 1-year follow-up (Lipsman et al, 2013b, 2017). While

short- and long-term clinical outcomes appear promising, DBS studies in general are plagued by

extremely low sample sizes. DBS, as previously discussed, is an invasive treatment, and is only

available in specialized academic neurosurgical centres, focusing on the most refractory clinical

cases. To date, no DBS studies have been published investigating its efficacy in BN.

As previously mentioned, one non-invasive and more accessible alternative to DBS is

tDCS. To date, there are few published randomized sham-controlled trials of tDCS for either AN

or BN. An overwhelming majority of the published tDCS literature has recruited healthy female

patients with abnormally high food cravings, as opposed to individuals with a formal DSM-5

diagnosis (for a review see (Val-Laillet et al, 2015)). In four independent studies, a single session

of right DLPFC anodal/left DLPFC cathodal tDCS significantly reduced cue-induced craving,

78

food take, and improved the participants’ ability to control cravings relative to sham tDCS

(Fregni et al, 2008; Goldman et al, 2011; Kekic et al, 2014; Lapenta et al, 2014).

Only one randomized controlled trial of tDCS in BN has been published to date. In this

study, 39 BN subjects received 3 different sessions of tDCS: anode right/cathode left, anode

left/cathode right, and sham tDCS. In all three tDCS sessions, BN patients experienced

significant improvements in binge-purge behaviours 24-hours post-stimulation, but mood only

improved in the anode right/cathode left active condition (Kekic et al, 2017). The first open-label

tDCS study, in 7 AN participants, involved 10 sessions of anodal left DLPFC tDCS, reporting

significant improvements of symptom severity on the Eating Disorders Inventory and Eating

Attitude Test (Khedr et al, 2014).

1.8.3.4 rTMS as a Treatment for AN and BN rTMS offers the advantages of being non-invasive, safe, focal, and more accessible than

many other neuromodulation techniques – as such, it is a potentially attractive alternative

treatment for AN and BN patients who do not respond to tDCS or do not consent to ECT or

DBS. The earliest reported study of rTMS as a treatment for BN is a case report of an MDD

patient with comorbid BN; the patient experienced an unexpected remission of BN symptoms

after 10 sessions of 20 Hz rTMS over the left DLPFC (Hausmann et al, 2004). Subsequent

studies using high frequency left dorsolateral stimulation report mixed clinical outcomes: two

studies reported that a single active rTMS session reduced the urge to eat, reduced the number of

binges 24-hours post-stimulation, and reduced salivary cortisol (Claudino et al, 2011; Van den

Eynde et al, 2010), while another reported no difference between active- and sham-stimulation

after 15-sessions of 20 Hz rTMS (Walpoth et al, 2008). More recently, a single session of high

frequency left DLPFC-rTMS reduced subjective craving immediately post-rTMS in a small BN

sample (Sutoh et al, 2016). However, the most recent randomized double-blind sham-controlled

trial of 10 Hz left DLPFC-rTMS in 47 BN patients reported no significant improvement in

binge-eating and purging relative to placebo (Gay et al, 2016). Recently, the dmPFC has been

discussed as a potential novel rTMS target for BN. As with the first case report of DLPFC-rTMS

for BN, our group reported an unanticipated remission of binge-eating and purging in an MDD

patient with comorbid BN following 20 sessions of 10 Hz dmPFC-rTMS (Downar et al, 2012).

79

This effect on binge-eating and purging was rapid and was maintained for 9-weeks post-

treatment in the absence of booster rTMS sessions or other further interventions. However, as

with OCD, the therapeutic mechanisms of dmPFC-rTMS in BN remain poorly characterized in

current literature.

In the case of AN, the first open-label case study of left DLPFC-rTMS reported that 19-

20 sessions of high frequency rTMS elicited significant symptom and mood improvements that

persisted at 6-months follow-up (McClelland et al, 2013). One pre-clinical study in 10 AN

subjects found that a single session of 10 Hz left DLPFC-rTMS elicited less anxiety and

alleviated feelings of fatness immediately post-rTMS (Van den Eynde et al, 2013). More

recently, a randomized sham-controlled trial of one session of left DLPFC-rTMS was performed

on 60 AN individuals, reporting no significant group-by-time interaction between treatment arm

and core AN symptoms (McClelland et al, 2016). It also appears that a follow-up study of 20

sessions of 10 Hz left DLPFC-rTMS is currently ongoing (Bartholdy et al, 2015).

Improving rTMS Response Recent meta-analyses of rTMS as a treatment for TRD report that clinical response and

remission rates are superior to sham, although even the most successful randomized trials have

struggled to surpass remission rates of 35-40% for active treatment (Levkovitz et al, 2015).

rTMS also shows promise as a treatment for other psychiatric disorders, including AN, BN, and

OCD, although the evidence base to date is more limited.

One factor that may limit current remission rates is that the ‘parameter space’ of possible

rTMS protocols is very large, encompassing setting such as stimulation site, intensity, frequency,

pattern, duration, interval, and course length. At present, the influence of most of these rTMS

parameters on clinical response and remission rates are currently unknown. In the current

absence of any validated preclinical biomarker for the neuroplastic effects of rTMS outside

motor cortex, efforts at parameter optimization are of necessity empirical in nature.

The following section outlines five rTMS parameters and two additional protocols that

have the potential to improve rTMS treatment response and remission rates, if properly

optimized. Since TRD is most heavily studied in clinical rTMS studies, all of the results

80

summarized hereafter pertain to TRD, although it is likely that these parameters would similarly

influence response and remission rates in OCD and eating disorders.

Number of Total Treatment Sessions Early rTMS trials for TRD used once daily rTMS, and applied short courses of as few as

5 sessions, compared to current norms of 20-36 sessions or longer in current practice. As such,

the total number of treatment sessions employed in these earlier trials is now considered perhaps

sufficient to separate active from sham effects, but insufficient to characterize the truw effect

size for an adequate course of rTMS treatment (Carpenter et al, 2012; Connolly et al, 2012;

McDonald et al, 2011). Notably, TRD-rTMS studies with methods that include longer treatment

courses have reported an increase in response and remission rates, around 50% and 33%

respectively (Carpenter et al, 2012; Ciobanu et al, 2013; Connolly et al, 2012; Downar et al,

2014; Fitzgerald et al, 2011; McDonald et al, 2011). A recent meta-analysis also reported a

significant positive correlation between the clinical response rate and the number of rTMS

treatment sessions (Teng et al, 2017).

Number of Daily Treatment Sessions A typical course of rTMS treatment involves a once-daily treatment regimen. Given that

the number of treatment sessions needed for response or remission is between 20-30 sessions,

this would mean that patients are required to come to the clinic every day, typically Monday to

Friday, for between four and six weeks. This is a serious logistical inconvenience for patients,

especially those who are able to work or must commute a long distance for daily treatment. Long

treatment courses also render rTMS less practical for certain settings in which it might otherwise

be useful, such as inpatient wards.

One possible solution to these challenges is to perform multiple rTMS treatment sessions

in a single day. The first accelerated rTMS study for TRD was reported by Holtzheimer and

colleagues in 2010. TRD patients received 15 sessions of left DLPFC-rTMS over 2 days, and a

subgroup experienced full remission from depression and anxiety that persisted at 3- and 6-

weeks follow-up in an open-label setting (Holtzheimer et al, 2010). Subsequent sham-controlled

81

crossover trials of 5-times daily left DLPFC-rTMS protocols have reported superior response

rates relative to sham, albeit at lower response rates (35-38%) relative to once-daily rTMS

protocols (Baeken et al, 2013; Duprat et al, 2016). One open-label trial of twice-daily left

DLPFC-rTMS reported that the response and remission rates are comparable to once-daily

treatment, with response and remission rates of 55.6% and 37.0%, respectively (McGirr et al,

2015). Open-label accelerated low frequency right DLPFC-rTMS was also recently reported to

be tolerable and possibly efficacious in TRD (Tor et al, 2016). Similarly, a chart review of

patients who received once- or twice-daily dmPFC-rTMS reported improvement trajectories that

tracked the number of sessions rather than the number of pulses, such that patients who received

their daily dose of pulses in 2 sessions 80 min apart rather than a single session improved in 10-

15 days rather than the typical 20-30 days (Schulze et al, 2018).

As a caveat, accelerated treatment regimens have not clearly achieved higher remission

rates than once-daily sessions – instead, those who respond merely seem to show more rapid

response. Thus, it is less clear that multiple daily session protocols will deliver large

improvements in remission rate over existing techniques.

Neuronavigation During rTMS Another proposed approach to improve rTMS outcomes is to develop better techniques to

position and orient rTMS coil during treatment, so as to more effectively induce activity in the

target brain region, and thereby improve treatment outcomes. For DLPFC-rTMS, the original

approach to neuronavigation employed a ‘5 cm rule,’ in which the rTMS operator localizes the

primary motor ‘hotspot’ requiring the least intense stimulation to elicit movements of the

contralateral abductor pollicis brevis muscle, and subsequently positions the rTMS coil 5 cm

anterior to that location as a simple heuristic to target the DLPFC (George et al, 1995; Pascual-

Leone et al, 1996b). However, one difficulty with the ‘5 cm’ heuristic is that it may not be

reliable for localizing the DLPFC: one study using MRI to verify coil placement found that the 5

cm rule placed the coil outside (posterior) to DLPFC in 1/3 of individuals (George et al, 2010).

More recent studies have taken advantage of image-guided neuronavigation, which aids

the technician in placing the rTMS coil based on a structural MRI (Herwig et al, 2001b; Krings

et al, 2001; Schönfeldt-Lecuona et al, 2005). This approach can help to ensure that the

82

stimulation is aimed at the desired DLPFC site, rather than more posterior or dorsal stimulation

sites outside the DLPFC that have been associated with lower chances of remission (Herwig et

al, 2001a). Recent studies investigating DLPFC targeting in MDD reported that the more ventral

and anterior DLPFC targets were associated with better response to rTMS (Fitzgerald et al,

2009; Fox et al, 2012a). However, purely anatomical MRI-based navigation techniques are also

potentially problematic, as anatomical localization still does not always align adequately with

functional localization. Other approaches suggest that the optimal DLPFC site is characterized

by activity that is anticorrelated to that of the sgACC on rs-fMRI, highlighting the potential

importance of functional localization of the target TMS site to downstream or monosynaptic

regions (Fox et al, 2012a; Herbsman et al, 2009).

Although functional MRI-guided targeting may prove fruitful for improving remission

rates in future, other studies suggest that there exist scalp heuristics that can nearly match the

accuracy of MRI-guided neuronavigation for targeting the DLPFC and DMPFC. These include

the ‘BeamF3’ heuristic for DLPFC stimulation and 25% nasion-to-inion measurement for

DMPFC stimulation, which may provide a reasonable approximation to MRI, while avoiding the

added cost, complexity, and accessibility issues that arise when MRI-guided neuronavigation is

employed (Mir-Moghtadaei et al, 2015, 2016). To date, the available literature on therapeutic

rTMS has not been able to demonstrate a clear advantage in remission rates for neuroimage-

guided vs. scalp-heuristic navigational methods.

Novel Stimulatory Protocols The majority of rTMS studies to date in MDD have used ‘conventional’ stimulatory

protocols, such as 1 Hz for low frequency and between 5 and 20 Hz for high frequency

stimulation. Although well-supported by large studies and meta-analyses over the last 25 years,

these conventional rTMS protocols carry some disadvantages. First, these protocols have

traditionally been quite long in duration; the U.S. Food and Drug Administration-approved

protocol of high frequency left DLPFC stimulation lasts approximately 37.5 minutes (O’Reardon

et al, 2007). Long stimulation protocols limit the capacity of rTMS treatment facilities to only a

few patients per device per day, which impedes rTMS from making a meaningful reduction in

the high (2%) population prevalence of TRD overall. Shorter stimulation protocols could

83

therefore increase rTMS clinic capacity, lower treatment costs, and lessen patient commitment

throughout treatment, while also enabling rTMS to potentially reduce the population prevalence

of TRD.

Theta-burst stimulation (TBS) is one such protocol that could achieve similar potency to

standard protocols, while requiring several-fold less time to administer. TBS applies rTMS

pulses in bursts of 3 pulses at 50 Hz with such bursts repeated at 5 Hz (Huang et al, 2005).

Excitatory stimulation, called intermittent TBS (iTBS), is achieved by cyclically applying 2

second TBS bursts followed by an inter-train interval of 8 seconds, while inhibitory stimulation,

called cTBS. involves repeating TBS bursts with no inter-train interval ((Huang et al, 2005; Di

Lazzaro et al, 2011), for a review see (Cárdenas-Morales et al, 2010)). Both protocols consist of

only 600 pulses total, with iTBS and cTBS thus requiring approximately 3 minutes versus 40

seconds, respectively.

A number of studies demonstrate that response and remission rates of iTBS are

comparable to that of longer conventional rTMS protocols for TRD. Most recently, our group

published a large, multi-site non-inferiority trial comparing left DLPFC 10 Hz and iTBS, finding

nearly identical response and remission rates for the two interventions (Blumberger et al, 2018).

Another recent meta-analysis of randomized and sham-controlled TBS trials for MDD reported

that the response but not the remission rates of active TBS protocols are superior to that of sham

TBS (35.6% vs. 17.5% for response, and 18.6% vs. 10.7% for remission, respectively) (Berlim et

al, 2017). Although protocols included left DLPFC iTBS (Duprat et al, 2016) and right DLPFC

cTBS (Chistyakov et al, 2015), it appeared that bilateral TBS (left iTBS, right cTBS) was the

most promising protocol relative to placebo (Berlim et al, 2017; Li et al, 2014a; Plewnia et al,

2014; Prasser et al, 2015). Considering targets other than DLPFC, bilateral dmPFC-iTBS also

appears to be as effective as 10 Hz bilateral dmPFC-rTMS, with no significant difference in

response and remission rates between these two protocols (Bakker et al, 2015). Neither iTBS nor

cTBS have yet been investigated in eating disorders and OCD.

A potential shortcoming of all rTMS protocols in clinical use to date is that their

physiological effects on brain plasticity are highly heterogeneous across individuals. In

preclinical studies using markers such as MEPs or rs-fMRI, a substantial fraction of individuals

show inhibition on nominally ‘excitatory’ protocols and excitation on nominally ‘inhibitory’

84

protocols. This heterogeneity is apparent not only for conventional protocols like 1 Hz and 10 Hz

rTMS (Eldaief et al, 2011; Maeda et al, 2000) but also in newer protocols like iTBS and cTBS

(Cárdenas-Morales et al, 2014; Hamada et al, 2013; Nettekoven et al, 2015). One potentially

more reliable novel protocol that has yet to be clinically assessed is quadripulse stimulation

(QPS). QPS involves four monophasic (rather than conventional biphasic) pulses, and when the

inter-stimulus interval (ISI) is manipulated QPS can be made more consistently excitatory or

inhibitory than other protocols. Generally, short ISIs (1.5-10ms) result in in MEP facilitation

(LTP), while longer ISIs (50-100ms) generate MEP inhibition (LTD) (Hamada et al, 2007,

2008). A recent replication study demonstrated that QPS-induced MEP effects are consistent in

direction across 80% of subjects (Nakamura et al, 2016), substantially higher than the

corresponding rates of as little as 50% for standard 1 Hz, 10 Hz, or TBS protocols. While QPS

appears to be safe and well-tolerated, clinical studies assessing the benefits of QPS have yet to be

performed in any psychiatric disorder population (Nakatani-Enomoto et al, 2011).

Novel rTMS Stimulation Targets in Psychiatric Illness An emerging area of therapeutic exploration is the use of alternative neurostimulatory

targets other than the DLPFC or dmPFC to treat psychiatric disorders. Given the transdiagnostic

abnormalities of regions such as the anterior cingulate cortex and anterior insula (Goodkind et al,

2015), and given that a host of other brain regions are structurally or functional abnormal in

psychiatric illnesses (Downar and Daskalakis, 2013), novel targets may offer treatment options

for subsets of psychiatric patients where dmPFC- or DLPFC-rTMS has failed. The frontopolar

cortex, ventromedial prefrontal cortex, and ventrolateral prefrontal cortex have, for example,

been proposed as MDD-rTMS targets in addition to the DLPFC and dmPFC based on the

evidence of lesion, stimulation, connectivity and neuroimaging studies (for a review see (Downar

and Daskalakis, 2013)). For example, one recent case report, one open-label case series, and one

sham-controlled trial have reported preliminary clinical findings using orbitofrontal /

ventrolateral prefrontal rTMS targets in MDD and OCD. All studies reported significant clinical

improvements following 1 Hz rTMS over the orbitofrontal cortex, in some cases where DLPFC_

and dmPFC-rTMS had failed; such observations support the potential future utility of novel

therapeutic rTMS targets for improving overall remission rates (Feffer et al, 2018; Fettes et al,

2017a; Nauczyciel et al, 2014).

85

Summary: Psychiatric Illness, IBNs, and rTMS Treatments

To summarize the key points reviewed thus far, MDD, AN, BN, and OCD are psychiatric

disorders characterized by substantial heterogeneity in terms of clinical presentation and

symptomatic improvement. Given that there is considerable overlap of these disorders, in both

comorbid symptoms and in the pharmaco- and psychotherapeutic options available, it is likely

that these disorders are related to common neurobiological or cognitive disruptions.

Neuroimaging studies indeed suggest that there are common, transdiagnostic domains of

abnormal function that may span multiple traditional diagnostic categories. The RDoC domains

of cognitive control, positive and negative valence systems offer one model of such

transdiagnostic abnormalities. Of note, these domains appear to be associated with specific

neural substrates in the form of specific IBNs, including the SN and VMN. Since rTMS appears

to exert an effect not only at the stimulation site but also across its associated IBN, this

intervention may be suitable for therapeutic use transdiagnostically across a variety of

psychiatric disorders. However, remission rates still fall below 50% in even the most successful

randomized trials, and further efforts are needed to improve treatment outcomes. Parameter

optimization may deliver some improvements in the consistency of rTMS effects; however,

given the bimodal outcomes observed in some rTMS studies, improving remission rates above

50% will likely require techniques for i) predicting treatment response from pre-treatment

neurobiological measures, and ii) characterizing the changes from pre- to post-treatment that are

associated with successful versus unsuccessful treatment outcomes. We now turn to these goals

in the next section of this Introduction.

Predicting Treatment Response

Why Predict Treatment Response? One potentially fruitful strategy for improving remission rates is to select an optimal

treatment strategy based on the individual’s underlying biological markers. Individualizing

patient care using reliable biological predictors, or ‘biomarkers,’ may deliver benefits in three

possible ways. First, identifying treatment responders from non-responders prior to treatment

86

may avoid having patients undergo logistically onerous but futile courses of treatment, and

thereby accelerate the heretofore trial-and-error process of finding an effective treatment for a

given patient. A significant time-commitment is required of rTMS patients, especially since

rTMS requires patients to receive treatment at a clinic or hospital daily over many weeks, and so

avoiding futile treatments by predicting responders and non-responders to a given rTMS protocol

is a reasonable first step. Second, identifying reliable biomarkers of treatment response may help

to inform the search for new treatments for those who do not respond. For example, a

neuroimaging biomarker may reveal brain regions whose baseline activity or connectivity differs

between responders and non-responders to a given treatment. Brain regions that are abnormally

hyper- or hypo-active in non-responders, or that have a different pattern of network connectivity

in responders versus non-responders, may present neurostimulatory targets worthy of future

clinical investigation. Finally, on a more basic and mechanistic level, biomarkers that distinguish

individual differences in response to treatment may help to parse the heterogeneity of illness

within the diagnostic entities of MDD and other psychiatric disorders, and may thereby further

the field in terms of understanding the various pathways by which individuals may develop

superficially similar forms of illness.

Clinical Predictors of Response to Conventional Interventions

1.10.2.1 Predicting Treatment Response in MDD

Pre-treatment clinical characteristics may provide insights on which MDD patients

respond best to certain first-line treatments (for a review, see (El-Hage et al, 2013)). For

example, the presence of bipolar depression is related to poorer outcomes in general (Gijsman et

al, 2004). Similarly, MDD with comorbid anxiety is associated with poorer pharmacotherapy

outcomes (Fava et al, 2008). Heightened somatic symptoms, including pain, or energy and

autonomic deficits, also predict a lower chance of remission: in individuals with no or limited

somatic symptoms, remission rates correspond with the STAR*D trial (68% remission rate),

while in individuals with severe somatic symptoms remission rates dropped to 29.3% (Novick et

al, 2013). However, age, sex, and MDD ‘subtype’ (e.g., atypical versus melancholic) do not

appear to moderate outcomes for MDD patients on psychotherapies or pharmacotherapies

(Cuijpers et al, 2014; Driessen and Hollon, 2010; Simon and Perlis, 2010), although one meta-

87

analysis has shown lower response rates in older patients (Tedeschini et al, 2011). In one study

assessing demographic and clinical predictors of citalopram (an SSRI) response, individuals who

were employed, Caucasian, female, and of higher socioeconomic or education status had higher

remission rates, while individuals with comorbid psychiatric or other medication conditions and

lower quality of life had poorer remission rates (Trivedi et al, 2006). Also, modest sex

differences have been found to the response rates of some other SSRI and tricyclic

antidepressants, including fluvoxamine and imipramine (Vermeiden et al, 2010). It is important

to note that, although these clinical predictors show significance at the group level, none has yet

achieved the much higher bar of clinically meaningful predictive value at the individual level.

1.10.2.2 Predicting Treatment Response in OCD The prognosis of pharmacotherapy and behavioural therapies in OCD is hampered by the

natural course of OCD psychopathology, often characterized by a ‘waxing and waning’ course of

symptom severity (Bandelow et al, 2008), and by the heterogeneity of OCD in terms of its

clinical presentation (Ravizza et al, 1997). These two factors make the identification of a reliable

pathopsychological or clinical markers more challenging (Kellner, 2010). However, many

clinical factors are correlated with Y-BOCS improvement on pharmaco- and psychotherapies.

For example, comorbid personality disorders, poor insight into obsessions, and the

symmetry/hoarding and contamination/washing aspects of OCD are associated with poorer

response to clomipramine (for a review see (Hazari et al, 2016)). One systematic review of 69

psychotherapy trials reported that hoarding, anxiety and OCD symptom severity, and

demographic factors such as unemployment and marital status were associated with poorer

treatment outcomes (Knopp et al, 2013). Individuals who suffer from comorbid depression,

early-onset OCD, and individuals who suffer from chronic OCD (at least 30 years) are unlikely

to respond to psychotherapy (Jakubovski et al, 2013). In one 7-year follow-up study of OCD

patients treated with a combination of CBT and a SSRI, Rufer and colleagues reported that over

95% of participants required additional treatment throughout follow-up, suggesting the chronic

nature of OCD and the need for treatment maintenance to achieve long-lasting improvements

(Rufer et al, 2005). Again, however, clinical predictors in OCD have generally not yielded

meaningful predictive value at the individual level thus far.

88

1.10.2.3 Predicting Treatment Response in AN and BN A number of clinical factors are associated with poorer treatment outcomes in AN. In one

13-year follow-up study of 484 patients, 8 factors predicted poor outcome at 2-years follow-up:

first, low BMI at hospital discharge; second, low food intake; third, high desire for excessive

exercise; fourth, high perfectionism; fifth, high interpersonal distrust; sixth, severe anxiety;

seventh, use of tube-feeding during treatment; and eighth, adherence to treatment (Rigaud et al,

2011). Two other studies have reported that outstanding functional impairment after treatment

predicted poorer long-term response to conventional interventions (Bryant-Waugh et al, 1996;

Lask and Bryant-Waugh, 1992). Additionally, age is negatively correlated with AN response

(Steinhausen, 1997; Strober et al, 1997). Low body weight at AN onset and lowest minimum

BMI are also predictive of poorer outcomes to conventional interventions (Herpertz-Dahlmann et

al, 1996). A more recent study of 680 women with AN reported that self-induced purging and

higher trait anxiety significantly predicted poorer outcomes to treatment, whereas higher

impulsivity predicted positive improvement (Zerwas et al, 2013). Another study reported that the

binge-eating/purging AN subtype are more susceptible to relapse (Carter et al, 2012).

As with AN, a number of clinical factors also predict response and relapse rates in BN. In

one study assessing predictors of treatment response following CBT, a history of obesity, poor

pre-treatment global functioning, BN symptom severity, and comorbid depression predicted

poorer outcomes to CBT (Bulik et al, 1998a). Furthermore, post-treatment binge-eating and food

restriction, and high cue-induced food craving predicted poor outcomes at 1-year follow-up

(Bulik et al, 1998a). Less severe BN patients tend to improve more quickly and remit for longer

periods following CBT, relative to high-severity BN patients (Bulik et al, 1999). Another

predictor of BN response is an early change in purging frequency – BN patients who reported no

purging at 4-weeks into treatment were more likely to be responders (risk = 0.25) (Fairburn et al,

2004). Another recent study reported that acute depression symptom change is predictive of long

term BN response at 6-months follow-up (Thompson-Brenner et al, 2015). Patients who had

severe binge-eating and purging before day treatment also tend to relapse faster (Olmsted et al,

2015). Again, however, few of these group-level clinical predictors are capable of delivering

reliable individual-level predictions of treatment outcome.

89

IBN Predictors of Response to Conventional Interventions

1.10.3.1 Predicting Treatment Response in MDD Several studies have identified predictors of antidepressant response based on the

functional connectivity of IBNs, including predictors of response to CBT (Forbes et al, 2010;

Ritchey et al, 2011; Siegle et al, 2006), pharmacotherapy (Admon et al, 2015; Davidson et al,

2003; Forbes et al, 2010; Korgaonkar et al, 2014; Mayberg et al, 1997; Pizzagalli et al, 2001;

Spies et al, 2017; Taylor et al, 2014), ECT (van Waarde et al, 2015) and even to placebo (Sikora

et al, 2016). A recent meta-analysis of structural and functional predictors of response to

pharmaco- and psychotherapies reported that greater antidepressant response is associated with

relatively higher activity in the rACC (DMN) and lower activity in the AI and striatum (SN) (Fu

et al, 2013).

Three of these studies reported differences in DMN function that were predictive of

response to pharmacotherapies. Lower activity of the PCC during an emotional discrimination

task was correlated with HAMD improvement after 2 weeks of escitalopram (Spies et al, 2017),

while hyperactivity and hypermetabolism of the rACC (BA 24) at rest correlated with

antidepressant response to other pharmacotherapies, including SSRIs and nortriptyline (Mayberg

et al, 1997; Pizzagalli et al, 2001). Of special note, the functional connectivity of dmPFC to

other regions including the DLPFC, OFC, and PCC appears to achieve a higher bar of predicting

ECT treatment responders and nonresponders at the individual-patient level, with a sensitivity

and specificity of 84% and 85%, respectively (van Waarde et al, 2015).

Pre-treatment SN functioning has successful distinguished responders from non-

responders across a range of interventions including pharmacotherapy, psychotherapy, and

placebo. First, one study reported that high baseline dACC resting-state connectivity to SN

regions predicted antidepressant response to 2 weeks of placebo and 10 subsequent weeks of

pharmacotherapy, suggesting that SN integrity might serve as a common biological marker of

response irrespective of intervention (Sikora et al, 2016). Second, venlafaxine responders were

characterized by baseline dACC hyperactivity to negative versus neutral images (Davidson et al,

2003). Finally, a third study reported baseline insular resting-state connectivity that correlated

with the degree of improvement on pharmacotherapy (Crowther et al, 2015). Frontostriatal

90

connectivity through dACC during reward processing of a monetary incentive delay task also

appears to be predictive of antidepressant response across two different studies (Admon et al,

2015; Walsh et al, 2017).

Finally, sgACC function has been studied extensively as a potential predictor of

treatment outcome, given the prominence of this region in many neuroimaging studies of the

neural correlates of depression (as reviewed above). Two studies have reported that baseline

sgACC hyperactivity during emotional processing is correlated with antidepressant response to

CBT (Ritchey et al, 2011; Siegle et al, 2006). Furthermore, the amplitude of low-frequency

fluctuations of the sgACC appears to correlate positively with ECT response (Argyelan et al,

2016). Resting-state connectivity between left and right sgACC also appears to distinguish MDD

individuals who relapse from those who continue to remit (Workman et al, 2017).

1.10.3.2 Predicting Treatment Response in OCD

Few studies have identified differences in structural and functional connectivity between

treatment responders and non-responders (for a review on this issue, see (Frydman et al, 2016)).

In one notable example, bilateral mOFC cortical thickness successfully distinguished OCD

treatment responders and non-responders to with 77% sensitivity, 81% specificity, and an overall

80% classification accuracy. Thicker left mOFC and thinner right mOFC was also associated

with a higher probability of being a responder to either CBT or an SSRI. However, this model

was only able to correctly classify roughly 70% of non-responders in a subsequent independent

sample (Hoexter et al, 2015).

1.10.3.3 Predicting Treatment Response in AN and BN

To date, only one neuroimaging study has reported structural or functional differences

between treatment responders and non-responders in eating disorders. In one study, caudate

hyperactivity during prediction errors predicted poor weight gain in adolescent AN (DeGuzman

et al, 2017). As of yet, there are no studies that investigate neurobiological predictors of binge or

purge behaviours, or predictors of improvements to other maladaptive thoughts of eating, weight,

91

or shape. This absence of studies likely relates to the lack of consistent definitions of eating

disorder response and remission, and the substantial inter-individual variability of neuroimaging

and clinical correlates of AN and BN, as discussed earlier.

Predictors of Response to rTMS

1.10.4.1 Clinical Predictors of rTMS for TRD

A number of demographic and clinical features have been identified as predictive of

response to DLPFC-rTMS in TRD, at the group level. A recent meta-analysis reported that

females had higher response rates than males (Kedzior et al, 2014). Age also appears to

negatively predict response to high frequency and low frequency DLPFC protocols and to

relapse following rTMS, such that older patients do not respond as well to DLPFC-rTMS

compared to younger individuals (Aguirre et al, 2011; Fregni et al, 2006; Pallanti et al, 2012),

and older TRD patients have higher relapse rates relative to younger individuals (Cohen et al,

2009). Episode duration and treatment refractoriness also negatively predict treatment response,

such that patients with longer depressant episodes or more treatment-resistant MDD show poorer

responses to DLPFC-rTMS (Brakemeier et al, 2007; Fregni et al, 2006; Wu and Baeken, 2017).

A note of caution is warranted in interpreting clinical trials using rTMS, however, as much of the

available literature report low sample sizes, or sub-optimal rTMS protocols (for example, too

few rTMS sessions). Furthermore, very little is known about the clinical predictors of dmPFC-

rTMS; only one study from our group found that high baseline anhedonia predicted poorer

response to open-label dmPFC-rTMS (Downar et al, 2014). Thus far, no reliable individual-level

predictors of rTMS response in TRD have been identified.

1.10.4.2 Structural and Functional Predictors of DLPFC-rTMS Response Early neuroimaging predictors of DLPFC-rTMS used positron emission tomography

(PET) to measure baseline differences and/or changes in cerebral blood flow or metabolism

between rTMS responders and non-responders. For example, one early study found lower

baseline glutamate concentrations in DLPFC in the rTMS responder group, that increased over

the course of treatment (Luborzewski et al, 2007). Higher baseline DLPFC, dmPFC, and rostral

92

ACC metabolism on fluorodeoxyglucose(FDG)-PET have been correlated with better response

to once-daily, and to accelerated 5-times-daily DLPFC-rTMS (Baeken et al, 2009; Li et al,

2010b). DLPFC and VMPFC regional cerebral blood flow (rCBF) have been reported as

predictive of outcomes for both high-frequency left and low-frequency right DLPFC-rTMS

response (Kito et al, 2012a, 2012b). Again, all reported predictors to date are for group-level

rather than individual-level outcomes.

In line with previous findings linking VMN and sgACC hyperactivity, and CEN

hypoconnectivity to antidepressant response, Fox and colleagues found that differences in

resting-state functional connectivity between the DLPFC and sgACC predicted treatment

response to DLPFC-rTMS. Critically, anticorrelated functional connectivity between these two

regions predicted better responses to DLPFC-rTMS (Fox et al, 2012a, 2013b, 2014). Such

anticorrelated functional connectivity between the DLPFC and sgACC has been independently

replicated by other groups as a predictor of outcome (Baeken et al, 2014; Philip et al, 2018).

Similarly, one recent study reported that higher baseline sgACC metabolic activity was

predictive of response to accelerated high frequency left DLPFC-rTMS, and that this

hypermetabolism decreased post-treatment (Baeken et al, 2015).

1.10.4.3 Predictors of Response to dmPFC-rTMS Much less information is available about biological markers of dmPFC-rTMS than is

available about DLPFC-rTMS. Two recent resting-state fMRI studies performed by our group,

however, provide insights into who may best respond to 10 Hz dmPFC-rTMS in TRD.

The first study investigated the response to open-label 10 Hz dmPFC-rTMS in 25

subjects with treatment-resistant MDD (Salomons et al, 2014). The primary aim of the study was

whether resting-state functional connectivity of the stimulation target predicted response to

treatment. A secondary aim of the study was whether resting-state functional connectivity

change was associated with treatment response.

25 treatment-resistant MDD patients (10 male; mean age = 42.6 years, range = 19-70

years) with treatment resistant unipolar or bipolar depression were recruited for the study. The

rTMS protocol consisted of 3000 pulses per hemisphere, at 10 Hz, in 60 trains daily over 20

93

sessions (4 weeks). High baseline resting-state functional connectivity of the dmPFC and a

ventromedial prefrontal and subgenual cingulate region was associated with better treatment

response. Likewise, lower baseline functional connectivity between the dmPFC and right

putamen, right thalamus and right hippocampus/amygdala was associated with better treatment

response. sgACC functional connectivity was also predictive of dmPFC-rTMS treatment

response. High baseline functional connectivity between the sgACC and DLPFC and low

baseline functional connectivity between the sgACC and parahippocampus/amygdala were also

associated with better improvements. Decreased connectivity between the dmPFC and the

bilateral insula, and increased functional connectivity between the dmPFC and the bilateral

thalamus accompanied treatment response. Decreased functional connectivity between the

sgACC and the ventral striatum and middle cingulate cortex also accompanied treatment

response.

As previously reviewed, the cortico-cortical connectivity relationships of the sgACC

appear to be a critical component of antidepressant response. In terms of the cortico-cortical

relationships between the sgACC and the dmPFC more specifically, we postulated that the

interplay between these two regions could be related to goal-directed cognitive control over

affective responses typically associated with these regions (Simpson et al, 2001a, 2001b). Given

the role of the dmPFC in facilitating goal-directed behaviour via reward and punishment

integration (Shackman et al, 2011), the relationship between the dmPFC, the sgACC and clinical

response could be related to a difference in the capacity for executive control over core

emotional functions.

A second study from our our group used a graph theoretical metric of ‘hubness’ in

networks (betweenness centrality) to characterize differences in IBN organization and

architecture between dmPFC-rTMS responders and non-responders (Downar et al, 2014).

Among 47 patients diagnosed with treatment-resistant MDD (unipolar or bipolar) who

underwent 20 sessions of neuronavigated 10 Hz bilateral dmPFC-rTMS, 24 of 47 patients

achieved clinical response, while, of those, 20 met the remission criterion. The distribution of

clinical response on the HAMD was bimodal, indicating a heterogeneous response to dmPFC-

rTMS. Regarding clinical predictors, only three items emerged significant post-correction: the

BDI-II pessimism subscale; the BDI-II loss of pleasure subscale; and the QIDS general interest

subscale. A composite anhedonia scale of these measures was strongly correlated with response

94

(r = -0.61). On graph theoretical analysis of the resting-state fMRI data, responders displayed

significantly lower betweenness centrality in the VMPFC, the only region to survive Bonferroni

correction. Non-responders also showed a distinctively abnormal pattern of whole-brain

connectivity form this VMPFC region to other regions of the SN and VMN, compared to

responders. We interpreted this difference in VMPFC betweenness centrality, in combination

with the clinical anhedonia predictor, as different depression subtypes – one with preserved

hedonic function and responsive to dmPFC-rTMS and another with abnormal hedonic function,

visualized via abnormal functional connectivity and betweenness centrality from the VMPFC.

Characterizing Mechanisms of Treatment Response

Why Characterize Mechanisms of Treatment Response? Identifying the biological mechanisms by which an individual achieves response or

remission from a psychiatric disorder is valuable knowledge for both scientific and therapeutic

reasons. From a scientific standpoint, it is helpful to clarify whether abnormal markers (for

example, IBN structure or function) are state-dependent markers of the current depressed mood

state, or are stable trait-like features that persist during response or remission. For example, a

(hypothetical) finding that low SN integrity pre-treatment does not normalize with either

successful or unsuccessful rTMS treatment would have implications for the mechanisms by with

rTMS treats depressed mood. Thus, understanding the mechanism of treatment response will

allow for the characterization of which structural or functional features normalize as MDD

symptoms remit, and which persist. On a related point, understanding the biological mechanism

of treatment remission could clarify the biological mechanisms in which an individual will

relapse, possibly leading to robust biological predictors of treatment relapse. Additionally, from

a therapeutic and translational standpoint, identification of neural markers that change with

successful by not unsuccessful treatment could help lead to new treatments that might be more

effective in treatment non-responders in future.

95

IBN Changes in Response to Conventional Interventions

1.11.2.1 Network Changes Associated with Improvement in MDD

From currently available literature, changes in abnormal DMN and SN function appear to

accompany antidepressant response to a variety of conventional pharmaco- and psychotherapies;

a relatively recent meta-analysis of PET and fMRI characterizing antidepressant response to

medications highlights this general finding. Delaveau and colleagues demonstrated that

antidepressant response to pharmacotherapies was associated with increases in dmPFC, DLPFC

and VLPFC activity, and decreases in the activity of the amygdala, hippocampus, OFC, and all

regions of the DMN (Delaveau et al, 2011). Using PET, MDD patients treated with either an

SSRI or psychotherapy experienced decreases in sgACC hypermetabolism and increases in

dorsal ACC hypometabolism that correlated with antidepressant improvements (Brody et al,

2001; Drevets et al, 2002). During task-based fMRI, patients who successfully responded to

pharmacotherapies showed a normalization of dmPFC hyperactivity during affective processing

(Bermpohl et al, 2009). Normalized SN hypoactivity during negative mood (Fitzgerald et al,

2008) and incentive reward cues (Stoy et al, 2012), and normalized DMN hyperactivity during a

negative emotion task (Cullen et al, 2016) have been also reported following successful

pharmacotherapy. At rest, increases in dACC activity (Pizzagalli et al, 2001), and normalization

of DMN activity (Andreescu et al, 2013; Posner et al, 2013; Wang et al, 2015) have also been

observed following successful antidepressant intervention. Furthermore, on spectral analysis of

resting-state fMRI, abnormally high low-frequency amplitude fluctuations of the dmPFC and

striatum decreases with response to escitalopram (Wang et al, 2017).

Studies on the mechanisms of response to novel interventions, including DBS and VNS,

report similar mechanisms of response to that of conventional pharmaco- and psychotherapies.

Just as with the correlates of antidepressant response for conventional interventions, sgACC-

DBS responders experience metabolic increases in the DLPFC and dACC, and metabolic

decreases in the sgACC and OFC (Lozano et al, 2008; Mayberg et al, 2005), and decreases in the

amplitude of low-frequency fluctuations in the sgACC (Argyelan et al, 2016). DMN connectivity

is also modulated by VNS, such that increases in ACC and PCC resting-state functional

connectivity correlated with clinical improvement in active, but sham VNS (Fang et al, 2016).

96

1.11.2.2 Network Changes Associated with Improvement in OCD

SN and VMN corticostriatal structural and functional connectivity changes have been

linked to symptomatic OCD improvement. Conventional pharmacotherapies normalize ACC

corticostriatal or basal ganglia hyperconnectivity (Anticevic et al, 2014; Beucke et al, 2013;

Rauch et al, 2006), and have been shown to alter striatal and thalamic gray matter volume

(Atmaca et al, 2016; Hoexter et al, 2012). CBT-related symptom improvement has been

correlated with increases in OFC gray matter volume (Huyser et al, 2013). Symptomatic

improvement in these cases has been attributed to striatal and thalamic reductions in glucose

metabolism (Zuo et al, 2013). In DBS cases, symptomatic improvement was correlated with

decreases of dACC-ventral striatal hyperconnectivity (Figee et al, 2013). Examining the

complementary causal evidence of lesion studies, disrupting the structural integrity of the dACC

via cingulotomy causes improvement in some, but not all, cases of treatment-resistant OCD

(Dougherty et al, 2002).

1.11.2.3 Network Changes Associated with Improvement in AN and BN

Few studies to date have examined neural correlates of treatment response in AN or BN.

One systematic review reported that nodes of abnormal gray matter volume may normalize with

successful interventions, although there is substantial heterogeneity in existing publications (Van

den Eynde et al, 2012). Many other studies report neural changes associated with successful

weight restoration. Such studies report normalized nucleus accumbens and anterior insula

activity in response to the unexpected receipt of rewards in weight-restored AN (DeGuzman et

al, 2017), and changes in visual processing regions for self-images during self-perception

(McAdams et al, 2016). Similarly, abnormal delay discounting performance and abnormal

ACC/striatal activity normalizes after weight restoration in AN (Decker et al, 2015). However, it

is unclear whether structural or functional changes in response to weight restoration alone reflect

true psychiatric improvement, or rather metabolic and tissue-composition changes arising from

restored weight gain from a state of starvation.

97

Mechanisms of Clinical Response to DLPFC-rTMS Identifying and describing the biological mechanisms of rTMS treatment response is

critical to determining how to best optimize treatment parameters to improve further treatment

outcomes for non-responders. Unfortunately, there is limited evidence on the biological effects

of a full course of DLPFC-rTMS (as opposed to a single experimental session) in healthy

controls and psychiatric populations. Compounding this problem, the substantial variability of

rTMS protocols across clinical trials further hinders the discovery of reliable mechanisms of

response. Two recent publications review current knowledge regarding the biological

mechanisms of high-frequency left DLPFC-rTMS in TRD (Anderson et al, 2016; Noda et al,

2015). To summarize their findings, a number of studies have investigated the mechanism of

rTMS-induced clinical response on the following items: molecular factors, including

neurotransmitters and BDNF, electrophysiological factors, cerebral blood flow and metabolic

factors, and brain structure and function. The effects of DLPFC-rTMS in TRD on these measures

are inconsistent across individuals, and there are very few studies investigating the effects of

rTMS that characterize response. The literature has not yet achieved consensus on which

biological effects of a full course of rTMS are critical to treatment response, and which are

present but peripheral the mechanisms of treatment response in TRD; still less information is

available regarding treatment mechanisms for disorders other than TRD. Findings to date on

these issues are summarized below:

Effects of DLPFC-rTMS on Molecular Factors. There is some consistent evidence that the mechanism of DLPFC-rTMS in TRD involves

changes in concentrations of glutamate. Three studies have reported that glutamate levels in the

left DLPFC and ACC increase in treatment responders following 10 Hz left DLPFC-rTMS

(Croarkin et al, 2016; Luborzewski et al, 2007; Yang et al, 2014). Yang et al. also reported

significantly increased choline levels in the DLPFC following DLPFC-rTMS that was specific to

responders (Yang et al, 2014). Regarding BDNF, responders experience significant increases in

plasma BDNF levels relative to non-responders following 10 sessions of 20 Hz left DLPFC-

rTMS. Taken together, it appears that TRD responders to rTMS experience increased local and

plasma levels of neurotransmitters and factors that are related to neuronal plasticity and LTP.

98

rTMS may also induce symptomatic improvements via the release of other

neurotransmitters, such as dopamine and serotonin. rTMS induces striatal dopamine in rats

(Kanno et al, 2004; Zangen and Hyodo, 2002), and both dmPFC- (Cho et al, 2015) and DLPFC-

rTMS (Strafella et al, 2001) release striatal dopamine in a localized fashion (in the striatal

regions most closely connected to the cortical site of stimulation) in PET studies of healthy

human volunteers. Some (Pogarell et al, 2006, 2007), but not all (Kuroda et al, 2006, 2010;

Miniussi et al, 2005), studies report that DLPFC-rTMS increases striatal dopamine in patients

with TRD.

Effects of DLPFC-rTMS on Electrophysiological Factors. Two studies have reported significant changes on electrophysiological factors due to

DLPFC-rTMS, as measured by EEG and MEP, in treatment responders versus non-responders.

rTMS responders experience significant decreases in motor excitability, as measured by TMS-

evoked MEPs, following 10 sessions of 20 Hz DLPFC-rTMS (Bajbouj et al, 2005). On EEG,

prefrontal theta cordance, a quantitative measure related to the absolute and relative power in a

frequency band, significantly increases in DLPFC-rTMS responders relative to non-responders

(Ozekes et al, 2014). In sum, the clinical benefit of DLPFC-rTMS for TRD may be related to

oscillatory power and synchrony in prefrontal theta rhythms or changes in motor-evoked

excitability.

Effects of DLPFC-rTMS on Cerebral Blood Flow. Three publications report significant changes on regional cerebral blood flow in rTMS

responders relative to non-responders. However, the reported results are inconsistent, and this is

likely due to the fact that the treatment stimulation protocols also vary. Teneback and colleagues

reported increases in regional cerebral blood flow in the bilateral IFG and cingulate, but

decreases in the right medial temporal lobe, in responders of 10 sessions of 20 Hz left DLPFC-

rTMS (Teneback et al, 1999). However, Nadaeu et al. reported significant decreases on this

measure in the cingulate, OFC, bilateral insula and right amygdala in high frequency left, or low

frequency right DLPFC rTMS responders (Nadeau et al, 2002). More recently, decreased

cerebral blood flow in the perirhinal cortex was associated with TRD response to 20 sessions of

99

10 Hz left DLPFC (Richieri et al, 2012). Similar to the effects of conventional antidepressants,

sgACC cerebral blood flow significantly decreases after high frequency left DLPFC-rTMS in

responders (Takahashi et al, 2013).

Effects of DLPFC-rTMS in TRD on Brain Structure and Function. An important question concerns the effects of a course of rTMS on brain structure and

function overall, setting aside for the moment whether these effects reliably correlate with

response to rTMS. A number of studies have investigated the average effects of DLPFC-rTMS

on brain structure (gray matter volume, or white matter integrity) and functional connectivity.

Kozel et al. and Peng. et al. recently reported an increase in left PFC white matter integrity in

TRD patients who received active high frequency left DLPFC versus those who received sham

rTMS (Kozel et al, 2011; Peng et al, 2012). However, an earlier publication reported no change

in left PFC volume between active and sham rTMS.

On functional neuroimaging, two studies have reported average changes in brain activity

following active high frequency left DLPFC. One study reported that rostral ACC activity and

mOFC-ventral striatum connectivity increased during a DLPFC-dependent verbal fluency task

following active rTMS (Shajahan et al, 2002). Finally, Liston and colleagues recently reported

that DLPFC-rTMS normalized sgACC hyperconnectivity at rest, and re-established an

anticorrelation in activity between the CEN and DMN (Liston et al, 2014).

Effects of DLPFC-rTMS on Brain Structure and Function. A follow-on question concerns which effects of rTMS on brain structure and function are

apparent in responders but not in non-responders to treatment in TRD. Regarding this question, a

number of studies have reported changes in gray matter volume or activity associated with

DLPFC-rTMS response. Furtado et al. reported that left amygdala gray matter volume increased

(at trend-level) in DLPFC-rTMS responders compared to non-responders (Furtado et al, 2013).

Lan et al., however, demonstrated that increases in ACC gray matter volume correlated with

DLPFC-rTMS response (Lan et al, 2016).

100

On functional neuroimaging, Fitzgerald and colleagues reported that, during a planning

task, activation the bilateral middle frontal gyrus and left precuneus decreased in low frequency

right DLPFC-rTMS responders, but increased in the left middle frontal gyrus and right IFG in

high frequency left DLPFC-rTMS responders (Fitzgerald et al, 2007). Finally, Baeken et al.

recently reported increases in resting-state functional connectivity between the sgACC and

ACC/dmPFC in responders to accelerated left DLPFC-rTMS (Baeken et al, 2014). A recent

study of DLPFC-rTMS for TRD with comorbid post-traumatic stress disorder found that

response was correlated with decreased connectivity between the sgACC and the CEN (DLPFC),

and between the hippocampus and SN, again signifying the importance of between-network

connectivity in mediating treatment response to rTMS (Philip et al, 2018).

MRI and Resting-State Functional MRI Methods As noted earlier, resting-state functional MRI carries several practical advantages as a

tool for examining the neural predictors and correlates of rTMS response, including the lack of

radiation, ready accessibility, suitability for use in psychiatric populations, and suitability for

examining the behavior of IBNs that are central to psychiatric pathology and rTMS mechanisms

of effect. Yet the validity of resting-state fMRI as a tool for these purposes depends upon what

types of neural activity it serves to index, and under what circumstances it may be considered a

reliable versus an unreliable metric of brain functioning. This next section therefore provides a

concise review of the basic principles of resting-state fMRI physiology, acquisition and analysis.

Blood-Oxygen Level Dependent fMRI Physiology

The brain is a highly-vascularized organ with high levels of metabolic activity. The adult

brain represents approximately 2% of an individual’s body weight, and yet it accounts for over

20% of the body’s metabolic expenditure at rest (Kety and Schmidt, 1948; Raichle and Gusnard,

2002). The brain’s metabolic need stays relatively constant over time, even given the varying

metabolic demands of ongoing mental activity (Shulman et al, 2004; Sokoloff et al, 1955).

However, task/stimulus-evoked fMRI and resting-state fMRI rely on the complex interplay

between brain activity and the ratio of oxygenated and deoxygenated hemoglobin in the blood.

101

When ensembles of neurons in a given brain region become active, local vasodilation occurs,

increasing local blood flow and altering the concentration gradients of oxygenated and

deoxygenated hemoglobin between the capillaries and local tissue. Deoxyhemoglobin is a

paramagnetic substance, whose presence in a strong magnetic field produces subtle local field

distortions and inhomogeneities. The resultant changes in the spin resonance properties of local

hydrogen nuclei, and thus in the brightness of the signal intensity on certain types of MRI image,

is known as the blood oxygen-level dependent (BOLD) hemodynamic response.

To examine this point in slightly more detail, BOLD fMRI capitalizes on the intrinsic

properties of hydrogen (protons in the nuclei, 1H) in the presence of an applied magnetic field. 1H nuclei possess a property known as ‘spin’, which although not necessarily equivalent to the

spin of a macroscopic object, can nonetheless be treated similarly, can be treated similarly from

the perspective of mathematics and physics. The nuclear spin of atoms differs by element, and so 1H is an ideal choice for the study of organic tissue given its abundance in water and fat (Huettel

et al, 2004). Protons also have a magnetic moment, meaning that in the presence of a strong

magnetic field (typically 3 T for MRI, called the B0), their spins will act as if precessing around

the local magnetic field lines, and thereby creating a net magnetization vector preferentially

aligning with the longitudinal plane of B0. The spin precession frequency is proportional to the

strength of the magnetic field, and corresponds to radio-range frequencies of electromagnetic

radiation (e.g., MHz range). For this reason, protons can readily absorb energy from a

radiofrequency pulse at the rate at which protons precess (known as the Larmor frequency),

causing the magnetization vector of protons to ‘flip’ to transverse plane, which is anti-parallel to

the magnetic field of the B0 and is subsequently of higher energy but more unstable. When the

radiofrequency pulse terminates, the protons begin to emit the energy absorbed from the

radiofrequency pulse back to the ‘spin lattice’ of surrounding nuclei, and return to precessing

about the more stable longitudinal plane. This means that the magnetization vector generated by

the protons in the presence of the B0 returns to the longitudinal plane from the transverse plane.

T1 relaxation refers to amount of time for 63% (1-1/e) of protons to revert to the longitudinal

plane, and is a constant that differs by tissue type due to the mobility of atoms within the

molecule it comprises, which determine the rate at which the RF-energized protons can shed

their energy back to the spin lattice of surrounding protons. T2 relaxation refers to the amount of

time for the magnetization vector in the transverse plane to return to 37% (1/e) of its original

102

value (before the radiofrequency pulse). The value of T2 relaxation depends on rate at which the

precession of protons begin to de-phase after the radiofrequency pulse, and is again tissue-

specific – depending in part on local field inhomogeneities that cause the Larmor frequencies of

neighbouring protons to fall out of phase with one another slowly or rapidly, depending on tissue

type.

fMRI images have a contrast sensitive to T2* decay, which relates not only to spin de-

phasing related to random interactions between neighbouring protons (T2), but also local field

in-homogeneities (referred to as T2’). BOLD fMRI capitalizes on the fact that oxygenated and

deoxygenated hemoglobin display different magnetic properties – oxygenated hemoglobin is

diamagnetic while deoxygenated hemoglobin is paramagnetic in the presence of an external

magnetic field (Pauling and Coryell, 1936). In other words, deoxygenated hemoglobin interferes

with the local field homogeneity of another magnetic field, such as the B0, which therefore

decreases the overall intensity of T2*-weighted signal from that voxel (Poldrack et al, 2011).

The end result is that more oxygenated blood (with low concentrations of deoxyhemoglobin)

appears brighter on T2* images, while less oxygenated blood appears darker. During neural

activity, local vasodilation creates a rush of oxygenated hemoglobin, which then increases the

local field homogeneity and the intensity of the local T2* signal collected. Fast MRI gradients

can acquire a low-resolution image slice in 60-80 ms, such that a single whole-brain BOLD

image requires 2-3 seconds to acquire (known as the repetition time; TR).

The time course of the BOLD response has been described in detail elsewhere (see

(Poldrack et al, 2011)). Briefly, the BOLD response function has three components: the initial

dip in signal intensity, a slow increase to peak signal intensity, and the post-stimulus undershoot.

First, the initial dip refers to an initial decrease in T2*-weighted intensity within the first 1-2

seconds of stimulus onset. Some, but not all, fMRI studies have reported an initial dip, and it is

thought to be related to initial metabolic consumption prior to vasodilation (Brown et al, 2007).

Because the initial dip is not consistently observed, most omit it from statistical analysis and

BOLD modeling (Poldrack et al, 2011). However, more recent analyses of BOLD response

variability from trial-to-trial reported that the initial dip is reliably observable (Watanabe et al,

2013), and can vary by cortical layer under high-field (7 Tesla) fMRI (Siero et al, 2015). Second,

a slow increase in local signal intensity occurs as vasodilation increases local blood flow. As a

result of this cerebral blood flow change, and an influx of oxygenated hemoglobin in local

103

capillaries, increases in local field homogeneities cause increases in T2* signal intensity,

reaching to a peak intensity within 4-6 seconds of stimulus onset (Poldrack et al, 2011). Some

studies have reported substantial variability in the time it takes reach peak signal intensity, with

BOLD response peaks varying between 6-11 seconds across regions within the same subject

(Kruggel and von Cramon, 1999). The slow rise to peak signal intensity is thought to be due to

the sluggish replenishment of cerebral blood flow from arteries to local capillaries in response to

a stimulus (Logothetis, 2003). This signal increase is very small in relative terms – typically the

difference in signal between task and control conditions is less than 5% for paradigms involving

sensorimotor paradigms, and between 0.1 and 0.5% for cognitive paradigms (Poldrack et al,

2011). After the peak, the signal intensity begins to drop, decreasing below baseline signal

intensity over 10-14 seconds. Interestingly, a recent analysis on the variability in the peak and

post-stimulus undershoot reported significant variability in the signal intensity change of the

peak and the duration of the post-stimulus undershoot dependent on the depth of the cortex and

the stimulus duration (Siero et al, 2015).

It is currently thought that the changes to local cerebral blood flow in response to

neuronal activity deliver a substantial oversupply of oxygen to the local tissue compared to its

actual increased demand (or else no localized increase in oxygen concentration could be

visualized). This suggests that the canonical BOLD response could be due to factors other than

an increase in metabolic need related to neuronal activity. Reviewed in detail elsewhere, ongoing

theories suggest that changes in regional blood flow are mediated by astrocytes in additional to

neurons, perhaps as in relation to non-neuronal metabolic processes, or as a means to remove

chemical or thermodynamic waste that results from brain activity (Attwell et al, 2010; Heeger

and Ress, 2002; Raichle and Mintun, 2006). It is also possible that increases in BOLD are related

other aspects of aerobic metabolism, including glycolysis, and glial recycling of glutamate

following neuronal activity (for a review, see (Heeger and Ress, 2002; Raichle and Mintun,

2006). For example, slow astrocytic calcium dynamics following neuronal activity are correlated

with BOLD activity in mice, rats, and ferrets (Gurden, 2013; Petzold et al, 2008; Schulz et al,

2012; Schummers et al, 2008).

Numerous studies over the last three decades have established that the BOLD response is

related to electrophysiological signals related to neuronal activation, both in mice (Ogawa et al,

1990) and in humans. One seminal study by Logothetis and colleagues compared the local field

104

potentials of spiking activity in primates with visual cortex fMRI signal change to a visual

stimulus. They found that the largest fMRI signal changes were observed in areas that displayed

local field potential changes (Logothetis et al, 2001). Later studies of the BOLD response in cat

visual cortex demonstrated that these BOLD and local field potential changes were specific to

increases in gamma power (Niessing et al, 2005). Early human fMRI studies report similar

results – in one study, occipital fMRI signal increases were correlated to a visual stimulus,

further corroborating the importance of the relationship between the BOLD response and

neuronal activity (Belliveau et al, 1991; Kwong et al, 1992). More recent studies have used

ECoG to characterize the neural basis of the BOLD response. For example, as changes to local

field potential in animal models, BOLD activity correlates with the ECoG gamma oscillatory

power (He et al, 2008; Nir et al, 2008) associated with local neuronal activity (Hermes et al,

2012).

While it appears that LFPs and other measures that quantify neuronal activity of

populations of neurons, the relationship between the neuronal activity of individual neurons and

BOLD activity is not well understood. One study demonstrated in macaque neurons that only a

small subpopulation (~16%) of cortical neurons correspond with the BOLD response, meaning

that fMRI and the BOLD response may not reflect all of the true neuronal activity induced by a

task condition (Heeger and Ress, 2002). However, optogenetically-induced activity in the mouse

cortical layer V neurons increases local BOLD signal (Kahn et al, 2011, 2013).

Resting-State Functional MRI

Conventional task-/stimulus-based fMRI studies rely on minute differences in signal

intensity between task-induced brain activity versus a control condition. rs-fMRI, however, aims

to capture spontaneous neural activity in the absence of overt and experimentally-induced

responses (Snyder and Raichle, 2012). rs-fMRI offers an easy-to-implement complementary

technique to task-based fMRI, in which fMRI image volumes are collected in the absence of a

task condition. In a rs-fMRI study, participants are typically asked to stay still, let their mind

wander, and to try to stay awake. Resting-state scans are either collected with the participants’

eyes open or closed; differences in connectivity strength can be observed in primary sensory

networks such as the auditory and visual networks, and cognitive networks including the default

105

mode and attention networks, depending on the resting condition (eyes open versus eyes closed

versus eyes fixated) (Patriat et al, 2013). rs-fMRI studies are often considered logistically

simpler to implement in patient populations, particularly in patients with psychiatric disorders

that may render accurate task performance more challenging. he ease of implementation for rs-

fMRI explains in part its widespread presence as an investigational tool for studying whole-brain

neural function in patient populations.

As discussed in Section 1.12.1, task-evoked fMRI activity has a long history evidencing

its electrophysiological underpinnings. Resting-state fMRI activity is also most closely

associated with gamma power (50-100 Hz) (Fox and Raichle, 2007; Schölvinck et al, 2010,

2013) and infraslow activity (<0.1 Hz) (Lu et al, 2016). Spontaneous gamma fluctuations in

monkey visual cortex, as measured by LFP, correlate with spontaneous fluctuations of rs-fMRI

signal (Shmuel and Leopold, 2008). In humans, spontaneous gamma oscillations correlate

between brain regions that are highly functionally connected on rs-fMRI (Nir et al, 2008);

conversely, anticorrelated networks on rs-fMRI, such as the DMN and CEN, show similar

anticorrelations in ECoG gamma power (Keller et al, 2013). Studies of intracranial EEG IBNs

rs-fMRI connectivity also suppor the idea that IBNs are sustained by synchronous increases in

gamma power that reflect increased local spiking in nodes of IBNs (Logothetis et al, 2001;

Manning et al, 2009; Mukamel et al, 2005; Nir et al, 2007; Ray and Maunsell, 2011). Resting-

state IBNs are also related to infraslow cortical potentials (<0.1 Hz) (He et al, 2008) and have

been found to be more tightly correlated to spontaneous fluctuations in brain activity relative to

gamma (Lu et al, 2016). Spontaneous fluctuations in infraslow LFPs using EEG and MEG have

been shown to be related to spontaneous fluctuations in rs-fMRI (Hiltunen et al, 2014).

Spontaneous changes in cerebral blood flow are also tightly coupled with infraslow

ECoG fluctuations (Golanov et al, 1994). Spontaneous neuronal infraslow activity is attenuated

by voltage-gated sodium channel or glutamate receptor antagonists (Chan et al, 2015).

Spontaneous fluctuations in infraslow activity are also associated with apical dendritic

(Birbaumer et al, 1990; Mitzdorf, 1985) depolarization, suggesting that such activity contributes

to neuronal activity and therefore the BOLD signal (as reviewed by (Raichle and Mintun, 2006)).

However, widespread spontaneous changes in BOLD signal have been correlated with recorded

activity from single neurons in some (Shmuel and Leopold, 2008) but not all studies (Schölvinck

et al, 2013). It has therefore been proposed that spontaneous BOLD fluctuations are of neuronal

106

origin, and that non-neuronal factors such as respiration and cardiovascular factors (Birn et al,

2006; Shmueli et al, 2007; Wise et al, 2004) contribute to variability in this signal. Interestingly,

patterns of aerobic and anerobic metabolism differ in sensorimotor versus association cortices,

which might contribute to differential properties of IBN coupling and BOLD connectivity

(Vaishnavi et al, 2010).

rs-fMRI Preprocessing

The data generated from resting-state fMRI is inherently noisy, and is therefore

associated with a low signal-to-noise ratio (Biswal et al, 1995) and variable test-retest reliability

that is dependent on the acquisition parameters (for example, amount of scan time or TR) and

preprocessing techniques (Bennett and Miller, 2010). The time series of a rs-fMRI scan acquired

from each voxel represents a combination of ‘true’ physiological/neural signal, noise related to

participant motion, physiologically-induced noise from cardiac and respiratory sources, scanner

instabilities and drift, and random noise fluctuations. Aspects of these confounds can create

spurious correlations or mask neuronally-derived functional connections in the time series of

brain regions. Further complicating this issue is the fact that there is no ‘ground truth’ from

which one can model resting-state signal, and that the changes in ‘true’ resting-state signal are

very small. Consequently, one objective of resting-state fMRI preprocessing is, as best we can,

to remove data-inherent cofounds.

One major challenge to rs-fMRI preprocessing is that a consensus ‘gold-standard’

preprocessing pipeline for resting-state fMRI does not yet exist (Caballero-Gaudes and

Reynolds, 2017; Craddock et al, 2013), and there is considerable debate on which preprocessing

steps to include, and in what order. The decisions made about which preprocessing steps to

include, and in what order, can potentially impact the test-retest reliability of functional

connectivity and other measures of network strength or architecture (Zuo and Xing, 2014).

Resting-state fMRI pipelines vary substantially from study to study, but a number of

preprocessing steps are typically performed.

One often-used initial preprocessing step is to account for the temporal lags inherent

during the acquisition of the slices in a functional volume. The TR, the time it takes to collect

107

one volume (3D image) of the brain, is in the order of seconds, and so data from slices collected

at the beginning of the stack of slices comprising an image volume represent a different time and

location in the brain than data collected from later slices. One technique to mitigate the impact of

slice timing is to acquire successive slices in an interleaved manner. Additionally, slice-timing

correction during preprocessing aims to address this mismatched timing of slice acquisition by

interpolating voxelwise signal intensity data of each slice given some reference slice (typically in

the middle of the volume) (Craddock et al, 2013). While slice-timing correction is often

recommended for fMRI data, one study investigating its effects on rs-fMRI found that this

preprocessing step had a minimal effect on functional connectivity values, especially if rs-fMRI

volume data is collected in an interleaved fashion (Wu et al, 2011).

Two other commonly-implemented preprocessing steps are to spatially normalize each

individual scan to some template, and to spatially smooth the data. Both steps aim to increase the

reliability and generalizability of results by allowing for individuals’ scans of to be statistically

compared despite inter-individual variations in neuroanatomy (Mikl et al, 2008). Functional

scans are often first linearly co-registered to the participant’s own high resolution anatomical

scan, and then non-linearly normalized to a standard atlas such as the Montreal Neurological

Institute’s (MNI) template brain, which is an average of 152 non-linearly transformed high

resolution anatomical scans (Maintz and Viergever, 1998). Furthermore, smoothing improves the

functional correspondence of brain areas across individuals, especially as image co-registration

and normalization distorts the anatomical location of voxels to conform to the template image

(Worsley et al, 1996). Spatial smoothing is commonly performed using a Gaussian kernel (often

full-width, half maximum [FWHM]) about 1.5-2 times the voxel size acquired (thus, typically 6-

12 mm) (Bennett and Miller, 2010).

One of the largest sources of noise is head motion (Caballero-Gaudes and Reynolds,

2017) which generated strong signal fluctuations that are correlated in time between distant areas

of the head; left uncorrected, these correlated fluctuations can create spurious patterns of

apparent functional connectivity between distant brain regions in rs-fMRI data (Van Dijk et al,

2012; Power et al, 2014; Satterthwaite et al, 2013). Head motion is problematic for three

reasons: first, motion misaligns the brain between volume acquisitions (Craddock et al, 2013);

second, fMRI signal intensity derived from different tissues will change if the slice composition

108

changes due to movement; and third, fMRI signal intensity could drastically change due to

residual magnetization from prior slice excitations (Friston et al, 1996).

The first step to removing confounds related to head motion is to compute affine

transformations of each functional volume to some reference volume (typically the middle

volume in the scan) (Cox and Jesmanowicz, 1999; Friston et al, 1995; Jiang et al, 1995). More

recent motion correction algorithms leverage the spatial resolution acquired in the T1 anatomical

scan (Bhagalia and Kim, 2008; Ferrazzi et al, 2014; Kim et al, 1999), or perform slice-wise

motion correction to improve accuracy (Beall and Lowe, 2010; Zotev et al, 2012). The resultant

transformation from initial motion correction outputs a time series of the magnitude and

direction of required motion adjustments for each of the 3 translational and 3 rotational

directions. These time series can be used subsequently to remove temporally associated signal

fluctuations, using a regression analysis (Caballero-Gaudes and Reynolds, 2017). Nonlinear

expansions of each of these six-time series are also often included in this regression, most

commonly the first derivative of each time series. This step serves to remove spin history related

aspects of motion (due to prior slice excitations) (Friston et al, 1996).

Unfortunately, modeling and regressing movement-related confounds does not

completely remove motion-generated confounds (Power et al, 2012; Satterthwaite et al, 2012),

particularly ‘head jerks’ (Caballero-Gaudes and Reynolds, 2017). One way to mitigate these

issues is to remove volumes corrupted by significant motion; this process is called volume

censoring, or ‘scrubbing.’ There are two different ways to accomplish this step: first, through

volume removal and data interpolation the adjacent to the corrupted volume; and second, by

adding null regressors to an individual-level (first-level) general linear model, specifying

volumes in the time course one wishes to censor from any subsequent statistical analysis

(Lemieux et al, 2007; Satterthwaite et al, 2013; Yan et al, 2013).

There are also different techniques to identify problematic volumes, such as computing

their Euclidean norm (Jones et al, 2010), or framewise displacement (Power et al, 2012). In these

techniques, volumes that reach some arbitrary threshold of movement are subsequently censored.

While scrubbing is found to increase the signal-to-noise ratio and improve clarity of results

(Aurich et al, 2015; Power et al, 2012; Satterthwaite et al, 2013; Yan et al, 2013), removing

volumes also results in a reduction in the temporal degrees of freedom (Caballero-Gaudes and

109

Reynolds, 2017). Because of this, and because scrubbing typically identifies high frequency

confounds like abrupt head movements (Caballero-Gaudes and Reynolds, 2017), these methods

do have the potential to bias results, and in particular to affect measures related to temporal

dynamics (Craddock et al, 2013). One recently developed alternative is a censoring algorithm

that identifies wavelets of motion-based noise, rather than single spikes or outliers (Patel et al,

2014).

Another source of noise stems from physiological sources, such as cardiac pulsations and

respiration (Birn et al, 2006; Biswal et al, 1996). Cardiac function imposes pulsatile motion in

the brain via its pervasive network of blood vessels, while respiration and the subsequent

movement of the chest and abdomen can cause additional movement artifact, as well as localized

changes in magnetic susceptibility in certain head regions, with resultant fluctuations in signal

intensity (for a review see (Birn, 2012)). These confounds can once again cause spurious

correlations and anticorrelations, or mask ‘true’ neuronal signal in functional connectivity results

(Shmueli et al, 2007).

The most rigorous way to remove such confounds is to simultaneously measure the

participant’s heart rate via a pulse oximeter and respiratory rate using a chest bellows during the

fMRI scan, and include these time-series as null regressors or process them through signal-

adjustment algorithms like “RETROICOR” (Birn et al, 2006; Biswal et al, 1996; Glover et al,

2000; Shmueli et al, 2007). However, one alternative to measuring and regressing out

physiological confounds is to use non-neuronal signal derived from white matter and

cerebrospinal fluid as proxies for cardiac and respiratory noise. Averaged time series from voxels

of these regions are used as null regressors (Anderson et al, 2011; Hallquist et al, 2013; Jo et al,

2010; Power et al, 2012; Weissenbacher et al, 2009; Yan et al, 2013). The temporal derivatives

of these regressors are also included to maximize the impact of denoising (Fox et al, 2005;

Power et al, 2014; Satterthwaite et al, 2013). One limitation of this technique is that it does not

account for spatial variability in the time course of signal fluctuations in white matter and

cerebrospinal fluid (Caballero-Gaudes and Reynolds, 2017). Consequently, principal

components-based approaches like ‘aCompCor’ have been implemented to identify orthogonal

components of noise to use as null regressors; this technique allows for different timecourses of

signal fluctuation in different regions of cerebrospinal fluid and white matter, and has been

110

shown to be superior to simply regressing the mean time series of the entire white matter and

cerebrospinal fluid compartments (Behzadi et al, 2007a; Chai et al, 2012).

“True” resting-state signals that is associated with neuronal/synaptic acitivty is typically

considered to have the highest power at low frequencies, estimated at approximately 0.01 to 0.1

Hz (Cordes et al, 2001). Bandpass filtering is commonly applied to remove frequencies outside

of this band, including sources of noise like scanner drift (< 0.01 Hz), respiratory (0.1 to 0.5 Hz)

and cardiac confounds (0.6 to 1.2 Hz), and ‘aliasing’ of neuronal signal given the low sampling

rate of fMRI (Cordes et al, 2001). More recently, some have suggested that true neuronal sources

of resting-state fMRI signal may also lie below the 0.01 Hz lower limit and above the 0.1 Hz

upper limit (for a review see (Chen et al, 2017)); however, less is known about the nature and

quality of these signals, and how they relate to the physiological underpinnings of rs-fMRI

(Boubela et al, 2013; Boyacioglu et al, 2013; Chen and Glover, 2015; Niazy et al, 2011).

Furthermore, the width of a given bandpass filtering kernel can drastically alter the test-retest

reliability of functional connectivity measures (Liang et al, 2012) For this reason, most

preprocessing algorithms currently employ a bandpass filter in the range ~0.01-0.1 Hz.

One preprocessing step subject to considerable debate is whether to implement global

signal regression (GSR). GSR involves taking the whole-brain average signal intensity time

series and using it as a null regressor; this technique is intended to improve the spatial specificity

of functional connectivity (Fox et al, 2009) and increase functional connectivity strength in

certain analyses (Chang and Glover, 2009; Weissenbacher et al, 2009). However, there are a

number of critical issues associated with GSR. First, the global signal is posited to consist of a

spatially non-homogeneous combination of motion and respiratory noise (Power et al, 2017), and

neural signal (Van Dijk et al, 2010; Murphy et al, 2009). For example, one electrophysiology

study found that the global signal significantly correlated with the neural signal of gray matter

(Schölvinck et al, 2010). Second, some studies have found that GSR negatively impacts results

by reducing the reliability of findings (Liang et al, 2012), altering individual variability on

functional connectivity measures (Saad et al, 2012), and making datasets more susceptible to

motion (Jo et al, 2013). Finally, applying GSR essentially de-means signal intensity at the

voxelwise level; this means that GSR imposes a negative bias on the distribution of functional

connectivity strengths by forcing the data to be zero-centred, and therefore runs the risk of

creating spurious anticorrelations (Fox et al, 2009; Keller et al, 2013; Murphy et al, 2009). Even

111

more troublingly, GSR can bias functional connectivity (FC) and resting-state functional

connectivity (rsFC) correlations downwards in an unpredictable and non-uniform way (Saad et

al, 2012, 2013). FC anticorrelations do exist and are reproducible, even without GSR, so there is

much debate on which reported anticorrelations on rs-fMRI are of true neural origin, and which

are artefacts of GSR (Van Dijk et al, 2010; Keller et al, 2013; Shehzad et al, 2009). The current

consensus is that the appropriateness and usefulness of GSR depends on the dataset and the

scientific question (Murphy and Fox, 2017).

Statistical Methods to Characterize IBNs

Once preprocessing is complete, the next step is to characterize each rs-fMRI scan in

terms of the functional relationships between voxels or anatomical regions. There are two main

approaches that can be taken to at this stage of analysis: hypothesis-driven and data-driven.

Some techniques, discussed below, identify patterns of spontaneous BOLD fluctuation in a data-

driven approach, while others attempt to establish the functional connections of regions of

interest (ROIs) based on a priori hypotheses. More recent advances in rs-fMRI analysis harness

multivariate approaches combined with supervised machine learning, while others tackle the

dynamics of spontaneous BOLD fluctuations instead of static correlations to establish

connectivity. Furthermore, analyses can involve univariate or multivariate statistics. The

following section will briefly review these rs-fMRI analysis methodologies.

1.12.4.1 Seed-to-Voxel-Based rs-fMRI Analysis

The simplest and most widely employed technique to establish individual-level statistical

maps of whole-brain functional connectivity is seed-to-voxel-based. This approach begins by

first identifying an ROI relevant to the research question; this “seed” is typically a set of

contiguous voxels that represents an anatomically-specific region. Next, a mean time series is

extracted from the ROI. This step typically involves averaging the time series data of all voxels

within the seed, although the first principal component is sometimes used instead. The resultant

time series for the ROI is subsequently used as a regressor of interest in a general linear model fit

at every other voxel (or ROI) within the brain (for a review see (Lee et al, 2013)). Next, the

112

resultant whole-brain statistical map can be converted to a standardized parametric map, where

the statistic at each voxel represents the degree to which that voxel’s time series is correlated or

anticorrelated to that of the region of interest. The values of this map are typically reported as

“beta” weights (also known as contrast of parameter estimates), or z-scores.

Biswal and colleagues applied such a method in the first rs-fMRI study ever published; in

this study, the time series of the sensorimotor cortex were extracted and found to have a high

temporal correlation with other regions associated with motor function (Biswal et al, 1995).

Seed-based approaches have also been used to identify resting-state functional networks of the

brain. For example, Greicius and colleagues extracted the resting-state time series of the

posterior cingulate cortex, a node of the default mode network (described below), to establish the

existence of a functional network that is consistently active during rest (Greicius et al, 2003).

This resting-state technique has subsequently been applied to identify homologous IBNs in other

mammalian species: a similar topographic pattern for the DMN has been established in the

anesthetized macaque monkey (Vincent et al, 2007). Networks representing sensorimotor and

visual function have also been characterized in the rat using seed-based approaches (Kannurpatti

et al, 2008; Pawela et al, 2008).

A major challenge of seed-to-voxel-based approaches is seed selection as the location,

size, and shape of the ROI can influence which voxels are included in the average time series of

the seed (Margulies et al, 2010). This variation can therefore bias results; however, seed

selection can be optimized using data-driven techniques to identify temporally homogeneous and

anatomically contiguous sets of voxels (Golestani and Goodyear, 2011).

Another major limitation of seed-to-voxel-based approaches is that they can only suggest

bidirectional, correlation-based ‘functional connections’ rather than establishing causal,

directional links of information flow between regions. This means that the directionality of

functional connectivity between brain regions cannot be characterized by seed-to-voxel based rs-

fMRI alone. In contrast, ‘effective connectivity’ describes the directionality of functional

connectivity, most notably by employing techniques like Granger causality (Granger, 1969) that

seek correlations in timeseries that are present at successive temporal lags in one direction but

not in the opposite direction . Effective connectivity techniques have been successfully used to

characterize directions of information flow in resting-state networks like the default mode

113

network (Sridharan et al, 2008; Uddin et al, 2009), and to demonstrate the impact of age on

information flow in those resting-state networks (Stevens et al, 2009).

1.12.4.2 Independent Component Analysis and Dual Regression

Data-driven methodologies offer a way to establish whole-brain connectivity without the

need for an a priori region of interest. An advantage of data-driven approaches is that they make

no a priori assumptions about the spatial or temporal characteristics of brain networks.

Independent components analysis (ICA) (Jutten and Herault, 1991) is one popular method to

decompose a high-dimensional dataset into a lower-dimensional set of statistically independent

(but not necessarily fully orthogonal) spatial or temporal components. These independent

components can be linearly mixed to generate the original data (Beckmann, 2012; Kiviniemi et

al, 2003; van de Ven et al, 2004).

ICA is often applied to resting-state fMRI data because the resultant components

accurately model the spatial distributions of functional brain networks (Beckmann et al, 2005;

Damoiseaux et al, 2006). ICA also requires relatively minimal preprocessing, as some of the

resultant components can be deemed to represent non-neuronal noise based on their spatial

distribution (Margulies et al, 2010). Because ICA can efficiently separates ‘true’ functional

networks from spurious noise based on different spatially-distributed patterns of signal, ICA can

also be used for artifact detection (Feis et al, 2015; Perlbarg et al, 2007; Pruim et al, 2015), and

is occasionally used for task-based fMRI (Long et al, 2013).

‘Group ICA’ is often applied by concatenating the rs-fMRI image series across all

subjects in the dataset, to identify the spatial or temporal distributions of functional networks

emerging across all subjects (Calhoun et al, 2009; Du et al, 2016; Perlbarg et al, 2007; Schöpf et

al, 2010). The components and functional networks generated by ICA have high test-retest

reliability across multiple scan sessions (Chen et al, 2008; Zuo et al, 2010a) (Franco et al, 2009;

Meindl et al, 2010), and the spatially distributed functional networks generated by ICA are

consistent across participants (Damoiseaux et al, 2006). Once group-level ICA-derived

components have been identified, a technique known as dual regression is commonly used to

project the group-level spatial components as ‘whole-brain seeds’ back onto each individual

114

subject’s timeseries to obtain a set of subject-specific components for each individual, which

then allows for statistical comparisons across individual subjects in the dataset as a whole.

Briefly, dual regression involves two steps: first, a spatial regression step to estimate individual-

level temporal dynamics for each ICA component; and second, a temporal regression step to

estimate individual-level functional connectivity for each voxel and for each component (Smith

et al, 2014).

ICA also has a number of drawbacks. First, temporal ICA is computationally intensive,

and as a result this technique is less frequently used in comparison to spatial ICA (Calhoun et al,

2001). Second, the spatial components generated by different ICA algorithms often have poor

agreement (Zuo et al, 2010a). Additionally, ICA is sensitive to model order selection (i.e., the

number of components identified) (Abou-Elseoud et al, 2010; Abou Elseoud et al, 2011), such

that too few components will not reflect true neuronal networks, and too many components

might inappropriately subdivide IBNs too finely into apparently separate networks. Finally, a

major limitation of ICA is the somewhat subjective task of determining which components to

include for subsequent analysis, or to reject as an artifacts (Margulies et al, 2010). More recently,

machine learning approaches such as support vector machines have been trained on ICA

component data in the attempt to automate the identification of meaningful resting-state

functional networks (for example, (Wang and Li, 2015)).

1.12.4.3 Regional Homogeneity

The rs-fMRI analysis techniques reviewed thus far generally rely on correlations between

signals in at least two voxels or regions, meaning that they are less well-suited to measuring

brain activity in a voxelwise manner (as is possible, for example, with FDG-PET). Recently, a

series of newer methods have been devised that enable voxelwise analysis of rs-fMRI timeseries.

One way to characterize local fluctuations of a single voxel or brain region is by assessing the

synchrony of the local BOLD signal, or regional homogeneity (ReHo). ReHo calculates the

temporal synchrony between voxels of a given cluster (Zang et al, 2004). Kendall’s coefficient

concordance is used to measure the intra-cluster temporal homogeneity (Baumgartner et al,

1999). ReHo is associated with BOLD oscillations between 0.02-0.04 Hz in the cortex, and a

wide range of higher or lower frequency BOLD oscillations in limbic and subcortical regions

115

(Song et al, 2014), although other studies have reported that ReHo is not associated with cortical

gamma-band oscillations but local field potentials of lower frequency bands (unlike the BOLD

response, although IBNs can be replicated in multiple fMRI frequency bands (Gohel and Biswal,

2015)) (Canolty and Knight, 2010; Siegel et al, 2012; Wang et al, 2012b).

From an IBN perspective, the voxels within a given node/brain region of an IBN will

tend to have high local temporal synchrony. For example, nodes of the DMN and other networks

are reported to have high ReHo, meaning that in the case of the DMN, the voxels within the

mPFC, PCC and IPL have high within-cluster temporal concordance (Long et al, 2008). ReHo

measures have also been linked to variations in cognition and behaviour. For example, ReHo in

regions related to food reward, including the OFC, AI and DLPFC, significantly differs between

healthy controls exhibit unrestricted versus restricted eating (dieting) (Dong et al, 2014). ReHo

also differs between happy and unhappy healthy controls, such that unhappy individuals had

significantly decreased ReHo in DMN regions and increased ReHo in CEN and SN regions (Luo

et al, 2014). ReHo is sensitive to the magnitude of spatial smoothing, as a larger Gaussian

smoothing kernel increases the spatial extent of local synchrony (Zang et al, 2004). ReHo is also

sensitive to temporal bandpass filtering, as limbic or subcortical ReHo may be related to BOLD

oscillations outside of bandpass filters used for other rsFC analyses (Song et al, 2014).

1.12.4.4 Amplitude of Low Frequency Fluctuations

Another method to characterize local fluctuations in activity in a voxelwise manner is to

characterize low frequency oscillations of spontaneous BOLD activity in each voxel’s timeseries

in terms of its amplitude (Amplitude of Low Frequency Fluctuations; ALFF). ALFF is proposed

to represent the ‘regional intensity’ of resting-state signal, and is measured voxelwise as the

square power root of the power spectrum in the low-frequency (BOLD) range (Zang et al, 2007).

However, it is believed that ALFF is susceptible to non-neuronal artifacts, as the power of low

frequency BOLD fluctuations decreases with respiratory hypercapnia (Biswal et al, 1997; Zuo et

al, 2010a). An alternative, fractional ALFF (fALFF), assesses ALFF relative to the entire fMRI

frequency range, and is the ratio of ALFF to the power spectrum of the entire computed

frequency-band (Zou et al, 2008). ALFF and fALFF have also been used to characterize IBNs:

for example, healthy controls have high ALFF and fALFF in regions of the DMN, including the

116

PCC, precuneus and mPFC (Zou et al, 2008; Zuo et al, 2010a). However, the relationship

between ALFF/fALFF and more traditional and physiologically well-grounded measures of brain

activity, such as metabolic activity on FDG-PET, remains unclear at present.

1.12.4.5 Graph Theoretical Approaches

Graph theoretical techniques can be used to assess patterns of brain connectivity within

or between networks. These measures assess the functional connectivity structure of the whole

brain in a single analysis, rather than examining the functional connectivity profile of individual

brain networks independently (for a review see, (Bullmore and Sporns, 2009)). In such

approaches, complex brain networks are modeled as a graph, with voxels or regions represented

by ‘nodes,’ and the relationship (e.g., correlation/anticorrelation) between these nodes as ‘edges.’

Nodes of the brain can be represented by ROIs or individual voxels (Martuzzi et al, 2011). Edges

can be bidirectional or directional, meaning that they can convey information about the

directionality of information transfer. The edges can also be un-weighted (meaning that all the

edges in the graph are considered to represent a relationship of equivalent strength and

direction), or weighted (meaning that edges of different values represent different connectivity

strengths or directions) (for a review, see (Wang et al, 2010)). From this graph, metrics can be

calculated that describe the relationships between nodes and to gain insight into the organization

of the entire brain (Reijneveld et al, 2007; Stam and Reijneveld, 2007).

Graph theoretical measures in common use can include: how many edges a node has

(degree); the shortest distance between two nodes (path length); how sparsely or densely clusters

of nodes and edges are organized and how well-connected are the nodes of a network to each

other versus nodes of other networks (sparsity and modularity); how nodes cluster ‘locally’

versus globally (small-worldness); how ‘important’ a node within a network is, based on how

efficiently it is positioned to propagate information between nodes (nodal centrality/efficiency);

or how efficient a community of nodes is based on the ability of the network to propagate

information within nodes of the same network (local efficacy) or across many networks (global

efficiency) (for a review, see (Wang et al, 2010)). Graph theoretical measures are often applied

to resting-state fMRI data to describe the topological architecture of the brain’s resting-state

networks at the spatial and temporal scale (Palaniyappan et al, 2016; Wang et al, 2010). For

117

example, work by He and colleagues has characterized the topological architecture of resting-

state networks (He et al, 2009), such that resting-state networks have higher connectivity to

nodes within their network than they do to nodes outside of their network (high ‘modularity’).

Furthermore, some networks have nodes of high centrality, called hubs, that appear to be

necessary for efficient information flow (Latora and Marchiori, 2001; Margulies et al, 2010).

Deficits in the efficiency of nodes with high centrality has implications in both healthy aging and

neurodegenerative diseases (Achard and Bullmore, 2007; Buckner et al, 2009).

A potential limitation to graph theoretical approaches is how nodes are defined. Nodes

for graph theoretical analyses are often defined by a whole-brain atlas or by parcellation, and as a

result could introduce unwanted biases and substantially alter the network architecture of results

(Butts, 2009). Indeed, many studies have reported that different node selection alters graph

theoretical properties of resting-state networks substantially(Hayasaka and Laurienti, 2010;

Wang et al, 2009; Zalesky et al, 2010). Fortunately, a recent systematic review of test-retest

reliability reported that the reproducibility of several graph theoretical measures was good,

stating that of the 23 studies included in the systematic review, 16 showed reproducibility in the

‘excellent’ range (intra-class correlation > 0.74) (Welton et al, 2015).

1.12.4.6 Pattern Classification Analyses

Pattern classification analyses, including multi-voxel pattern analysis, harness

multivariate statistics to simultaneously take into account the connectivity patterns of multiple

voxels (for a review see (Haynes and Rees, 2006; Norman et al, 2006)). Pattern analyses

typically require the dataset to be divided into a ‘training’ and ‘test’ dataset, and a machine

learning algorithm is subsequently used to identify patterns (features) of brain connectivity that

accurately discriminate between cognitive states and/or a control condition, correlate with

continuous measures, or discriminate patient and control groups (Margulies et al, 2010). One

common type of machine learning algorithm used in multi-voxel pattern analysis is known as a

support vector machine; compared to univariate methods, SVM methods can sometimes achieve

higher prediction accuracy regarding the brain state, while being less sensitive to voxelwise noise

(Chen et al, 2006; LaConte et al, 2005; Mourão-Miranda et al, 2005). In the study of psychiatric

118

disorders, multi-voxel pattern analysis has been successfully used to accurately distinguish MDD

patients from healthy controls using distinct patterns of rsFC (Craddock et al, 2009).

1.12.4.7 Brain-Wide Association Analyses

Brain-wide association studies aim to characterize differences in resting-state networks

between phenotypes from whole-brain rsFC maps of individual voxels instead of ROIs, using an

analytic approach that is similar to that of genome-wide association studies (Cheng et al, 2015).

Given that the brain is comprised of tens of thousands of voxels, techniques that perform voxel-

wise statistics (including brain-wide association analysis) require robust multiple comparisons

correction to reduce the incidence of Type I (false-positive) error. A recent study addressed this

issue in detail, with the authors presenting a multiple comparisons correction tool for voxel-wise

functional connectivity using Random Field Theory (discussed in Section 1.12.6) (Gong et al,

2018). Similar to genome-wide association studies, brain-wide association analyses require a

large sample size to identify brain-wide rsFC variants (Cheng et al, 2015).

In the psychiatric context, brain-wide association analysis has been used to characterize

differences in rsFC in MDD relative to controls (Cheng et al, 2016). In this study, the authors

demonstrate the dissociable patterns of mOFC and lOFC differ in MDD relative to controls, and

that mOFC rsFC related to deficits in reward processing, while lOFC rsFC related to deficits in

non-reward processing (Cheng et al, 2016).

1.12.4.8 Dynamic Functional Connectivity

The aforementioned techniques to characterize brain connectivity (i.e. excluding ReHo,

ALFF and fALFF as typical applications of these techniques quantify regional dynamics) are

‘static’ measures of functional connectivity. This means that they ignore temporal fluctuations in

FC that occur during the scan. Such ‘static’ techniques might not be the optimal approach,

depending on the research question at hand, because the brain’s functional connectivity is

constantly changing. Thus, ‘dynamic’ FC approaches aim to quantify the temporal variability in

FC (Bassett et al, 2011; Cabral et al, 2011; Madhyastha et al, 2015).

119

A common approach of many dynamic FC techniques is know as the sliding window

technique. For this approach, the BOLD timeseries is subdivided into time segments

(“windows”) (Hindriks et al, 2016) of a pre-defined number of volumes. Functional connectivity

is then measured within the first window, often using a bivariate correlation, and then the

measurement is done over and over again, each time sequentially shifted by one TR at a time , to

compute a new time series representing the fluctuations in FC itself over the duration of the scan.

Other methods investigate time-frequency or phase synchronization between pairs of rs-fMRI

timeseries (Chang and Glover, 2010; Pan et al, 2013). As noted earlier, currently available

dynamic FC techniques are sensitive to bias related to preprocessing and residual physiological

noise, such that rigorous preprocessing algorithms are essential to avoid spurious observations

due to motion or physiological noise (Hutchison et al, 2013; Leonardi and Van De Ville, 2015).

A number of statistical strategies have been proposed to quantify dynamic FC, including

seed-based dynamic FC variability, ROI-to-ROI-based FC dynamic variability, and dynamic

ICA (as reviewed by (Preti et al, 2017). ROI-to-ROI-based dynamic FC is a method to establish

dynamic FC between a priori ROIs (Lindquist et al, 2014) and such variability has been recently

shown to reflect human behaviour. For example, Kucyi and Davis reported that FC variability

between nodes of the DMN was associated with the degree of mind wandering in healthy

individuals (Kucyi and Davis, 2014). Seed-based dynamic FC evaluates dynamic FC of a ROI to

every other voxel in the brain; such a technique been demonstrated to replicate IBNs generated

by static FC (Tagliazucchi et al, 2012, 2016). Finally, dynamic ICA incorporates information

generated from sliding windows and spatial-ICA to identify IBN FC variability, and was recently

shown to map DMN variability (Kiviniemi et al, 2011).

Regarding applications in the psychiatric research context, a dynamic FC approach was

recently used to characterize functional abnormalities in MDD relative to healthy controls. In this

study, Kaiser and colleagues demonstrated, using a seed-based approach, that MDD was

associated with less variable dynamic rsFC between the mPFC ROI and the parahippocampal

gyrus and more variable dynamic rsFC between the mPFC and the insula. Resting-state

variability between the mPFC and DLPFC also positively correlated with MDD severity (Kaiser

et al, 2016).

120

1.12.4.9 Inter-hemispheric Connectivity Analyses

Inter-hemispheric connectivity attempts to characterize the FC between corresponding

regions in each hemisphere. One such technique, voxel-mirrored homotopic connectivity,

assesses the FC of anatomically-corresponding brain regions by transforming one hemisphere of

the brain to be anatomically symmetrical and quantifying the rsFC of one voxel and its

contralateral, mirrored counterpart (Gee et al, 2011; Zuo et al, 2010b).

As an example of an application of this method to psychiatric research, Canna and

colleagues applied this technique in a recent study to examine differences in inter-hemispheric

connectivity in AN and BN relative to controls. In that study, AN exhibited less inter-

hemispheric connectivity between mirrored regions of the insula, while BN exhibited less inter-

hemispheric connectivity in the DLPFC and OFC (Canna et al, 2017).

Considerations for Selecting the Optimal rsFC Method

With a large toolbox of statistical analyses for rs-fMRI data, how should one select the

appropriate technique? As proposed by Dunlop and Downar (2017), a number of considerations

arise when selecting an rs-fMRI technique. First, the selected technique should be appropriate to

the research question; for example, in a study with a well-defined region of interest (e.g., a study

applying rTMS to a specific brain region), seed-to-voxel based approaches may be most

appropriate. Second, any inherent assumptions to the technique that could impact the validity of

findings should be discussed; for example, if the technique is highly sensitive to head motion,

and populations A and B show different degrees of head motion, then any reported differences

should be discussed in this context and compensatory analytical strategies should be deployed.

Third, the selected measure should reflect underlying neuronal physiology. For example, the

neurophysiological basis of BOLD power spectral analyses is not well-established, and so the

significance of findings using this measure may be difficult to interpret in a meaningful way.

Fourth, the technique chosen should be ideally a priori shown to have good construct validity

(i.e., the rs-fMRI technique should accurately measure brain function), as well as be reproducible

and reliable. For example, the reproducibility of graph theoretical measures has recently been

reported (Welton et al, 2015) and such measures have been related to neurophysiology via

electrophysiological recordings during seizure activity (Niso et al, 2015; Schmidt et al, 2014).

121

Finally, the technique should ideally demonstrate utility. In other words, the rs-fMRI technique

selected should be able to provide some novel or useful information about human behaviour, its

underlying neurobiological underpinnings, or about a particular clinical population. For example,

simply reporting that a given region shows weaker inter-hemispheric connectivity in population

A versus population B may not be particularly informative, unless that information can be

offered in a context that sheds light on pathophysiology, clinical symptoms, diagnostic or

prognostic markers, or potential treatment strategies.

Assessing Group-Level Comparisons Across Subjects

Once a metric has been chosen, the final stage of rs-fMRI analysis is to assess for group-

level differences between the individual subjects whose scans were acquired in the study.

Individual-level analyses typically generate a statistical parametric map that represents the FC or

other metric of interest for that subject. A second-level (group-level) general linear model is

typically used to generate voxel-wise statistics that could represent the more generalizable

average activation or functional connectivity seen across a group of individuals. Such voxel-wise

statistics could also be used to identify regional differences in activation or functional

connectivity between different groups (such as treatment responders and non-responders), or

correlations in activation to a continuous measure, such as participants’ percent improvement

from their baseline score. In more complex circumstances, a group-level general linear model

incorporating covariates of interest (or non-interest) can also be used to model and/or control for

the effects of confounding categorical or continuous variables (e.g. ANCOVA), such as gender

and sex. The resultant voxel-wise maps from these group-level analyses are commonly

represented as z-scores or beta-weights.

Whole-brain fMRI analysis techniques require some form of correction \to compensate

for the very large number of statistical tests performed across the parametric map as a whole, and

the resultant likelihood of false-positive (Type I) errors even when conservative p-value

thresholds are applied (Benjamini and Hochberg, 1995; Genovese et al, 2002). Such stringent

multiple comparisons correction is essential for the generalizability of fMRI results, because the

whole brain maps used in rs-fMRI typically consist of tens of thousands of voxels, and individual

122

statistical tests are performed at each voxel. At a Type I error rate of p < 0.05, 1 in 20 voxels

would be show apparent significance under the null hypothesis.

A number of methods can be used to correct for multiple comparisons, while at the same

minimizing Type II error (false negatives), including steps to account for the familywise error

(FWE; including Bonferroni correction and random field theory) or using the false discovery rate

(FDR) (Friston et al, 1994). The FWE rate is the probability of making a Type I error despite

accounting for the number of tests performed. The most conservative approach to minimizing

FWE is Bonferroni correction, in which a p-value is considered significant only if the value is

less than the base significance level (α) divided by the number of statistical tests performed.

Bonferroni correction is usually too conservative a multiple comparisons correction

technique for routine use in rs-fMRI, because tens of thousands of statistical tests are required to

compute voxelwise statistics, and spatial smoothing during preprocessing results in voxels and

clusters that are not statistically independent of each other. Consequently, random field theory

and other cluster-based approaches have been developed to address the fact that adjacent voxels

are not statistically independent of each other. The more alike voxels are, the less severe the

Type I error problem, and so correction for multiple comparisons can be performed with a less

stringent divisor. Random field theory corrects whole-brain analyses accounting for the

probability that volumetrically large activation clusters could have occurred by chance, based on

statistical significance of individual voxels (height threshold) and the size of the cluster in

question (extent), under the assumption that noise will follow a Gaussian distribution in space

(thus providing an additional rationale for applying Gaussian smoothing during preprocessing)

(Forman et al, 1995; Friston et al, 1994; Woo et al, 2014). There is an ongoing debate as to

whether random field theory adequately corrects for false positives, and at present, this issue is

not yet resolved (see (Eklund et al, 2016; Kessler et al, 2017; Slotnick, 2017)). Nonparametric

approaches to multiple comparisons correction (e.g., bootstrapping or permutation testing) have

been proposed as an alternative to Gaussian random field theory, thereby potentially offering a

less overly conservative multiple comparisons correction method that to control the false positive

rate (Hayasaka and Nichols, 2003; Nichols and Hayasaka, 2003; Nichols and Holmes, 2002).

Though more computationally intensive, non-parametric methods may become more practical

and even preferred as the cost of computation falls in future years.

123

Summary

In order to address the suboptimal response and remission rates associated with

conventional pharmacotherapy and psychotherapy for mental illness, there is a need for new

therapeutic interventions, as well as reliable, biologically-grounded tools to select the optimal

treatment for achieving remission. As a novel intervention, rTMS has recently emerged to offer

another chance at remission for those who have not responded to conventional psychotherapy

and pharmacotherapy. However, despite 20 years of research into rTMS treatments, response and

remission rates for rTMS still fall well below those of more invasive neuromodulation therapies

for treatment-resistant psychiatric illness, such as ECT. If our ultimate aim is to understand the

biological underpinnings of psychiatric disorders and thereby improve overall remission rates,

then a central research problem is to identify neurobiological predictors and correlates of

response for treatments such as rTMS. fMRI and rs-MRI are uniquely well-positioned to address

this problem, as these techniques are capable of characterizing the neurobiological effects of

rTMS over a course of treatment (the neural mechanisms or correlates of response), and also

characterizing neurobiological differences that might distinguish responders from non-

responders prior to treatment (the neural predictors of response).

When examined via fMRI, brain activity organizes into a set of fairly robustly

reproducible whole-brain functional networks or IBNs, each consisting of regions that are highly

functionally connected, meaning that they activate and deactivate in a correlated fashion, both

during task-evoked and during unconstrained ‘rest’ activity (Beckmann et al, 2005; Sporns,

2011). While the architecture, electrophysiology and brain-wide hierarchy of IBNs are not yet

fully understood, the neuroanatomy of IBNs is thought to reflect the underlying white matter

structure of the brain, and the correlated activity of their component regions is thought to be

sustained by high- and low-frequency electrophysiological oscillations in the coordinated activity

of neuronal populations. Fluctuations in IBN coherence have been linked to specific components

of human perception, cognition, and behaviour, including performance on cognitive control tasks

(Kelly et al, 2008) and impulsivity (Shannon et al, 2011).

Four IBNs may be of particular interest for the study of psychiatric disorders. First, the

DMN is related to internally-generated cognition and self-referential processing (Crittenden et al,

2015; Dixon et al, 2014), including mind wandering (Mason et al, 2007), hippocampal-

124

dependent autobiographical memory retrieval (Addis et al, 2007; Svoboda et al, 2006), spatial

navigation, and thinking about the thoughts of others (Spreng et al, 2009). Second, the SN is

related to cognitive control during goal-directed behaviour, such that nodes of the SN operate in

concert to identify, compare, and filter relevant external sensory information to support one

action choice over another (Pleger et al, 2006; Ploran et al, 2007), before projecting to motor and

premotor regions (often via the dACC) for action initiation (Lamichhane and Dhamala, 2015;

Landmann et al, 2007). Third, the CEN is associated with working memory and cognitive

control, and serves to identify changes in the environment that necessitate the inhibition of an

action or a behavioural adjustment to a different strategy (Dosenbach et al, 2008; Posner and

Petersen, 1990). Finally, the VMN contributes to reward processing, prediction, and reward-

based reversal learning (Boorman et al, 2013; Fellows, 2007; Kringelbach, 2005; Montague and

Berns, 2002), and in generating negative affective states and during emotional appraisal

(Johnstone et al, 2007; Wager et al, 2008). All four of these IBNs possess reciprocal

corticocortical connections between IBNs, as well as CSTC circuits to prepare and implement

planned, goal-directed behaviour (Haber, 2016).

SN dysfunction and its associated cognitive control deficits appear to be a central and

transdiagnostic abnormality across many psychiatric disorders (McTeague et al, 2016). Three

large neuroimaging studies have reported that structural abnormalities of the dACC and AI are

common to many diverse psychiatric disorders (Chang et al, 2018; Goodkind et al, 2015; Wise et

al, 2017). Alongside structural abnormalities of the SN, functional abnormalities of the SN

appear to be a transdiagnostic feature of many forms of mental illness, although there is

substantial heterogeneity in terms of whether this abnormality involves higher versus lower

activity or connectivity of the SN, across individuals and disorders (Kaczkurkin et al, 2017;

McTeague et al, 2017; Shanmugan et al, 2016). In light of this emerging evidence, multiple

‘unifying’ theories of SN-derived psychopathology have been proposed, generally agreeing on

the idea that SN dysfunction may lead to aberrant activity in other networks, resulting in

difficulties integrating salient external or physiological events, as well as difficulties accessing

networks for attention and working memory or self-referential/autobiographical thought. These

structural, functional, and behavioural deficits affecting cognitive control are consequently

apparent across a diverse set clinical disorders, including MDD, OCD and ED.

125

Given the transdiagnostic involvement of the SN in psychiatric illness, therapeutic

techqniues targeting the SN have the potential to achieve clinical benefits across multiple

categories of illness in psychiatry. rTMS is uniquely well-positioned for use as a focal

therapeutic intervention to remodel activity in the SN, as it is thought to induce its therapeutic

effects via synaptic plasticity and neuronal excitability at the local scale, and via modifications to

downstream cortico-cortical, cortico-limbic or cortico-striatal connections at the macroscopic,

whole-brain scale of IBNs. By altering the activity of one node of an IBN, rTMS may be capable

of achieving therapeutic effects via anatomical and functional connections with other nodes of

the same IBN, or connections with related IBNs.

As a network-probe, TMS can focally alter the activity of cortical brain regions to

causally test brain structure and function relationships. TMS and rTMS have been shown to

modulate monosynaptic downstream targets, allowing for the study of not only the stimulated

brain region, but also the network interconnected to that region (Ruff et al, 2009). rTMS

targeting the dmPFC can therefore be expected to modulate the activity of connected regions

such as the dACC and the SN circuitry in general. In healthy controls, excitatory dmPFC-rTMS,

reduces dACC rCBF, releases striatal dopamine in areas connected to medial frontal regions, and

improves cognitive control on behavioural measures (Cho et al, 2015), and (Hayward et al,

2007), thus demonstrating engagement of the target network and its functions. Conversely,

disrupting dACC function using single-pulse TMS or inhibitory rTMS impairs conflict

monitoring and increases errors during a cognitive control task (Duque et al, 2012; Taylor et al,

2007). Such preclinical findings provide a rational for therapeutic use of dmPFC-rTMS in

clinical populations with SN dysfunction.

To date, however, the dmPFC is not the best-studied rTMS target. Instead, the DLPFC is

the most common therapeutic target of rTMS, although this neurostimulatory target may be less

clearly beneficial for a wide range of clinical disorders other than MDD. DLPFC-rTMS does

appear to be most successful for TRD, with current response and remission rates between 50-

55% and 30-35%, respectively (Brunelin et al, 2014; Fitzgerald et al, 2011). However, current

rTMS protocols are not superior to ECT for response or remission, and a number of meta-

analyses have found that the clinical efficacy of rTMS for TRD is inferior to ECT (Micallef-

Trigona, 2014; Xie et al, 2013). For patients with OCD, randomized sham-controlled trials of 1

Hz right DLPFC-rTMS (Prasko et al, 2006) and 20 Hz left DLPFC-rTMS (Sachdev et al, 2007)

126

have reported that active stimulation is not superior to placebo. The most recent randomized

double-blind sham-controlled trial of 10 Hz left DLPFC-rTMS in 47 BN patients reported no

significant improvement in binge-eating and purging relative to placebo (Gay et al, 2016).

dmPFC-rTMS might be a promising alternative to DLPFC-rTMS for individuals with

diverse or complex clinical symptomatology for three reasons. First, numerous meta-analyses

have reported that behavioural, structural and functional deficits of the dACC and SN proper are

present across patients with a wide range of neuropsychiatric symptoms (Chang et al, 2018;

Goodkind et al, 2015; Kaczkurkin et al, 2017; McTeague et al, 2017; Shanmugan et al, 2016;

Snyder et al, 2015; Wise et al, 2017). Second, TMS and rTMS studies targeting the dmPFC have

shown to modulate behaviour on cognitive control tasks related to SN function (Cho et al, 2015;

Duque et al, 2012; Taylor et al, 2007) and alter brain function in the dACC (Hayward et al,

2007) and striatum (Cho et al, 2015). Third, early studies of dmPFC-rTMS have shown that

response and remission rates for TRD are comparable to that of DLPFC (Bakker et al, 2015).

One early study also reported that dmPFC-rTMS also appears to be beneficial for BN, suggesting

that this neurostimulatory target might be a better therapeutic site for a wider range of symptoms

(Downar et al, 2012). Nonetheless, dmPFC-rTMS still appears to be ineffective in a substantial

percentage of patients, leaving much room to further improve the rates of remission.

One potentially fruitful strategy for improving remission rates is to identify which

patients are most likely to benefit from treatment, using reliable biological predictors, or

predictive ‘biomarkers.’ A reliable method for distinguishing treatment responders from non-

responders prior to treatment may spare patients from having to undergo logistically onerous but

futile courses of treatment. Furthermore, if we consider dmPFC-rTMS as a probe as well as an

intervention, then biomarkers that distinguish responders from nonresponders may also shed

light on the heterogeneity of illness within the diagnostic entities of MDD and other psychiatric

disorders. This knowledge, in turn, may help to clarify the various subpopulations that are

suspected to exist in the heterogeneity of nominally unitary diagnostic entities such as MDD. To

date, however, there has been limited progress in identifying reliable predictive biomarkers for

rTMS, or indeed any kind of intervention, for MDD or any other psychiatric disorder.

Identifying the biological mechanisms by which an individual achieves response or

remission from a psychiatric disorder is also valuable knowledge for both scientific and

127

therapeutic reasons. From a scientific standpoint, it is helpful to clarify whether abnormal IBN

structure or function are state-dependent markers or are stable trait-like features that persist

during response or remission. On a related point, understanding the biological mechanism of

treatment remission could clarify the biological mechanisms in which an individual will relapse,

possibly leading to robust biological predictors of treatment relapse. Additionally, from a

therapeutic and translational standpoint, identification of neural markers that change with

successful by not unsuccessful treatment could help lead to new treatments that might be more

effective in treatment non-responders in future.

Grounded in the broader effort to improve remission rates for psychiatric illnesses, a

central problem in current research is to identify reliable biologically-based predictors and

correlates of treatment outcome for our available interventions. It is this central problem that

informs the specific aims, hypotheses, and methods of the studies described in this thesis.

128

Rationale, Specific Aims & Hypotheses

General Aims & Approach The goal of this thesis is to identify transdiagnostic predictors and mechanisms of

treatment response to rTMS over the dorsomedial prefrontal cortex across three categories of

psychiatric disorders. These three aims will be pursued through a set of three studies, in TRD,

OCD, and binge-purge behaviours in eating disorders. In each case, functional MRI and

standard clinical outcome scales will be used to assess the therapeutic effects of dmPFC-rTMS.

In addition, for each of these three patient populations, in addition to the assessment of

therapeutic effects, seed-to-voxel analyses will be conducted on rs-fMRI data acquired pre- and

post-treatment in order to identify neural predictors and correlates of treatment response, to

assess their similarities and differences across diagnostic categories.

Aim 1. To assess the clinical efficacy of dmPFC-rTMS as a treatment for TRD. A

secondary aim is to conduct preliminary assessments of therapeutic effects of dmPFC-

rTMS in OCD and eating disorders.

Approach: Clinical efficacy will be tested using a triple-blind, sham-controlled trial of dmPFC-

rTMS as a treatment for treatment-refractory MDD, and will be assessed using patient-rated

questionnaires and assessor-conducted interviews that quantify MDD symptom severity.

Additionally, two open-label trials of dmPFC-rTMS will be used to test the clinical efficacy of

this intervention for treatment-resistant OCD and BN/AN-BP.

Aim 2. To identify differences in pre-treatment resting-state functional connectivity

between dmPFC-rTMS responders and non-responders across 3 categories of mental

illness (TRD, OCD, and eating disorders).

Approach: Baseline rs-fMRI will be analyzed using a seed-to-voxel based approach to establish

differences in dmPFC functional connectivity between treatment responders and non-responders,

and regions whose baseline rsFC correlate with symptom improvement.

129

Aim 3. To identify pre- to post-treatment changes in resting-state functional connectivity

that accompany dmPFC-rTMS response for TRD, OCD, and eating disorders.

Approach: rs-fMRI will be conducted and the data analyzed at the pre- and post-treatment time

points. A seed-to-voxel based approach will be used to either identify differences in rsFC

change between dmPFC-rTMS responders and non-responders, and to identify regions whose

rsFC change correlates with symptom improvement.

Rationale for a Seed-to-Voxel-Based rsFC Approach As noted above, a variety of methods can be used to analyze resting-state fMRI data.

This thesis will use a seed-to-voxel-based rs-fMRI approach, with the general aim of identifying

transdiagnostic predictors of, and mechanisms of response to, dmPFC-rTMS across three

diagnostic categories of psychiatric illness. In light of the desiderata outlined in Section 1.12.5

for selection of interpretable, physiologically grounded, and clinically meaningful rs-fMRI

metrics (Dunlop and Downar, 2017), the following rationale is provided for using seed-to-voxel-

based analyses as the most appropriate rs-fMRI analysis technique to adequately address the

general aims of this thesis:

1. IBNs identified via seed-to-voxel-based analyses are physiologically well-grounded.

In other words, such rsFC patterns are established to be related to brain structure,

neuronal activity, and electrophysiology.

a. IBNs and rsFC generated by rs-fMRI relate to brain structure (Section 1.2.3).

Task-evoked and rs-FC are correlated to white matter connectivity (Greicius et al,

2009; van den Heuvel et al, 2009), and the rsFC strength between two brain

regions is correlated with the integrity of the white matter tracts between them

(Hermundstad et al, 2013). Some rsFC may be related to indirect structural

connections via an intermediary synapse, but nonetheless such rsFC reflects

anatomical pathways or key circuitry in the brain (Baek et al, 2016; Honey et al,

2009).

130

b. IBNs and rsFC generated by rs-fMRI relate to electrophysiology (Sections 1.2.5

and 1.12.2). Resting-state fMRI activity is also associated with gamma power (50-

100 Hz) (Fox and Raichle, 2007; Schölvinck et al, 2010, 2013) and infraslow

activity (Lu et al, 2016). The rsFC strength between two regions correlates with

their spontaneous gamma synchrony (Nir et al, 2008), and spatial patterns of

gamma oscillations differentiate between anticorrelated networks (Keller et al,

2013). Spontaneous gamma fluctuations in monkey visual cortex, as measured by

LFP, correlate with spontaneous fluctuations of rs-fMRI signal (Shmuel and

Leopold, 2008). Spontaneous fluctuations in infraslow LFPs using EEG and MEG

have been shown to be related to spontaneous fluctuations in rs-fMRI (Hiltunen et

al, 2014).

c. IBNs and rsFC generated by rs-fMRI relate to neuronal activity (Sections 1.2.5

and 1.12.2). rsFC is generated by synchronous increases in gamma power that

reflect increased local spiking (Logothetis et al, 2001; Manning et al, 2009;

Mukamel et al, 2005; Nir et al, 2007; Ray and Maunsell, 2011). Spontaneous

changes in cerebral blood flow are tightly coupled with infraslow ECoG

fluctuations (Golanov et al, 1994), and infraslow activity is attenuated by voltage-

gated sodium channel or glutamate receptor antagonists (Chan et al, 2015).

However, widespread spontaneous BOLD activity has been correlated with

recorded activity from single neurons in some (Shmuel and Leopold, 2008) but

not all studies (Schölvinck et al, 2013), and it is likely the case that all measures

of rs-fMRI are influenced by non-neuronal factors (Birn et al, 2006; Shmueli et

al, 2007; Wise et al, 2004) that contribute to variability in this signal.

2. rsFC maps generated via seed-to-voxel-based analyses provide meaningful

information about human behaviour. Section 1.2.4 provides evidence that rsFC and its

IBNs reflect aspects of human behaviour. First, rsFC-generated IBNs closely resemble

task-evoked IBNs (Krienen et al, 2014). Second, variability in the functional coupling of

IBNs has been related to behavioural variability, as measured by response time variability

during a Flanker task (Kelly et al, 2008). Such inter-individual variability has also been

related to behavioural variability in other tasks; for example, stronger within-network

IBN coupling in cortical regions responsible for motor and language function were

131

associated with higher performance on a reading task (Koyama et al, 2011), and higher

rates of behavioural impulsivity were related to the coupling of attentional and executive

control regions to premotor regions (Shannon et al, 2011). Therefore, differences in seed-

to-voxel-based rsFC maps are likely to be associated with the same cognitive or

behavioural processes that are associated with similar IBNs identified during task-based

fMRI studies.

3. rsFC maps identified via seed-to-voxel-based analyses provide meaningful

information about psychiatric disorders (Sections 1.3, 1.4, 1.5 and 1.6). Advances in

the understanding of corticocortical and corticostriatal rsFC across many disorders have

led to transdiagnostic, network-based models of the domains of pathology that underlie

psychiatric disorders. In other words, abnormalities in behaviour and clinical symptoms

are being now described in terms of dysfunction of specific IBNs, as opposed to more

general psychosocial constructs or biochemical imbalances. Section 1.3.2 highlights three

extant theories of such rsFC deficits extending across conventionally defined diagnostic

disorders, and all three specifically highlight the SN and its corticocortical/corticostriatal

connectivity as a key transdiagnostic substrate of illness (Marsh et al, 2009a; McTeague

et al, 2016; Menon, 2011; Menon and Uddin, 2010). Dysfunction in these transdiagnostic

IBNs is also linked to deficits in behaviour, cognition, and clinical symptom severity, and

has been shown to predict, and/or change with, response to clinical intervention (Sections

1.10 and 1.11).

4. rsFC maps identified via seed-to-voxel-based analyses are replicable. rsFC maps

generated from seed-based connectivity analyses show a close resemblance to maps

generated via other rs-fMRI techniques, such as ICA (Hale et al, 2015) and offer good

reproducibility (Franco et al, 2013). The neuroanatomical features of IBNs are also

highly replicable, with general topology congruent both across subjects and across

analysis techniques (Fox et al, 2005; Greicius et al, 2003).

5. Assumptions of seed-to-voxel-based rs-fMRI analysis are addressed. A major

assumption of seed-based connectivity analyses is ROI selection, as ROI location and

size can affect rsFC maps and therefore bias results (Margulies et al, 2010). A recent

systematic comparison of rsFC ROI methods demonstrated that there currently is no

132

optimal method to define ROIs for seed-based rsFC analyses, although ROIs derived

from resting-state data appear to more reliably capture rsFC compared to ROIs derived

from structural connectivity or cytoarchitecture (Arslan et al, 2018). Consequently, the

ROIs used in this thesis are derived from cortical and striatal parcellations of resting-state

fMRI connectivity (Craddock et al, 2012a; Di Martino et al, 2008). Another assumption

of seed-based analyses is that ROIs should ideally be defined a priori to avoid circularity

of analysis methods, and to minimize the impact of multiple comparisons across large

numbers of ROIs. Fortunately, in the setting of rTMS, both the stimulation site in dmPFC

(Avissar et al, 2017; Cárdenas-Morales et al, 2014), and the network of regions activated

by rTMS at the specific site (Cho et al, 2015; Dowdle et al, 2018; Hanlon et al, 2013;

Liston et al, 2014) are known in advance from previous work, and can therefore be

defined a priori as ROIs for analytical purposes. Similarly, ROI selection from DBS

targets is another evidence-driven technique that results in robust results (Fox et al,

2014), and this approach is consequently used to generate exploratory seeds in Study II.

6. rsFC observations from seed-to-voxel-based approaches in this thesis will generate

novel information about neurobiological processes, therapeutic interventions, and

clinical populations. Prior to the execution of the studies in this thesis, the existing

literature has offered very little evidence about the neurobiological predictors of response

to novel and conventional interventions for OCD and eating disorders (Section 1.11.2).

To date, there are no functional predictors of response for either conventional

interventions or rTMS in the setting of OCD and BN. Furthermore, there are no fMRI or

structural MRI studies that report correlates/mechanisms of response to any intervention

that correlates to binge/purge improvements in BN (Section 1.11.2), and very little

evidence on the neurobiological mechanisms of novel neurostimulatory interventions. For

TRD, dmPFC-rTMS is a relatively novel intervention, and consequently only two

previous studies have reported rsFC predictors and mechanisms of response to this

intervention in TRD (Downar et al, 2014; Salomons et al, 2014), both of which were

conducted under open-label rather than sham-controlled conditions and neither of which

can therefore inform us about whether observed changes were treatment-specific rather

than representative of nonspecific effects or regression-to-mean over time in the treated

population. Therefore, these studies to identify seed-based rsFC predictors and correlates

133

of response to dmPFC-rTMS in all three of these clinical populations will provide novel

information of potential importance for both basic-science and translational purposes.

7. rsFC observations from seed-to-voxel-based approaches in this thesis will

demonstrate useful information about neurobiological processes, therapeutic

interventions, and clinical populations.

a. Identifying baseline seed-based rsFC maps that distinguish between rTMS

responders and non-responders has utility for three reasons. First, reliable

methods for differentiating treatment responders from non-responders could spare

patients from undergoing futile treatments, and enable more individualized

approaches to treatment selection. A significant time commitment is required of

rTMS patients, especially since rTMS requires patients to receive treatment at a

clinic or hospital daily over many weeks, so avoiding futile treatments is an

important clinical objective. Second, identifying reliable biomarkers of treatment

response may yield insights into how we may develop new treatments for those

who do not respond to dmPFC-rTMS. For example, a fMRI-based rTMS

biomarker may reveal brain regions that differ between responders and non-

responders to a given treatment. Brain regions that are abnormally hyper- or

hypoactive in a group of non-responders may be promising candidate targets for

neurostimulation in future studies. Finally, biomarkers distinguishing individual

differences in response to treatment may help with the broader goal of

understanding more clearly the clinical and neurbiological heterogeneity that

underlies MDD and other psychiatric disorders; such biomarkers could also prove

informative about the etiology and pathophysiology of these disorders.

b. Identifying the biological mechanisms by which an individual achieves response

or remission from a psychiatric disorder is useful knowledge for both scientific

and therapeutic reasons. First, there is limited available evidence to date regarding

the biological effects of DLPFC- and dmPFC-rTMS in healthy controls and

psychiatric populations. Identifying changes in seed-based rsFC in patients treated

with excitatory dmPFC-rTMS may clarify the biological mechanisms of this

intervention, and may help us to understand the heterogeneity in its effects. Such

134

knowledge could help to develop more consistently effective rTMS protocols, or

help to develop new interventions for those who do not respond to dmPFC-rTMS.

Second, it is unknown whether alterations in rsFC are state-dependent markers of

current mood state or stable, trait-like features that persist during response or

remission. Thus, understanding the mechanisms of treatment response will allow

for a better understanding of which structural or functional features normalize as

MDD symptoms remit, and which persist. Finally, understanding the biological

mechanisms of treatment remission could help to clarify the biological

mechanisms by which an individual patient suffers relapse, possibly leading to

clinically useful biomarkers for predicting treatment relapse.

Study I. Baseline Predictors and Mechanisms of dmPFC-rTMS response in Anorexia and Bulimia Nervosa

Rationale of Study I AN and BN carry a disproportionately high burden of illness, and a high mortality rate.

Approximately 10% of AN sufferers will die within 10 years of disease onset (Sullivan, 1995).

According to a recent meta-analysis, the overall standard mortality ratio for AN is 5.86,

substantially higher than schizophrenia (2.8), MDD (1.6) and bipolar disorder (2.1) (Arcelus et

al, 2011). To date, there is no single well-established treatment for eating disorders; treatment or

treatment-combinations vary according to eating disorder type, severity, and clinician experience

(for a review, see (Halmi, 2005)). Further compounding this issue, only a third of AN patients

will recover within 4 years of disease onset (Berkman et al, 2007; Steinhausen, 2002), and

clinical studies of BN report relapse rates between 25-63% (Grilo et al, 2012; Halmi et al, 2002;

Herzog et al, 1999; Keel et al, 2005; Olmsted et al, 2005). Consequently, a substantial

proportion of patients are not adequately served by current treatments.

To date, previously published studies of DLPFC-rTMS in medically-unresponsive BN

have reported modest improvements (Section 1.8.3). However, there is one recently published

case of a patient with TRD and comorbid BN who unexpectedly remitted from binge eating and

purging following a course of 10 Hz dmPFC-rTMS for comorbid TRD (Downar et al, 2012). It is

possible that dmPFC-rTMS improved binge eating and purging symptoms via normalizing

135

effects on structural and functional abnormalities that are common to both TRD and eating

disorders, centring on the dmPFC and adjacent dACC (Sections 1.3.2, 1.4.56 and 1.6.6); a

follow-on case series in the same population, this time incorporating pre- and post-treatment

neuroimaging, would be required to clarify the mechanism of action.

There are also no studies to date that report pre-treatment neurobiological predictors of

binge eating or purge improvement, and no studies that report pre- to post-treatment changes in

neurobiological measures that accompany improvements in eating eating or purging (Sections

1.9.3 and 1.10.2). However, in the setting of TRD, two recent rs-fMRI publications have

reported baseline differences between dmPFC-rTMS responders and non-responders and

changes that correspond with clinical improvement in TRD symptoms (Section 1.9.4). These

studies reported that higher functional connectivity between the dmPFC and sgACC was

associated with better response to dmPFC-rTMS, and that TRD responders displayed lower

baseline dmPFC resting-state functional connectivity with the thalamus compared to rTMS non-

responders (Salomons et al, 2014). Given the common elements among the structural and

functional brain abnormalities seen for TRD and eating disorders, it could be the case that similar

rs-fMRI predictors and mechanisms of response would be associated with successful dmPFC-

rTMS treatment in BN and TRD. Consequently, the general aim of Study I is to establish

baseline predictors of, and neural correlates/mechanisms of, improvements in binge-eating and

purging behaviors with dmPFC-rTMS treatment, in BN and AN-BP.

Specific Aims of Study I 1. To determine whether 20-30 sessions of open-label 10 Hz dmPFC-rTMS improves binge

eating and purging in patients with AN-BP or BN.

2. To identify patterns of baseline dmPFC resting-state functional connectivity that

significantly differ between AN-BP/BN dmPFC-rTMS responders and non-responders.

3. To identify pre- to post-treatment changes in resting-state functional connectivity that

correlate with binge eating and purging frequency in patients with AN-BP and BN.

136

Hypotheses of Study I 1. Mean binge eating and purging frequency will significantly improve in AN-BP/BN

patients who receive 20-30 sessions of open-label 10 Hz dmPFC-rTMS.

2. Lower baseline dmPFC-CSTC and higher baseline dmPFC-sgACC resting-state

functional connectivity will be associated with better response to 10 Hz dmPFC-rTMS.

3. Increases in pre- to post-treatment CSTC-dmPFC rsFC after 10 Hz dmPFC-rTMS will

correlate with improvements in binge-eating and purge frequency.

Study II. Baseline Predictors and Mechanisms of dmPFC-rTMS response in Obsessive-Compulsive Disorder

Study Rationale OCD is a particularly challenging psychiatric disorder from a therapeutic perspective. It

is estimated that between 30-60% of OCD patients do not respond to conventional behavioural or

pharmacological treatments (Pallanti et al, 2002; Simpson et al, 2006), and even those who

initially respond to conventional pharmaco- and psychotherapies show persistent functional

impairment (Steketee, 1997), and substantial future relapse rates (Simpson et al, 2005). Novel

therapeutics are therefore critical to address the substantial population of OCD sufferers for

whom conventional interventions have limited efficacy.

Recent randomized, sham-controlled trials have demonstrated that DBS of the ventral

striatum is an effective treatment for medically-refractive OCD (Section 1.8.2). In these DBS

cases, symptomatic improvement was correlated with decreases of abnormal dACC frontostriatal

hyperconnectivity (Figee et al, 2013) (Section 1.10.2). However, DBS is the least accepted and

least preferred novel treatment compared to other novel behavioural therapies, possibly reflecting

a lack of evidence for this intervention and indication, in addition to its invasive nature and the

requirement for a costly device as well as specialized neurosurgical expertise for successful

implantation (Patel et al, 2017).

rTMS over the dACC is a more economical and less invasive intervention that can

modulate dACC function and frontostriatal connectivity (Sections 1.7.2). Previously published

studies have shown that other medial frontal sites may be effective rTMS targets in medically-

137

refractory OCD (Mantovani et al, 2010) (Section 1.8.2). However, high-frequency rTMS over

the dmPFC and dACC as a treatment for OCD has never before been studied, despite structural

and functional evidence that this region and its associated IBN is abnormal in this population

(Section 1.5.6). Furthermore, there are also no studies that report pre-treatment neurobiological

predictors of OCD symptom improvement to rTMS, and no studies that report pre- to post-rTMS

changes in neurobiological measures that accompany improvements in OCD symptom severity

(Sections 1.9.3 and 1.10.2). Due to the hypothesized neurobiological effects of dmPFC-rTMS on

frontostriatal connectivity, it could be the case that similar rs-fMRI predictors and mechanisms

of VS-DBS response would be associated with successful dmPFC-rTMS treatment in OCD.

Consequently, the general aim of Study II is to establish baseline predictors of, and neural

correlates/mechanisms of, OCD symptom improvements following 20-30 sessions of 10 Hz

dmPFC-rTMS.

Specific Aims of Study II 1. To determine whether 20-30 sessions of open-label 10 Hz dmPFC-rTMS improves the

severity of obsessive thoughts and compulsive behaviours in OCD.


significant differ between OCD dmPFC-rTMS responders and non-responders.


correlate with improvements in OCD symptom severity.

Hypotheses of Study II 1. The severity of obsessive thoughts and compulsive behaviours in OCD patients will

significantly improve in response to 20-30 sessions of open-label 10 Hz dmPFC-rTMS.

2. As with DBS, OCD patients with higher dmPFC-CSTC resting-state functional

connectivity will show better response to open-label 10 Hz dmPFC-rTMS.

3. As with DBS, OCD patients who show decreases in pre- to post-treatment dmPFC-CSTC

resting-state functional connectivity after dmPFC-rTMS will show more improvement in

OCD symptom severity.

138

Study III: Baseline Predictors and Mechanisms of High- and Low-Frequency dmPFC-rTMS in TRD under triple-blind sham controlled settings

Rationale of Study III

Three open-label case series from our research group have previous reported

improvements in TRD symptom severity following 20-30 sessions of once daily 10 Hz dmPFC-

rTMS (Bakker et al, 2015; Downar et al, 2014; Salomons et al, 2014). Only one other

independent study of dmPFC-rTMS in TRD currently exists. A preliminary randomized control

trial reported significant improvement following dmPFC-rTMS;however this study included a

very small number of patients (n = 40) who received only 15 sessions of rTMS in an inpatient

setting with concurrent medication regimen adjustments(Kreuzer et al, 2015). Given the modest

sample size and abbreviated dmPFC-rTMS treatment course (Section 1.8.1), further investigation

is warranted to determine whether the antidepressant effects of this intervention are superior to

those seen with placebo dmPFC-rTMS.

Two of the open-label case series of dmPFC-rTMS for TRD cited above (Downar et al,

2014; Salomons et al, 2014) examined rs-fMRI differences at baseline that predicted treatment

response. Baseline resting-state neuroimaging findings revealed differences in functional

connectivity between responders and non-responders to dmPFC-rTMS in TRD (Section 1.9.4).

Changes in dmPFC resting-state functional connectivity correlated with response to open-label

dmPFC-rTMS (Section 1.10.2). Generally, connectivity of the stimulation site with frontostriatal

CSTC circuitry and cortico-cortical VMN regions at baseline predicted response to dmPFC-

rTMS; connectivity change from pre- to post-treatment scans in these regions also correlated

with response to dmPFC-rTMS in TRD. However, the therapeutic effects of dmPFC-rTMS and

its rs-fMRI correlates were highly heterogeneous across individuals. To address this

heterogeneity and improve outcomes, rTMS protocols that theoretically elicit different cortical

responses could be compared in terms of both clinical efficacy and effects on neural activity;

such comparisons could help to identify mechanisms of response, and to optimize treatment

parameters (Section 1.8.4). Low-frequency (1 Hz) and high-frequency (20 Hz) protocols have

both been used in numerous studies of rTMS in TRD, comparing treatment outcomes in order to

139

optimize the treatment protocol at a given stimulation site, and determine mechanisms of

antidepressant response (e.g., Kim et al, 2014). In light of these findings, the general aim of

Study III is to establish baseline predictors of and mechanisms of improvements in TRD

symptom severity following 30 sessions of 20 Hz dmPFC-rTMS under placebo-controlled

conditions.

Specific Aims of Study III 1. To determine whether the clinical efficacy of active 20 Hz dmPFC-rTMS in TRD is

superior to that of 1 Hz active dmPFC and sham dmPFC-rTMS.

2. To replicate previously identified patterns of baseline dmPFC resting-state functional

connectivity that significantly differ between TRD dmPFC-rTMS responders and non-

responders.

3. To replicate previously identified patterns of pre- to post-treatment changes in resting-

state functional connectivity that correlate with improvements in TRD severity.

Hypotheses of Study III 1. Improvements in depression symptoms with 20 Hz dmPFC-rTMS will be significantly

higher than that achieved with 1 Hz dmPFC-rTMS or placebo dmPFC-rTMS.

2. TRD patients with lower dmPFC-CSTC and higher dmPFC-VMPFC resting-state

functional connectivity will show greater symptom improvement with active 20 Hz

dmPFC-rTMS, but not with active 1 Hz or placebo dmPFC-rTMS.

3. TRD patients showing greater increases in pre- to post-treatment dmPFC-CSTC resting-

state functional connectivity will show greater improvement in clinical symptoms, after

active 20 Hz dmPFC-rTMS but not after active 1 Hz or placebo dmPFC-rTMS.

140

General Methods

Project Overview This project was conducted with three independent cohorts of patients (n = 28 for Study I; n = 20

for Study II; and n = 123 for Study III), and two independent cohorts of healthy individuals (n =

40 for study II and n = 41 for Study III).

All study patients (n = 151) who met study inclusion and exclusion criteria (described in detail

below) underwent the following general experimental protocol:

1. Pre-rTMS and Post-rTMS Brain Imaging Session: This session included acquisition

of structural MRI and rs-fMRI scans with parameters detailed below. Patients’ scans

were acquired 1-week prior to a course of dmPFC-rTMS. For Study III, all patients also

completed a baseline battery of clinical questionnaires assessing symptomatology beyond

core TRD symptoms.

2. rTMS Treatment: All patients received 20-30 sessions of dmPFC-rTMS. All patients

recruited to Studies I and II received 20-30 sessions of once-daily, open-label dmPFC-

rTMS. All patients recruited to Study III were randomized assigned to 30 twice-daily

sessions (15 dayss) of either active 20 Hz dmPFC-rTMS, active 1 Hz dmPFC-rTMS, or

placebo dmPFC-rTMS using a sham rTMS coil customized for dmPFC stimulation.

All study controls who met study inclusion and exclusion criteria (n = 80) underwent two

experimental sessions:

1. Clinical Assessment and Baseline Questionnaires: All participants completed a battery

of neuropsychiatric questionnaires and clinical interview with a trained assessor as

detailed below.

2. Brain Imaging Session: This single session included acquisition of structural MRI and

rs-fMRI sequences that were identical that acquired in the patient cohort.

141

Subject Recruitment For all studies, male and female patients and healthy controls between the ages of 18 and

65 were recruited. All participants provided informed consent and study procedures were

approved by the University Health Network Research Ethics Board.

All patients were referred to MRI-Guided rTMS Clinic at Toronto Western Hospital

(typically by their general practitioner or psychiatrist) and were recruited into the aforementioned

studies by psychiatrists and trained clinical assessors at the Clinic. Patients’ psychiatric

diagnoses were verified by a Canadian Royal College-certified psychiatrist at the time of their

initial consultation assessment for rTMS treatment. Inclusion criteria in all three studies

included: competency to give voluntary, informed consent to treatment and participate in the

research study; and met DSM-5 diagnostic criteria for MDD, BN, AN-BP, or OCD at the time of

their consultation for rTMS. Patients were excluded from any of the three studies if they: were

pregnant; had a lifetime MINI diagnosis of bipolar I or II disorder, schizophrenia, schizoaffective

disorder, schizophreniform disorder, delusional disorder, or current psychotic symptoms; had a

MINI diagnosis of post-traumatic stress disorder (current or within the last year), anxiety

disorder (including generalized anxiety disorder, social anxiety disorder, panic disorder), or

dysthymia, assessed by a study investigator to be primary and causing greater impairment than

the primary diagnosis; had a diagnosis of any personality disorder, as assessed by a study

investigator to be primary and causing greater impairment than the primary diagnosis; had

previously received rTMS for any previous indication due to potential compromise of subject

blinding; had any significant neurological disorder or insult including, but not limited to: any

condition likely to be associated with increased intracranial pressure, space occupying brain

lesion, any history of seizure except those therapeutically induced by ECT, cerebral aneurysm,

Parkinson's disease, Huntington's chorea, multiple sclerosis, significant head trauma with loss of

consciousness for greater than or equal to 5 minutes; had an intracranial implant (e.g., aneurysm

clips, shunts, stimulators, cochlear implants, or electrodes) or any other metal object within or

near the head, excluding the mouth that could not be safely removed; or had a non-correctable

clinically significant sensory impairment (i.e., could not hear well enough to cooperate with

interview). If a patient was participating in psychotherapy, they were required to have been in

stable treatment for at least 3 months prior to entry to the study, with no anticipation of change in

the frequency of therapeutic sessions, or the therapeutic focus over the duration of the study. If a

142

patient was taking antidepressant or other psychotropic medication, they were required to have

been on a stable regimen for at least 4 weeks prior to entry to the study. Patients were also

required not to take more than 2 mg daily of lorazepam (or equivalent dosage of other

benzodiazepines), or any dose of an anticonvulsant due to the potential to limit rTMS efficacy.

All control participants were recruited using Research Ethics Board-approved poster

advertisements at Toronto Western and Toronto General Hospitals and word of mouth.

Exclusion criteria for control subjects included: a lifetime history of neurological or psychiatric

disease, including substance abuse or dependence, or cardiovascular disease; currently use of any

medications that could significantly affect brain perfusion or activity, including antidepressants,

mood stabilizers, neuroleptics, anxiolytics, hypnotics, stimulants, anticonvulsants, anti-migraine

agents, cognitive enhancing agents, opioids, anti-nausea agents or beta-blockers; any

contraindications to MRI (e.g., implanted medical devices, metallic foreign objects, epilepsy), or

a HAMD score ≥ 8.

MRI and rs-fMRI Acquisition For all studies, the structural and resting-state functional MRI scans of all participants

were acquired on the Toronto Western Hospital’s 3 Tesla GE HDx MRI system (General

Electric, Milwaukee, WI) with an eight-channel phased-array head coil. Scans were acquired

with participants positioned supine on the MRI table and their head was padded to reduce

movement.

A high-resolution, whole-brain, three-dimensional anatomical scan was acquired using a

T1-weighted fast spoiled gradient-echo sequence: echo time (TE) = 12 ms, inversion time (TI) =

300 ms, flip angle = 20°, 116 sagittal slices, thickness = 1.5 mm, no gap, 256 x 256 matrix, 0.78

x 0.78 x 1.5 mm3 voxels; field of view (FOV) = 240 mm.

Resting-state functional MRI was acquired for 10 minutes in the eyes-closed condition.

All patients were instructed to closed their eyes and ‘let their mind wander; do not fall asleep.’

rs-fMRI was acquired using a T2*-weighted echo-planar sequence: TE = 30 ms, TR = 2000 ms,

143

flip angle = 85°, 32 axial slices, thickness = 5 mm, no gap, 64 x 64 matrix, FOV = 220mm, 300

TRs/volumes).

Motor Threshold Assessment Resting motor thresholds for all patients were identified using stimulation of the medial

M1 and activation of the extensor hallucis longus muscle (resulting in movement of the great

toe). The focus of a Cool-DB80 rTMS coil (MagVenture, Farum, Denmark) was placed over the

interhemispheric fissure with lateral TMS coil orientation to preferentially stimulate the motor

cortex of the hemisphere contralateral to the direction of current flow, and thus the great toe

ipsilateral to the coil handle. Resting motor threshold was determined visually by a staff

psychiatrist according to a previously published motor threshold estimation procedure (Schutter

and van Honk, 2006). Briefly, this procedure involves four steps. First, the patient is seated in an

rTMS treatment chair and asked to remove footwear, and relax the legs. Second, the optimal site

for motor threshold is identified by five consistent motor-evoked hallux extensions. Identifying

the motor threshold is performed using the staircase method – TMS machine output begins at a

typically low, subthreshold intensity and is titrated upwards in increments of 1-5% to identify the

location at which minimal energy is needed to elicit motor responses. Third, after identifying the

this optimal M1 site, the intensity is gradually decreased to the lowest TMS output that evokes a

visibly detectable twitch in five of ten pulses. Fourth, the motor threshold is further optimized by

systematically searching for whether there is any an alternative M1 site that induces a response

for at least six of ten pulses; if such a site is identified, then the third and fourth steps are

repeated finalize the location and intensity of the resting motor threshold. This process is then

repeated with the coil handle oriented laterally in the opposite direction, to establish a motor

threshold for the other hemisphere’s lower limb M1 hotspot.

dmPFC-rTMS Treatment The intended dmPFC rTMS target on the cortex corresponded to the Talairach and

Tournoux stereotaxic coordinate (X+0, Y+30, Z+30) (Talairach and Tournoux, 1988). MRI-

144

guided neuronavigation was performed for Studies I and II. The nearest scalp point, at the

coordinate (X+0, Y+60, Z+60), was marked in the neuronavigation software (Visor, ANT Neuro,

Enschede, Netherlands) and used as the scalp site for stimulation. This scalp site was later

determined to correspond closely to a point at 25% of the distance of from the subject’s nasion to

inion (Mir-Moghtadaei et al, 2016), and so this scalp heuristic was used to orient the coil vertex

during all dmPFC-rTMS treatment sessions in Study III.

Neuronavigation involved a number of steps. The Visor 2.0 navigation system was used

to preprocess the patient’s T1 high-resolution anatomical scans for neuronavigation. Three MRI

processing steps were included: first, segmentation of the image into 2 components (scalp and

brain) and registration into the Talairach and Tournoux stereotactic space (Talairach and

Tournoux, 1988); second, landmarking of the tragus of the left and right ear, the anterior

commissure, posterior commissure, the interhemispheric point (between the two cortical

hemispheres), the anterior-most point of the brain, and the posterior-most point of the brain; and

finally, reconstruction of the surfaces of the patient’s brain and scalp into stereotactic space and

the overlaying of the dmPFC scalp coordinate to identify the stimulation target.

Neuronavigation involved a number of preparatory steps prior to stimulation. First,

patients were seated in a treatment chair that was positioned in clear view of the neuronavigation

camera. Next, a headband with a marker clip was attached around the patient’s head and placed

laterally (so as not to obstruct coil placement); the neuronavigation camera detected the marker

clip as a reference point to locate the patient’s head. Using the Visor system’s neuronavigation

pen, each of the scalp targets were highlighted on the patient (bilateral tragus, nasion, inion, and

scalp circumference) for calibration with the preprocessed MRI scan. After calibration, a second

marker clip was equipped to the rTMS head coil to ensure placement of the vertex of the coil at

(X+0, Y+60, Z+60).

For Study I and II, all open-label rTMS over the dmPFC was performed using a MagPro

R30 rTMS system (MagVenture, Farum, Denmark) and a Cool-DB80 TMS coil. Stimulation was

delivered under the following parameters: 10 Hz at 120% of resting motor threshold intensity, 5

second trains with an inter-train interval of 10 seconds off, 60 trains (3000 pulses) per

hemisphere . As during motor thresholding, the inductor coil was oriented laterally to achieve

145

preferential stimulation of the right and then the left hemisphere. For safety purposes, all patients

and the rTMS operator wore hearing protection during stimulation.

For Study III, all dmPFC-rTMS treatments (both active and sham) were performed using

a MagPro R30 rTMS System (MagVenture, Farum, Denmark) and a custom active/placebo

version of the Cool-DB80 TMS double-cone coil. Patients received 30 weekday sessions of

bilateral rTMS over the dmPFC, twice daily (60 minutes apart), over 15 days. For safety

purposes, all patients and the rTMS operator wore hearing protection during stimulation. Patients

randomized to excitatory active rTMS received 20 Hz stimulation at 120% of the resting motor

threshold, 2 s on 12 s off, 38 trains, for 1520 pulses per hemisphere, for a total of 3080 pulses per

session delivered over 8 min 40 s per hemisphere. Patients randomized to inhibitory active

stimulation received 1 Hz stimulation over the bilateral dmPFC at 120% resting motor threshold,

60 s on and 30 s off, for 360 total pulses, 8.5 minutes per hemisphere (Brunelin et al, 2014).

Sham rTMS used a custom active/placebo version of the DB80 coil, designed to maintain

blinding in clinical trial settings (MagVenture, Farum, Denmark). The double-sided stimulation

coil contained a set of windings on the active side, but not on the identically marked sham side.

Preprogrammed software controlled a switch within the rTMS device that instructed the

technician on which (externally indistinguishable) side of the coil to place over the scalp using

the instruction “flip coil” when appropriate. The allocation to active vs sham stimulation was

based on the patient’s study ID code, entered into the device by the technician and thereby

maintaining technician blinding. Patients randomized to the sham arm were randomly allocated

to receive either the 1 Hz or 20 Hz dmPFC-rTMS pattern above, again to maintain blinding; this

pulse pattern was delivered with the active coil in the upper position (oriented away from the

scalp), thus avoiding stimulation of the target region. The sensory and nociceptive side effects

were mimicked using adhesive electrodes placed bilaterally above the participant’s eyebrows.

These electrodes were connected to the custom DB80 coil, which synchronized superficial pulses

of electrical stimulation with the timing of rTMS pulses (at either 1 or 20 Hz). The pulse

intensity escalated with the rTMS stimulation intensity so as to preserve the impression of actual

stimulation with rTMS even in the sham condition.

The stimulation intensity for all treatments were titrated adaptively upward within the

limits of the patient’s maximum pain tolerance without distress, in order to ensure adequate

146

tolerability of dmPFC-rTMS. During the first week of treatment, patients continuously rated their

pain between rTMS trains on a verbal analogue scale from zero to ten (zero corresponding to no

pain, and ten corresponding to their maximal pain tolerability without emotional distress). The

intensity of stimulation was gradually titrated upwards, while maintaining a verbal analogue

score of nine or less, until the stimulator intensity output reached 120% of the patient’s resting

motor threshold.

During treatment, all patients in all studies were asked to defer any medication changes

throughout the course of rTMS to avoid confounding effects. Patients were also assessed for the

emergence of any adverse events (e.g., headache, seizure, worsening of symptoms) at each

session of treatment and at each follow-up assessment. Patients were withdrawn from the study if

they experienced a worsening in depression, defined as an increase in the HAMD from baseline

of >25% during two consecutive assessments, or the development of active suicidal intent, or

attempted suicide.

147

Study I: Baseline Predictors and Mechanisms of dmPFC-rTMS response in Anorexia and Bulimia Nervosa

Introduction This chapter contains the following manuscript: Dunlop, K., Woodside, B., Lam, E.,

Olmstead, M., Colton, P., Giacobbe, P., Downar, J. “Increases in frontostriatal connectivity are

associated with response to dorsomedial repetitive transcranial magnetic stimulation in refractory

binge/purge behaviors” NeuroImage: Clinical (2015) 8, 611-8.

Recurrent episodes of binge eating and purging behavior occur in both AN−BP and

bulimia nervosa BN. In women, the lifetime prevalence of AN and BN is 0.9% and 1.5%

respectively (Hudson et al, 2007). Eating disorders (ED), in particular AN, have among the

highest mortality rate of all psychiatric disorders (Sullivan, 1995) and treatment options are

limited for severe forms of AN and BN, particularly for AN−BP (Arcelus et al, 2011).

Conventional treatments include psychotherapy, pharmacotherapy and inpatient treatments, but

rates of treatment dropout, limited treatment response and relapse rates are substantial (Carter et

al, 2012; Hay et al, 2012; Mitchell et al, 2007; Olmsted et al, 2005; Shapiro et al, 2007). New

treatment options are urgently needed.

Therapeutic brain stimulation is a novel approach in treatment-refractory ED, as recently

demonstrated in a pilot study using DBS in AN (Lipsman et al, 2013b). Non-invasive techniques

such as rTMS would offer greater accessibility and lower medical risk than DBS, if suitable

stimulation targets could be identified. Neuroimaging research has identified a variety of

neuroanatomical substrates of ED pathophysiology. For example, on structural imaging, ED

patients show reductions in gray matter volume in regions involved in reward, impulse control,

and emotion regulation: the caudate nucleus, VS, ACC, and OFC (Van den Eynde et al, 2012;

Friederich et al, 2012; Schäfer et al, 2010; Titova et al, 2013). Likewise, fMRI studies in ED

reveal abnormal patterns of resting-state connectivity in the default-mode network (Cowdrey et

al, 2014) and other intrinsic brain networks incorporating the ACC and insula (Amianto et al,

2013). ED patients also show abnormal VS activation in response to rewarding and aversive

stimuli (Wagner et al, 2007, 2010). Patients with BN show hyperactivity in medial frontal lobe

148

regions during inhibition of prepotent actions (Lock et al, 2011), and hypoactivity in response to

food cues (Joos et al, 2011).

Among the various neural substrates implicated in ED pathology, a target of particular

interest that is accessible to rTMS is the dmPFC. The dmPFC plays an important role in forms of

self-control, including self-inhibition of movements (Brass and Haggard, 2007), self-cessation of

loss-chasing in pathological gamblers (Campbell-Meiklejohn et al, 2008), self-suppression of

emotional responses (Kühn et al, 2011), and impulse control (Cho et al, 2013). Likewise, non-

invasive stimulation of medial prefrontal areas, via rTMS, enhances inhibitory control over

prepotent responses (Obeso et al, 2013), improves subjective choice for delayed rewards in a

delayed discounting task, and interferes with striatal dopamine (Cho et al, 2015). In sum,

stimulation of the medial prefrontal cortex using rTMS may alter the top-down executive control

of the dmPFC to striatal regions associated with the urge to binge and purge, thereby improving

symptom severity.

Considering this literature on the role for the DMPFC in self-regulation, one potential

implication is that DMPFC-rTMS might be worth exploring as a therapeutic approach for

addressing self-regulatory deficits in ED. To date, most previous rTMS studies in ED have

focused on the conventional target, the left DLPFC. Although DLPFC-rTMS reduces cue-

induced food craving in BN (Van den Eynde et al, 2010; Uher et al, 2005), double-blind trial of

DLPFC-rTMS for binge−purge symptoms found no significant improvement over sham

(Walpoth et al, 2008).

There are few reported studies of dmPFC-targeted rTMS to date. However, we recently

investigated dmPFC-rTMS in TRD, finding efficacy rates comparable to DLPFC-rTMS, but with

sharply dichotomous outcomes of improvement (Downar et al, 2014). Using resting-state

functional MRI (rs-fMRI), we found that low dmPFC–subcortical connectivity predicted

successful outcome, and that symptomatic improvements were associated with increased dmPFC

connectivity. We also reported the serendipitous finding of full remission from binge−purge

symptoms during dmPFC-rTMS for comorbid depression, in a patient with treatment-refractory

BN (Downar et al, 2012). These observations suggested that dmPFC-rTMS might treat a subset

of BN patients, and that enhanced frontal−subcortical connectivity on rs-fMRI might accompany

symptomatic improvement.

149

Aims & Hypotheses The general aim of Study I is to establish baseline predictors of, and neural

correlates/mechanisms of, improvements in binge-eating and purging behaviors with dmPFC-

rTMS treatment, in BN and AN-BP.

Specific Aims of Study I 1. To determine whether 20-30 sessions of open-label 10 Hz dmPFC-rTMS improves binge

eating and purging in patients with AN-BP or BN.


significantly differ between AN-BP/BN dmPFC-rTMS responders and non-responders.


correlate with binge eating and purging frequency in patients with AN-BP and BN.

Hypotheses of Study I 1. Mean binge eating and purging frequency will significantly improve in AN-BP/BN

patients who receive 20-30 sessions of open-label 10 Hz dmPFC-rTMS.

2. Lower baseline dmPFC-CSTC and higher baseline dmPFC-sgACC resting-state

functional connectivity will be associated with better response to 10 Hz dmPFC-rTMS.

3. Increases in pre- to post-treatment CSTC-dmPFC resting-state functional connectivity

after 10 Hz dmPFC-rTMS will correlate with improvements in binge-eating and purge

frequency.

150

Methods

Subjects

28 subjects (26 female, age range = 20 − 56 years, mean = 31.0 ± 9.5 years) meeting

DSM-5 criteria for AN−BP (n = 11), BN (n = 17) participated in this proof-of-concept, open-

label study. Subjects included individuals with bingeing and purging behavior, both at a normal

and lower than normal weight. BMI ranged from 14.5–28.8. AN−BP participants displayed both

bingeing and purging behaviors. While there are certainly important differences in the clinical

features of these disorders (e.g., failure to maintain normal body weight, or the presence of

restricting symptoms), binge and purge behaviors were the specific target symptom for the

purposes of this pilot study. Subjects engaged in ≥ 2 cumulative objective binge/purge episodes

weekly at baseline to be included in the study. Diagnoses were established through interviews

conducted by independent, Canadian Royal College-certified psychiatrists. In recognition of the

high prevalence of psychiatric comorbidities in this population, and in order to characterize

rather than minimizing sample heterogeneity, common comorbidities were not excluded. This

inclusion of comorbidities is in line with our previous work in TRD (Downar et al, 2014).

Comorbid diagnoses included MDD (n = 16), OCD (n = 6), post-traumatic stress disorder (n =

8), and bipolar disorder (n = 6). Patients with a history of a psychotic disorder, neurological

disorder, active substance abuse, or contraindications to MRI or rTMS were excluded. All

subjects had no improvements in bingeing/purging to or were unable to tolerate at least 1

previous medication trial (n = 27), and had also not responded to previous inpatient/outpatient

treatment courses in terms of an improvement of binges and purges (n = 25). On average,

patients had not respond to 2.61 ± 2.44 SD inpatient/outpatient treatments, and 3.00 ± 2.37 SD

medication trials. Current medications included SSRIs (n = 11), antipsychotics (n = 11),

benzodiazepines (n = 8), trazodone (n = 5), and SNRIs (n = 3). As routinely stipulated (Salomons

et al, 2014), subjects were required to refrain from any medication changes during and for ≥4

weeks before rTMS.

151

Clinical Outcomes

The clinical outcome of interest, weekly frequency of binge and purge episodes, was

monitored via a structured clinical interview, the EDE (Fairburn and Cooper, 1993),

administered 1 week before treatment, at each week during treatment, and at 4-weeks post-

treatment. For the purposes of stratification, response was defined as ≥ 50% decrease in objective

binge (≥ 1000 calories per binge) and purge episode frequency from pre-rTMS to 4 weeks post-

rTMS (at follow-up). Pre-treatment EDE assessments acquired binge/purge frequency for the 4

weeks prior to treatment, and the post-treatment EDE acquired frequency for the 4 weeks

following treatment, and so the pre- and post-treatment scores were obtained from weeks outside

of the rTMS treatment sessions. Data on bingeing and purging frequencies derived from the EDE

was divided by 4 to generate a weekly frequency for bingeing and purging. To obtain weekly

scores, EDE frequencies (which measure severity over 4 weeks) were divided over the 4 weeks.

Although a more conservative criterion is sometimes used in clinical efficacy studies, our

primary aim was to characterize and distinguish the neural activity of subpopulations of patients

showing improvement versus non-improvement in this population. A set of clinical and

psychometric data, including the 17-item HAMD (Hamilton, 1960), BDI-II (Beck et al, 1961,

1996) and the Beck Anxiety Inventory (BAI) (Beck et al, 1988) were also collected as secondary

metrics. To assess the distribution of outcomes across the study sample, we employed kernel

density estimation by applying the Epanechnikov kernel (Epanechnikov, 1969) in Stata13

(StataCorp). Two-tailed t-tests (Bonferroni-adjusted) were performed to determine the

significance of differences in clinical measures between groups and between timepoints. The

nonparametric Mann–Whitney U test was performed for comparisons of binge and purge

frequency do to the markedly non-normal distribution of these data.

Neuroimaging Acquisition

All subjects underwent 3 Tesla MRI the week prior to and the week after rTMS, with a

T1-weighted (TE = 12 ms, TI = 300 ms, flip-angle = 20°, 0.94 × 0.94 × 1.5 mm voxels) and a

10-min eyes-open resting state functional MRI scan (TE = 30 ms, TR = 2000 ms, flip-angle =

85°, 3.4 × 3.4 × 5 mm voxels).

152

Neuronavigation and rTMS Treatment

Neuronavigation employed the Visor 2.0 system (Advanced Neuro Technologies,

Enschede, Netherlands) to target the stereotaxic coordinate (X = 0, Y = +30, Z = +30) in

Talairach space (Talairach and Tournoux, 1988). rTMS employed the MagPro-R30 system

(MagVenture, Farum, Denmark) and a Cool-DB80 coil. Motor thresholds were determined via

contractions of the extensor hallucis longus (Hayward et al, 2007). Stimulation of the dmPFC

was delivered at 120% resting motor threshold, at 10 Hz, 5s on, 10s off, 3000 pulses/

hemisphere, with left then right lateralized coil orientation (Terao et al, 2001). Unilateral

stimulation was achieved by orienting the coil vertex at the stereotactic target laterally, with

current flow oriented towards the desired hemisphere (Harmer et al, 2001). If a patient missed

an rTMS session for logistical reasons, an additional session was added at the end of the course

of treatment (no patient required more than 4 such sessions). Patients underwent 20 sessions of

dmPFC-rTMS on weekdays; responders with any residual binge/purge symptoms were extended

to 30 sessions (mean = 21.2 ± 3.7 sessions, range = 18–30).

MRI Preprocessing, Seed Selection & Statistical Analysis

Preprocessing of resting-state fMRI data from patients employed the FMRIB Software

Library (FSL) (Jenkinson et al, 2012). Preprocessing included: the removal of the first 5 volumes

to account for scanner instabilities; interleaved slice-timing correlation; segmentation; motion

correction via a linear affine transformation of each volume of the time-series to the middle

volume (reference) of the scan ; spatial smoothing using a 6 mm FWHM Gaussian kernel,

correction for white matter and cerebrospinal fluid signal artifacts using linear regression,

temporal bandpass filtering (0.009–0.09 Hz), and nonlinear co-registration to the MNI-152

template.

ROIs were defined a priori from the parcellation atlas of Craddock et al. (2012) for the

dmPFC and adjacent dACC based on proximity to the stimulation target, in line with previous

analyses of dmPFC-rTMS in TRD (Craddock et al, 2012a; Salomons et al, 2014). ROI centroids

were MNI = −4, 44, 42 for dmPFC and MNI = 0, 38, 24 for dACC. For first-level analysis, each

153

seed ROI was co-registered to each subjects’ brain via nonlinear transformation, then applied as

masks to extract the mean ROI time series. These time series were used to generate whole-brain

maps of positively and negatively correlated voxels with the seed ROI before and after treatment,

via a linear regression analysis using a fixed effects model at the individual-subject level.

Group level analysis was performed using FSL’s FLAME mixed effects model

(Beckmann et al, 2003). We first identified regions where the degree of correlation at baseline to

the ROIs differed significantly between responders and non-responders. To assess whether

differing changes in resting-state connectivity over treatment would be associated with different

outcomes, we then compared pre- and post-treatment scans to identify regions where the pre- to

post-treatment change in correlation to the seed ROIs differed significantly between responders

and non-responders. The results of these analyses were transformed into z-score maps, correcting

for multiple comparisons using a Gaussian random field theory cluster-based correction (z-score

> 1.96, cluster significance p < 0.05, corrected). Finally, the clusters previously obtained from

group-level baseline predictor and change analyses were co-registered to patients’ individual

scans, and connectivity values were extracted for each subject.

Results

Primary Clinical Outcomes

No serious or treatment-limiting adverse effects occurred, and subjects reported only the

localized scalp discomfort and transient headache routinely associated with rTMS treatment

sessions.

154

Table 4-1: Descriptive Statistics for all patients, rTMS responders and non-responders. U

indicates nonparametric significance testing.

All Subjects (n=28)

rTMS Responders (n=16)

rTMS Non-responders (n=12)

Responder vs. Non-Responder t(p)

Women (Men) 26 (2) 15 (1) 11 (1) 0.04(0.97)

Age 31.04 (±9.48) 30.75(±8.23) 31.42(±11.32) 0.04(0.97)

Duration of ED (Years) 14.75 (±10.19) 12.75(±7.51) 17.42(±12.83) 0.84(0.41)

# Prior Treatments 2.61 (±2.44) 2.63(±2.06) 2.58(±2.97) 0.04(0.97)

# Prior Medications 3.00(±2.37) 2.38(±2.00) 3.83(±2.66) 0.99(0.33)

# of Hospitalizations 3.57(±4.20) 5.00(±3.14) 6.42(±5.35) 0.65(0.52)

Baseline BMI 19.03(±5.33) 19.81(±3.68) 18.05(±3.22) 1.35(0.19)

AN-BP (BN) 12(16) 5(11) 7(5) 0.87(0.39) Weekly Binge Frequency (pre-rTMS) 11.14(±18.59) 7.48(±5.41) 20.55(±26.82) U 0.91(0.36)

Weekly Purge Frequency (pre-rTMS) 17.57(±31.72) 8.75(±6.93) 36.68(±44.30) U 1.21(0.23)

Weekly Binge Frequency (post-rTMS) 8.32(±15.09) 1.39(±1.88) 18.25(±19.41) U 4.12(<0.0001)

Weekly Purge Frequency (post-rTMS) 19.63(±53.96) 1.34(±2.01) 54.53(±76.93) U 3.51(0.0004)

Abbreviations: AN-BP = Anorexia Nervosa Binge/Purge Subtype; BMI = Body Mass Index; BN

= Bulimia Nervosa; ED = Eating Disorders; rTMS = Repetitive Transcranial Magnetic

Stimulation. Values indicate means, with standard deviations reported in brackets.

Baseline binge and purge episode frequency per week was 11.1 ± SD 18.6 and 17.6 ± SD

31.7 (Table 4-1). Combining both subpopulations, there was no significant overall change in

binge frequency (post-rTMS mean = 8.6 ± 2.9, mean percent improvement = 20.4 ± 77.0,

Wilcoxon signed-rank W27 = 1.29, p = 0.20) but a significant decrease in purge frequency was

found (post-rTMS mean = 20.33 ± 10.2, mean percent improvement = 35.7 ± 62.1, Wilcoxon

signed-rank W27 = 2.20, p = 0.03). 16 of 28 subjects achieved ≥ 50% reduction in binge and

purge frequency from baseline to follow-up. However, outcomes were widely divergent across

individuals, ranging from full remission to marked worsening of symptoms (Figure 4-1). The

degree of improvement across individuals followed a non-normal distribution (Shapiro–Wilk W27

= 0.85, p = 0.001). For this reason, subsequent analyses were performed to characterize

responder and non-responder subpopulations separately.

155

Figure 4-1: Probability distribution function of binge and purge percent improvement

across all patients following dmPFC-rTMS.

Among responders, symptom frequency improved from 7.5 ± SD 5.4–1.4 ± SD 1.9 binge

episodes/week (Wilcoxon signed-rank W15 = 3.39, p = 0.0007) and from 8.7 ± SD 6.9–1.3 ± SD

2.0 purge episodes/week (W15 = 3.49, p = 0.0005). Among non-responders, symptom frequency

before and after treatment showed a non-significant worsening from 20.5 ± SD 26.8–18.3 ± SD

19.41 binge episodes per week (W11 = −1.30, p = 0.20) and 36.7 ± SD 44.3–54.5 ± SD 76.9

purge episodes per week (W11 = −0.59, p = 0.55). There was no significant difference between

responders and non-responders at baseline for either binge (Mann–Whitney U26 = 0.91, p = 0.36)

or purge (U26 = 1.21, p = 0.23) frequency.

156

Secondary Clinical Outcomes

Global EDE scores significantly decreased from 4.1 ± 1.3 to 3.2 ± 1.3 4-weeks post-

rTMS (t27 = 2.32, p = 0.03) for all subjects and from 4.1 ± 1.3 to 3.1 ± 1.1 for responders (t15 =

2.55, p = 0.02). Global EDE scores for non-responders decreased non-significantly from 4.1 ±

1.5 to 3.3 ± 1.8 post-rTMS. Responders and non-responders did not significantly differ at

baseline or post-rTMS on global EDE and subscale scores. There were no significant changes in

eating and shape concern scores overall, but there was a significant improvement on Restraint

(from 3.4 ± 1.5 to 2.7 ± 1.5, t15 = 2.75, p = 0.01). Responders significantly improved on Restraint

(from 3.3 ± 1.7 to 2.3 ± 1.6, t15 = 3.06, p = 0.009) and weight concern subscales (from 4.6 ± 1.5

to 3.2 ± 1.8, t15 = 2.58, p = 0.02).

Among all subjects, the mean baseline HAMD was 19.3 ± 7.5, improving to 12.1 ± 8.4

post-treatment (t20 = 3.33, p = 0.003). Among binge/purge responders, HAMD did not show

significant improvement (18.9 ± 8.2 and 11.6 ± 9.1 pre- to post-rTMS, t11 = 1.99, p = ns);

however, among non-responders, HAMD did significantly improve from a mean score 19.8 ± 6.9

to 12.6 ± 7.9, respectively (t9 = 2.80, p = 0.02). On BDI-II, subjects had a baseline mean of

35.8±13.8, which significantly improved to 25.3±13.3 (t25=5.29, p<0.0001. In responders, BDI-II

significantly improved from 37.6±12.7 to 21.8±11.8 (t14=7.95, p<0.0001). However, in non-

responders, BDI-II did not change significantly, from 33.5 ± 15.4 to 30.2 ± 14.2 (t10 = 1.50, p =

0.16). On BAI, all subjects significantly improved, from 26.7 ± 15.5 to 13.5 ± 12.3 (t25 = 6.76, p

< 0.0001). Marked improvement was evident in both responders (BAI, 29.4 ± 15.8 to 14.2 ±

12.3; t15 = 5.11, p = 0.0002) and non-responders (BAI, 22.6 ± 14.9 to 12.3 ± 13.6; t9 = 5.91, p =

0.0004).

The relationship between the following pre-treatment clinical measures and binge/purge

percent improvement was also investigated: baseline BMI, illness duration, baseline binge/purge

severity, age, ED diagnosis (AN-BP or BN), presence of co-morbidity (MDD, OCD, post-

traumatic stress disorder, bipolar disorder), current medication (benzodiazepine, selective

serotonin reuptake inhibitor, antipsychotic), and baseline severity and percent improvement of

secondary measures (HAMD, BDI-II, and BAI). No clinical measures, diagnoses, or mediations

predicted rTMS treatment response, or percent improvement in binge/purge frequency, either

157

before or after Bonferroni correction for multiple comparisons (Spearman’s rank correlation

coefficient ρ < 0.30, p > 0.12 for each comparison above).

Functional Connectivity Predictors of Response

Pre-treatment resting-state fMRI revealed significant differences between responders and

non-responders from the dmPFC and dACC regions of interest (Figure 4-2). From the dmPFC,

responders displayed significantly lower pre-treatment connectivity to bilateral temporal pole,

OFC, and right posterior insula, and higher connectivity to bilateral lateral and medial occipital

cortex (Table 4-2). Pre-treatment functional connectivity from dmPFC to left and right lateral

OFC was slightly positive in responders (z = 1.12 ± 0.56), and positive in both non-responders (z

= 4.75 ± 0.56). Parameter estimates also revealed that pre-treatment functional connectivity from

dmPFC to right insula was negative in responders (z = −3.36 ± 0.39) and non-responders (z =

−0.20 ± 0.73) (Fig. 3-2B). Both dmPFC–insula and dmPFC–OFC connectivity at baseline were

significantly anti-correlated to binge/purge percent improvement (r = −0.42 p = 0.03 and r =

−0.46 p = 0.01, respectively).

From the dACC seed, responders had significantly lower pre-treatment connectivity to

the right posterior insula, putamen, hippocampus, and middle temporal gyrus, and higher

connectivity to superior parietal and medial occipital cortices and precuneus (Table 4-2; Figure

4-2C). Parameter estimates revealed that pre-treatment functional connectivity from dACC to

right hippocampus was negative in responders (z = −3.69 ± 0.42), non-responders (z = −0.44 ±

0.46). Pre-treatment functional connectivity from dACC to right insula and thalamus was

negative in responders (z = −3.15 ± 0.74) but positive in non-responders (z = 0.51 ± 0.53)

(Figure 4-2D). dACC–insula connectivity at baseline significantly anti-correlated to binge/purge

percent improvement (r = −0.38 p = 0.01).

158

Figure 4-2: Baseline functional connectivity differences between dmPFC-rTMS Responders

and Non-Responders. A: Regions with lower pre-treatment functional connectivity to the

dmPFC seed (red) in responders versus non-responders to dmPFC-rTMS. B: Parameter estimates

for pre- treatment functional connectivity between the dmPFC seed and the bilateral orbitofrontal

cortex (OFC) and right posterior insula in responders and non-responders prior to treatment. C:

Regions with lower pre-treatment functional connectivity to the dACC seed (red) in responders

versus non-responders to dmPFC-rTMS. D: Parameter estimates for functional connectivity

between the dmPFC seed and right hippocampus and posterior insula in responders and non-

responders prior to treatment.

159

Table 4-2: Brain regions where pre-treatment functional connectivity to dmPFC and

dACC seeds differed significantly between rTMS responders and non-responders.

Seed Brain Region Brodmann Area

MNI coordinates z

X Y Z

dACC

Non-responders > Responders

R Posterior Insula 13 44 -6 4 3.70

R Hippocampus 32 -28 -8 3.61

R Posterior Superior/Middle Temporal Gyrus 21 / 22 50 -18 -8 3.78

Responders > Non-responders R Precuneus 7 10 -58 58 3.64

L Precuneus 7 -20 -58 60 4.00

R Intracalcarine Cortex 18 8 -80 2 3.04

B Cuneus 18 8 -90 22 3.11

dmPFC Non-responders > Responders

R Lateral OFC 25 / 47 34 24 -14 2.83

L Lateral OFC 47 -34 22 -24 3.87

R Posterior Insula 13 38 -10 4 3.79

L Temporal Pole 28 / 38 -38 -10 -20 3.84

R Temporal Pole 28 / 38 52 4 -24 3.35

Responders > Nonresponders

B Intracalcarine Cortex, Lingual Gyrus 18 4 -80 2 4.06

Functional Connectivity Changes Associated with Response

There were also significant differences between responders and non-responders in terms

of the observed changes in functional connectivity to the dACC and dmPFC before and after

rTMS treatment. Compared to non-responders, responders underwent significantly greater

increases in functional connectivity between the dACC and bilateral caudate nucleus and VS,

anterior insula, inferior frontal gyrus and adjacent OFC, and right putamen (Table 4-3, Figure 4-

3A). Responders also showed greater increases in functional connectivity between the dmPFC

and right middle temporal gyrus, and greater decreases in functional connectivity between the

dmPFC and bilateral thalamus (Table 4-3).

Inspection of parameter estimates revealed marked differences between responders and

non-responders in how dACC–VS functional connectivity changed following treatment (Figure

4-3B). The responder group showed lower functional connectivity between the dACC and VS

160

prior to treatment, increasing significantly following successful treatment (t12 = 2.37, p = 0.035).

Conversely, connectivity in non-responders decreased with a trend to significance (t8 = 2.316, p

= 0.05), becoming significantly lower than treatment responders post-treatment (t20 = 2.409, p =

0.03). A similar pattern of significant increases in functional connectivity in responders (t12 =

3.314, p = 0.006), but decreases in functional connectivity in non-responders, trending to

significance (t8 = 2.078, p = 0.07), was observed between the dACC and left anterior insula over

the course of treatment (Figure 4-3B). Binge/purge percent improvement was significantly

correlated with the change in dACC-left anterior insula connectivity (r = 0.45, p = 0.04) and

trended to significant correlation with the change in dACC-VS connectivity (r = 0.39, p = 0.07).

161

Figure 4-3: Changes in functional connectivity in dmPFC-rTMS Responders and Non-

Responders. A: Regions showing significantly greater increases in functional connectivity to the

dACC seed from pre- to post-treatment in responders versus non-responders to dmPFC-rTMS.

B: Parameter estimates for functional connectivity between the dACC seed and the ventral

striatum (right) and insula (left) in responders and non-responders before and after undergoing

dmPFC-rTMS.

162

Table 4-3: Brain regions where the change in functional connectivity to dACC and dmPFC

seed from pre- to post-treatment differed significantly between rTMS responders and non-

responders.

Seed Brain Region Brodmann Area

MNI coordinates z

X Y Z dACC

Increased connectivity in responders > non-responders

L Caudate/Ventral Striatum

-14 18 2 3.89

R Caudate/Ventral Striatum

16 20 0 4.02

R Insula, Putamen 13 32 -18 -2 4.03

R Anterior Insula/OFC 13 / 47 36 18 -8 3.49

L Anterior Insula/OFC 13 / 47 -38 14 2 3.16

L Inferior Frontal Gyrus 47 -34 22 -14 2.66 B Thalamus 14 -18 14 3.13

dmPFC Increased connectivity in responders > non-responders

R Middle Temporal Gyrus 21 / 22 50 -26 -10 4.27

Decreased connectivity in responders > non-responders

B Thalamus -4 -26 6 3.08

Discussion & Conclusion To our knowledge, this is the first report of neuroimaging findings in ED patients

undergoing non-invasive brain stimulation. As an intervention-probe, dmPFC-rTMS divided

patients into treatment-responsive and nonresponsive groups, with widespread differences in

resting-state connectivity apparent between these groups at baseline. rTMS-responsive patients

showed baseline hypoconnectivity from the stimulation target to other cortical and subcortical

regions, relative to non-responders. In responders, frontostriatal connectivity was enhanced

following dmPFC-rTMS, in association with improvement in binge and purge frequency.

Conversely, in patients with higher baseline connectivity, the same intervention produced the

opposite effect, reducing frontostriatal connectivity, in association with non- improvement or

worsening of symptoms.

163

Of particular interest were the divergent trajectories of outcome across individuals. While

many subjects showed marked improvement, a large proportion of subjects did not improve, or

even showed deterioration, when exposed to the same intervention. The observation of divergent

responses to dmPFC-rTMS in ED is in keeping with our previous report of a sharply bimodal

distribution of responses to dmPFC-rTMS in TRD (Downar et al, 2014). Of note, similarly

dichotomous trajectories of improvement have been identified in clinical trials of duloxetine

(Gueorguieva et al, 2011) and escitalopram (Thase et al, 2011).

The possibility of a neurobiological basis for the responder–non-responder dichotomy in

the present study is supported by differences in resting-state functional connectivity observed

between the two groups before treatment. In keeping with our first hypothesis, responders had

significantly lower functional connectivity from the dmPFC and adjacent dACC to subcortical

and cortical areas involved in emotion generation and regulation: hippocampus, lateral

orbitofrontal cortex, and insula. These findings are consistent with several previous studies in

MDD patients, linking baseline resting-state hypoconnectivity to better rTMS response (Fox et

al, 2012b, 2013a; Liston et al, 2014; Salomons et al, 2014).

The observation of baseline hypoconnectivity to the target as a predictor of rTMS

response is notable for being consistent with recent findings from three independent groups:

better response to rTMS in trials using DLPFC targets that were more negatively correlated with

sg-ACC (Fox et al, 2013a), better response to DLPFC-rTMS in patients with baseline

hypoconnectivity from DLPFC and dmPFC to sg-ACC (Fox et al, 2012b), and better response to

dmPFC-rTMS in patients with baseline hypoconnectivity in cortico-striatal-thalamic circuits via

the stimulation target (Salomons et al, 2014). Given the consistency of this observation across

multiple rTMS targets in multiple populations, hypoconnectivity of the stimulation target to

limbic cortical or subcortical regions may be worth further investigation as a biomarker of rTMS

response in future trials with larger sample sizes.

Another key finding in the present study is that the same 10 Hz intervention had widely

divergent effects on functional connectivity across individuals: patients with high baseline

cortico-cortical and fronto-insular connectivity saw significant reductions after 10 Hz rTMS,

while patients with low baseline frontostriatal and fronto-insular connectivity saw significant

increases after stimulation at the same frequency. This phenomenon is unlikely to represent a

164

regression to the mean, as the groups showed divergence rather than convergence in both

symptomatology and functional connectivity over treatment. The results are also consistent with

our previously reported observations of markedly dichotomous effects on striatal, thalamic and

cortical functional connectivity from 10 Hz dmPFC-rTMS in patients with MDD (Salomons et

al, 2014). The results are also consistent with previous studies of DLPFC-rTMS in depression:

while some improved with high- frequency rTMS, others worsened on this regimen, and instead

responded better to low-frequency stimulation of the same target (Kimbrell et al, 1999; Speer et

al, 2000, 2009). These findings are consistent with mounting evidence that the physiological

effects of a given rTMS protocol can vary widely across individuals, not just in magnitude but

also in direction (Cárdenas-Morales et al, 2014; Eldaief et al, 2011; Hallett, 2007; Maeda et al,

2000).

The divergent effects of 10 Hz stimulation on fronto-insular and fronto-striatal

connectivity could potentially reflect pre-existing differences in impulsivity and cognitive

flexibility between responders and non-responders. For example, in healthy populations, higher

resting-state dACC-insula and dACC-striatal connectivity is correlated with better performance

on decision-making tasks, cognitive flexibility, and lower measures of impulsivity (Jung et al,

2014; Müller et al, 2015). Furthermore, impulsivity when experiencing negative affect is

positively correlated with bulimia severity (Fischer et al, 2003), and BN is associated with

deficits in executive functioning (as assessed, for example, on a gambling task) (Brand et al,

2007). Thus, significant increases in dACC-striatal functional connectivity in responders

following rTMS could potentially lead to improvements in impulsivity and executive

functioning, which in turn lead to symptom improvement. Conversely, symptom worsening in

non-responders may potentially be a result of the observed decreases in fronto-insular and

frontostriatal connectivity and concomitant worsening of impulsivity and executive functioning.

The present study unfortunately did not measure these cognitive functions directly. However, a

future study including behavioral measures of cognitive control and impulsivity would be useful

in testing a hypothesized mechanism of action in which rTMS-induced increases in dACC-

striatal/insular functional connectivity lead to improved impulse control and executive function,

and thus to symptomatic improvement.

The heterogeneity of observed rTMS effects at the neural and clinical level, speaks to the

unresolved issue of individual variability in the physiological effects of rTMS. Classically, 10 Hz

165

rTMS is considered to have an excitatory effect on neural activity, while low frequency rTMS is

thought to have an inhibitory effect (Hallett, 2007). However, marked inter-individual

variability in rTMS effects has been long been observed in electrophysiological, and

neuroimaging studies. For example, a large minority of subjects show paradoxical inhibition of

motor evoked potentials after 10 Hz rTMS, or enhancement from 1 Hz rTMS, and the effects of

the different frequencies are not reliably opposite within a given individual (Maeda et al, 2000).

On fMRI, 1 Hz rTMS enhances functional connectivity from the target to co-activating cortical

regions in some individuals, but reduces functional connectivity in others, with no consistent

direction of overall effect (Eldaief et al, 2011). Conversely, patterns of network connectivity to

the stimulation target have been reported to influence the magnitude of rTMS-induced motor

facilitation (Cárdenas-Morales et al, 2014). In light of such findings, it has been proposed that rs-

fMRI might provide a useful marker for individualizing the target of rTMS (Fox et al, 2013a).

The findings of the present study suggest that fMRI may also be useful for individualizing the

pattern of rTMS, as has previously been shown in studies using positron emission tomography

(Kimbrell et al, 1999; Speer et al, 2000, 2009). Once again, the potential utility of neuroimaging

biomarkers for rTMS parameter selection is an important topic for future study.

One limitation of this preliminary study involves its relatively small sample size. Though

the sample was slightly larger than those used in several recent rTMS-fMRI studies in more

prevalent disorders such as TRD (Baeken et al, 2014; Liston et al, 2014; Salomons et al, 2014),

the present study does not allow a more detailed characterization of potentially important clinical

markers that could help distinguish treatment subpopulations. Although the present investigation

did not find statistically significant pre-treatment differences between responders and non-

responders in terms of duration of illness, BMI, and binge/purge frequency, future studies with

higher power will be necessary to better assess whether those with more severe illness in general

are less likely to respond to treatment. Another caveat relates to the inclusion of multiple DSM-5

diagnostic categories in the current sample, which embraced ED patients with binge−purge

behaviors alongside common comorbidities. On this point, we note that the sample composition

was chosen to be representative of clinical populations presenting with refractory binge−purge

symptoms, and that no DSM-5 diagnosis significantly correlated with rTMS response. More

generally, it is unclear that current categorical nosologies map particularly well on to rTMS

responsiveness in general; the distinction between unipolar and bipolar depression had no

166

bearing on outcome in a recent meta-analysis of rTMS efficacy (Berlim et al, 2014). We are

aware that some of our subjects were underweight, and that improvements in bingeing and

purging represent a limited area of improvement for such individuals. However, given the

exploratory nature of this trial, and the notorious difficulty in treatment individuals with AN-BP,

we felt it was important to include such individuals in the trial. It is clear that the question of

whether dmPFC-rTMS aids other forms of disordered eating (including restrictive-type

behaviors) is still an important area for future investigation.

Another limitation is the lenient multiple comparisons correction. As more recent

publications have recommended stringent correction for multiple comparisons of resting-state

functional connectivity (Eklund et al, 2016), the risk for Type I errors in these studies is

potentially high. However, a number of factors mitigate concern regarding this methodological

limitation. Most importantly, the neuroimaging results across these three independent samples

are consistent. Second, the baseline frontostriatal predictor of treatment response is consistent

with Study III (which used more stringent correction). Finally, as evidenced in Sections 1.4, 1.5,

1.6, 1.8 and 1.9 of this thesis, the neuroimaging results are consistent with disorder etiology and

mechanisms of treatment response.

Another potential criticism relates to the use of an open-label design without sham

stimulation. Since no previous studies have examined dmPFC-rTMS in ED, and since the

primary aim of the study was to apply an intervention-probe and characterize neural predictors

and correlates of response (as in our recent work in depression), we opted for the present

approach in order to identify sources of heterogeneity that might confound a future sham-

controlled trial. The results suggest that three potentially critical confounds will need to be

accommodated in future trials: first, the presence of distinct neural endophenotypes, not readily

apparent on standard diagnostic criteria, but with differential responses to the intervention at both

the neural and the clinical level; second, the need to tailor rTMS parameters to the individual

patient in order to avoid paradoxical effects; third, whether predictors of improvement relate to

illness mechanism, or a capacity to change with rTMS treatment, or both. Our results suggest

that resting-state fMRI techniques help address both of these issues in any future randomized

controlled study of rTMS in ED.

167

In conclusion, dmPFC-rTMS shows promising therapeutic effects in a subset of ED

patients with refractory binge and purge behaviors. As in TRD, therapeutic effects are associated

with increases in initially low levels of frontostriatal functional connectivity, detectable on pre-

treatment rs-fMRI. However, a significant proportion of patients have initially higher baseline

frontostriatal connectivity, and in these patients, the same intervention produces paradoxical

decreases in frontostriatal connectivity after treatment, alongside non-response or worsening of

clinical symptoms. A randomized controlled trial of dmPFC-rTMS in ED would be a reasonable

next step. However, the results of this study join a growing body of evidence, suggesting that

future trials of rTMS will need to take into account both the heterogeneity of individual patients’

neural activity, and the heterogeneity of effects ensuing from the stimulation itself. Pre-treatment

neuroimaging may eventually play a crucial role in optimizing stimulation parameters, to

maximize the chances of success in each patient presenting for treatment.

168

Study II: Baseline Predictors and Mechanisms of dmPFC-rTMS response in OCD

Introduction The research in this chapter was originally published as Dunlop, K., Woodside, B.,

Olmsted, M., Colton, P., Giacobbe, P., Downar, J. “Reductions in Cortico-Striatal

Hyperconnectivity Accompany Successful Treatment of Obsessive-Compulsive Disorder with

Dorsomedial Prefrontal rTMS,” Neuropsychopharmacology. 41, 1395-403.

OCD is a severely disabling psychiatric disorder with a lifetime prevalence of 1–3%

(Abramowitz et al, 2009). OCD is characterized by intrusive, anxiety-provoking, ego-dystonic

thoughts (obsessions), and associated repetitive behaviors (compulsions) (American Psychiatric

Association, 2013). Of the OCD patients, 40– 60% are refractory to conventional

pharmacological and behavioral therapies (Pallanti et al, 2002). It is therefore crucial to develop

novel therapies through a better understanding of the pathophysiology of OCD and the

mechanisms of successful treatment.

Previous human and animal studies suggest that abnormalities in the CSTC circuitry may

be central to OCD pathophysiology (Admon et al, 2015; Ahmari et al, 2013; Harrison et al,

2009; Menzies et al, 2008). In healthy humans, specific CSTC loop circuits are important for

self-regulation of affect, cognition, and behavior (van den Heuvel et al, 2010; Lipsman et al,

2013a; Marsh et al, 2009a). In OCD, these circuits display structural abnormalities relative to

controls: volumetric gray matter reductions and reduced white matter integrity in the anterior

cingulate cortex (Kühn et al, 2013), gray matter reductions in the orbitofrontal cortex (Rotge et

al, 2010), and gray matter increases in thalamus and ventral striatum (Hou et al, 2013). On

fMRI, abnormal cortical-ventral striatal hyperconnectivity has been observed in OCD during a

monetary incentive delay task (Beucke et al, 2012), during symptom provocation and during rest

in many studies (Anticevic et al, 2014; Cocchi et al, 2012; Figee et al, 2013; Harrison et al,

2009). One study has shown the opposite, corticostriatal hypoconnectivity, in a group of

unmedicated patients (Posner et al, 2014). In addition, altered anterior cingulate cortex metabolic

activity has been observed in OCD on fluorodeoxyglucose-positron emission tomography and

magnetic resonance spectroscopy (Ebert et al, 1997; O’Neill et al, 2013; Perani et al, 1995;

169

Saxena et al, 2004, 2009). Taken together, these observations delineate a possible

neuroanatomical substrate for OCD symptomatology.

Neuromodulation treatments offer a novel, anatomically targeted approach to refractory

psychiatric conditions. DBS has shown promising effects for OCD in recent studies (Greenberg

et al, 2010); several CSTC targets have been explored, including the anterior limb of the internal

capsule (Abelson et al, 2005), subthalamic nucleus (Mallet et al, 2008), and ventral striatum/

NAc (Denys et al, 2010; Figee et al, 2013). Regarding therapeutic mechanisms, a recent fMRI

study found that NAc-DBS normalized excessive functional connectivity between NAc and

dorsomedial and DLPFC in OCD patients; the degree of reduction correlated to the degree of

symptomatic improvement (Figee et al, 2013).

Although DBS has shown promising effects in severe, refractory OCD cases, noninvasive

forms of neuromodulation could be offered to a much wider range of patients. rTMS could

present a noninvasive alternative to DBS in OCD, if directed at a suitable stimulation target.

rTMS to the DLPFC, although successful in major depression (Berlim et al, 2014; O’Reardon et

al, 2007), has shown minimal clinical benefit over sham in double-blind trials for OCD (Alonso

et al, 2001; Sachdev et al, 2007). However, medial prefrontal targets appear more promising: 1

Hz rTMS of the SMA and pre-SMA has achieved substantial symptom improvement in case

reports and randomized controlled trials (Mantovani et al, 2006, 2010). Likewise, with

transcranial direct current stimulation, cathodal but not anodal stimulation of the SMA has been

reported to improve OCD symptoms (D’Urso et al, 2016).

A neighboring potential target is the dmPFC, just anterior to the pre-SMA. Abnormally

high resting state functional connectivity and gray matter volume reductions in dmPFC (Radua

et al, 2010) have been observed in OCD patients relative to controls. With NAc-DBS (Figee et

al, 2013), therapeutic efficacy correlated to reduction in excessive frontostriatal connectivity

through the dmPFC regions. dmPFC-rTMS has not yet been studied in OCD. However, in major

depressive disorder (MDD), a recent open-label study of 10Hz dmPFC-rTMS for treatment-

resistant MDD achieved ⩾50% symptom improvement in approximately half of the patients

(Downar et al, 2014). Notably, therapeutic effects correlated to change in CSTC connectivity

through the dmPFC on fMRI, as with NAc- DBS in MDD (Salomons et al, 2014). These

170

observations raise the possibility that dmPFC-rTMS might be able to engage a similar

therapeutic mechanism through noninvasive means in OCD.

In the current study, we used resting-state fMRI to characterize predictors and correlates

of treatment response to 20-30 sessions of open-label 10 Hz dmPFC-rTMS for treatment-

refractory OCD.

Aims & Hypotheses The general aim of Study II is to establish baseline predictors of, and neural

correlates/mechanisms of, OCD symptom improvements following 20-30 sessions of 10 Hz

dmPFC-rTMS.

Specific Aims of Study II 1. To determine whether 20-30 sessions of open-label 10 Hz dmPFC-rTMS improves the

severity of obsessive thoughts and compulsive behaviours in OCD.


significant differ between OCD dmPFC-rTMS responders and non-responders.


correlate with improvements in OCD symptom severity.

Hypotheses of Study II 1. The severity of obsessive thoughts and compulsive behaviours in OCD patients will

significantly improve in response to 20-30 sessions of open-label 10 Hz dmPFC-rTMS.

2. As with DBS, OCD patients with higher dmPFC-CSTC resting-state functional

connectivity will show better response to open-label 10 Hz dmPFC-rTMS.

3. As with DBS, OCD patients who show decreases in pre- to post-treatment dmPFC-CSTC

resting-state functional connectivity after dmPFC-rTMS will show more improvement in

OCD symptom severity.

171

Methods

Subjects

Twenty patients with a DSM-IV diagnosis of OCD participated in the study (males = 4;

mean age = 37.3 ± [standard deviation] 15.5; age range = 21 – 63 years). OCD and co-morbid

Axis I and Axis II disorders were diagnosed by a board-certified psychiatrist using the

aforementioned DSM-IV criteria during a semi-structured clinical psychiatric interview

incorporating the MINI. All patients reported at least one failed medication trial on clinical

interview, meaning that the individual was unable to tolerate or was clinically unresponsive to

the treatment (mean number of previously failed medication = 5.7 ± 4.1 trials). 19 of 20

participants reported at clinical interview that they had failed at least one attempt at cognitive or

behavioural invention, meaning that they did not complete the intervention, or that the individual

was nonresponsive clinical intervention. The mean OCD illness duration was 24.2 ± 15.2 years

(OCD illness duration range = 8 – 54), indicating that all patients enrolled in the study exhibited

chronic OCD. No patients exhibited hoarding symptomatology. Comorbidities including MDD

(n = 16), bipolar disorder (n = 2), anorexia nervosa (n = 5), bulimia nervosa (n = 4), post-

traumatic stress disorder (n = 4), and Tourette’s syndrome (n = 1). No patients had a history of

tics.

Patients were required to maintain a stable medication regimen for at least 4 weeks prior

to and throughout rTMS treatment. Current medications included neuroleptic agents (n = 12),

selective serotonin reuptake inhibitors (n = 8), serotonin-norepinephrine reuptake inhibitors (n =

3), trazodone (n = 2), lithium (n = 1) and benzodiazepines (n = 10). The maximum daily

benzodiazepine doses were 2mg clonazepam; 4.5mg bromazepam, and 0.75mg alprazolam.

Forty healthy controls (males = 17, mean age = 34.88 ± 11.76 years, age range = 18 – 66

years) were recruited as a comparator group for resting-state connectivity analyses. All controls

were screened by trained research staff. Participants were screened for current and previous

psychiatric illness including current substance abuse or dependence, contraindications to MRI,

and any psychotropic medications. There were no differences in age and sex between the OCD

and healthy control group (age t58 = 1.85; p = n.s.; sex χ2 = 2.97, p = n.s.). Healthy participants

underwent MRI, but did not undergo a course of rTMS.

172

All patients and healthy controls provided informed consent, and the study was approved

by the University Heath Network Research Ethics Board.

Clinical Measures

The primary clinical outcome measure was the Y-BOCS (Goodman et al, 1989), and was

collected at baseline (prior to rTMS treatment) and at 2-weeks post-treatment. In this study,

treatment response was defined as ⩾50% improvement on the Y-BOCS from baseline to 2-

weeks post-treatment.

Secondary clinical scales assessing comorbid depressive and anxiety symptomatology

were also collected, and included the 17-item HAMD (Hamilton, 1960), the BDI-II (Beck et al,

1961), and the BAI (Beck et al, 1988). Additional clinical variables collected included the

duration of illness, number of previous hospitalizations and outpatient treatment programs,

number of previous medication trials, and the current medication type and dosage. Kernel

density estimates of the distribution of clinical response on Y-BOCS was generated using

Stata13 (College Station, TX, USA).

Intervention

rTMS was delivered under MRI guidance using the MagPro R30 system equipped with a

Cool-DB80 coil (MagVenture, Farum, Denmark) and the Visor 2.0 neuronavigation system

(Advanced Neuro Technologies, Enschede, Netherlands). The Visor software was used for

anatomical landmarking and co-registration of the brain into the standard stereotaxic space of

Talairach and Tournoux, followed by segmentation of T1 images into scalp and brain

compartments, rendering of 3D surfaces for the scalp and brain, and co-registration of these

surfaces to the patient’s head for coil placement during treatment. As in our previous studies of

dmPFC-rTMS, we selected a dmPFC target region at the stereotaxic coordinate (X 0 Y+30 Z

+30) for stimulation and the scalp point closest to the coordinate (X 0 Y+60 Z+60) was used as

the focal point for coil placement during treatment. This point corresponds to approximately

25% of the distance from nasion to inion (slightly anterior to Fz according to the international

173

10-20 EEG electrode positioning system). By comparison, the dmPFC target is slightly anterior

to the location specified by previous authors targeting the pre-SMA for rTMS in OCD, which

was at 35% of the nasion-inion distance (Mantovani et al, 2010). The coil vertex was placed over

the interhemispheric fissure, immediately anterior to the central sulcus, under MRI guidance.

Preferential stimulation of left and right hemisphere was achieved by orienting the coil laterally,

with the current flow directed towards the hemisphere to be stimulated. Resting motor

thresholds were established based on activation of the contralateral extensor halluces longus,

determined by visual inspection.

During treatment, rTMS was delivered over the dmPFC scalp coordinate above, with

lateral coil orientation for preferential stimulation of the left then right dmPFC (Harmer et al,

2001; Terao et al, 2001) at 120% of the EHL resting motor threshold, at a stimulation frequency

of 10 Hz, with a duty cycle of 5 s on and 10s off, for 60 trains, resulting in 3000 pulses per

hemisphere in each session. Stimulation was delivered on weekdays for 20 daily sessions on

weekdays, with non-remitters offered extension to 30 sessions.

Neuroimaging Acquisition and Analysis

Patients underwent MRI sessions one week prior to and one week after rTMS treatment,

on a 3T GE Signa HDx scanner equipped with an 8-channel phased-array head coil, using a

protocol we have previously reported (Salomons et al, 2014). This comprised a T1-weighted fast

spoiled gradient-echo anatomical scan (TE = 12ms, TI = 300ms, flip angle=20°, 116 sagittal

slices, thickness = 1.5mm, no gap, 256 x 256 matrix, FOV = 240mm), followed by a 10-min

resting-state, eyes-closed T2* series (TE=30ms, TR = 2000ms, flip angle=85°, 32 axial slices,

thickness=5mm, no gap, 64x64 matrix, FOV=220mm, 300 TRs, 2 s temporal resolution).

MRI data preprocessing and seed-based region-of-interest analyses was implemented in

the FSL software package (Jenkinson et al, 2012), using FEAT (Beckmann et al, 2003) for the

following steps: discarding the first 5 volumes for signal stabilization, interleaved slice-timing

correction, tissue segmentation using BET (Jenkinson et al, 2002), motion correction using

MCFLIRT (Smith, 2002), and spatial smoothing (6 mm FWHM Gaussian kernel). A nuisance

linear regression analysis was conducted using 6 motion parameters and FAST (Zhang et al,

174

2001), extracted white matter and cerebrospinal fluid mean time series. Functional data was then

bandpass filtered (0.009-0.009 Hz) and co-registered to the MNI-152 standard atlas.

One a priori cortical and five exploratory subcortical ROIs were selected as seeds for

whole-brain functional connectivity analysis (Table 5-1, Figure 5-1). In cortex, a dmPFC ROI

were defined from a resting-state connectivity-based atlas (Craddock et al, 2012b), based on

proximity to the rTMS stimulation coordinate. Two other seeds were selected based on CSTC

connectivity in OCD and their therapeutic use in DBS: the medial dorsal thalamus (MDT), from

an thalamic connectivity atlas (Behrens et al, 2003), and the STN, from a probabilistic STN atlas

(Forstmann et al, 2012). Three exploratory bilateral ventral striatal sites (superior ventral

striatum [VSs], inferior ventral striatum [VSi] and ventral rostral putamen [VRP]) (Di Martino et

al, 2008) were selected to identify striatal connectivity associated with response. A medial dorsal

thalamus (MDT) ROI was created using the Oxford Thalamic Connectivity Atlas (Behrens et al,

2003), and a subthalamic nucleus (STN) ROI was created using a probabilistic subthalamic

nucleus atlas (Forstmann et al, 2012).

Each seed ROI was co-registered from standard space to each subject’s anatomical MRI

using the transformation matrix from the original registration to standard space in FLIRT, and its

mean time series was then extracted and used as a regressor in a first-level analysis to generate a

whole-brain map of voxels positively and negatively correlated to each ROI for each subject. To

localize regions where pre-treatment functional connectivity correlated to treatment response,

FSL’s FLAME mixed effects model (Beckmann et al, 2003) was then applied for the group-level

analysis using the responder/non-responder status of each subject as a categorical, group-level

regressor.

175

Table 5-1: Centre of gravity coordinates for regions of interest created from parcellation

atlases, and MNI coordinates for sphere-based regions of interest. The dmPFC ROIs was

used a priori, and the subcortical ROIs corresponding to DBS targets in OCD were used to

identify predictors and correlations in an exploratory analysis.

Seed Type of Seed Citation MNI

X Y Z dmPFC Parcellation Craddock, 2012 0 38 24

STN Parcellation Forstmann, 2012 0 -12 -8 MDT Parcellation Behrens, 2003 2 -22 10 VSs Coordinate di Martino, 2008 10 15 0 Vsi Coordinate di Martino, 2008 9 9 -8

VRP Coordinate di Martino, 2008 20 12 -3

dmPFC = dorsomedial prefrontal cortex; MDT = medial dorsal thalamus; MNI = Montreal

Neurological Institute; STN = subthalamic nucleus; VRP = ventral rostral putamen; VSi =

inferior ventral striatum; VSs = superior ventral striatum

176

Figure 5-1: Regions of interest created from parcellation atlases and coordinates. (A) The a

priori dmPFC parcellation-based region of interest derived from the Craddock (2012) atlas. (B)

The VSs (above) and VSi (below) regions of interest from the di Martino (2008) striatal

parcellation. (C) The MDT parcellation-based region of interest derived from the Behrens (2003)

thalamic parcellation. (D) The STN parcellation-based region of interest derived from the

Forstmann (2012) probabilistic parcellation of the STN. (E). The VRP region of interest from the

di Martino (2008) striatal parcellation.

To localize regions where the changes in functional connectivity from pre- to post-

treatment correlated to the degree of treatment response, a within-subjects, fixed-effects general

linear model analysis was performed for each subject and seed region. This resulted in a set of

first-level statistical parametric maps for each subject and each seed indicating regional increases

and decreases in functional connectivity to the seed from pre- to post-treatment. For group-level

analysis, these individual-subject change maps were then entered into a between-subjects mixed-

177

effects linear regression analysis, using the responder/non-responder status of each subject as a

categorical, group-level independent variable. Corrections for multiple comparisons were

performed using Gaussian random field theory (Z>1.96, cluster significance p<0.05 corrected).

Parameter estimates for individual subjects’ functional connectivity values (mean z-

score) of our seed and cluster regions were then extracted for post hoc analysis. An 8 mm sphere

centered at the peak cluster voxel (masked by the relevant cluster to ensure anatomical

specificity) was registered from standard space to each individual subject using the

transformation matrix from the original registration, extracting the mean z-score values from

relevant connectivity maps. Baseline z-scores were extracted from individual subjects’ baseline

functional connectivity maps. Change z-scores were extracted from individual subjects’ pre-

versus-post-treatment contrast maps.

Results

Clinical Outcomes

Subjects completed, on average, 21.3 ± 4.1 sessions of 10 Hz dmPFC-rTMS (range = 14

– 30 sessions). Treatment was well-tolerated, with no serious or treatment-limiting adverse

effects occurring. One subject discontinued treatment after 14 rTMS owing to non-response, and

was subsequently analyzed as a non-responder using baseline measures.

Across all subjects, baseline Y-BOCS scores significantly decreased from 30.5 ± 4.3 to

18.4 ± 10.8 (Wilcoxon rank-sum test, W18 = 3.41, p = 0.007) (Table 4-2). However, kernel

density estimation revealed a sharply bimodal response distribution, with distinct responder and

non-responder subpopulations (Figure 5-2); these subpopulations are therefore considered

separately hereafter. Ten of twenty subjects met the response criterion of greater than or equal to

50% of Y-BOCS improvement from baseline to 2-weeks post-treatment. Among responders, Y-

BOCS scores significantly decreased 67.2%, from a mean baseline score of 29.3 ± 4.6 to 9.6 ±

3.9 2 weeks post-treatment (W9 = 2.81, p = 0.005). Among non-responders, Y-BOCS scores

decreased non-significantly by 11.4%, from a mean baseline score of 31.7 ± 4.1 to 28.1 ± 7.8

178

(W8 = 1.45, p = 0.15). Of note, responders and non-responders did not differ in Y-BOCS severity

at baseline (29.3 ± 4.6 versus 31.7 ± 4.1; Mann-Whitney U18 = 1.21, p = 0.22).

Figure 5-2: Probability distribution function (a) and ranked individual-patient plot (b) of

treatment outcomes for dmPFC-rTMS in OCD. A bimodal distribution of treatment outcomes

is evident, suggesting distinct responder and non-responder subpopulations within the patient

sample. Outcomes are calculated as percent improvement in YBOCS scores from pre- to post-

treatment. One subject was a non-completer (14 rTMS sessions performed).

179

Table 5-2: Summary of Primary and Secondary Clinical Measures in All Patients, and in

Y-BOCS Responders and Non-Responders.

All OCD Patients Y-BOCS Responders

Y-BOCS Non-Responders

Baseline Y-BOCS 30.5 (4.3) 29.3 (4.6) 31.7 (4.1) Follow-up Y-BOCS 18.4 (10.8) 9.6 (3.9) 28.1 (7.8)

Baseline HAMD 17.7 (7.7) 15.8 (5.8) 18.0 (6.8) Follow-up HAMD 9.9 (7.3) 5.8 (4.0) 13.7 (7.6)

Baseline BDI 29.9 (15.8) 26.9 (18.4) 30.6 (16.3) Follow-Up BDI 19.4 (15.6) 8.7 (9.6) 29.7 (13.7) Baseline BAI 28.7 (15.0) 29.8 (18.3) 25.5 (13.1)

Follow-Up BAI 15.9 (12.4) 10.4 (10.9) 20.2 (13.4)

BAI = Beck Anxiety Inventory; BDI = Beck Depression Inventory; HAMD = Hamilton Rating

Scale for Depression; OCD = Obsessive-Compulsive Disorder; Y-BOCS = Yale-Brown

Obsessive Compulsive Scale. Note that all numbers indicate the mean (standard deviation).

Regarding secondary measures, across all subjects, depression severity significantly

improved (Table 5-2). HAMD severity significantly improved across all subjects, from a mean

baseline score of 17.7 ± 7.7 to 9.9 ± 7.3 (t18 = 3.08, p = 0.008). Similarly, BDI scored also

significantly improved across all subjects, from a mean baseline score of 29.9 ± 15.8 to 19.4 ±

15.6 (t18 = 3.27, p = 0.005). Again, outcomes were sharply dichotomous. Y-BOCS responders

improved significantly on HAMD from a mean baseline score of 15.8 ± 5.8 to 5.8 ± 4.0 at

follow-up (t6 = 3.64, p = 0.01). BDI scores among Y-BOCS responders followed a similar

pattern of change, with mean baseline scores significantly improving from 26.9 ± 18.4 to 8.7±9.6

post-treatment (t9 = 5.51, p = 0.0006). Y-BOCS non-responders showed no significant

improvement on HAMD, from 18.0 ± 6.8 to 13.7 ± 7.6 (t9 = 1.24, p = 0.25), or on BDI, from

30.6 ± 16.3 to 29.7 ± 13.7 (t8 = 0.37, p = 0.72). There was no difference in baseline depressive

severity between Y-BOCS responders and non-responders on both the HAMD (15.8 ± 5.8 versus

18.0 ± 6.8, t18 = 0.696, p = 0.49) and the BDI (29.6 ± 18.4 versus 30.6 ± 16.3, t18 = 0.476, p =

0.64).

Anxiety symptomatology on BAI also significantly improved in all subjects (Table 5-2),

from a mean baseline score of 28.7 ± 15.0 to 15.9 ± 12.4 at follow-up (t17 = 4.01, p = 0.0009),

180

and in Y-BOCS responders (from 29.8 ± 18.3 to 10.4 ± 10.9, t8 = 5.11, p = 0.0009). Y-BOCS

non-responders showed a non-significant improvement on BAI, from a mean baseline score of

25.5 ± 13.1 to 20.2 ± 13.4 (t8 = 1.47, p = 0.18). There was no difference in baseline anxiety

severity between Y-BOCS responders and non-responders on the BAI (29.8 ± 18.3 versus 25.5 ±

13.1, t18 = 0.602, p = 0.55).

A Spearman’s rank correlation coefficient was performed on baseline psychometric and

clinical measures to determine whether any clinical factors predicted treatment outcome. No

baseline factors, including number of rTMS sessions, age, baseline severity, duration of illness,

number of failed medications or behavioural therapies, co-morbidities, current medications, or

baseline HAMD, BDI or BAI showed a significant correlation to treatment outcome, either

before or after Bonferroni correction for multiple comparisons.

Resting-State fMRI Predictors of Treatment Response

Resting-state functional connectivity from the dmPFC ROI selected a priori did not

significantly differ between treatment responders and non-responders at baseline. The following

section reports results from exploratory subcortical ROIs.

From the VRP seed, responders were characterized by significantly higher pre-treatment

functional connectivity to the left dorsolateral and dorsomedial prefrontal cortex (Figure 4-3),

and significantly lower functional connectivity to the right posterior insula, superior temporal

gyrus, supramarginal gyrus, pre-central gyrus and post-central gyrus (Table 5-3). Baseline VRP-

dmPFC connectivity was significantly higher in responders than in non-responders (t18 = 2.2, p =

0.04), or healthy controls (t48 = 2.02, p = 0.05) (Figure 5-3B). Of note, higher baseline VRP-

dmPFC connectivity significantly predicted a greater percent improvement in YBOCS score

following treatment (Pearson r17 = 0.51, p = 0.03). Baseline VRP-dmPFC connectivity was

positive in responders (z-statistic = 2.81 ± 0.80), slightly negative in non-responders (z = -0.62 ±

1.33) and positive in healthy controls (z = 1.20 ± 0.34).

From the thalamic (MDT) seed, responders were characterized by significantly higher

pre-treatment functional connectivity to the bilateral pregenual cingulate, posterior cingulate and

181

precuneus, and medial/lateral orbitofrontal cortex, and lower connectivity to the bilateral insula,

temporal pole, left hippocampus and pre-central gyrus (Table 5-3, Figure 5-3C). Baseline MDT-

right OFC functional connectivity was significantly higher in responders versus non-responders

(t18 = 2.53, p = 0.02) (Figure 4-3D), and significantly predicted percent improvement in Y-BOCS

score following treatment (r17 = 0.56, p = 0.01). Likewise, baseline MDT-left OFC connectivity

was also significantly higher in responders versus non-responders (t18 = 2.58, p = 0.02) and

significantly predicted percent Y-BOCS improvement (r17 = 0.63, p = 0.004). MDT-left OFC

connectivity was positive in responders (z = 0.70 ± 0.75) and negative in non-responders (z = -

2.13 ± 0.36). Neither group differed significantly from healthy controls in MDT-left OFC or

right OFC connectivity.

For the STN seed, responders were characterized by significantly higher pre-treatment

functional connectivity to the bilateral thalamus and caudate nucleus, and lower functional

connectivity to the bilateral insula, supplementary motor area, right pre-central gyrus and left

temporal pole (Table 5-3, Figure 5-3E). Baseline STN-thalamus functional connectivity was

significantly higher in responders versus non-responders (t18 = 2.36, p = 0.03) (Figure 5-3F) and

significantly predicted percent Y-BOCS improvement (r17=0.57, p=0.01). STN connectivity to

the thalamus was positive in both responders (z = 6.35 ± 0.73) and non-responders (z = 2.61 ±

0.84). Likewise, baseline STN-caudate connectivity also significantly predicted percent Y-BOCS

improvement (r17 = 0.48, p = 0.04). STN connectivity to the caudate nucleus was positive in

responders (z = 1.73±1.08) and negative in non-responders (z = -1.65±0.56). Neither group

differed significantly from healthy controls in STN-caudate and STN-thalamus connectivity.

For the VSi seed, responders had significantly higher pre-treatment functional

connectivity to the bilateral precuneus and posterior cingulate gyrus, and lower functional

connectivity to the superior posterior cingulate gyrus, right supramarginal gyrus and temporal

pole relative to non-responders (Table 5-3). For the VSs seed, responders had significantly

higher functional connectivity to the precuneus and posterior cingulate gyrus, and lower

connectivity to the superior temporal gyrus and superior posterior cingulate gyrus (Table 5-3).

182

Figure 5-3: High baseline frontal-striatal-thalamic-subthalamic connectivity predicts

response to dmPFC-rTMS in OCD. Bar graphs are intended to convey the absolute magnitudes

of the parameter estimates in each group, as complementary information for the difference maps.

A. Regions of higher baseline functional connectivity (orange) to the bilateral VRP seed (green)

in rTMS-responders vs. non-responders. B. Parameter estimates between the VRP and dmPFC

for baseline healthy control, responder and non-responder groups. C. Regions of higher baseline

functional connectivity (orange) to the bilateral MDT seed (green) in rTMS-responders vs. non-

183

responders. D. Parameter estimates between the MDT and OFC for baseline healthy control,

responder and non-responder groups. E. Regions of higher baseline functional connectivity

(orange) to the bilateral STN in rTMS responders vs. non-responders. F. Parameter estimates

between the STN and Thalamus for baseline healthy control, responder and non-responder

groups.

184

Table 5-3: Brain regions where baseline resting-state functional connectivity to exploratory

seed regions differed significantly between responders and non-responders. All activations

are whole-brain Gaussian random field theory corrected for multiple comparisons at a cluster

threshold p<0.05, but do not meet across-seed FDR-correction.

Seed Region Brodmann Area

MNI Coordinate Peak Z Score X Y Z

MDT Response > Nonresponse

B Pregenual Cingulate 9

-2 46 12 3.05

L Medial, Lateral OFC 10, 11, 25

-12 26 -22 3.35

R Lateral OFC 10, 11

40 50 -6 3.38

B Posterior Cingulate Gyrus 23, 30, 31

-2 -66 12 3.70

B Precuneus 7, 31

8 -70 38 4.18

Nonresponse > Response

R Temporal Pole, Superior/Middle Gyrus

52 -14 -8 3.50

L Temporal Pole 22

-56 8 -2 3.65

L Precentral Gyrus 3, 4

-54 -8 40 3.77

L Insula 13

-36 -8 -8 3.91

R Insula 13

36 6 -4 2.51

L Hippocampus

-34 -12 -20 2.90

STN Response > Nonresponse

R Thalamus

16 -22 4 3.99

L Thalamus

-4 -22 14 4.06

R Caudate Nucleus

14 10 14 3.36

L Caudate Nucleus

-18 16 10 3.13


L Temporal Pole, Inferior Frontal Gyrus

-42 18 16 4.02

L Insula 13

-48 2 -8 3.94

B SMA

-10 8 64 3.50

R Precentral Gyrus

34 -12 58 4.53

R Insula 13

44 -12 8 3.06

VRP Response > Nonresponse

L Dorsomedial Prefrontal Cortex 9, 32 -10 32 26 3.19

L Dorsolateral Prefrontal Cortex 6 -30 16 42 3.26


R Posterior Insula 13

38 -8 8 3.77

R Posterior Superior Temporal Gyrus 41

58 -20 8 3.62

R Supramarginal Gyrus 40

62 -20 22 3.20

R Pre/Postcentral Gyrus 3, 4

60 12 26 3.91

VSi Response > Nonresponse

B Precuneus 23, 31

-4 -58 28 3.05

B Posterior Cingulate Gyrus 30, 31

-4 -42 18 3.10


R Supramarginal Gyrus 40

64 -20 28 5.60

B Dorsal Posterior Cingulate Gyrus 31 6 -20 40 4.44

R Temporal Pole 22

62 10 -2 4.40

VSs Response > Nonresponse

B Precuneus, PCC 23, 30, 31

-6 -56 16 3.24


R STS 40, 42

58 -30 10 3.97

L STS 40, 42

-56 -32 14 3.20

R Posterior Cingulate Gyrus 31

6 -16 50 3.79

185

MDT = medial dorsal thalamus; MNI = Montreal Neurological Institute; OFC = orbitofrontal

cortex; SMA = supplementary motor area; STN = subthalamic nucleus; VRP = ventral rostral

putamen; VSi = inferior ventral striatum; VSs = superior ventral striatum.

186

Resting-State fMRI Correlates of Treatment Response

Comparisons of pre-treatment and post-treatment resting-state fMRI revealed significant

differences in how functional connectivity to the seed regions changed over the course of

treatment (Table 5-4). For the dmPFC seed, successful treatment response was associated with

increased functional connectivity to the bilateral pre- and post-central gyrus and left precuneus,

and decreased functional connectivity to the bilateral caudate nucleus, midbrain, thalamus,

superior frontal gyrus, and right hippocampus (Table 5-4 Figure 5-4A). Responders showed

significant decreases in dmPFC functional connectivity to the bilateral caudate (pre-treatment z =

4.81 ± 1.27, post-treatment z = 1.65 ± 1.16, t8 = 2.61, p = 0.03), and thalamus (pre-treatment z =

0.64 ± 1.24, post-treatment z = -3.37 ± 1.01, t8 = 2.57, p = 0.03). Conversely, non-responders

showed significant increases in functional connectivity from the dmPFC to the hippocampus

(pre-treatment z = -3.19 ± 0.83, post-treatment z = -0.86 ± 0.88, t6 = 3.24, p = 0.01) and midbrain

(pre-treatment z = -3.71 ± 1.00, post-treatment z = -1.20 ± 0.86, t6 = 2.50, p = 0.04). Again,

across all subjects, the percent Y-BOCS improvement correlated significantly to the degree of

reduction in dmPFC functional connectivity to the caudate (r17 = -0.56, p = 0.02) and

hippocampus (r17 = -0.58, p = 0.02).

Compared with healthy controls, responders’ dmPFC-caudate connectivity was

significantly higher at baseline (control dmPFC-caudate z = 3.04 ± 0.85, baseline responders z =

4.81 ± 1.27, t47 = 2.05, p = 0.05) and non-significantly higher compared with non-responders

(baseline non-responder dmPFC-caudate z = 2.34 ± 1.16, t15 = 2.01, p = 0.06) (Figure 5-4B).

The results for the remaining exploratory seeds are found in Table 5-5. For the VRP seed,

increases in functional connectivity to the bilateral brainstem, thalamus and insula and decreases

in functional connectivity to the bilateral dmPFC were associated with successful response to

treatment (Table 5-5). Compared to healthy controls, responders’ VRP-dmPFC connectivity was

significantly higher at baseline (control z-score = 0.57 ± 0.31, t48 = 2.12, p = 0.04) but

significantly lower after treatment (t47 = 2.61, p = 0.01), with significant reductions in VRP-

dmPFC functional connectivity over treatment (pre-treatment z = 2.47 ± 0.76, post-treatment z =

-2.05 ± 0.81, t9 = 3.99, p = 0.004). In contrast, non-responders did not show significantly higher

VRP-dmPFC connectivity than healthy controls at baseline (t45 = 0.50, p = 0.68), and did not

show any significant change in VRP-dmPFC connectivity over the course of treatment (pre-

187

treatment z = 0.45 ± 1.09, post-treatment z = 1.35 ± 0.71, t6 = 0.88, p = 0.41). However, non-

responders did continue to show significantly higher VRP-dmPFC connectivity than both healthy

controls and responders after the course of treatment (t45 = 2.45, p = 0.02). Across all subjects,

the degree of reduction in VRP-dmPFC functional connectivity significantly correlated with

percent Y-BOCS improvement (r15 = -0.64, p = 0.008)

Table 5-4: Brain regions where the pre-to-post treatment change in functional connectivity

to the dmPFC that differed significantly between rTMS responders and non-responders.

Seed Region Brodmann Area MNI Coordinate Peak Z

Score X Y Z dmPFC

FC increase in Resp > Nonresp

L Precuneus 7

-6 -48 62 3.93

L Postcentral Gyrus 7

-4 -44 70 4.23

L Precentral Gyrus 6

-16 -30 54 3.83

R Precentral Gyrus 6

4 -20 60 2.87

R Postcentral Gyrus 7

14 -36 48 2.74

FC reduction in Resp > Nonresp

L Superior Frontal Gyrus 6

-10 14 64 4.84

R Superior Frontal Gyrus 6

12 26 58 3.33

L Caudate Nucleus

-8 14 6 2.72

R Caudate Nucleus

6 14 8 3.40

R Thalamus

22 -4 8 3.07

L Thalamus/Putamen

-6 -14 2 3.02

R Hippocampus

30 -22 -8 3.39

B Dorsal Midbrain

-6 -28 -10 2.57

B Ventral Midbrain

-2 -16 -16 2.61

dmPFC=dorsomedial prefrontal cortex; FC=functional connectivity; MNI=Montreal

Neurological Institute; Nonresp=non-responder; Resp=responder

188

Figure 5-4: Reductions in cortical-striatal-thalamic connectivity correlate to improvements

in OCD symptoms following dmPFC-rTMS. Bar graphs are intended to convey the absolute

magnitudes of the parameter estimates in each group, as complementary information for the

difference maps. (a) Regions of significant reduction in functional connectivity to the dmPFC

seed (green) in rTMS-responders vs non-responders are shown in blue. (b) Parameter estimates

of functional connectivity between the dmPFC and caudate nucleus for healthy controls,

responders, and nonresponders. Time 1, pre-treatment; Time 2, post-treatment.

189

Table 5-5: Brain regions where the pre-to-post treatment change in functional connectivity

to exploratory ROIs that differed significantly between rTMS responders and non-

responders. All activations are Gaussian random field theory corrected for multiple comparisons

at a cluster threshold p<0.05, but do not meet across-seed FDR-correction.

Seed

Region BA MNI Coordinate Peak

Z Score X Y Z

MDT


L Putamen

-32 -16 10 2.82

L Amygdala, Hippocampus

-26 -4 -16 2.8

L Parahippocampal Gyrus

-18 -24 -22 3.36


L Caudate Nucleus

-20 10 16 3.57

B Cuneus, Precuneus

0 -72 26 3.33

L Intracalcarine Cortex 30

-10 -70 16 3.29

STN


R Precuneus, PCC 31

12 -62 32 3.85

VRP


B Brainstem

-2 -22 -24 3.1

L Thalamus

-8 -26 0 2.72

R Thalamus

12 -26 0 3.08

L Insula

-36 -4 2 3.04

R Insula

42 2 -10 3.57


B Dorsomedial Prefrontal Cortex 32

-10 18 42 3.62

L Supracalcarine Cortex 17

-4 -90 10 3.86

R Occipital Pole 19

-4 -98 -2 3.75

VSi


R Middle Temporal Gyrus (Posterior) 12 62 -36 -10 3.76

R Posterior Insula 13

40 -4 -10 3.06


L Occipital Pole 18

-6 -96 -6 4.4

L Cuneus 7, 19

-8 -88 38 3.67

VSs


L Insula 13

-38 0 -8 2.95

L Temporal Pole, Amygdala 34

-38 4 -20 3.64

R Temporal Pole, Inferior Temporal Gyrus 20 48 -32 -20 3.22

R Insula 13

40 2 -10 3.64


B Dorsomedial Prefrontal Cortex 32

-8 18 42 4.51

B Dorsomedial Prefrontal Cortex 9, 32 -10 30 30 4.35

FC=functionalconnectivity;MDT=medialdorsalthalamus;MNI=MontrealNeurologicalInstitute;OFC=orbitofrontalcortex;PCC=posteriorcingulatecortex;SMA=supplementary

190

motorarea;STN=subthalamicnucleus;VRP=ventralrostralputamen;VSi=inferiorventralstriatum;VSs=superiorventralstriatum

191

For the VSs seed, increased functional connectivity to the bilateral insula and temporal

pole and decreased functional connectivity to the bilateral dmPFC were associated with

successful treatment response (Table 5-5). In responders, VSs-dmPFC connectivity decreased

significantly from pre- to post-treatment (pre-treatment z=1.33±1.29, post-treatment z=-

1.92±0.80, t8=2.75, p=0.03), and was significantly lower than in healthy controls following

treatment (control z=1.03±0.39, t47=2.87, p=0.01).

Conversely, in non-responders, VSs-dmPFC functional connectivity showed a non-

significant trend toward increasing from pre- to post-treatment (pre-treatment z=0.21±1.05, post-

treatment z=1.79±0.60, t6=1.40, p=0.21), and did not differ significantly from healthy controls

following treatment (t45=0.54, p=0.61). Once again, across all subjects, the degree of reduction in

VSs-dmPFC functional connectivity significantly correlated to percent YBOCS improvement

(r15=-0.65, p=0.007).

For the VSi, treatment response was associated with increased connectivity to the right

middle temporal gyrus and posterior insula and decreased connectivity to the left occipital pole

and cuneus (Table 5-5). For the MDT, treatment response was associated with increased

functional connectivity to the left putamen, amygdala, hippocampus, and parahippocampal

gyrus, and decreased functional connectivity to the left caudate and bilateral cuneus and

precuneus (Table 5-5). For the STN, treatment response was associated with decreased

functional connectivity to the right precuneus and posterior cingulate cortex (Table 5-5).

Discussion & Conclusion To our knowledge, this is the first case series using fMRI to identify neural predictors and

correlates of response to any form of non-invasive brain stimulation in OCD. Previous studies of

rTMS in OCD have encountered little therapeutic benefit with stimulation of lateral prefrontal

targets such as the DLPFC (Alonso et al, 2001; Prasko et al, 2006; Sachdev et al, 2007; Sarkhel

et al, 2010), but somewhat more success with medial prefrontal targets such as the SMA or pre-

SMA (Kumar and Chadda, 2011; Mantovani et al, 2006, 2010). The results of the present study

are congruent with this latter literature in suggesting that rTMS of a slightly more anterior medial

prefrontal target, the dmPFC, can also yield substantial symptom reduction in a proportion of

192

OCD cases, even when multiple previous medication trials have failed. However, as previously

observed in patients with MDD (Downar et al, 2014), dmPFC-rTMS outcomes appear sharply

bimodal in OCD.

In keeping with our first hypothesis, responders showed significant differences from non-

responders in baseline dmPFC-striatal resting-state functional connectivity. Furthermore, in

keeping with our second hypothesis, the amount of symptomatic improvement correlated to the

degree of reduction in functional connectivity through a very similar frontal-striatal-thalamic-

subthalamic circuit connecting the dmPFC to VS, caudate nucleus, thalamus, and midbrain. In

relation to healthy controls, responders initially showed hyperconnectivity from dmPFC to

caudate before treatment, and this hyperconnectivity normalized following treatment; no such

pattern was evident in non-responders.

These observations are remarkably consistent with the recently reported results of an

fMRI study of therapeutic mechanisms in OCD, using ventral striatum-DBS rather than rTMS of

the dmPFC (Figee et al, 2013). In that study, OCD patients also showed baseline functional

hyperconnectivity from ventral striatum to a dmPFC region just immediately anterior to the

region identified in Figure 4-4 of the present study. DBS reduced functional connectivity from

ventral striatum to dmPFC, and the degree of reduction correlated well with the amount of

symptomatic improvement. The results of the present study suggest that similar therapeutic

effects in OCD might also be achieved via noninvasive stimulation of the more superficial

dmPFC, via a similar mechanism. Notably, the circuit identified in the present study also

includes nodes in lateral orbitofrontal cortex, ventral striatum, medial thalamus, and subthalamic

nucleus, each of which has shown promise as a therapeutic target for either DBS (Abelson et al,

2005; Denys et al, 2010; Lipsman et al, 2013a; Mallet et al, 2008), or in the case of lateral

orbitofrontal cortex, rTMS (Ruffini et al, 2009).

More generally, the observations of the present study support the growing body of

evidence that OCD pathophysiology may arise from functional hyperconnectivity through

specific cortico-striatal loop circuits projecting from ventral striatum to the medial PFC (Beucke

et al, 2013; Harrison et al, 2009; Sakai et al, 2011). Recent studies, in larger patient samples,

have even suggested that different OCD symptom dimensions may map on to anatomically

distinct corticostriatal pathways (Harrison et al, 2013). Animal studies also suggest that the

193

dynamics of the hyperactivation may be important. For example, a recent study showed that

repeated exposures to brief optogenetic hyperactivation of an orbitofrontal cortex-ventral striatal

circuit in mice led to progressively increased compulsive grooming behaviors that were

reversible by fluoxetine, a standard pharmacotherapy for OCD (Ahmari et al, 2013). These more

nuanced techniques could lead to a better understanding of the variability of response to current

neuromodulation therapies in OCD.

We also note that the fMRI predictors and correlates of successful 10 Hz dmPFC-rTMS

for OCD appear opposite to those we have previously reported for the same intervention in major

depression not directly but indirectly, by relieving intrusive thoughts, as suggested by a recent

study (Salomons et al, 2014). In the MDD patients, low baseline connectivity from dmPFC to the

putamen and thalamus predicted better response to treatment, and the degree of increase in

frontal-striatal-thalamic connectivity correlated to symptomatic improvement. The results of the

present study suggest that dmPFC-rTMS may exert therapeutic effects via a similar CSTC

pathway in both MDD and OCD, but via opposite mechanisms (ie, via reduction of a

pathologically high baseline CTSC connectivity in OCD, rather than via strengthening of a

pathologically low baseline CTSC connectivity in MDD). These seemingly contrary findings

could potentially be reconciled if dmPFC-rTMS exerts its therapeutic effects on OCD and MDD

not directly but indirectly, by relieving intrusive thoughts, as suggested by a recent study (Carew

et al, 2013). However, this issue will require further study.

One important difference in technique between the present study and previous studies of

rTMS in OCD targeting the medial wall is that the previous studies used 1 Hz rather than 10 Hz

stimulation. Classically, 1 Hz stimulation is considered inhibitory, and 10 Hz stimulation

excitatory (Hallett, 2007); suppression of overactive regions in SMA and pre-SMA provided a

rationale for using 1Hz stimulation in these previous studies (Mantovani et al, 2006, 2010). Yet

it is increasingly recognized that the effects of many rTMS protocols, including both 1 and 10 Hz

stimulation, can be quite heterogeneous: a substantial proportion of individuals show

‘paradoxical’ excitatory responses to 1 Hz stimulation or inhibitory responses to high-frequency

stimulation, both on motor evoked potentials (Hamada et al, 2013; Maeda et al, 2000) and on

resting-state fMRI (Eldaief et al, 2011). In the present study, as well as in our previous study of

dmPFC- rTMS in MDD, the effects of 10 Hz stimulation on cortical- striatal-thalamic activity

were in fact quite variable across individuals, with ‘paradoxical’ reductions in functional

194

connectivity appearing in up to 40% of MDD patients in the previous study and in more than half

of the OCD patients in the present study (Figure 3). Thus, it is possible that 1 Hz SMA-rTMS

and 10 Hz dmPFC-rTMS are achieving successful OCD treatment outcomes via similar

mechanisms, notwithstanding the differences in technique. Alternatively, individual patients may

require different sites or patterns of stimulation to achieve a therapeutic effect, as has been

reported for DLPFC-rTMS in depression (Speer et al, 2009). The question of whether 1 Hz

SMA-rTMS and 10 Hz dmPFC-rTMS treat similar or different subpopulations of OCD patients

will be an important topic for future study. In either case, resting-state fMRI is likely to have an

important role both in stratifying patients and in clarifying therapeutic mechanisms.

One limitation of the present study is use of an open-label design, leaving open the

possibility that the observed symptomatic improvements were due to non-specific or placebo

effects. However, it should be noted that in the setting of refractory OCD, previously reported

effects of sham rTMS are relatively minimal, ranging from 1% to ~20% improvement in

YBOCS scores; indeed, even active stimulation of lateral targets such as DLPFC has consistently

achieved less than 25% YBOCS improvement across several independent studies (Alonso et al,

2001; Mantovani et al, 2010; Sachdev et al, 2007; Sarkhel et al, 2010). Thus, it is unlikely that

placebo effects can fully account for the present observations of a ~ 40% overall improvement in

the present study, a bimodal outcome distribution, distinct patterns of functional connectivity

through the dmPFC-ventral striatal target circuit in responders vs non-responders (who were not

otherwise distinct on clinical measures), distinct patterns of change in this circuit in responders

and non-responders, and the concordance of these results with the previous findings of an

independent study using DBS rather than rTMS. Nonetheless, replication of the present findings

under sham-controlled conditions would be an important next step in this line of research.

Another limitation is the lenient multiple comparisons correction performed. As more

recent publications have recommended stringent correction for multiple comparisons of resting-

state functional connectivity (Eklund et al, 2016), the risk for Type I errors in these studies is

potentially high. However, a number of factors mitigate concern regarding this methodological

limitation. Most importantly, the neuroimaging results across these three independent samples

are consistent. Second, the baseline frontostriatal predictor of treatment response is consistent

with Study III (which used more stringent correction). Finally, as evidenced in Sections 1.4, 1.5,

195

1.6, 1.8 and 1.9 of this thesis, the neuroimaging results are consistent with disorder etiology and

mechanisms of treatment response.

Another potential criticism of the current study relates to the small sample size. Although

the sample size here is comparable with or larger than that in other dmPFC rTMS-MRI and

DBS-MRI studies (Figee et al, 2013; Salomons et al, 2014), open-label and randomized control

rTMS treatment studies for OCD (Mantovani et al, 2006, 2010) and neuroimaging studies for

OCD (Admon et al, 2012), it is nonetheless underpowered to capture the full heterogeneity of

OCD symptomatology, comorbidity, and treatment types. Hence, the present observations are as

yet insufficient to properly address the question as to whether dmPFC-rTMS response could also

be predicted on the basis of clinical features such as OCD subtype (eg, hoarding vs checking),

presence or absence of comorbid symptoms (eg, major depression, tics), or adjunctive treatment

types (eg, use of SSRIs or neuroleptics). They also cannot yet address the question of whether

dmPFC-rTMS might selectively improve some OCD symptom dimensions but not others, as

might be suspected if different symptom clusters map reliably on to distinct neural substrates.

Resolution of such issues must await a larger sample of patients.

In summary, 10 Hz dmPFC-rTMS may offer a promising, noninvasive therapeutic option

for medication-refractory OCD, achieving ⩾ 50% reductions in YBOCS scores in 50% of the

patients in the sample. In agreement with recent findings with ventral striatal DBS for OCD,

rTMS may be most effective in patients with greater hyperconnectivity between dmPFC and

ventral striatum on resting-state fMRI. Also in agreement with DBS findings, therapeutic effects

of dmPFC-rTMS correlated with reductions in dmPFC-ventral striatum functional connectivity.

A randomized controlled trial incorporating a sham rTMS arm would be a logical next step in

evaluating dmPFC-rTMS as a noninvasive alternative to DBS in medication-refractory OCD.

Future studies of rTMS in OCD may also benefit from using fMRI to characterize cortico-striatal

connectivity prior to treatment, in order to predict response and/or tailor the stimulation target

and parameters in individual patients (Fox et al, 2013a). Properly optimized, rTMS could evolve

into a potent, novel treatment option for patients faced with this challenging and crippling illness.

196

Study III: Baseline Predictors and Mechanisms of High- and Low-Frequency dmPFC-rTMS in TRD under triple-blind sham controlled settings

Introduction TRD is associated with high rates of disability and mortality (Furman et al, 2011;

Kerestes et al, 2015). Conventional treatment efficacy is also limited; approximately one third of

MDD patients will fully respond to an initial antidepressant medication treatment, another third

will respond to a subsequent pharmacotherapy intervention, while the remaining third of MDD

sufferers are considered treatment-resistant (Ferrari et al, 2013; Murray and Lopez, 1997). In

sum, further research must be done to identify targeted treatments for both MDD and treatment-

resistant MDD.

One novel, targeted treatment option is repetitive transcranial magnetic stimulation

(rTMS); rTMS stimulates specific cortical targets addressing pathological circuitry in MDD. One

proposed cortical target for MDD is the dmPFC. The dmPFC is active during cognitive control

(Rush et al, 2006), emotion regulation (Brass and Haggard, 2007; Campbell-Meiklejohn et al,

2008; Cho et al, 2013) and the inhibition of prepotent responses (Kühn et al, 2011). Furthermore,

non-invasive stimulation of the medial prefrontal wall enhances inhibition on impulsivity-related

tasks in healthy controls (Cho et al, 2015; Jung et al, 2014; Müller et al, 2015). Taken together,

the therapeutic mechanism across psychiatric disorders involving dmPFC dysregulation may

modulate cognitive or behavioural phenotypes of impulsivity or emotional regulation, thereby

improving clinical symptoms and quality of life.

Structural and functional MRI studies reveal that the volume of the dmPFC is reduced

relative to healthy controls in MDD (Cho et al, 2015; Obeso et al, 2013), bipolar disorder (Bora

et al, 2012a), OCD (Bora et al, 2012b), binge eating disorder (Radua et al, 2010), and post-

traumatic stress disorder (Schäfer et al, 2010). The neuroanatomical overlap provides a potential

explanation for the frequent comorbidity between AN/BN and OCD, and mood disorders. It also

suggests that neuromodulatory treatments targeted towards the dmPFC could be more effective

for a wider range of symptoms (Kühn and Gallinat, 2013).

197

In two open-label studies and two case series, we recently showed that 20-30 sessions of

high-frequency dmPFC-rTMS for MDD was a promising treatment for TRD (Bakker et al, 2015;

Downar et al, 2014; Salomons et al, 2014; Schulze et al, 2018). Our first case series reported on

dmPFC-rTMS as a treatment for 185 treatment-resistant MDD patients; the response and

remission rate on the Beck Depression Inventory (BDI-II) was 40.6% and 29.2%, respectively

(Bakker et al, 2015). Our more recent case series in 130 patients showed that the clinical efficacy

of twice-daily and once-daily sessions of open-label dmPFC-rTMS did not significantly differ

(Schulze et al, 2018). Clinical factors, such as primary diagnosis, age, antidepressant medication,

and comorbidity, did not significantly predict response in TRD.

We then proceeded to examine resting-state functional magnetic resonance imaging (rs-

fMRI) correlates at baseline that predicted treatment response. Baseline resting-state

neuroimaging findings revealed differences in functional connectivity between responders and

non-responders to dmPFC-rTMS in MDD. In MDD, higher baseline dmPFC-dgACC and lower

dmPFC-dorsomedial thalamus functional connectivity was associated with response to treatment

(Salomons et al, 2014). In addition, altered dorsomedial and ventromedial functional

connectivity on resting-state graph theory measures predicted treatment response and levels of

anhedonia (Downar et al, 2014). Baseline differences in responders and non-responders to

dmPFC-rTMS show functional differences in regions previously implicated in reward

processing, and emotion regulation.

We have also examined how changes in rs-fMRI connectivity correlated with response to

treatment in MDD. We found that increased dmPFC-mediodorsal thalamus connectivity and

decreased sgACC-caudate connectivity correlated treatment response (Salomons et al, 2014).

Generally, connectivity alterations from the stimulation site with sites in cortico-basal ganglia-

thalamic-cortical circuits potentially reflect increased top-down cognitive control over regions

associated with reward and emotion processing, and impulsivity.

The therapeutic effects of dmPFC-rTMS and its rs-fMRI correlates were highly

heterogeneous across individuals in the aforementioned studies. To address this heterogeneity

and improve outcomes, rTMS protocols that theoretically elicit different cortical responses could

be compared in terms of both clinical efficacy and effects on neural activity; such comparisons

could help to identify mechanisms of response, and to optimize treatment parameters.

198

Low-frequency (1 Hz) and high-frequency (20 Hz) protocols are conventionally

considered to have inhibitory and excitatory effects on neural activity, respectively (although a

substantial proportion of individuals show effects in the opposite direction: Maeda et al., 2000;

Eldaief et al, 2011). These protocols have both been used in numerous studies comparing

treatment outcomes in order to optimize the treatment protocol at a given stimulation site, and

determine mechanisms of clinical response (e.g., Kim et al, 2014). Functional MRI can also be

used to identify baseline differences that distinguish treatment response from non-response on a

given protocol. Here, we plan to use these two ‘opposite’ rTMS protocols to better characterize

the predictors and mechanisms of dmPFC-rTMS treatment response first identified in our pilot

findings.

This study also includes a sham arm, to build upon the open-label findings of our

previous work by assessing clinical efficacy and determining whether the neural predictors and

correlates of response are treatment-specific or non-specific. Currently, there is only one double-

blind placebo-controlled trial of dmPFC-rTMS, which included only 45 patients over 3 arms

(DLPFC-rTMS, dmPFC-rTMS, and sham), and applied only 15 sessions of treatment (Kreuzer et

al, 2015), which is insufficient to achieve full effect based on other case series of longer duration

(e.g., Bakker et al, 2015).

Aims & Hypotheses The general aim of Study III is to establish baseline predictors of and mechanisms of

improvements in TRD symptom severity following 30 sessions of 20 Hz dmPFC-rTMS under

placebo-controlled conditions.

Specific Aims of Study III 1. To determine whether the clinical efficacy of active 20 Hz dmPFC-rTMS is superior to

that of 1 Hz active dmPFC and sham dmPFC-rTMS.

199

2. To replicate previously identified patterns of baseline dmPFC resting-state functional

connectivity that significantly differ between TRD dmPFC-rTMS responders and non-

responders.

3. To replicate previously identified patterns of pre- to post-treatment changes in resting-

state functional connectivity that correlate with improvements in TRD severity.

Hypotheses of Study III 1. Improvements in depression symptoms with 20 Hz dmPFC-rTMS will be significantly

higher than that achieved with 1 Hz dmPFC-rTMS or placebo dmPFC-rTMS.

2. TRD patients with lower dmPFC-CSTC and higher dmPFC-VMPFC resting-state

functional connectivity will show greater symptom improvement with active 20 Hz

dmPFC-rTMS, but not with active 1 Hz or placebo dmPFC-rTMS.

3. TRD patients showing greater increases in pre- to post-treatment dmPFC-CSTC resting-

state functional connectivity will show greater improvement in clinical symptoms, after

active 20 Hz dmPFC-rTMS but not after active 1 Hz or placebo dmPFC-rTMS.

Methods In overview, this study compared the efficacy of two rTMS protocols over the dmPFC for

treatment-resistant MDD. Patients were randomized to 1 Hz, 20 Hz, or sham dmPFC-rTMS,

administered twice daily for 15 days. Patients who received sham-rTMS were subsequently

offered active rTMS upon trial withdrawal or completion. Structural and functional MRI were

acquired before and after treatment, along with a battery of clinical and psychometric measures

to characterize symptom severity and cognitive control. Age-, and sex-matched healthy controls

to the TRD cohort were recruited as a comparator group.

200

Patient Recruitment

The study screened 123 male and female patients who were referred from the community

to the MRI-Guided rTMS Clinic at Toronto Western Hospital. MDD diagnosis was verified by a

Canadian Royal College-certified psychiatrist at the time of their initial consultation assessment

for rTMS treatment. Patients were included in the study if: they were competent to give

voluntary consent to treatment and agreed to participate; met DSM-5 diagnostic criteria for

MDD at the time of their consultation for rTMS, and had a HAMD score of or at least 18 at the

time of their screening. Patients were excluded if they: were pregnant; had a lifetime MINI

diagnosis of bipolar I or II disorder, schizophrenia, schizoaffective disorder, schizophreniform

disorder, delusional disorder, or current psychotic symptoms; had a MINI diagnosis of OCD,

post-traumatic stress disorder (current or within the last year), anxiety disorder (including

generalized anxiety disorder, social anxiety disorder, panic disorder), or dysthymia, assessed by a

study investigator to be primary and causing greater impairment than MDD; had a diagnosis of

any personality disorder, as assessed by a study investigator to be primary and causing greater

impairment than MDD; had previously received rTMS for any previous indication due to

potential compromise of subject blinding; had any significant neurological disorder or insult

including, but not limited to: any condition likely to be associated with increased intracranial

pressure, space occupying brain lesion, any history of seizure except those therapeutically

induced by ECT, cerebral aneurysm, Parkinson's disease, Huntington's chorea, multiple sclerosis,

significant head trauma with loss of consciousness for greater than or equal to 5 minutes; had an

intracranial implant (e.g., aneurysm clips, shunts, stimulators, cochlear implants, or electrodes)

or any other metal object within or near the head, excluding the mouth that cannot be safely

removed; or had a non-correctable clinically significant sensory impairment (i.e., could not hear

well enough to cooperate with interview). If a patient was participating in psychotherapy, they

must have been in stable treatment for at least 3 months prior to entry to the study, with no

anticipation of change in the frequency of therapeutic sessions, or the therapeutic focus over the

duration of the study. If a patient was taking antidepressant medication, they must have received

stable dosage for at least 4 weeks prior to entry to the study. Patients were also required not to

take more than 2 mg daily of lorazepam (or equivalent dosage of other benzodiazepines), or any

dose of an anticonvulsant due to the potential to limit rTMS efficacy. Inclusion and exclusion

criteria were assessed in a pre-treatment interview administered by trained staff with clinical

supervision.

201

All randomized patients provided informed consent to participate, and the study was

approved by the University Heath Network Research Ethics Board.

Primary Clinical Measures

The HAMD (Kimbrell et al, 1999; Kreuzer et al, 2015; Padberg et al, 1999) and the BDI-

II (Hamilton, 1960) were the two primary clinical outcome measures for the study, as in previous

studies of dmPFC-rTMS in TRD (Bakker et al., 2015). Trained research staff blinded to

treatment arm administered these measures at pre-treatment, weekly (every 5 treatment days)

throughout treatment, immediately post-treatment, and at 1,- 4-, and 12-weeks post-treatment.

The BAI (Beck et al, 1988) was also included as a secondary outcome measure. The BAI

is a widely used and validated measure of anxiety symptoms. It is a 21-item self-report

questionnaire. This measure was assessed at pre-treatment, weekly throughout treatment,

immediately post-treatment, and at 1-, 4, and 12-weeks post-treatment.

For continuous outcome measures, a mixed effects model was used to assess the

significance of any differences in clinical ratings over the course of treatment and follow-up

across the 120 subjects in the group. The mixed effects model offers advantages in terms of

handling missing values, when compared to the fixed effects ANOVA with a last observation

carried forward (LOCF) approach. Time (baseline, weekly during treatment, and 1-week post-

treatment) was used as the explanatory variable in a mixed-effects, linear regression model for

the continuous outcomes (HAMD, BDI, BAI).

Randomization and rTMS Treatment

Upon entry into the study, all patients were randomized into one of three treatment arms:

active 1 Hz dmPFC-rTMS, active 20 Hz dmPFC-rTMS, or sham rTMS. Prior to the first session

of rTMS, each subject’s motor threshold was determined using single pulse TMS over the

medial primary motor cortex corresponding to the foot and toes (Hayward et al, 2007; Schutter

and van Honk, 2006).

202

Subjects underwent bilateral rTMS of the dmPFC either using the 1 Hz (inhibitory) or 20

Hz (excitatory) protocol. rTMS employed the MagPro R30 stimulator and a custom active/sham

variant of the DB80 fluid-cooled coil described below (MagVenture, Farum, Denmark). Patients

received 30 weekday sessions of bilateral rTMS over the dmPFC, twice daily (60 minutes apart),

over 3 weeks. Patients randomized to excitatory active rTMS received 20 Hz stimulation at

120% of the resting motor threshold, 2s on 12 s off, 38 trains, for 1520 pulses per hemisphere,

for a total of 3000 pulses per session delivered over 8 min 40 s per hemisphere. Patients

randomized to inhibitory active stimulation received 1 Hz stimulation over the bilateral dmPFC

at 120% resting motor threshold, 60s on and 30 s off, for 360 total pulses, 8.5 minutes per

hemisphere (Brunelin et al, 2014).

Sham rTMS employed a custom active/placebo version of the DB80 coil, designed to

maintain blinding in clinical trial settings (MagVenture, Farum, Denmark) (Figure 6-1). The

double-sided stimulation coil was positioned with one coil in contact with the scalp for active,

the other for sham. Preprogrammed software controlled a switch within the rTMS device that

determines which (externally indistinguishable) side of the coil was placed over the scalp, based

on the patient’s study ID code, thereby maintaining technician blinding. Patients randomized to

the sham arm were randomly allocated to receive either the 1 Hz or 20 Hz dmPFC-rTMS pattern

above, again to maintain blinding; this pulse pattern was delivered with the active coil in the

upper position (oriented away from the scalp), thus avoiding stimulation of the target region.

During treatment, patients were asked to defer any medication changes throughout the

course of rTMS to avoid confounding effects. Patients were also assessed for the emergence of

any adverse events (e.g., headache, seizure, worsening of symptoms) at each session of treatment

and at each follow-up assessment. Patients were withdrawn from the study if they experienced a

worsening in depression, defined as an increase in the HAMD from baseline of >25% during two

consecutive assessments, or the development of active suicidal intent, or attempted suicide.

203

Figure 6-1: The custom MagVenture active/placebo coil.

MRI Acquisition and Analysis

An MRI scan was acquired one week prior to and one week following treatment. Subjects

underwent a 30-minute MRI protocol consisting of a 6-minute high-resolution T1 anatomical

sequence, and a 10-minute T2* BOLD fMRI sequence during the resting state for assessment of

whole-brain functional connectivity. MRIs were performed using 3T GE Signa HDx scanner

equipped with an 8-channel phased-array head coil. This comprised a T1-weighted fast spoiled

gradient-echo anatomical scan (TE = 12 ms, TI = 300 ms, flip angle = 20°, 116 sagittal slices,

thickness = 1.5 mm, no gap, 256 x 256 matrix, FOV = 240 mm), followed by a 10-min resting-

state, eyes-closed T2* series (TE = 30 ms, TR = 2000 ms, flip angle = 85°, 32 axial slices,

thickness = 5 mm, no gap, 64 x 64 matrix, FOV = 220mm, 300 TRs, 2 s temporal resolution).

204

Resting-state fMRI was preprocessed and analyzed using SPM12

(http://www.fil.ion.ucl.ac.uk/spm/doc/) and the CONN FC toolbox (Whitfield-Gabrieli and

Nieto-Castanon, 2012), executed in Matlab v8.5.0.197613

(https://www.mathworks.com/products/matlab/). Before rs-fMRI preprocessing, the first five

volumes of each subject’s resting state fMRI scan were removed to account for scanner

instabilities. Preprocessing steps included the following: interleaved slice-time correction; affine

realignment and volume censorship (using the ARtifact detection Tools software package,

removing volumes exceeding the 95th percentile); linear affine registration of the functional

image to the anatomical T1 image and non-linear registration of the functional and anatomical

images to the MNI standard brain; and spatial smoothing with a 6 mm full-width half-maximum.

Skull-stripped anatomical T1 images were additionally segmented into gray matter, white matter,

and cerebrospinal fluid for the removal of physiological noise.

Motion (outputs from affine realignment), physiological noise and other non-neuronal

noise were removed via linear regression. To remove physiological noise, aCompCor (Behzadi et

al, 2007b) was performed using the segmented white matter and cerebrospinal fluid T1

anatomical images. The first five principal components of each tissue segment were removed in

addition to the 12 motion parameters, their first derivative, and the linear motion trend. Finally,

all resting-state functional data were then bandpass filtered at 0.008–0.09 Hz.

We applied a region-of-interest approach, using the dmPFC, VMPFC and striatal regions

of interest that were previously predictive of dmPFC-rTMS (Dunlop et al, 2015, 2016a;

Salomons et al, 2014). In total, 6 regions were used (Figure 6-2): two dmPFC ROIs (dmPFC and

anterior mid-cingulate cortex), a VMPFC ROI, and three striatal seeds (superior and inferior

ventral striatum, and the dorsal caudate). All cortical ROIs were extracted from the Craddock

and colleagues’ resting-state whole-brain parcellation (Craddock et al, 2012a), while striatal

ROIs were generated using coordinates described by Di Martino and colleagues (Di Martino et

al, 2008). Whole-brain seed-to-voxel-based FC was performed for each ROI for each subject.

This analysis step resulted in a bivariate correlation map for each subject; these maps were then

Fisher-transformed to normalized z-scores and were entered into a second-level mixed-effects

general linear model to identify group effects.

205

Figure 6-2: Regions of interest used for seed-to-voxel based analyses. A-F refer to the

dmPFC, dACC, VMPFC, dorsal caudate, superior ventral striatum, and inferior ventral striatum

seed, respectively.

At the group-level, we performed the following five baseline analyses. First, we

compared the baseline rsFC maps of all TRD patients versus healthy controls in order to localize

any differences between our patient group and healthy controls. Second, we compared the

baseline rsFC maps between the 3 TRD treatment arm groups, to ensure that the patients

randomized to the three arms of the study did not significantly differ prior to treatment. Third,

we performed a continuous covariate interaction analysis on baseline rsFC in order to find any

differences between treatment arms’ baseline rsFC and its correlation to subsequent

improvement on the primary outcome measure (% improvement in HAMD score at 1-week post

treatment, versus baseline). Fourth, we performed a correlation analysis on baseline rsFC,

pooling all groups, in order to identify any nonspecific predictors of clinical improvement,

regardless of the intervention. Finally, we repeated this baseline correlational analysis on the

subgroup of patients randomized to the 20 Hz arm, as a replication analysis to compare to our

previously published findings with open-label 10 Hz dmPFC (Downar et al, 2014).

206

We also completed the following two analyses to characterize changes in functional

connectivity associated with response to rTMS. First, we compared differences in the correlation

between rsFC change (baseline to post-treatment) and clinical improvement (% improvement in

HAMD score at 1-week post treatment, versus baseline) using a 3-way continuous covariate

interaction analysis. Second, we identified changes in rsFC that correlated with clinical

improvement across all 3 arms, regardless of treatment intervention.

Each of the resultant whole-brain group-level analyses were thresholded at p < 0.001

(cluster-corrected using the false discovery rate [FDR], with a height threshold of p < 0.05). An

adjusted peak p-value of 0.001/6 = 0.0002 was used to account for the number of ROIs included

for analysis. The resulting clusters were then displayed on the MNI brain. To determine any

significant statistical differences between groups, post hoc parameter estimates for the mean

effect size for each resulting cluster were extracted. For regions of interest arising in these

analyses, we also extracted the rsFC parameter estimates from the preprocessed control scans, as

a comparator group.

Supplementary Clinical Measures

The following subject-reported supplementary clinical measures were collected at

baseline and post-treatment: The Difficulties in Emotion Regulation Scale (DERS) (Gratz and

Roemer, 2004); the Rumination Responses Scale (RRS) (Treynor et al, 2003); the

Multidimensional Perfectionism Scale (MPS-II) (Hewitt et al, 1991); the UPPS Impulsive

Behaviour Scale (UPPS-P) (Whiteside and Lynam, 2001); the Behavioural Inhibition

Scale/Behavioural Approach Scale (BIS/BAS) (Carver and White, 1994); the Barratt

Impulsiveness Scale (BIS-11) (Patton et al, 1995); the Monetary-Choice Questionnaire (MCQ)

(Kirby and Marakovic, 1996); and the Revised NEO Questionnaire (NEO) (Costa and R., 1992).

The DERS is a self-report rating scale of 36 items measuring six separate dimensions of

emotion regulation: Nonacceptance of emotional responses; Difficulty in engaging in goal-

directed behaviour; Deficits in Impulse control; Poor emotional awareness; Limited access to

emotion regulation strategies; and Lack of emotional clarity. The measure has high internal

consistency, reliability, and validity (Gratz and Roemer, 2004). The RRS is a 22-item self-

207

reported scale that is a measure of rumination severity; higher scores indicate more severe

rumination. The MPS-II is a 45-item self-report questionnaire that assesses three dimensions of

perfectionism: self-oriented perfectionism, other-oriented perfectionism, and social prescribed

perfectionism. The UPPS-P is a 59-item self-report questionnaire that assesses five personality

subscales that are associated with impulsive behaviours: First, negative urgency, or the tendency

to act impulsively under negative emotions; Second, positive urgency, or the tendency to act

impulsive under positive emotions; Third, lack of premeditation, or the inability to think before

acting; Fourth, lack of perseverance, or difficulties in sustained focus on a task; and Finally,

sensation seeking, or the tendency to seek out novel or thrilling experiences (Whiteside and

Lynam, 2001). The BIS/BAS is a 24-item scale that assesses behavioural responses to reward

and punishment. This questionnaire groups questions into 4 categories: Drive, Fun-Seeking,

Reward Responsiveness, and Behavioural Avoidance. The BIS-11 is a 30-item self-report

questionnaire that assesses three constructs of impulsivity: Attentional, Motor, and Nonplanning

Impulsivity. The MCQ is a 27-item self-administered questionnaire. For each item, the

participant chooses either a lower value, immediate reward or a higher value, delayed reward.

Scores are calculated by fitting the respondent’s answers on reference discounting curves.

Placement on steeper discounting curves is indicative of higher impulsivity in decision-making

among rewards. The revised NEO personality inventory involves 240 questions that measure the

‘big five’ personality traits: extraversion, agreeableness, conscientiousness, neuroticism, and

openness to experience.

The same analyses as described above for clinical data were performed for these

supplementary psychometric measures. These analyses identify any group (20 Hz responder/20

Hz non-responder/1 Hz responder/1 Hz non-responder/sham/healthy control) by time

interactions that reflect changes in cognitive measures of impulsivity or perfectionism over the

course of treatment.

Healthy Control Recruitment & Study Visits

Forty-one healthy controls were recruited as a comparator group for TRD patients (n = 1

did not meet inclusion/exclusion criteria, n = 1 did not complete study visits). Healthy subjects

underwent an initial screening to assess medical and medication history, suitability for MRI, and

current/past psychiatric symptomatology. Healthy participants were excluded if they had: a

208

lifetime history of neurological or psychiatric disease, including substance abuse or dependence,

or cardiovascular disease; currently used any medications that may affect brain perfusion or

activity, including antidepressants, mood stabilizers, neuroleptics, anxiolytics, hypnotics,

stimulants, anticonvulsants, anti-migraine agents, cognitive enhancing agents, opioids, anti-

nausea agents or beta-blockers; had contraindications to MRI (e.g., implanted medical devices,

metallic foreign objects, epilepsy), or a HAMD score of at least 8. Healthy controls underwent

one visit in which all primary and secondary clinical measures were collected, and a second MRI

visit identical to that described above.

All controls provided informed consent, and the study was approved by the University

Heath Network Research Ethics Board.

Results

Demographic and Primary Clinical Results

Of the 123 patients referred to the study, n = 103 met the eligibility for the study (Figure

6-3, Table 6-1) and were randomized to one of the 3 study arms. Of these 103 subjects, 2 were

subsequently excluded due to potential unblinding during treatment, and 10 patients withdrew

prematurely from treatment (Figure 6-3). Consequently, of those that completed a full course of

treatment, 31 had been randomized into 1 Hz rTMS, 30 into 20 Hz rTMS, and 30 into placebo

rTMS (total n = 91 TRD patients). 39 healthy controls were also screened for the study; 38 met

the inclusion criteria and of those N = 37 completed the study. Healthy controls and the three

treatment groups did not significantly differ in terms of age (F (3, 124) = 0.540, p = 0.656) or sex

(χ2 = 0.604, p = 0.895). These controls and completers were included in the analyses below.

Comorbidities and concomitant medication of patients are included in Tables 6-2, and 6-

3. To summarize, only 4 patients had previously received ECT; all of those who received ECT

displayed full or partial benefit from that intervention. Concomitant medications, comorbidities,

and antidepressant history varied widely. An overwhelming majority of TRD participants were

stably on at least one concomitant psychotropic medication, most notably SRRIs, SNRIs,

atypical antipsychotics and benzodiazepines. Treatment groups did not significantly vary in

terms of the proportion of patients taking a particular medication class. TRD patients also

209

displayed a high level of non-response to conventional antidepressant pharmacotherapies. All

TRD patients did not respond or could not tolerate an average of 6.87±3.81 antidepressant

interventions in the current major depressive episode. The number of previous trials did not vary

by treatment intervention (F (2,86) = 0.66, p = 0.52; mean number 1 Hz = 7.40±4.34; 20 Hz =

6.93±3.31, Placebo = 6.27±3.75). Most patients had a current or remitted psychiatric

comorbidity.

At baseline, healthy controls displayed significantly lower HAMD, BAI and BDI-II

scores relative to all three treatment groups (Figures 6-4, 6-5 and 6-6, Table 6-1). On average,

controls displayed subclinical levels of depressive and anxiety symptoms (mean HAMD =

1.38±1.48 SD; BDI-II = 2.73±3.12; BAI = 3.00±4.51). On all three measures, all patient groups

displayed severe or very severe depressive and moderate-to-severe anxiety scores. The TRD

treatment groups did not differ in terms of depression or anxiety severity at baseline or at any

other time-point (F < 1.196, p > 0.307). Among the three TRD treatment arms, a mixed effects

model using HAMD revealed a significant effect of time (F = 27.85, p < 0.0000001) but no

group by time interactions (F = 0.36, p = 0.96). A similar effect was found in the BDI-II (effect

of time: F = 7.26, p = 0.000004; group x time: F = 0.17, p = 0.998), and BAI (effect of time: F =

3.18, p = 0.01; group x time: F = 0.12, p = 0.9996). This indicates that while the patients of all

three treatment arms exhibited significant improvements over the course of treatment and into

follow-ups, there was no significant difference in the degree of improvement on any primary

clinical measures between active versus the placebo arms of the study. Age, sex and baseline

severity did not significantly predict treatment response either across the entire sample, or within

each treatment group.

210

Figure 6-3: CONSORT diagram of patients enrolled in the study.

AssessedforEligibility(MDDn=123)

Excluded(MDDn=20)

• Declinedtoparticipate(MDDn=7)• Inclusioncriteria(MDDn=13)

• Unabletodemonstrateinformedconsent (MDDn=1)• UnstablePsychotherapy (MDDn=1)• Didnotmeetseveritycutoff(MDDn=7)• Historyofmania/bipolar(MDDn=4)

Enrollment(MDDn=103)

MDD1Hz(n=36)

MDD20Hz(n=32)

MDDPlacebo(n=35)

CompletedTx(n=31)

Withdraw(n=5)• Schedule(n=1)• Tolerance(n=2)• Unblinding (n=2)

CompletedTx(n=30)

Withdraw(n=2)• Schedule(n=1)• Tolerance(n=1)

CompletedTx(n=30)

Withdraw(n=5)• Schedule(n=2)• Tolerance(n=1)• Other(n=2)

Follow-Up1(n=30)

Follow-Up1(n=28)

Follow-Up1(n=29)

Follow-Up2(n=29)

Follow-Up2(n=26)

Follow-Up2(n=27)

211

Table 6-1: Summary of demographic and primary clinical outcomes for healthy controls

and TRD treatment arms.

Controls 1 Hz 20 Hz Placebo Statistical Test

N (# Female) 37 (25) 31 (19) 30 (20) 30 (18) χ2 = 0.604, p = 0.895 Age 37.89 (14.74) 40.74 (11.68) 41.53 (11.86) 39.23 (11.80) F (3, 124) = 0.540, p = 0.656

HAMD Baseline* 1.38 (1.48) 22.68 (3.24) 22.30 (3.27) 22.37 (3.07) F(3,124) = 494.38, p < 0.001

Week 1 - 19.23 (4.30) 19.40(3.40) 18.27 (5.04) F (2, 88) = 0.598, p = 0.552 Week 2 - 17.81 (5.43) 18.40 (5.54) 16.57 (5.80) F (2, 88) = 1.196, p = 0.307

Final - 17.48 (5.43) 16.80 (5.54) 16.60 (6.39) F (2, 88) = 0.195, p = 0.824 Follow-Up 1 - 16.37 (5.44) 16.89 (5.95) 17.17 (5.64) F (2, 84) = 0.154, p = 0.858 Follow-Up 2 - 17.90 (5.45) 16.00 (6.96) 17.19 (6.37) F (2, 79) = 0.154, p = 0.559

BDI-II Baseline* 2.73 (3.12) 38.45 (10.29) 37.87 (10.80) 35.47 (11.43) F (3,124) = 121.80, p < 0.001

Week 1 - 32.45 (11.91) 34.63 (11.71) 30.27 (13.01) F (2, 88) = 0.958, p = 0.388 Week 2 - 30.55 (10.28) 31.50 (12.38) 29.03 (13.59) F (2, 88) = 0.315, p = 0.730

Final - 29.52 (11.69) 29.43 (15.24) 27.37 (15.58) F (2, 88) = 0.232, p = 0.794 Follow-Up 1 - 28.60 (12.61) 28.96 (12.85) 28.79 (14.05) F (2, 84) = 0.006, p = 0.994 Follow-Up 2 - 30.45 (13.10) 28.88 (16.40) 28.78 (14.86) F (2, 78) = 0.332, p = 0.718

BAI Baseline* 3.00 (4.51) 22.03 (10.06) 22.10 (11.82) 21.10 (11.55) F (2,124) = 31.58, p < 0.001)

Week 1 - 18.26 (8.91) 18.67 (11.53) 18.21 (12.20) F (2, 86) = 0.015, p = 0.985 Week 2 - 16.17 (9.12) 18.20 (12.60) 15.76 (9.12) F (2, 85) = 0.414, p = 0.662

Final - 16.00 (9.70) 16.43 (11.48) 15.93 (10.70) F (2, 86) = 0.019, p = 0.981 Follow-Up 1 - 16.33 (10.12) 15.96 (12.47) 16.83 (11.32) F (2, 84) = 0.042, p = 0.959 Follow-Up 2 - 17.25 (10.50) 15.62 (13.51) 16.32 (11.43) F (2, 76) = 0.129, p = 0.879 *All TRD groups displayed significantly higher HAMD, BDI-II, and BAI scores relative to controls (p < 0.0000001). BAI = Beck Anxiety Inventory; BDI-II = Beck Depression Inventory II; HAMD = 17-item Hamilton Rating Scale for Depression.

212

Table 6-2: Summary of concomitant medications by treatment intervention.

All TRD 1 Hz 20 Hz Placebo Statistical Test

SSRI 28 (30.7) 11 (35.5) 10 (33.3) 7 (23.3) χ2 = 1.20, p = 0.55

SNRI 24 (26.5) 5 (16.1) 7 (23.3) 12 (40.0) χ2 = 4.69, p =0.10

MAOI 2 (2.2) 2(6.5) 0 (0.0) 0 (0.0) χ2 = 3.96, p = 0.14

TCA 9 (9.8) 5 (16.1) 3 (10.0) 1 (3.3) χ2 = 2.80, p = 0.25

Atypical Antidepressant 36 (39.5) 15 (48.4) 11 (36.7) 10 (33.3) χ2 = 1.60, p = 0.45

Typical Antipsychotic 1 (1.1) 1 (3.2) 0 (0.0) 0 (0.0) χ2 = 1.96, p =0.38

Atypical Antipsychotic 13 (14.3) 5 (16.1) 3 (10.0) 5 (16.7) χ2 = 0.68, p = 0.71

Stimulant 12 (13.2) 4 (12.9) 3 (10.0) 5 (16.7) χ2 = 0.59, p = 0.75

Z-Drug 6 (6.6) 1 (3.2) 2 (6.7) 3 (10.0) χ2 = 1.14, p = 0.57

Lithium 3 (3.3) 1 (3.2) 2 (6.7) 0 (0.0) χ2 = 2.09, p = 0.35

Benzodiazepine 27 (29.8) 6 (19.4) 11 (36.7) 10 (33.3) χ2 = 2.48, p = 0.29

Concomitant selective serotonin reuptake inhibitors (SSRI) included: escitalopram, fluoxetine,

paroxetine, and sertraline. Concomitant selective noradrenergic reuptake inhibitors (SNRI)

included: desvenlafaxine, duloxetine, levomilnacipran, and venlafaxine. Concomitant MAO-A

uptake inhibitors (MAOI) included moclobemide and phenelzine. Concomitant tricyclic

antidepressants (TCA) included amitriptyline, desipramine, and nortriptyline. Concomitant

atypical antidepressants included: bupropion, buspirone, mitazapine, trazodone, and vortioxetine.

Concomitant typical and atypical antipsychotics included: aripiprazole, bexipiprazole, loxapine,

lurasidone, and quetiapine. Concomitant stimulants included lisdexamphetamine and

methyphenidate. Concomitant Z-drugs included zolpidem and zopiclone. Concomitant

benzodiazepines included: clonazepam, lorazepam and oxazepam.

213

Table 6-3: Summary of psychiatric comorbidities (currently ongoing and remitted).

All MDD 1 Hz 20 Hz Placebo Statistical Test

ED-BP 21 (23.1) 7 (22.6) 5 (16.7) 9 (30.0) χ2 = 1.51, p = 0.47

ED-R 2 (2.2) 0 (0.0) 2 (6.7) 0 (0.0) χ2 = 4.16, p = 0.13

Anxiety Disorders 57 (62.6) 17 (54.8) 23 (76.7) 17 (56.7) χ2 = 3.79, p = 0.15

PTSD 15 (16.5) 6 (19.4) 7 (23.3) 2 (6.7) χ2 = 3.31, p = 0.19

OCD 11 (12.1) 4 (12.9) 4 (13.3) 3 (10.0) χ2 = 0.19, p = 0.91

ADD/ADHD 18 (19.8) 9 (29.0) 3 (10.0) 6 (20.0) χ2 = 3.48, p = 0.18

Chronic Pain 11 (12.1) 6 (19.4) 4 (13.3) 1 (3.3) χ2 = 3.75, p = 0.15

AUD/SUD 8 (8.8) 2 (6.5) 2 (6.7) 4 (13.3) χ2 = 1.15, p = 0.56

BPD 6 (6.6) 3 (9.7) 1 (3.3) 2 (6.7) χ2 = 1.00, p = 0.61 ADD/ADHD = attention-deficit disorder/attention-deficit hyperactivity disorder; AUD/SUD = alcohol/substance use disorder; BPD = borderline personality disorder; OCD = obsessive-compulsive disorder; PTSD = post-traumatic stress disorder. ED-BP refers to eating disorders with a binge and/or purge phenotype, including binge-eating disorder, bulimia nervosa and anorexia nervosa with the binge/purge subtype. ED-R refers to eating disorders with a food restriction subtype, including anorexia nervosa with the restriction subtype and eating disorders not otherwise specified. Anxiety disorders include generalized anxiety disorder, social anxiety disorder and panic disorder. Chronic pain refers to both chronic pain and fibromyalgia.

Figure 6-4: Mean HAMD scores across the three treatment interventions.

12

14

16

18

20

22

24

Baseline Week 1 Week 2 Final F/U 1 F/U 2

Mea

n H

AMD

Sco

re

1 Hz 20 Hz Placebo

214

Figure 6-5: Mean BDI-II scores across the three treatment interventions.

Figure 6-6: Mean BAI scores across the three treatment intervention.

24

26

28

30

32

34

36

38

40

42


Mea

n BD

I-II

Scor

e

1 Hz 20 Hz Placebo

12

14

16

18

20

22

24

26


Mea

n BA

I Sco

re

1 Hz 20 Hz Placebo

215

Table 6-4: Summary of Baseline Secondary Clinical Outcomes

Controls 1 Hz 20 Hz Placebo F-Test p Post Hoc BIS/BAS Drive 11.84 (2.15) 8.94 (2.58) 9.40 (3.48) 8.77 (2.45) < 0.0001 HC > A, C

Fun Seeking 11.54 (2.28) 8.77 (2.92) 9.50 (2.27) 9.43 (2.82) 0.0001 HC > A, C

Reward Sensitivity 17.05 (2.19) 13.53 (2.70) 15.03 (2.75) 13.60 (3.06) < 0.0001 HC > A, C

BIS Total 18.97 (4.37) 24.42 (3.52) 24.50 (2.99) 24.20 (2.43) < 0.0001 HC < A, B, C BIS-11 Attention 13.19 (4.10) 20.00 (5.21) 20.47 (4.56) 19.07 (3.50) < 0.0001 HC < A, B, C Motor 18.00 (4.47) 19.79 (2.69) 19.57 (3.33) 19.14 (5.09) 0.27 Non-plan. 22.08 (4.21) 25.70 (5.53) 25.37 (5.82) 25.77 (6.41) 0.02 Total 53.67 (10.06) 65.96 (7.87) 65.40 (10.44) 64.52 (11.33) 0.0001 HC < A, B, C DERS Non-Accept. 11.19 (5.82) 19.32 (8.42) 18.70 (7.58) 18.23 (6.26) 0.0001 HC < A, B, C

Goals 12.00 (4.46) 20.90 (3.25) 18.93 (4.47) 19.23 (3.65) < 0.0001 HC < A, B, C Impulse 9.84 (4.63) 16.55 (6.16) 15.83 (5.33) 14.03 (5.56) < 0.0001 HC < A, B

Awareness 13.59 (5.33) 16.06 (5.40) 18.38 (4.81) 18.10 (5.42) 0.001 Strategies 14.32 (7.09) 28.87 (6.15) 27.79 (6.70) 27.53 (6.57) < 0.0001 HC < A, B, C Clarity 9.38 (4.11) 12.97 (4.56) 15.63 (4.46) 14.43 (4.76) < 0.0001 HC < B, C Total 70.32 (23.00) 114.96 (19.49) 116.39 (24.27) 111.57 (19.82) < 0.0001 HC < A, B, C MPS Self 63.51 (18.41) 82.03 (16.30) 69.13 (21.62) 69.92 (17.56) 0.001 HC < A Other 59.57 (13.23) 58.16 (13.38) 54.93 (14.81) 53.31 (12.96) 0.24 Social 51.23 (14.27) 65.81 (19.07) 62.73 (18.44) 62.87 (13.67) 0.002 MCQ (log k) -2.17 (0.80) -2.21 (0.54) -2.16 (0.82) -2.14 (0.59) 0.98

RRS - Total 33.56 (10.26) 64.80 (11.09) 64.59 (11.61) 60.93 (12.39) < 0.0001 HC < A, B, C

UPPS-P Negative Urgency 29.08 (9.67) 30.84 (4.80) 29.67 (4.98) 31.90 (5.43) 0.34

Pre-meditation 21.72 (3.95) 20.71 (4.53) 21.83 (4.16) 20.97 (6.10) 0.73

Per-severance 20.54 (3.32) 22.47 (2.97) 22.43 (3.65) 23.55 (3.98) 0.01

Sensation Seeking 27.86 (7.10) 32.32 (8.67) 32.10 (6.52) 33.53 (8.77) 0.02

Positive Urgency 38.80 (12.78) 40.55 (12.36) 39.83 (11.04) 46.27 (8.90) 0.05

NEO Neuro-ticism 142.06 (19.95) 149.72 (8.46) 146.79 (8.93) 147.00 (7.23) 0.12

Extra-version 139.19 (18.35) 148.03 (10.31) 147.78 (9.75) 145.88 (9.93) 0.03

Openness 150.44 (19.61) 154.66 (9.41) 154.93 (11.56) 150.39 (7.57) 0.36 Agreeable-

ness 146.00 (19.49) 141.27 (12.09) 143.57 (8.73) 143.39 (10.37) 0.59

Conscientiousness 151.41 (20.73) 155.63 (10.16) 154.21 (10.94) 151.19 (8.12) 0.53

216

BIS/BAS = Behavioural activation/inhibition system scale; BIS-11 = Barratt Impulsivity Scale, version 11, DERS = Difficulties in Emotion Regulation Scale; MCQ = Monetary Choice Questionnaire; MPS: Multidimensional Perfectionism Scale; RRS = Rumination Responses Scale; NEO = Revised NEO Five-Factor Inventory; UPPS-P = UPPS-P Impulsive Behaviour Scale

Supplementary Clinical Results

Baseline/pre-treatment secondary clinical measures are summarized in Tables 6-4 and 6-

5. In brief, all patient groups displayed abnormal scores in avoidance behaviours on the

BIS/BAS, impulsivity on the BIS-11, emotion regulation on the DERS, and rumination on the

RRS. After Bonferroni correction, only 1 Hz and Placebo treatment arms displayed deficits in

approach behaviours on the BIS/BAS, and only 20 Hz and Placebo treatments arms displayed

deficits in the impulsivity and lack of clarity subscales of the DERS. Only patients randomized to

1 Hz rTMS displayed an abnormal ‘self’ subscale of the MPS at baseline. At baseline, the three

treatment groups did not differ on any secondary clinical measures after Bonferroni correction

(p-cutoff = 0.0016). Across the entire TRD sample and in the 20 Hz and Placebo treatment arms

alone, none of the baseline secondary clinical measures predicted symptom improvement (p >

0.08, uncorrected). In the 1 Hz treatment arm group, both the total score of the BIS-11 (r = 0.50,

p = 0.008), the Fun Seeking Subscale of the BIS/BAS (r = 0.37, p = 0.04) and the lack of

perseverance subscale of the UPPS-P (r = 0.37, p = 0.04) was positively correlated before

multiple comparisons correction. In other words, individuals treated with active 1 Hz rTMS

exhibited with higher baseline self-reported impulsivity and sensation seeking, and poorer

perseverance, exhibited greater improvements in TRD symptom severity.

From baseline to post-treatment, improvements on the impulsivity and goal-setting

subscales of the DERS significantly correlated with BDI-II improvement across all three groups

(both r = -0.35, p = 0.001), even after correcting for multiple comparisons. Furthermore, the pre-

to post-treatment change in the total DERS score was also negatively correlated with subsequent

BDI-II improvement (r = -0.41, p = 0.0003). Patients whose depression scores improved over the

course of treatment also exhibited greater reductions (improvements) on self-reported emotion

regulation on the DERS (Table 5-6, Figure 5-7). The only correlation that survived multiple

comparisons correction was present in the active 20 Hz rTMS arm: a significant negative

correlation between the global DERS improvement and both BDI-II and HAMD improvement

(both r = -0.62, p = 0.001), such that those who saw the greatest clinical response had greater

reductions in emotion dysregulation on the DERS (Figure 5-8).

217

Figure 6-7: DERS improvement is significantly correlated with response to all three

treatment arms on the BDI-II.

218

Figure 6-8: DERS improvement is significantly correlated with response to active 20 Hz

dmPFC-rTMS on the BDI-II.

Baseline Differences in rs-fMRI functional connectivity

The first contrast we investigated was to identify rs-fMRI differences between the healthy

control group and all TRD patients. Only two seeds revealed a significant difference between

patients and controls. First, relative to healthy controls, MDD patients displayed significantly

lower frontostriatal rs-FC between the inferior ventral striatal seed and the right orbitofrontal

cortex (Figure 6-9) (Control mean ß = 0.48±0.03, MDD mean ß = 0.35±0.01, t (121) = 4.17, 157

voxels, cluster p-FDR = 0.0495, peak p = 0.00006). Second, MDD patients displayed

significantly lower rs-FC between the VMPFC and both the right (Control mean ß = 0.35±0.03,

MDD mean ß = 0.19±0.01, t (121) = 5.20, 260 voxels, cluster p-FDR = 0.009, peak p = 0.00001)

and left orbitofrontal cortex (Control mean ß = 0.34±0.03, MDD mean ß = 0.15±0.02, t(121) =

5.06, 179 voxels, cluster p-FDR = 0.028, peak p < 0.00001) (Figure 6-10).

219

Second, an ANOVA was performed between the TRD treatment arms to determine

whether there were any significant differences in baseline rsFC (Figure 6-11). The only

significant difference between MDD groups was frontostriatal connectivity between the dorsal

caudate and right middle frontal gyrus (F (2,83) = 14.98, 147 voxels, cluster p-FDR = 0.033,

peak p = 0.00001). Further post-hoc testing revealed that this difference was largely driven by

patients in the 20 Hz treatment arm, such that they displayed significantly higher frontostriatal

rsFC relative to patients treated with active 1 Hz (20 Hz mean ß = 0.20±0.02, 1 Hz mean ß =

0.05±0.02, LSD-corrected p = 0.000001), placebo (Placebo mean ß = 0.08±0.01, LSD-corrected

p = 0.00009), and healthy controls (Control mean ß = 0.09±0.02, LSD-corrected p = 0.0001).

220

Table 6-5: Summary of Post-Treatment Secondary Clinical Outcomes.

1 Hz 20 Hz Placebo BIS/BAS

Drive 9.26 (3.05) 9.60 (2.55) 8.73 (2.53) Fun Seeking 9.07 (2.89) 9.50 (2.08) 9.31 (2.48)

Reward Sensitivity 14.16 (2.31) 14.33 (2.09) 13.40 (3.23) BAS Total 33.14 (6.22) 33.72 (4.30) 31.71 (6.23) BIS Total 24.53 (3.62) 23.93 (3.62) 23.48 (3.19)

BIS-11 Attention 19.03 (3.54) 20.00 (4.04) 18.80 (3.64)

Motor 19.42 (3.64) 19.10 (3.09) 19.07 (4.76) Nonplanning 24.35 (5.26) 26.79 (5.11) 25.17 (6.16)

Total 62.41 (8.25) 65.42 (8.45) 63.03 (12.05) DERS

Nonacceptance 18.65 (6.24) 16.87 (8.00) 17.47 (6.23) Goals 19.19 (4.39) 18.00 (4.07) 18.37 (4.12)

Impulse 14.48 (5.28) 15.10 (4.84) 13.31 (5.31) Awareness 15.57 (5.50) 17.17 (4.84) 16.60 (6.01) Strategies 26.04 (6.57) 25.52 (6.80) 25.47 (7.07)

Clarity 12.55 (4.21) 14.34 (4.58) 14.87 (5.54) Total 105.96 (19.79) 108.67 (25.21) 105.45 (23.82)

MPS Self 79.10 (16.83) 74.69 (17.82) 68.13 (19.43)

Other 57.22 (18.89) 54.76 (15.77) 52.46 (12.38) Social 62.97 (17.69) 62.75 (16.93) 62.03 (16.54)

MCQ - k -2.12 (0.54) -2.16 (0.80) -2.13 (0.56) RRS - Total 59.13 (9.63) 64.96 (12.53) 56.96 (11.35) UPPS-P

Negative Urgency 31.35 (5.89) 29.75 (5.32) 32.07 (6.16) Premeditation 20.50 (4.49) 21.50 (4.57) 20.70 (4.36) Perseverance 21.58 (2.93) 22.44 (3.67) 23.30 (3.93)

Sensation Seeking 33.17 (8.45) 32.17 (6.00) 32.27 (8.41) Positive Urgency 41.76 (11.20) 39.14 (11.68) 45.90 (8.49)

NEO Neuroticism 147.14 (7.85) 142.30 (29.25) 146.10 (8.69)

Extraversion 147.32 (12.45) 148.40 (9.68) 145.41 (10.48) Openness 154.45 (9.07) 152.89 (9.29) 151.27 (11.36)

Agreeableness 144.97 (12.55) 142.85 (6.76) 142.20 (10.68) Conscientiousness 154.62 (10.96) 155.48 (10.80) 152.34 (8.28)

BIS/BAS = Behavioural activation/inhibition system scale; BIS-11 = Barratt Impulsivity Scale,

version 11, DERS = Difficulties in Emotion Regulation Scale; MPS: Multidimensional

Perfectionsim Scale; MCQ = Monetary Choice Questionnaire; RRS = Rumination Responses

Scale; NEO = Revised NEO Five-Factor Inventory; UPPS-P = UPPS-P Impulsive Behaviour

Scale

221


inferior VS and the right OFC in all TRD patients relative to controls.

DiagnosisMDDControls

Mea

n In

ferio

r VS

-Rig

ht O

FC P

aram

eter

Est

imat

e .60

.50

.40

.30

.20

.10

.00

Page 1

R

A) B)

*

222


VMPFC seed in controls versus all TRD patients.

*

*

223

Figure 6-11: A) Regions showing significantly higher resting-state functional connectivity

to the dorsal caudate seed in the 20 Hz versus other MDD treatment arms.

Next, a three-way continuous covariate interaction analysis was performed to determine

whether there were any significant differences between rTMS treatment arms in the linear

association between symptomatic improvement and rsFC. While no regions reached significance

at the aforementioned multiple comparisons threshold criteria, there was a significant difference

in BDI-II improvement and frontostriatal connectivity from the superior ventral striatum to right

frontal pole and inferior frontal gyrus at an uncorrected p height threshold < 0.005, and a cluster

p-FDR < 0.05 (Figure 6-12) (389 voxels, cluster p-FDR = 0.008, peak p = 0.00002). Extracting

the mean parameter estimates of superior VS connectivity to the cluster showed that there was no

association to BDI-II improvement in the 1 Hz active arm (r = 0.17, p = 0.36). rsFC between the

superior VS and the frontal pole and inferior frontal gyrus in the 20 Hz arm displayed a

significant anticorrelation to BDI-II improvement (r = -0.68, p = 0.00009), such that those with

lower baseline rsFC from right ventrolateral prefrontal cortex to superior VS displayed the

greatest clinical response following treatment. Finally, rsFC between these regions in the placebo

arm revealed a significant positive correlation to BDI-II improvement (r = 0.55, p = 0.002), such

*

* *

224

that those with higher rsFC connectivity mean parameter estimates displayed the greatest

improvements in TRD symptom severity. However, this association did not survive for the

placebo group after adjusting for the number of ROIs (FDR-p = 0.0002). Healthy controls

displayed slightly positive rsFC between the ventral striatum and inferior frontal gyrus (mean ß =

0.10±0.02).

Figure 6-12: Regions where baseline rsFC to ventral striatum showing significantly

different correlations to BDI-II improvement across the three treatment groups;

Scatterplot of mean parameter estimates for functional connectivity between the superior

VS seed and the right frontal pole/inferior frontal gyrus.

The final two baseline analyses were performed to: first, identify any rsFC correlations

between symptomatic improvement across the entire MDD group regardless of intervention; and

second, identify frontostriatal rsFC predictive of response to dmPFC-rTMS. Only rsFC between

the VMPFC seed and the left cerebellum was positively associated with BDI-II improvement

across all patients (Figure 6-13) (r = 0.52, 199 voxels cluster p-FDR = 0.023, peak p < 0.00001).

Again, healthy controls displayed slightly positive rsFC between the cerebellum and sgACC

A) B)

225

(mean ß = 0.03±0.02). No other seeds revealed any significant correlations in rsFC and response

to treatment at the a priori-defined multiple comparisons correction threshold. At a voxel-p <

0.005, low frontostriatal connectivity from two striatal seeds predicted clinical response to 20 Hz

dmPFC-rTMS. First, low ventral striatum connectivity to the left dorsal ACC and frontal pole

(Figure 6-14) (r = -0.77, 308 voxels, cluster p-FDR = 0.017, peak p = 0.0049), and high ventral

striatum connectivity to the cerebellum (r = 0.72, 293 voxels, cluster p-FDR = 0.017, peak p =

0.00003) predicted HAMD improvement. Second, low ventral striatum connectivity to the left

(Figure 6-15) (r = -0.67, 372 voxels, cluster p-FDR = 0.011, peak p = 0.00499) and right frontal

pole (r = -0.72, 308 voxels, cluster p-FDR = 0.011, peak p = 0.00499) predicted percent BDI-II

improvement.

Figure 6-13: Regions where baseline rsFC to VMPFC showed significant correlations to

BDI-II improvement in all three treatment groups; Scatterplot of mean parameter

estimates for functional connectivity between the VMPFC seed and the left cerebellum.

A) B)

226

Figure 6-14: Regions where baseline rsFC to Superior VS showed significant correlations

to HAMD improvement to 20 Hz dmPFC-rTMS; Scatterplot of mean parameter estimates

for functional connectivity between the VS seed and the left dorsal ACC/frontal pole

cluster.

A) B)

227

Figure 6-15: Regions where baseline rsFC to Inferior VS showed significant correlations to

BDI-II improvement to 20 Hz dmPFC-rTMS; Scatterplot of mean parameter estimates for

functional connectivity between the VS seed and the left frontal pole cluster.

Changes in rs-fMRI

Two analyses were performed to identify changes in rsFC from baseline to post-treatment

associated with response to dmPFC-rTMS. The first analysis performed was a 3-group

continuous covariate interaction to identify differences in rsFC change that correlated

differentially with clinical response between the three treatment interventions. Only the

correlation between anterior midcingulate cortex-right parietal operculum rsFC change and

HAMD improvement significantly differed between treatment arms (170 voxels, cluster p-FDR

= 0.012, peak p = 0.00001) (Figure 6-16). Further analysis revealed that there was no significant

association between dACC rsFC change and HAMD response in the 1 Hz rTMS arm (r = -0.21,

p = 0.31). A significant positive correlation was found in the 20 Hz rTMS arm, such that those

with the greatest HAMD improvement exhibited the greatest increase in dACC-parietal

operculum rsFC (r = 0.52, p = 0.005; this correlation does not survive multiple comparisons

correlation). Finally, a significant negative correlation was found in the placebo treatment arm,

A) B)

228

such that those with the greatest HAMD improvement exhibited the greatest decrease in dACC-

parietal operculum rsFC (r = -0.67, p = 0.0001).

The final change analysis performed was aimed at identifying any significant changes in

rsFC that correlated with symptomatic improvement, across all treatment groups. While no

region reached significance at the aforementioned multiple comparisons threshold criteria, there

was a significant positive correlation between BDI-II improvement and VMPFC connectivity

change to bilateral dorsal anterior cingulate cortex at an uncorrected p height threshold < 0.005,

and a cluster p-FDR < 0.05 (321 voxels, cluster p-FDR = 0.035, peak p = 0.00006) (Figure 6-17).

Post-hoc extractions of mean parameter estimates revealed that treatment responders (those with

a >33% improvement on the BDI-II) exhibited significantly lower baseline VMPFC-dACC rsFC

relative to both non-responders (F (2,120) = 7.32, p = 0.001; Responders mean ß = 0.18±0.02,

Non-Responders mean ß = 0.25±0.02, LSD-corrected p = 0.01) and healthy controls (Control

mean ß = 0.30±0.02, LSD-corrected p = 0.0002). The baseline VMPFC-ACC rsFC of the non-

responders did not significantly differ from that of controls (LSD-corrected p = 0.11). At post-

treatment, the VMPFC-dACC rsFC of responders had significantly increased (Effect of Time F =

2.93, p = 0.13, Group [Responders, Non-Responders] by Time Interaction F = 8.29, p = 0.005,

Post-treatment Responder mean ß = 0.27±0.02, t(30) = -3.51, p = 0.001) such that there was no

significant difference relative to controls (LSD-corrected p = 0.35) or non-responders (LSD-

corrected p = 0.17).

229

Figure 6-16: Regions of dACC rsFC change showing significantly different correlations to

HAMD improvement across the three treatment groups; Scatterplot of mean parameter

estimates for functional connectivity change between the dACC seed and the right parietal

operculum.

A) B)

230

Figure 6-17: Regions of VMPFC rsFC change showing a significant positive correlation to

BDI-II improvement in all three treatment groups; Scatterplot of mean parameter

estimates for functional connectivity change between the VMPFC seed and the dorsal

ACC. Pre- and post-treatment mean parameter estimates for treatment responders (>33%

improvement), non-responders, and healthy controls.

Discussion & Conclusion To summarize, some but not all of the results of Study III supported the three outlined

hypotheses. The first hypothesis was that improvements in depression symptoms with 20 Hz

dmPFC-rTMS would be significantly higher than that achieved with 1 Hz dmPFC-rTMS or

placebo dmPFC-rTMS. The results do not support this hypothesis as neither 1 Hz nor 20 Hz

achieved significantly greater reductions in symptom severity when compared to ‘sham’ dmPFC-

rTMS. The second hypothesis was that TRD patients with lower dmPFC-CSTC and higher

dmPFC-VMPFC resting-state functional connectivity would show greater symptom

improvement with active 20 Hz dmPFC-rTMS, but not with active 1 Hz or placebo dmPFC-

rTMS. Partially confirming this hypothesis, low CSTC rsFC from the lOFC and dACC predicted

response to 20 Hz dmPFC-rTMS, as was the case for 10 Hz dmPFC-rTMS in our previous work

A)

B) C)

*

* *

231

(Salomons et al, 2014). The final hypothesis was that TRD patients showing greater increases in

pre- to post-treatment dmPFC-CSTC resting-state functional connectivity would show greater

improvement in clinical symptoms, after active 20 Hz dmPFC-rTMS but not after active 1 Hz or

placebo dmPFC-rTMS. This hypothesis was not supported by the results, and an increase in

corticocortical connectivity between the dmPFC and VMPFC across all three interventions was

significantly correlated with treatment arm-nonspecific clinical response. Improvements in

emotion regulation, particularly in goal setting and impulsivity, were also significantly correlated

with symptom improvement across all patients.

To date, this study is the second randomized controlled trial of mPFC rTMS for TRD

(Kreuzer et al, 2015). The first mPFC-rTMS trial, the ‘AC/DC’ study, recruited 40 subjects who

completed 15 once daily sessions (3 weeks) of either standard DLPFC-rTMS, active mPFC-

rTMS, or sham rTMS. While Kreuzer et al. (2015) reported a significant treatment group-by-

time interaction on the primary clinical measure of their study, post hoc comparisons of 21-item

HAMD improvement revealed that both active mPFC-rTMS and DLPFC-rTMS were not

superior to sham rTMS. Despite the findings from randomized controlled trials, open-label

studies have found that roughly half of patients respond and one third of patients remit following

dmPFC-rTMS; these rates are comparable to recent randomized controlled trials of DLPFC-

rTMS (Bakker et al, 2015; Downar et al, 2014; Salomons et al, 2014).

As with the ‘AC/DC’ randomized controlled trial, the clinical results of Study III support

the idea that active dmPFC-rTMS is not superior over ‘sham’ stimulation. However, there are

two methodological differences between the current study and previous trials of open-label and

double-blind sham-controlled dmPFC-rTMS that could warrant further study of dmPFC-rTMS

for TRD. First, unlike the ‘AC/DC’ trial and our previous open-label studies, the current study

employed 30 twice-daily 20 Hz rTMS sessions instead of 20-30 once daily 10 Hz sessions. If the

twice-daily stimulation did not achieve a twofold acceleration of effect in most patients, then the

overall course length may be inadequate at 15 days. Regarding this point, however, one recent

study from our group of once-daily 10 Hz versus twice-daily 20 Hz open-label dmPFC-rTMS

showed similar rates of improvement session-by-session, and no significant difference in

response or remission rates (Schulze et al, 2018). In light of that finding, it could be the case that

the rate of response to active rTMS in the current sample was otherwise influenced by genetic or

molecular factors related to neuronal plasticity, such as BDNF, thereby limiting the clinical

232

efficacy of active accelerated 20 Hz rTMS in some patients in the randomized controlled trial

(Hwang et al, 2015).

A second potential confound is whether the ‘sham’ stimulation electrodes, placed at

regions between the DLPFC and frontal pole, might have inadvertently delivered therapeutic

doses of transcranial electrical stimulation, or trigeminal nerve stimulation. Trigeminal nerve

stimulation is an emerging novel non-invasive brain stimulation treatment for major depression

that uses electrodes placed over the V1 subregion of the trigeminal nerve (forehead); backwards

propagation of trigeminal stimulation to the cortex is thought to elicit improvements in MDD

symptom severity. Three open-label trials (Cook et al, 2013; Schrader et al, 2011; Shiozawa et

al, 2014) have reported significant clinical benefit of trigeminal nerve stimulation. One

randomized controlled trial has also shown that trigeminal nerve stimulation is superior to

placebo stimulation; in this trial patients received 10 sessions of stimulation over 2 weeks, with

electrodes placed above the supraorbital foramen (Shiozawa et al, 2015). Although the

stimulation protocol is different from that of our placebo stimulation, it could be the case that by

mimicking the sensory and nociceptive effects of active rTMS, we have inadvertently induced a

clinical effect in the placebo arm via V1 trigeminal nerve stimulation from the electrodes.

Along similar lines, one other study has reported that ‘sham’ rTMS with simultaneous

electrical stimulation achieved similar effects to active rTMS, with therapeutic efficacy

dependent on the site of stimulation and not on whether stimulation was active or ‘sham’ (Triggs

et al, 2010). Patients who received right DLPFC rTMS (active or sham) experienced greater

clinical response than those who received rTMS over the left DLPFC (Triggs et al, 2010). In

light of this finding, it could be the case that different patterns of electrical stimulation (1 Hz and

20 Hz) of ‘sham’ dmPFC-rTMS also have differing clinical efficacy. Future work should test the

possibility that certain ‘sham’ electrical scalp stimulation protocols delivers therapeutic doses of

stimulation over the 30-session, 15-day course. Further research is also warranted to clarify the

clinical benefit of trigeminal nerve stimulation, or stimulation of the DLPFC/frontopolar region.

Future work is also necessary to quantify the clinical effects of 20 Hz dmPFC-rTMS using

another placebo method (for example, vertex stimulation with a shielded coil).

Consistent with previous research, TRD patients displayed significantly lower baseline

frontostriatal rsFC to the OFC, and lower cortico-cortical rsFC between the VMPFC and the

233

OFC, compared to healthy controls. In healthy controls, the ventral striatum, VMPFC, and OFC

all play an active role the ventromedial affective network function, including during dopamine-

dependent reward processing, and affective processing (Tremblay et al, 2005; Zhang et al,

2013). In MDD, lower VMN activity has also been observed during reward-related task

performance; for example, lower VMN activation has been related to anhedonic symptoms

during the unexpected receipt of rewards (Segarra et al, 2016).

Low baseline frontostriatal connectivity between the ventral striatum bilateral inferior

frontal gyrus, frontal pole and dACC predicted TRD response to active 20 Hz dmPFC-rTMS but

not to active 1 Hz dmPFC-rTMS or sham dmPFC-rTMS. Consistent with previous findings from

our group, frontostriatal connectivity between the ventral striatum and dACC predicted response

to 20 Hz dmPFC-rTMS (Dunlop et al, 2015, 2016a). Furthermore, previous studies from our

group have also reported the role of other frontostriatal circuits in predicting dmPFC-rTMS

response. For example, our group recently reported that baseline ventromedial and ventrolateral

PFC frontostriatal rsFC predicted response to dmPFC-rTMS in TRD, ED and OCD (Dunlop et

al, 2015, 2016a; Salomons et al, 2014).

Baseline frontostriatal rsFC has also shown promise in predicting response to DLPFC-

rTMS and conventional interventions in MDD. For 10 Hz DLPFC-rTMS, greater improvement

in TRD symptom severity is associated with higher baseline frontostriatal connectivity in dorsal

PFC regions (Avissar et al, 2017). For conventional pharmacotherapies and psychotherapies,

VLPFC CSTC connectivity differentially predicts response to CBT or antidepressant medication

(Dunlop et al, 2017a). Taken together, baseline frontostriatal connectivity shows promise as a

biological marker of clinical response to active 20 Hz dmPFC-rTMS, and further research is

therefore warranted to characterize the individual-level prediction accuracy of such biomarkers

in a large sample size.

Contrary to previous studies of dmPFC-rTMS response, pre- to post-treatment CSTC rFC

change did not significantly correlate with response. Previous studies reporting the mechanisms

of DPFC- and dmPFC-rTMS response support the idea that 20 Hz rTMS normalizes functional

connectivity between fronto-limbic regions and striatal, pallidal and thalamic components of a

CSTC loop (Anderson et al, 2016; Noda et al, 2015). The results of study III do not support the

234

idea that the therapeutic mechanism of dmPFC-rTMS response is related to CSTC rsFC change,

contrary to previous studies of dmPFC-rTMS response in TRD, ED, and OCD.

Instead, clinical response irrespective of rTMS treatment arm was correlated with

improvement in emotion regulation, particularly related to goal-directed behaviour and impulse

control on the DERS. Nonspecific improvements in TRD symptom severity was also associated

with increases in dACC-VMPFC rsFC from pre- to post-treatment. Previous research in healthy

controls indicates that the role of the VMPFC and its structural and functional connections to the

dmPFC and dorsal ACC in emotion regulation, the regulation of negative affect, and social

cognition processes like emotion recognition and theory-of-mind (Kühn et al, 2011). Low

VMPFC rsFC connectivity appears to be a hallmark of MDD (Murrough et al, 2016).

Although MDD-related abnormalities in VMPFC-dACC connectivity are consistent with

previous studies, increases in dmPFC-VMPFC connectivity are not consistent with previous

literature investigating changes in rsFC related to response in MDD and TRD. Instead,

significant reductions in sgACC resting-state connectivity to the ventral affective network has

been associated with response to either active or placebo DLPFC-rTMS (Taylor et al, 2018). The

authors attribute non-specific changes in sgACC activity as ‘common pathway of improvement

in MDD,’ as decreases in sgACC activity and rsFC accompanies response to conventional

interventions and to sgACC-DBS (Delaveau et al, 2011). Instead, the results of Study III indicate

that treatment nonspecific symptom improvement was associated with significant increases of

the VMPFC to the dorsal ACC, a node of the SN. It could be the case that increases in ventral

affective and dorsal salience network connectivity reflect improved top-down control of VMN-

related emotion processing and cognition.

To conclude, we found no clinical superiority in the active versus sham treatment in a three-arm

randomized controlled trial of 1 Hz, 20 Hz and placebo dmPFC-rTMS, although the possibility

that the ‘sham’ electrodes in fact delivered therapeutic doses of stimulation bears future

investigation. While there were no significant baseline clinical predictors of improvement,

response across all treatment arms was associated with improvements in emotion regulation, and

in particular the goal-setting and impulsivity subscales of the DERS. On rs-fMRI, baseline

frontostriatal connectivity differentially predicted treatment response to sham and 20 Hz rTMS.

Replicating our previous findings (Dunlop et al, 2015; Salomons et al, 2014), low baseline

235

frontostriatal rsFC predicted treatment response to 20 Hz rTMS. Increases in cortico-cortical

connectivity between the dmPFC and the VMPFC was correlated with intervention non-specific

improvements. A logical next step would be to implement another form of sham rTMS in a

randomized controlled trial to avoid trigeminal/transcranial electrical stimulation in the placebo

arm, and to further characterize the effects of dmPFC-rTMS using TMS-EEG or interleaved

TMS-fMRI. Properly optimized and tested under an adequate placebo condition, dmPFC-rTMS

may yet provide a new treatment option in the challenging setting of TRD.

236

General Discussion

Summary of Results and Comparison to Hypotheses The goal of this thesis was to identify transdiagnostic neural predictors and mechanisms

of treatment response to rTMS over the dmPFC. To this end, this thesis has presented three rs-

fMRI studies to identify predictors and mechanisms of response to dmPFC-rTMS in TRD, OCD,

and eating disorders (AN-BP and BN). All studies acquired resting-state fMRI scans at pre- and

post-treatment in order to identify baseline differences in dmPFC rsFC between rTMS

responders versus non-responders, and pre- to post-treatment changes in dmPFC rsFC that were

associated with symptom improvement.

The first general aim of this thesis was to assess the clinical efficacy of dmPFC-rTMS as a

treatment for TRD, and to conduct preliminary assessments of therapeutic effects of

dmPFC-rTMS in OCD and eating disorders.

The first hypothesis of Study I was that binge eating and purging frequency would

significantly improve in AN-BP/BN patients who received 20-30 sessions of open-label 10 Hz

dmPFC-rTMS. In line with this hypothesis, 16 of 28 subjects with AN-BP or BN experienced a

50% or greater reduction in weekly binges and purges, four weeks following 20-30 sessions of

10 Hz dmPFC-rTMS. These patients also had significant improvements in depressive and

anxiety symptoms, as measured by the HAMD, BDI-II and BAI. Symptomatic improvements in

mood and anxiety were not significantly associated with improvements in disordered eating and

purging, meaning that open-label dmPFC-rTMS modulated disordered eating behaviours

independent of its effect on depression severity.

The first hypothesis of Study II was that the severity of obsessive thoughts and

compulsive behaviours in OCD patients would significantly improve following 20-30 sessions of

open-label 10 Hz dmPFC-rTMS. In line with this hypothesis, 10 of 20 subjects with treatment-

refractory OCD experienced a 50% or greater reduction in obsessive thoughts and compulsive

behaviours, as measured by the Y-BOCS, following 20-30 sessions of open-label 10 Hz dmPFC-

rTMS. As in our ED case series, all OCD patients also had significant improvements in

depressive and anxiety symptom severity, but these improvements did not significantly differ

between rTMS responders and non-responders as defined by OCD symptom improvement.

237

The first hypothesis of Study III was that improvements in depression symptoms with 20

Hz dmPFC-rTMS would be significantly greater than that achieved with 1 Hz dmPFC-rTMS or

placebo dmPFC-rTMS. The clinical outcomes in this study showed a significant effect of time

for all primary clinical measures (HAMD, BDI-II and BAI), such that patients on average

experienced a significant improvement in symptom severity across all rTMS interventions.

However, there was no significant group by time interaction on any primary clinical measure,

meaning that the degree of symptom improvement from baseline to the first follow-up visit did

not significantly differ between treatment interventions. Thus, this study was not able to

demonstrate clinical efficacy for either 20 Hz or 1 Hz active dmPFC-rTMS over the sham

intervention employed. Baseline demographic factors, and primary clinical factors did not

significantly predict who would respond to dmPFC-rTMS. Furthermore, none of the secondary

clinical measures assessing emotion regulation, perfectionism, impulsivity, reward/punishment

processing, rumination, or the five personality factors predicted response to 20 Hz, 1 Hz or

placebo dmPFC-rTMS in Study III.

The second general aim of this thesis was to identify significant differences in pre-

treatment resting-state functional connectivity differences between dmPFC-rTMS

responders and non-responders for TRD, OCD, and eating disorders.

The second hypothesis of Study I was that lower baseline dmPFC-CSTC and higher

baseline dmPFC-sgACC resting-state functional connectivity would be associated with better

response to dmPFC-rTMS. In line with this hypothesis, AN-BP and BN dmPFC-rTMS

responders had significantly lower dACC rsFC to the right putamen, as well as the hippocampus

and posterior insula (Figure 7-1 A. and B.). However, contrary to this hypothesis, AN-BP and

BN dmPFC-rTMS responders also had significantly higher dmPFC rsFC to the and bilateral

lateral OFC, bilateral temporal pole and right posterior insula.

The second hypothesis of Study II was that, as with DBS, OCD patients with higher

dmPFC-CSTC resting-state functional connectivity would show better response to open-label 10

Hz dmPFC-rTMS. In line with this hypothesis, dmPFC-rTMS responders had displayed

significantly higher frontostriatal connectivity between the ventral rostral putamen ROI and

238

dACC relative to dmPFC-rTMS non-responders and healthy controls in an exploratory analysis

(Figure 7-1 C.). Furthermore, OCD dmPFC-rTMS responders also exhibited significantly higher

baseline rsFC between the dmPFC ROI and the ventral striatum in an a priori analysis. In

further exploratory analyses, OCD rTMS responders also displayed increased rsFC between the

medial dorsal thalamus ROI and lateral OFC (Figure 7-1 D.), and between the STN ROI,

thalamus and striatum relative to non-responders.

The second hypothesis of Study III was that TRD patients with lower dmPFC-CSTC and

higher dmPFC-VMPFC resting-state functional connectivity would show greater symptom

improvement with active 20 Hz dmPFC-rTMS, but not with active 1 Hz or placebo dmPFC-

rTMS. In line with this hypothesis, the results of a 3-way continuous covariate interaction

analysis showed that response specifically to the 20 Hz active dmPFC-rTMS intervention was

anti-correlated with baseline frontostriatal rsFC. Lower baseline frontostriatal connectivity

between the inferior VS ROI and right IFG and OFC was correlated with a greater response to

active 20 Hz dmPFC-rTMS (Figure 7-1 F.). Similarly, an analysis of baseline rs-fMRI predictors

focusing specifically on the 20 Hz active dmPFC-rTMS study arm found that clinical

improvement was correlated with lower baseline frontostriatal rsFC between the inferior VS and

the bilateral IFG and OFC, and between the superior VS and the dACC and dmPFC (Figure 7-1

E.).

239

Figure 7-1: Summary of differences in pre-treatment resting-state functional connectivity

between dmPFC-rTMS responders and non-responders for AN-BP/BN (Study I, A. and

B.), OCD (Study II, C. and D.), and TRD (Study III, E. and F.). Green and red clusters

indicate regions of interest; blue clusters indicate regions where lower baseline rsFC correlates

with greater subsequent dmPFC-rTMS response; orange clusters indicate regions where higher

baseline rsFC correlates with greater subsequent dmPFC-rTMS response.

240

The third aim of this thesis was to identify pre- to post-treatment changes in resting-state

functional connectivity that accompany dmPFC-rTMS response for TRD, OCD, and eating

disorders.

The third hypothesis of Study I was that increases in pre- to post-treatment CSTC-

dmPFC resting-state functional connectivity after dmPFC-rTMS would correlate with

improvements in binge-eating and purge frequency. Consistent with this hypothesis, increased

dmPFC resting-state functional connectivity to the bilateral dorsal striatum was correlated with

improvements in weekly binge and purge frequency at 4-weeks follow-up (Figure 7-2 A.).

Furthermore, response to dmPFC-rTMS in AN-BP and BN was correlated with increases in rsFC

between the dACC and the bilateral AI, IFG and OFC.

The third hypothesis of Study II was that, as with DBS, OCD patients who show

decreases in pre- to post-treatment dmPFC-CSTC resting-state functional connectivity after

dmPFC-rTMS would show more improvement in OCD symptom severity. Consistent with this

hypothesis, decreases in rsFC between the dACC and the bilateral caudate nucleus and

mediodorsal thalamus was correlated with OCD symptom improvement (Figure 7-2 B.). OCD

dmPFC-rTMS responders had abnormally high baseline dmPFC-caudate rsFC relative to a

sample of healthy controls, and this hyperconnectivity significantly decreased with successful

treatment. We also observed that successful OCD rTMS response was associated with increased

dmPFC connectivity to the bilateral pre- and post-central gyrus and left precuneus, and decreased

connectivity to the midbrain, superior frontal gyrus, and right hippocampus.

The third hypothesis of Study III was that TRD patients showing greater increases in pre-

to post-treatment dmPFC-CSTC resting-state functional connectivity would show greater

improvement in clinical symptoms, after active 20 Hz dmPFC-rTMS but not after active 1 Hz or

placebo dmPFC-rTMS. A 3-group continuous covariate interaction analysis was used to identify

significant differences in the correlation between rsFC change and symptom improvement across

the three treatment arms. Contrary to this hypothesis, the between-arms differences in the change

in rs-FC associated with HAMD improvement were found not in the connectivity between

dmPFC and its associated CTSC regions, but rather in the connectivity between dACC and right

241

parietal operculum. For this functional connection, in the 20 Hz study arm, those with the

greatest response exhibited the greatest increase in dACC-parietal operculum rsFC. However,

there was no such association observed in the 1 Hz study arm. Finally, a negative correlation was

found in the sham treatment arm, such that those with the greatest improvement exhibited the

greatest decrease in dACC-parietal operculum rsFC. Thus, although between-arms differences

in the neural correlates of improvement were detected in Study III, they were found in a circuit

different from the expected dmPFC-CTSC circuit.

Regarding common correlates of change across all three study arms, there was a positive

correlation between symptom improvement and change in rsFC between the VMPFC and the

bilateral dACC and rostral ACC. Treatment responders, regardless of rTMS treatment arm,

displayed lower baseline VMPFC-dACC rsFC relative to both non-responders and healthy

controls (Figure 7-2 C.), and the rsFC between these regions increased significantly post-

treatment in a manner correlated with response. Post-treatment, among those who responded,

there was no longer a significant difference relative to controls or non-responders in this

VMPFC-dACC circuit, suggesting normalization of rsFC with treatment response in this circuit,

across all three study arms.

Regarding supplementary clinical measures, improvements on the impulsivity and goal-

setting subscales of the DERS and the total DERS score significantly correlated with symptom

improvement across all three groups of the randomized controlled trial. Across all three

interventions, patients who improved over the course of treatment exhibited greater

improvements in self-reported emotion regulation on the DERS. Examining outcomes within

each treatment arm separately, the only correlation that survived multiple comparisons correction

was present in the 20 Hz dmPFC-rTMS arm: a significant correlation between the global DERS

improvement and both BDI-II and HAMD improvement, such that those who saw the greatest

response had greater improvements in emotion dysregulation.

242

Figure 7-2: Summary of pre- to post-treatment changes in resting-state functional

connectivity that accompany dmPFC-rTMS response in AN-BP/BN (Study I, A.), OCD

(Study II, B), and TRD (Study III, C.). Green and red clusters indicate regions of interest; blue

clusters indicate decreases rsFC accompanying dmPFC-rTMS response; orange clusters indicate

increases in rsFC accompanying dmPFC-rTMS response.

243

Implications for the Treatment of Psychiatric Disorders Patients with treatment-resistant MDD, AN-BP/BN, and/or OCD commonly suffer a

profound impairment in day-to-day function. As noted in the Introduction, MDD is the leading

cause of years lived with a disability (Ferrari et al, 2013), and accounts for roughly US$2 billion

in workplace productivity losses per year (Birnbaum et al, 2010). In Ontario, the MDD-related

burden of illness was greater than the combined burden of breast, colorectal, lung, and prostate

cancer (Ratnasingham et al, 2013). Along similar lines, even OCD symptom scores in the ‘mild’

range on standard scales can result in severe deficits in a patient’s quality of life (Moritz et al,

2005). For patients with AN or BN, comprehensive treatment programs are costly, onerous, and

difficult to access, limiting their impact on the prevalence and overall burden of disease

(Bamford and Sly, 2010). Eating disorders also have a significant mortality rate: the overall

mortality rate for AN is substantially higher than that of schizophrenia and MDD (Arcelus et al,

2011). Effective interventions are therefore required to address the burden and mortality of these

illnesses.

Unfortunately, MDD, OCD and eating disorders are typically chronic disorders, with

many patients showing little or no response to conventional interventions. Only half of patients

in a current major depressive episode will remit in the first 3 months (Patten et al, 2015), and

another third will not remit following four sequential pharmacotherapy interventions (Rush et al,

2006). For OCD, between 30 and 60% of patients will not adequately respond to standard

pharmacotherapies or psychotherapies, and even those who show symptomatic improvement on

such treatments often continue to suffer persistent impairment in daily function (Steketee, 1997).

AN and BN are also chronic disorders; only a third of patients will recover within 4 years of

disease onset (Berkman et al, 2007; Steinhausen, 2002). Approximately 25% of AN patients will

continuously relapse or develop a chronic form of the illness (Berkman et al, 2007; Steinhausen,

2002). As with AN, BN shows relapse rates between 25-63%, depending on the definition of

relapse and follow-up length (Grilo et al, 2012; Halmi et al, 2002; Herzog et al, 1999; Keel et al,

2005; Olmsted et al, 2005). Novel interventions are therefore required for each of these

challenging illnesses.

244

As noted in the Introduction, neuromodulation techniques are emerging as promising new

treatments for cases resistant to medications and psychotherapy. In TRD, the oldest and most

potent of these is ECT. Yet despite its high remission rate, only a small proportion of treatment-

resistant psychiatric patients receive ECT. Less than 1% of TRD patients are treated with ECT

(Delva et al, 2011) due in part to capacity limitations, but equally due to other factors such as the

risk of cognitive impairment, the need for general anesthesia, the stigma associated with the

treatment, and logistical barriers to accessing treatment (Andrade and Thyagarajan, 2007; Delva

et al, 2011). ECT also offers limited benefit for OCD and eating disorders.

DBS is another invasive alternative for patients with severe intractable TRD, OCD or

AN. Studies of long-term treatment outcomes for sgACC-DBS show remission rates of

approximately 40-50% 2 years post-implantation (Holtzheimer et al, 2012; Kennedy et al, 2011).

DBS for OCD has also been clinically promising in preliminary studies: two-thirds of patients

achieved response after 12-months of active DBS compared to sham (Goodman et al, 2010), and

those who achieve initial benefit exhibit the same treatment response over up to 9-years of

follow-up (Fayad et al, 2016). Again, however, DBS is not a widely accessible or well-accepted

intervention: one recent survey of adults with self-reported OCD showed that DBS is the least

accepted and least preferred novel treatment (Patel et al, 2017). In addition, the evidence base to

date is still rather limited. Although some DBS studies are double-blind, most have extremely

small sample sizes. Finally, the resources and expertise needed to perform DBS are generally

confined to academic centres of research excellence, thereby limiting both the generalizability of

clinical results and the ability of patient access to care. Although it can be highly effective in

those individual cases where resources and expertise are available, in terms of population health,

DBS is not well-positioned to achieve large reductions in the overall prevalence or societal

burden of TRD, OCD or eating disorders.

Given the limited accessibility and acceptability of DBS and ECT, alternative

interventions are needed to serve the substantial population of patients who do not respond to

conventional psycho- and pharmacotherapies. rTMS is one such intervention that non-invasively

stimulates the brain, has minimal serious side effects, has high patient acceptability and

tolerability superior to antidepressant medications (Milev et al., 2016). rTMS is more scalable

than DBS, as it is non-invasive and does not require neurosurgical expertise; concurrently-run

treatment rooms can be operated by trained technicians under the supervision of a psychiatrist,

245

increasing capacity. rTMS is also an increasingly accepted tool for treatment-refractory

psychiatric cases given its advantages over ECT: less associated stigma, no need for general

anesthesia, and an absence of cognitive side-effects (Schulze et al, 2016). From an efficacy

standpoint, the main drawback to rTMS compared to ECT is a lower overall rate of remission.

Improving remission rates, by identifying predictors of outcome and mechanisms of treatment

effect, is thus an important goal for current rTMS research.

This thesis reports the first open-label series of dmPFC-rTMS for medically intractable

AN-BP/BN and OCD. dmPFC-rTMS showed promising therapeutic effects in a subset of

patients who report treatment-refractory obsessional thoughts and compulsive behaviours, or

binge and purge behaviours. As DLPFC stimulation has shown lesser efficacy for patients with

OCD or an eating disorder, dmPFC-rTMS appears to be a viable alternative worthy of future

study. Neural predictors and correlates appear to involve CSTC circuitry, and specifically

implicate the SN, which may help to explain the transdiagnostic therapeutic effects of dmPFC-

rTMS given the role of the SN in cognitive control, a domain in which deficits are seen

transdiagnostically in psychiatric disease.

This thesis also reports the second randomized controlled trial of dmPFC-rTMS for TRD,

and the first dmPFC-rTMS trial with a sample size >100 individuals. Although Study III did not

show clinical superiority of either the active 20 Hz or 1 Hz dmPFC-rTMS arm over the sham

intervention employed, some commonalities were observed among the neural correlates of

improvement in the sham and the active arms. As such, it remains to be clarified whether the

electrodes used to generate scalp sensations in the ‘sham’ intervention might actually have

delivered a therapeutic form of stimulation (transcranially or via the trigeminal nerve) over the

30 sessions of treatment. In future, additional double-blind randomized controlled trials with an

alternative form of sham intervention should be used to further assess the efficacy of dmPFC-

rTMS, and to optimize rTMS stimulatory parameters to maximize its therapeutic efficacy in

psychiatric disorders. Properly optimized, dmPFC-rTMS has the potential to become a potent

treatment for patients with debilitating and treatment refractory illnesses like TRD, BN or OCD.

Identifying and optimizing new treatments such as rTMS is crucial, given that treatment-resistant

MDD alone afflicts some 2% of the entire Canadian population. Effective, scalable treatment

options in TRD would also reduce health system and economic costs, such as those associated

with disability, caregiver burden, and direct costs of inpatient and outpatient care.

246

Clinical Implications of Identifying Baseline Predictors of dmPFC-rTMS Response.

To date, Study I and Study II are the first case series to report pre-treatment

neurobiological differences that differentiate dmPFC-rTMS responders from non-responders in

patients with OCD and eating disorders. Study III is third to report baseline biomarkers of

response to dmPFC-rTMS for TRD, and the first to report such biomarkers in a sham-controlled

clinical trial.

As previously discussed, identifying baseline biomarkers has two treatment implications

for patients suffering from mental illness. First, distinguishing treatment responders from non-

responders prior to treatment may avoid having patients undergo logistically onerous but futile

courses of treatment, and thereby accelerate the trial-and-error process of finding an effective

treatment for a given patient. A significant time-commitment is also required of rTMS patients,

especially since rTMS requires patients to receive treatment at a clinic or hospital every weekday

over many weeks; consequently, avoiding futile treatment and improving outcomes among

treated patients by predicting responders and non-responders to a given rTMS protocol is an

important goal.

A second implication of identifying reliable biomarkers of treatment response is that such

biomarkers may help to inform the search for new treatments for those who do not respond. For

example, a neuroimaging biomarker may reveal brain regions whose baseline activity or

connectivity differs between responders and non-responders to a given treatment. Brain regions

that are hyper- or hypo-active, or that have a different pattern of network connectivity in

responders versus non-responders, may present neurostimulatory targets worthy of future clinical

investigation. As an example, a recent ‘biotyping’ study in MDD found that one group of

patients who do not respond well to dmPFC-rTMS show a ‘nexus’ of abnormal connectivity in

the right lateral orbitofrontal cortex (lOFC) – an area amenable to rTMS intervention.

The neuroimaging biomarkers of dmPFC-rTMS reported in Study III likewise revealed

potential neurostimulatory targets worthy of future clinical investigation. In Study III, high

baseline frontostriatal connectivity from the right lOFC and IFG was associated with non-

247

response to active 20 Hz dmPFC-rTMS, but not to active 1 Hz dmPFC-rTMS or sham dmPFC-

rTMS. Consequently, the results of Study III provide an independent line of evidence supporting

the premise that inhibitory right lOFC stimulation could be a potential alternative intervention

for those who do not respond to dmPFC-rTMS.

Consistent with this premise, two preliminary studies from our group have recently

reported that inhibitory right lOFC rTMS may be an effective alternative for those who are non-

responsive to dmPFC-rTMS. First, we reported a case in which a TRD patient who had

previously failed both dmPFC- and DLPFC-rTMS achieved remission with 1 Hz right lOFC-

rTMS (Fettes et al, 2017a). A subsequent case series from our group in N=47 patients found that

patients who were initially non-responsive to dmPFC-rTMS and then went on to receive a course

of 1 Hz right OFC-rTMS achieved response and remission rates of 35.7% and 23.8%,

respectively (Feffer et al, 2018). Thus, preliminary clinical findings support the hypothesis

emerging from the findings of Study III that right OFC-rTMS may be worth pursuing in

dmPFC_rTMS non-responders. This implication from Study III is a direct demonstration of the

high potential translational value offered by baseline biological markers of response for the

identification of novel therapeutic strategies that may be useful in treatment non-responders.

Clinical Implications of Identifying the Neurobiological Mechanism of dmPFC-rTMS Response.

As noted earlier, Study I and II are the first case series to report pre- to post-treatment

changes in rsFC that accompany 10 Hz dmPFC-rTMS response in treatment refractory OCD and

eating disorders. Study III is also the first randomized sham-controlled trial of 20 Hz dmPFC-

rTMS for TRD to characterize pre- to post-treatment rsFC changes correlated with improvements

in depression severity. Understanding the biological mechanism of an intervention has

implications in two clinical areas. First, it is helpful to clarify whether abnormal markers (for

example, IBN structure or function) are state-dependent markers of the current depressed mood

state, or are stable trait-like features that persist during response or remission. In OCD, for

example, dACC frontostriatal hyperconnectivity normalized following 20-30 sessions of 10 Hz

dmPFC-rTMS. This mechanism of response was not only consistent with a previous VS-DBS

248

study in OCD, but also allows for characterization of which functional features normalize as

symptoms remit (i.e. which features represent states rather than traits).

On a related point, understanding the biological mechanism of treatment remission could

help clarify the biological mechanisms by which an individual may suffer relapse, and this

knowledge in turn may lead to robust biological predictors of treatment relapse. Hypothetically,

if dACC frontostriatal hyperconnectivity re-appears on a subsequent follow-up fMRI scan, it

could predict an impending relapse or worsening of obsessive-compulsive symptomatology.

Second, from a therapeutic and translational standpoint, the identification of neural

markers that change with successful but not unsuccessful treatment could help lead to new

treatments that might be more effective in treatment non-responders in future. For example, the

evidence from Studies I-III collectively suggests that dmPFC-rTMS acts to modulate the

integrity of the SN as a cohesive IBN, and additionally acts to modulate the connectivity in the

CSTC circuits serving the dmPFC site of stimulation. The former finding suggests that future

treatments for non-responders may need to target mechanisms other than cognitive control, such

as abnormal signalling in the positive or negative valence systems – a premise supported by

some preliminary evidence on rOFC-rTMS, as above. The latter finding suggests that treatment

effects might be enhanced by co-interventions that affect the activity and plasticity of CSTC

circuitry: for example, active engagement in a cognitive task during stimulation (Isserles et al,

2013), or co-administration of a dopaminergic medication during treatment, as supported by

recent evidence (Enomoto et al, 2015).

Taken together, Studies I-III have three main clinical implications for patients suffering

from treatment-resistant mental illness. First, properly optimized, dmPFC-rTMS has the potential

to become a potent treatment for patients with debilitating and treatment refractory illnesses.

Effective, non-invasive, scalable treatment options in TRD such as dmPFC-rTMS could also

reduce health system and economic costs, such as disability, caregiver burden, and direct costs of

inpatient and outpatient care. Second, reliable baseline biomarkers of dmPFC-rTMS response

have the potential to improve outcomes by individualizing treatment selection and circumventing

the trial-and-error process of finding an optimal treatment. Third, characterizing the biological

mechanisms of dmPFC-rTMS response has the potential to inform the development of new

treatment strategies, and to predict and avoid illness relapse. In the longer term, both the

249

biological predictors and neurobiological mechanisms of response may provide useful

information to help to develop new interventions that are more effective for treatment non-

responders.

Comparing the Clinical Efficacy of dmPFC-rTMS to DLPFC-rTMS The clinical results of this thesis suggest that dmPFC-rTMS might be beneficial for a

wide array of psychiatric symptoms, including disordered eating, obsessions, compulsions,

depression and anxiety. This finding diverges from the poorer outcomes previously reported for

OCD and ED in studies using DLPFC-rTMS rather than dmPFC-rTMS. For OCD, both open-

label and sham-controlled trials of DLPFC-rTMS have had inconsistent results in improving

clinical symptoms (Sachdev et al, 2001). Early studies showed that OCD patients experienced

some clinical benefit with 20 Hz left DLPFC-rTMS in an open-label setting (Greenberg et al,

1997), while other sham-controlled studies of 1 Hz right DLPFC-rTMS showed no clinical

difference relative to sham (Alonso et al, 2001). Subsequent randomized sham-controlled trials

of 1 Hz right DLPFC-rTMS (Prasko et al, 2006) and 20 Hz left DLPFC (Sachdev et al, 2007)

have confirmed that active DLPFC-rTMS is not clinically superior to placebo in OCD.

Consequently, Study I provide early evidence that dmPFC-rTMS might be a more favourable

treatment for OCD relative to DLPFC-rTMS.

As with OCD, individuals who present with binge eating and purging do not appear to

show consistent benefit from DLPFC-rTMS. Studies using high frequency left DLPFC-rTMS

show inconsistent clinical efficacy in eating disorders. Two open-label studies reported that a

single session reduced the urge to eat, reduced the number of binges 24-hours post-stimulation,

and reduced salivary cortisol (Claudino et al, 2011; Van den Eynde et al, 2010), while another

found no difference in efficacy between active- and sham-stimulation after 15-sessions of 20 Hz

DLPFC-rTMS (Walpoth et al, 2008). Furthermore, the only two randomized double-blind sham-

controlled trials of 10 Hz left DLPFC-rTMS found that this intervention was not superior to

placebo for BN (Gay et al, 2016) or AN (McClelland et al, 2016).

250

Relative to DLPFC-rTMS, dmPFC-rTMS may represent a favourable treatment strategy

for BN. An early case report from our group reported an unanticipated remission of binge-eating

and purging in a TRD patient with comorbid BN following 20 sessions of 10 Hz dmPFC-rTMS

(Downar et al, 2012). This effect on binge-eating and purging was robust and was maintained for

9-weeks post-treatment. The clinical results of Study II confirms the initial finding from our case

report, showing that 10 Hz dmPFC-rTMS can significantly reduce the frequency of binge and

purge episodes. Given this, a reasonable next step would be to assess the efficacy of 10 Hz

dmPFC-rTMS under double-blind, placebo-controlled conditions.

For TRD, the clinical efficacy of 20 Hz dmPFC-rTMS was not superior to that of sham in

Study III, suggesting that dmPFC may not be a preferable target for rTMS compared to left or

right DLPFC in overall terms. These results were consistent with one recent sham-controlled trial

of daily 10 Hz dmPFC-rTMS. In this study, Kreuzer and colleagues (2015) reported that there

was a significant group-time interaction on improvements in depression severity when

comparing active DLPFC-rTMS, active dmPFC-rTMS and sham stimulation (Kreuzer et al,

2015). However, subsequent post hoc tests assessing the difference in symptom improvement

between active dmPFC-rTMS and placebo did not show that active dmPFC-rTMS was clinically

superior to placebo rTMS. Thus, one interpretation of the clinical results of Study III is that the

improvements seen with dmPFC-rTMS are due to non-specific factors and not do the

neurobiological effects of the intervention. However, given that the sham arm in Study III did

indeed show some changes in rsFC similar to those in the active arms, another interpretation is

that the electrical stimuli delivered to patients in the sham arm actually did stimulate the brain in

a therapeutic manner, either transcranially or via the trigeminal nerve. This possibility requires

further investigation in future work.

Of note, despite the null findings of sham-controlled trials of dmPFC-rTMS, three open-

label studies of dmPFC-rTMS show comparable efficacy rates to DLPFC-rTMS. First, our group

published a chart review of 98 patients who received 10 Hz dmPFC-rTMS in MDD, finding

response and remission rates comparable to that of DLPFC-rTMS (50.6% and 38.5%,

respectively) (Bakker et al, 2015). Two other open-label case series from our group have also

reported significant improvements in depressive symptomatology following 20 sessions of 10 Hz

dmPFC-rTMS, with sharply bimodal rates of response and substantial improvement in

individuals who achieved response to treatment (Downar et al, 2014; Salomons et al, 2014).

251

In sum, Studies I and II provide early evidence that dmPFC-rTMS could be a preferable

and more clinically effective intervention relative to DPFC-rTMS for OCD and ED. Clinical

trials of dmPFC-rTMS for these indications is therefore warranted to confirm the efficacy of this

intervention under sham-controlled settings. Despite null findings in early sham-controlled trials

of dmPFC-rTMS for TRD, the results of open-label studies suggesting comparable response and

remission rates to DLPFC-rTMS support the need for further research, possibly employing an

alternative sham stimulation technique, to examine the clinical efficacy of this intervention for

TRD.

dmPFC-rTMS as a Transdiagnostic Intervention in Psychiatric Illness The results reported in Studies I, II and III provide early evidence that rTMS targeting the

dmPFC could serve as a transdiagnostic intervention with efficacy across a range of traditional

psychiatric diagnostic categories. 10 Hz dmPFC-rTMS appears to achieve significant

improvements in OCD symptom severity and binge-eating and purging in ED. Notably, OCD

and ED patients with comorbid depressive or anxiety symptoms also achieved significant

improvements in these clinical dimensions following dmPFC-rTMS. However, these

improvements in depression and anxiety symptom severity did not correlate with improvements

in obsessive thoughts and compulsive behaviours in OCD or with improvements in disordered

eating and purging in ED. This finding suggests that these clinical improvements following

dmPFC-rTMS occur independently of TRD response, and that dmPFC-rTMS might act as a

transdiagnostic intervention.

From a clinical perspective, why might dmPFC-rTMS induce benefit across a diverse

range of symptoms? For one, this transdiagnostic clinical efficacy is not unique to dmPFC-rTMS

as many interventions, including SSRIs and psychotherapy, have proven beneficial for all of

these disorders. One potential account for such transdiagnostic efficacy draws fron the

observation that MDD, AN, BN, and OCD have common comorbidity. MDD is rarely the only

or primary psychiatric disorder as 72.1% of lifetime and 78.5% of 12-month cases of MDD have

a comorbid DSM diagnosis (Kessler et al, 2003). Similarly, OCD is often comorbid with other

psychiatric disorders, and common comorbidities include anxiety disorders, MDD, bipolar

252

disorder (Ruscio et al, 2010), and obsessive-compulsive personality disorder (Eisen et al, 2010).

Another common disorder comorbid with OCD is tic disorder, present in 30% of OCD cases

(Leckman et al, 2010). 56.2% of patients with AN and 94.5% of patients with BN also meet

criteria for another psychiatric disorder (Hudson et al, 2007). Mood disorders like depression and

dysthymia are found in 42.1% of AN and 70.7% of BN patients (Hudson et al, 2007).

This overlap in comorbidity between MDD, OCD and ED is perhaps suggestive of sub-

dimensions of psychiatric illness that exist irrespective of DSM-5 diagnosis. Such sub-

dimensions that go beyond the DSM diagnosis are likely related to genetic, cognitive or

biological factors. Such similarities make sense given that many interventions, including SSRIs

and psychotherapy, have proven beneficial for all of these disorders. The possibility of non-

congruence between biological mechanisms and conventional diagnoses also suggests that

studies recruiting patients purely on the basis of DSM-5 or ICD-10 criteria may fail to achieve

homogeneity in the patient sample, and may instead capture a rather heterogeneous set of

subpopulations from a neurobiological perspective.

Transdiagnostic Efficacy of dmPFC-rTMS: A Biological Perspective The observation of transdiagnostic efficacy for dmPFC-rTMS across many psychiatric

disorders is suggestive of sub-dimensions of psychiatric illness that likely reflect cognitive or

neurobiological endophenotypes common across a wide spectrum of mental illness. In particular,

abnormalities in the domain of cognitive control may be a common element across many

psychiatric disorders (as reviewed by (McTeague et al, 2016)), as broad deficits in tasks related

to cognitive control are observed irrespective of psychiatric disorder class (Snyder et al, 2015).

Additionally, large phenotypic epidemiological studies across disorders have demonstrated that

specific patterns of clinical symptoms are strongly related to an underlying ‘general

psychopathology’ factor that may correspond in part to cognitive control deficits (Carragher et

al, 2016; Krueger, 1999; Lahey et al, 2012). However, the relationship between these broad

cognitive control deficits and the specific forms of psychopathology encountered in conventional

psychiatric disorders such as major depression, OCD, or eating disorders is not well understood.

253

Taken together, a common deficit in cognitive control might be the behavioural factor that

mediates the therapeutic effects of dmPFC-rTMS across multiple disorders.

As a notable point in support of the premise that cognitive control might be a

transdiagnostic factor of response, rTMS over the dmPFC has been previously shown to

modulate cognitive control in healthy controls. Cho et al. demonstrated that, relative to sham,

active medial PFC excitatory rTMS releases striatal dopamine and improves cognitive control on

a delay discounting delay (Figure 1-11) (Cho et al, 2015). Similarly, disrupting mPFC function

using single-pulse TMS or inhibitory rTMS impairs conflict monitoring and increases errors

during cognitive control tasks (Duque et al, 2012; Taylor et al, 2007). These observations

suggest that dmPFC-rTMS may act to enhance cognitive control in general, and that this effect

may have therapeutic value in some individuals with psychiatric illness, regardless of the specific

diagnostic category with which they present. On the other hand, Study III included secondary

clinical questionnaires assessing impulsivity and cognitive control, and no secondary clinical

outcomes of these domains predicted response to active dmPFC-rTMS at baseline, or

significantly changed with response to active rTMS. Further research in psychiatric populations

is therefore warranted to elucidate the cognitive and behavioural mechanisms of transdiagnostic

dmPFC-rTMS response.

The transdiagnostic structural and functional abnormalities of the dACC and SN

reviewed in the Introduction support the premise that dmPFC-rTMS could achieve clinical

benefit over a wide range of psychiatric symptoms. Regarding transdiagnostic structural

abnormalities, three large neuroimaging studies have reported that abnormalities of dACC are

common to many diverse disorders. Goodkind and colleagues demonstrated that dACC gray

matter loss relative to healthy controls was a common feature across six disorders, and that this

gray matter loss corresponded with poor cognitive control (Goodkind et al, 2015). This result

was replicated by two independent meta-analyses, reporting that dACC gray matter loss in the

dACC was common to both unipolar and bipolar depression (Wise et al, 2017), and

schizophrenia (Chang et al, 2018). Specific to the psychiatric disorders of the present study,

dACC volumetric abnormalities are reported in MDD, OCD and ED across a number of studies

in the literature. For MDD, dACC and SN volumetric reductions have been related to

behavioural deficits on such cognitive tasks relative to healthy controls(Li et al, 2010a). ACC

gray matter volume is also reduced relative to healthy controls in OCD (Gilbert et al, 2008;

254

Kühn et al, 2013; Reess et al, 2016), and in ED (Coutinho et al, 2015; Frank et al, 2013;

Friederich et al, 2012; Schäfer et al, 2010; Titova et al, 2013).

Regarding transdiagnostic functional abnormalities, functional SN and dACC

abnormalities appear to be associated with general psychopathology, although there is substantial

heterogeneity in terms of whether this abnormality is related high versus low activity or

connectivity. General psychopathology is associated with SN hypoactivity during a working

memory task (Shanmugan et al, 2016), as well as elevated cerebral blood flow and dACC

frontostriatal hypoconnectivity (Kaczkurkin et al, 2017). However, McTeague and colleagues

recently that SN hyperactivity was also present in a variety of disorders during various cognitive

control paradigms relative to controls (McTeague et al, 2017). Thus, the common element may

be more generally an abnormal functioning of the SN across disorders, rather than a universal

etiology of hypo- or hyper-functioning in this IBN across disorders.

Studies that separately examine MDD, OCD, and ED have also reported abnormal SN

and dACC activity relative to healthy controls. For OCD, dACC hyperactivity and

hyperconnectivity is observed related to controls: for example, dACC is abnormally hyperactive

during symptom provocation and while committing errors in cognitive control tasks (Bourne et

al, 2012). Frontostriatal hyperconnectivity from the dACC to other brain regions has also been

observed in OCD patients relative to controls (Bourne et al, 2012; Rauch et al, 2006).

In MDD, both SN hypo- and hyperactivity has been reported, albeit for different types of

cognition. SN and dACC hypoactivity is associated with maladaptive rumination, (Alexopoulos

et al, 2012; Bartova et al, 2015; Belleau et al, 2015; Hamilton et al, 2011b; Lemogne et al, 2012;

Rzepa and McCabe, 2016; Sheline et al, 2009; Zhu et al, 2017), deficits in incentive salience,

and with anhedonia (Admon et al, 2015; Chantiluke et al, 2012; Knutson et al, 2008; Manelis et

al, 2016; Nestler and Carlezon, 2006; Pizzagalli et al, 2009; Robinson et al, 2012; Schiller et al,

2013; Smoski et al, 2011; Tremblay et al, 2005). Conversely, MDD patients also exhibit dACC

hyperactivity on tasks that assess working memory and cognitive control, including the n-back

(Harvey et al, 2005) and Stroop (Holmes and Pizzagalli, 2008; Wagner et al, 2006). These

observations highlight the earlier point that psychopathology may emerge from abnormal SN

activity in either direction, rather than specifically hypo- or hyper-functioning.

255

Relative to controls, BN patients reliably have abnormal SN activity related to deficits in

cognitive control and impulsivity. Specifically, SN hypoactivity is associated with food

impulsivity. For example, lower dACC activity in response to food reward anticipation predicts

how much a BN patient will overeat (Bohon and Stice, 2011; Frank et al, 2006). Also, SN

hypoactivity during a cognitive control (Stroop) task is associated with poor dietary restraint in

patients who binge eat (Balodis et al, 2013). BN patients also have frontostriatal hypoactivity

circuitry during cognitive control tasks such as the Simon Spatial Incompatibility task (Celone et

al, 2011; Marsh et al, 2009b, 2011), and the go/no-go task (Skunde et al, 2016). Behavioural

deficits and frontostriatal hypoconnectivity are also reported in AN patients during cognitive

control tasks, including during response inhibition (Oberndorfer et al, 2011; Wierenga et al,

2014), the Wisconsin Card Sorting Task (Lao-Kaim et al, 2015), and delay discounting (Decker

et al, 2015; Wierenga et al, 2014).

Taken together, previous studies reporting transdiagnostic abnormalities of brain

structure and function in psychiatric illness provide a possible explanation for observations that

rTMS over the dmPFC may elicit improvements across a wide array of disorder-specific

symptoms, including TRD, OCD and ED. This transdiagnostic clinical benefit may be related to

illness-associated deficits in cognitive control, and structural/functional abnormalities localized

to the SN and dACC that are common across many clinical phenotypes. Further supporting this

interpretation, studies recruiting healthy controls have found that dmPFC-rTMS causally

modulates dACC function and behaviour during cognitive control paradigms.

Frontostriatal Functional Connectivity as a Potential Transdiagnostic Biomarker of dmPFC-rTMS Response

In all three studies, baseline frontostriatal connectivity differed between responders and

non-responders to excitatory dmPFC-rTMS. For patients with AN-BP, BN, or TRD, dmPFC-

rTMS responders displayed significantly lower frontostriatal rsFC relative to non-responders.

For patients with OCD, however, dmPFC-rTMS responders displayed significantly higher

frontostriatal rsFC.

256

These baseline frontostriatal rsFC markers identified as predictive of response are

somewhat similar to those in our first open-label case series of 25 TRD patients treated with 20

once-daily sessions of 10 Hz dmPFC-rTMS (Salomons et al, 2014). In this early study, we found

that high baseline resting-state functional connectivity between the dmPFC and the VMPFC and

sgACC was associated with better treatment response to dmPFC-rTMS. Likewise, lower baseline

rsFC between the dmPFC and right putamen, right thalamus and right hippocampus/amygdala

was associated with better treatment response. While Study III did not replicate our finding that

high dmPFC-sgACC connectivity was predictive of response in TRD (Salomons et al., 2014), it

is encouraging that frontostriatal connectivity was predictive of response in four independent

samples of patients treated with excitatory dmPFC-rTMS.

The connectivity patterns predictive of dmPFC-rTMS response were also qualitatively

similar to those reported in another recent publication identifying baseline rsFC predictors of

response in TRD to 10 Hz and iTBS dmPFC-rTMS (Drysdale et al, 2017). In this study, the

authors reported that TRD patients can be reliably classified into one of four biological subtypes,

on the basis of unique profiles of clinical symptoms alongside specific patterns of abnormal

resting-state functional connectivity. These subtypes were stable over time, and were associated

with depressive symptoms in another DSM-5 diagnosis (in other words, showed transdiagnostic

validity even in the presence of comorbidities). More importantly, these biological subtypes were

able to predict treatment outcomes among patients who received excitatory dmPFC-rTMS. The

most discriminant regions of abnormal resting-state functional connectivity in predicting

treatment response in the study of Drysdale and colleagues aligned well with the regions reported

in the present studies as predictive of response to dmPFC-rTMS, including rsFC of dACC,

globus pallidus, NAc, OFC and VLPFC (Drysdale et al, 2017).

There is also some preliminary evidence supporting the idea that pre-treatment dACC

frontostriatal connectivity may be predictive of response to other, non-rTMS interventions in

MDD, OCD, and AN. For example, in MDD, frontostriatal connectivity from the dACC during

reward processing of a monetary incentive delay task was predictive of antidepressant response

in two studies (Admon et al, 2015; Walsh et al, 2017). In OCD patients undergoing VS-DBS

implantation, pre-implantation resting-state hyperconnectivity was present in individuals who

responded (Figee et al, 2013). In the setting of ED, caudate hyperactivity during prediction errors

predicted poor weight gain in adolescent AN (DeGuzman et al, 2017).

257

Unlike with dmPFC-rTMS, response to DLPFC-rTMS and other interventions appears

more consistently associated with baseline functional activity of the sgACC. A series of

published reports by Fox and colleagues found that anti-correlated baseline rsFC between the left

DLPFC and sgACC predicted favourable response to DLPFC-rTM in TRDS (Fox et al, 2012a,

2013b, 2014). Furthermore, this resting-state predictor of DLPFC-rTMS outcome has been

independently replicated by other groups (Baeken et al, 2014; Philip et al, 2018). For non-rTMS

interventions, two studies have also showed that baseline sgACC hyperactivity during emotional

processing is correlated with response to CBT in MDD (Ritchey et al, 2011; Siegle et al, 2006),

and that baseline sgACC fALFF appears to correlate positively with ECT response (Argyelan et

al, 2016). This difference in patterns of dmPFC and sgACC connectivity predictive of response

is suggestive of the possibility that there may exist TRD subtypes that preferentially respond to

either conventional interventions, ECT, dmPFC- or DLPFC-rTMS.

Why might baseline frontostriatal rsFC differentiate treatment responders from non-

responders to dmPFC-rTMS? Studies spanning last 40 years have reported that frontostriatal

circuitry from regions of the DMN, SN, CEN, and VMN have a key role in motivational,

affective, and cognitive functions that are related to goal-directed action (for example,

(Alexander et al, 1986)). While there is substantial integration between CSTC loops, the caudate

also has a general rostrocaudal hierarchy to its functional architecture, such that rostral striatal

neurons that have DLPFC, dmPFC and VMPFC afferents are more active during reward-based

learning and actions that require conscious attention, while caudal striatal neurons that have

sensorimotor afferents are more active during actions that do not require conscious attention

(Kim and Hikosaka, 2015).

CSTC circuits serving the SN specifically are thought to play an etiological role in

cognitive control deficits consistent across a wide array of psychiatric disorders (as discussed in

(Marsh et al, 2009a)). Marsh and colleagues (2009a) reported that deficits in self-regulatory

control common across many psychopathologies, including OCD and ED, are related to both

dysregulated parallel and integrated frontostriatal circuits stemming from medial and lateral

prefrontal cortex and orbitofrontal cortex. Given the possible existence of parallel CSTC circuits

that give rise to different clinical phenotypes, it could therefore be the case that in dmPFC-rTMS

non-responders, the intervention was targeting the wrong frontostriatal circuit for the patient’s

underlying pathology.

258

dmPFC-rTMS Response Correlates with CSTC Connectivity Change In Studies I and II, CSTC connectivity change was associated with OCD and ED

symptom improvement following 20-30 sessions of open-label dmPFC-rTMS. However, the

while the circuit was similar, the direction of therapeutic effect was different: OCD treatment

responders saw a significant decrease in dACC frontostriatal connectivity associated with

improvement, whereas eating disorder treatment responders saw a significant increase in

connectivity with improvement. These observations support the premise (noted earlier above)

that psychopathology may be associated with either hypo- or hyper-functioning of SN regions.

Directionality aside, the brain regions in which change was associated with improvement

in Studies I and II align well with the findings in our first open-label case series of 25 TRD

patients treated with 20 once-daily sessions of 10 Hz dmPFC-rTMS (Salomons et al, 2014). In

this earlier study, pre- to post-treatment increases in functional connectivity between the dACC

and the bilateral medial dorsal thalamus accompanied dmPFC-rTMS response. Similarly,

decreases in resting-state functional connectivity between the sgACC, the ventral striatum and

middle cingulate cortex also accompanied treatment response to dmPFC-rTMS. Taken together,

it appears that changes in SN CSTC rsFC following dmPFC-rTMS are associated with symptom

improvements across a wide variety of disorders, signifying a possible transdiagnostic

mechanism of response to the same neurostimulatory intervention.

Previously studies that report neurobiological changes associated with improvements in

binge eating and purging are difficult to find in the literature. It is known, however, that bulimia-

related deficits in impulse control are related to SN corticostriatal circuitry. As previously

discussed, SN frontostriatal hypoactivity is associated with food impulsivity (Bohon and Stice,

2011; Frank et al, 2006) and deficits in cognitive control tasks such the Simon Spatial

Incompatibility task (Celone et al, 2011; Marsh et al, 2009b, 2011), and the go/no-go task

(Skunde et al, 2016). Increases in SN connectivity and its associated frontostriatal connections

following dmPFC-rTMS are perhaps a neural substrate for improvements in cognitive control

and impulsivity surrounding binge episodes. As negative affect and negative urgency are

significant predictors of binge eating (Fischer et al, 2003; Racine and Martin, 2017), it could also

259

be the case that dmPFC-rTMS caused improvements in emotion regulation and cognitive control

over negative urgency, which in turn allowed reductions in binge eating and purging.

For patients with OCD, dmPFC-rTMS responders had baseline dACC frontostriatal

hyperconnectivity that normalized following treatment. This finding is in line with previous

studies reporting that SN corticostriatal structural and functional connectivity changes

accompany symptomatic OCD improvement with both conventional and novel therapeutics.

Conventional OCD pharmacotherapies have been shown to normalize dACC corticostriatal

hyperactivity (Anticevic et al, 2014; Beucke et al, 2013; Rauch et al, 2006), and have been

shown to alter striatal and thalamic gray matter volume (Atmaca et al, 2016; Hoexter et al,

2012). Given that frontostriatal hyperconnectivity has been associated with symptom

provocation and deficits in emotion regulation in OCD, is it likely that normalized frontostriatal

hyperconnectivity is related to improvements in these cognitive abnormalities and associated

OCD symptoms.

Notably, the rsFC changes observed in OCD dmPFC-rTMS responders overlap closely

with the results reported by one study using rs-fMRI in patients treated with VS-DBS. In an

observation nearly identical to that reported for neural correlates of improvement in Study II,

Figee and colleagues (2013) reported that dACC resting-state hyperconnectivity normalized in

OCD VS-DBS responders (Figee et al, 2013). Given the close anatomical overlap of our findings

and those of Figee and colleagues, it is likely that the mechanisms of symptom improvement

may be similar for both VS-DBS and dmPFC-rTMS for OCD. This finding also agrees well with

previously reported rsFC cortical networks of DBS targets for OCD (Fox et al, 2014). In light of

these observations, it may be reasonable to conclude that dmPFC-rTMS achieves therapeutic

effects via a similar, or perhaps the same, mechanism as that of more invasive VS-DBS.

It is also worth noting that the direction of frontostriatal rs-fMRI predictors and correlates

of OCD response to excitatory dmPFC-rTMS are opposite to what we have previously observed

for TRD (Salomons et al, 2014) and eating disorders (Study I) (Dunlop et al, 2015). The

opposite-direction correlate of therapeutic effect could be related to inter-individual variability of

functional abnormalities in the SN across many disorders, or related to the variability reported in

the general neurobiological mechanisms of rTMS as reviewed in the Introduction.

260

Regarding the first possibility, the direction of rTMS-induced effect on rsFC might be

state-dependent. In other words, the direction of effect for 10 Hz rTMS might depend on whether

baseline rsFC of this circuit is abnormally hyper- or hypoactive at rest. The observations in Study

I of different directions of effect for 10 Hz stimulation in responders and non-responders,

depending on whether pre-treatment rsFC was high or low, supports this hypothesis. Similarly,

as previously discussed, while transdiagnostic neuroimaging studies report consistent volumetric

reductions in the dACC and AI, there is considerable heterogeneity of functional abnormalities in

the SN. Furthermore, studies report hyper- or hypoactivity in the SN depending on the

neuroimaging paradigm employed (McTeague et al, 2016, 2017).

Regarding the second possibility, variability in the direction of rTMS effect could be a

function of other genetic or neurobiological factors that influence synaptic plasticity. The actual

direction of effect for nominally ‘excitatory’ and ‘inhibitory’ rTMS protocols is extremely

heterogeneous across individuals, as a substantial proportion of healthy controls exhibit increases

in MEP amplitude on nominally inhibitory protocols such as 1 Hz rTMS, or decreases in MEP

amplitude on nominally excitatory protocols at 10, 15, and 20 Hz (Maeda et al, 2000). Likewise,

on functional MRI, there is substantial inter-individual variability in the direction of connectivity

change in response to excitatory and inhibitory rTMS (Eldaief et al, 2011). To date, however, the

neurobiological underpinnings of this heterogeneity in rTMS’ neuroplastic effects is not well-

understood. Consequently, it is possible that the direction of therapeutic effect of dmPFC-rTMS

may depend on either state-dependent SN abnormal function, or other neurobiological factors

that contribute to inter-individual variability in rTMS-induced synaptic plasticity.

Treatment Non-Specific Clinical Response in TRD is Related to Emotion Regulation and Impulsivity In Study III, changes in neural activity that were associated with clinical improvement

across all three study arms included increased rsFC between the VMPFC and the dACC. These

brain areas are divergent from those identified in previous similar studies of the rsFC correlates

of response for other interventions such as DLPFC-rTMS, or antidepressant medications. Both

conventional antidepressant medications and high frequency left DLPFC-rTMS have been

reported to increase rsFC in nodes of the CEN and SN, and decrease rsFC from the sgACC (as

261

reviewed in Section 1.11.2 and 1.11.3). For example, a relatively recent meta-analysis of PET

and fMRI studies reported that response to pharmacotherapies was associated with increases in

dmPFC, DLPFC and VLPFC activity, and decreases in the activity of all regions of the DMN

(Delaveau et al, 2011). Correlates of improvement for DBS of the sgACC were also very similar

to those for conventional interventions: sgACC-DBS responders show metabolic increases in the

DLPFC and dACC, and metabolic decreases in the sgACC and OFC (Lozano et al, 2008;

Mayberg et al, 2005). Reductions in sgACC-medial prefrontal hyperconnectivity have also been

associated with nonspecific response to either active or placebo DLPFC-rTMS as a ‘final

common pathway’ of improvement in TRD (Taylor et al, 2018).

Consistent with previous findings, increases in dACC rsFC accompanied clinical

improvement across all three of the interventions of Study III. This observation agrees with some

previously reported findings. For example, on FDG-PET, MDD patients treated with either an

SSRI or psychotherapy showed increases in ACC metabolism correlated with symptom

improvement (Brody et al, 2001). During task-fMRI, normalization of SN hypoactivity during

induced negative mood (Fitzgerald et al, 2008) and during incentive reward cues (Stoy et al,

2012) has been reported following successful pharmacotherapy.

At the same time, in contrast to these previous findings, the results of Study III indicate

that treatment non-specific clinical response was associated with significant increases of the rsFC

from the VMPFC to the dorsal ACC, a node of the SN. For DLPFC-rTMS, non-specific

correlates may be different. One previous study of mechanisms of treatment response reports

similar increases in functional connectivity between dorsal and ventral cortical structural with

response to rTMS. Baeken et al. (2014) reported that responders had increased rsFC between the

sgACC and perigenual ACC in responders to accelerated left DLPFC-rTMS (Baeken et al,

2014).

Study III also reported that intervention-nonspecific response was correlated with

improvements in self-reported emotion regulation. TRD responders showed significant

reductions on the impulsivity and goal-setting subscales, and the total score of the DERS.

Consequently, a reasonable interpretation is that a common pathway of improvement in

depression severity is related to an improved impulsive, dysregulated, persistent negative

thoughts and emotions, apparent across all three study arms. This behavioural change would

262

presumably occur via normalizing top-down dorsal prefrontal connectivity to ventral affective

regions, with resultant enhancements of cognitive control over negative thoughts and emotions.

Menon’s ‘Triple Network Theory’ may be helpful in interpreting the meaning of Study

III’s observed intervention-nonspecific increases in SN function that correlated with

improvements in depression symptoms and emotion regulation. Menon hypothesizes that the SN

has a possible role as a functional ‘switch’ between the CEN during externally-driven cognition,

and DMN during internally-driven cognition (as discussed in (Menon, 2011; Menon and Uddin,

2010)). Consequently, SN dysfunction may permit aberrant activity in these other networks,

leading to difficulties integrating salient external or physiological events and accessing networks

for attention and working memory or self-referential/autobiographical thought. Given that the

VMPFC ROI used in Study III overlaps with the mPFC node of the DMN, as well as the

VMPFC node of the VMN, it is therefore possible that increases in dACC-VMPFC functional

connectivity following treatment reflect a restored functional regulatory ‘switch’ in IBN network

connectivity between the SN and the DMN, accompanied clinically by an enhanced capacity for

emotional self-regulation.

Limitations and Challenges Several important limitations and challenges for the present Studies I-III bear discussion.

Of greatest note, the results of Study III did not find either 20 Hz or 1 Hz active dmPFC-rTMS to

be superior to the sham technique in reducing clinical symptoms. The lack of significant

difference was driven not by a lack of effect in the active arms but by an equal degree of

improvement in patients who received the ‘sham’ form of stimulation using scalp electrodes.

The most straightforward conclusion from this observation would be that there is no

specific therapeutic effect of active dmPFC-rTMS over the non-specific effects of attending

treatment sessions. However, there are some caveats that may render such a conclusion may be

premature. First, the response and remission rates of Study III’s active arms, as with previous

open-label dmPFC-rTMS, are comparable to that of DLPFC-rTMS in studies that did

demonstrate efficacy over sham(Bakker et al, 2015), and the degree of clinical improvement in

the sham arm of Study III is superior to that previously found in meta-analyses that report sham

263

response rates for DLPFC-rTMS as low as 10% in TRD cases (Berlim et al, 2013). Furthermore,

healthy control studies of excitatory medial PFC have reported significant behavioural and

neurobiological changes in response to active rTMS versus placebo, indicative of a potential

mechanism of dmPFC-rTMS in non-depressed individuals (Cho et al, 2015). The higher than

expected response to sham stimulation in Study III is therefore somewhat aberrant when

compared to previous literature, and further study may therefore be warranted to test the clinical

efficacy of dmPFC-rTMS for TRD, ED, and OCD.

Sham treatment technique is a major challenge in rTMS trial design. Many of the earlier

major randomized controlled trials assessing rTMS efficacy did not effectively blind the rTMS

technician during treatment, as separate coils or different placement techniques were used for

active and sham treatment. Older models of sham rTMS also did not effectively mimic the

somatosensory or nociceptive aspects of active rTMS during treatment by using a shielded rTMS

coil or placing an active rTMS coil at an angle that would insufficiently apply stimulation. While

more recent rTMS trials sometimes attempt to mimic the somatosensory and nocieptive

sensations of rTMS, the electrical stimuli used to do so may run the risk of being in themselves a

neurophysiological stimulus with therapeutic properties. In one notable sham-controlled study of

right versus left DLPFC-rTMS, ‘sham’ rTMS with simultaneous electrical stimulation via scalp

electrodes achieved similar effects to active rTMS at both sites, and the therapeutic efficacy

depended on the site of stimulation (right > left DLPFC efficacy) rather than whether the

stimulation was delivered by scalp electrodes or the ‘active’ rTMS coil (Triggs et al, 2010).

This observation may be pertinent to the sham techniques employed in the present Study

III. This trial of dmPFC-rTMS for TRD is the first randomized controlled trial to employ a

custom active/placebo double-sided double-cone coil that attempts to mimic the sensation of

rTMS, while at the same time attempting to blind the rTMS technician to the treatment being

given. As we previous described, however, it could also be the case that by mimicking the

sensory and nociceptive effects of active dmPFC-rTMS, the sham electrodes inadvertently

delivered a therapeutic form of electromagnetic stimulation to the brain, either transcranially

(since the electrodes were positioned between the DLPFC and frontal pole) or else via V1

trigeminal nerve stimulation. The observation of some commonalities in the change in pre- to

post-treatment rsFC for the active and sham arms supports this premise. Further research would

be required to elucidate the clinical effects of trigeminal nerve stimulation, and perhaps to

264

evaluate the clinical efficacy of excitatory dmPFC-rTMS using some other placebo method. For

example, other rTMS studies have used active rTMS over the vertex as an alternative control

intervention (Razza et al, 2018).

As a second issue affecting all 3 studies, relatively small sample sizes limit the

generalizability of clinical and neuroimaging findings, and could also contribute to the null

clinical finding of the randomized controlled trial in Study III. Despite this issue, the sample

sizes of each study are comparable to other neuromodulation studies involving rs-fMRI (Downar

et al, 2014; Figee et al, 2013; Salomons et al, 2014; Taylor et al, 2018). Furthermore, the sample

sizes of both open-label case series (Studies I and II) are comparable to other open-label rTMS

treatment studies for investigational indications such as OCD (Mantovani et al, 2006, 2010) and

eating disorders (Walpoth et al, 2008), and are also comparable to other non-interventional

neuroimaging studies in these populations (Admon et al, 2012; Cowdrey et al, 2011). Finally, the

sample size of our randomized controlled trial is substantially larger than the only other trial of

dmPFC-rTMS for TRD, which randomized N=45 patients (Kreuzer et al, 2015). Irrespective of

previous studies, the sample sizes of the present study are still somewhat underpowered to

adequately capture the clinical and neurobiological heterogeneity of these psychiatric

populations, and to identify generalizable biomarkers of treatment response. It is encouraging,

however, to see transdiagnostic agreement across the different disorders of Studies I-III in the

findings that dmPFC-rTMS achieved significant therapeutic effects, and that baseline

frontostriatal connectivity may differentiate treatment responders from non-responders.

Another limitation arises from the inherent clinical and neurobiological heterogeneity of

MDD, OCD and ED, which might obscure the efficacy of dmPFC-rTMS if only a subpopulation

of each study group had pathology congruent with the underlying mechanism of the intervention.

As noted earlier, the results suggest that dmPFC-rTMS may modulate the integrity of the SN and

its CSTC components, regardless of disorder. If only a subgroup of patients in each disorder

group actually have abnormalities of SN function, then the observed remission rate may be

proportional to the size of that subpopulation rather than the potency of the intervention itself.

The problem then shifts from enhancing the potency of the treatment to identifying the

subpopulation for which the treastment is appropriate.

265

If this premise is correct, then neurobiologically-based subtypes of psychiatric disorders,

irrespective of DSM-5 diagnosis, may be necessary to identify optimal candidates for dmPFC-

rTMS. To that end, as noted earlier, an important recent study developed a novel method to

define MDD subtypes by clustering subjects by distinct patterns of abnormal rsFC in

frontostriatal and limbic brain networks (Drysdale et al, 2017). We showed in a large multisite

sample that MDD patients could be subdivided into four rsFC-based subtypes that predicted

treatment responsiveness in individuals receiving dmPFC-rTMS. Furthermore, patients with

anxiety with comorbid depressive symptoms, but no formal MDD diagnosis, were also

successfully categorized into MDD subtypes based on rsFC. These findings suggest that these

endophenotypes are both useful for treatment prediction and potentially applicable

transdiagnostically; they may also help explain why there is considerable overlap in the set of

effective treatment interventions used across MDD, OCD, and ED. In the longer term,

identifying subtypes of psychiatric disorders may prove critical to optimizing rTMS clinical

efficacy.

A fourth challenge to the study and therapeutic use of dmPFC-rTMS is the inherent

heterogeneity in the effects of rTMS itself. At present, we have limited information about how

how manipulating individual rTMS parameters (such as stimulus frequency, pattern, or target

site) influences clinical response rates, or how these parameters interact with neurobiological

sources of variation among patients. The results from the present study indicate that the observed

neuroplastic changes following excitatory dmPFC-rTMS are not consistent across protocols,

disorders, or individuals: when given the same 10 Hz dmPFC-rTMS intervention, OCD

responders saw significant decreases in frontostriatal connectivity, while eating disorder

responders saw significant increases in the same frontostriatal connectivity.

Such heterogeneity is consistent with previous literature. For example, there is substantial

inter-individual variability in the magnitude and even direction of the neuroplastic effects of

rTMS at 1, 10, 15, and 20 Hz, as measured on motor evoked potentials (Maeda et al, 2000). On

functional MRI, there is significant inter-individual variability in both the magnitude and

direction of connectivity change in response to ‘excitatory’ 20 Hz and ‘inhibitory’ 1 Hz rTMS

(Eldaief et al, 2011). Furthermore, the two rTMS protocols engaged slightly different

neuroanatomical networks of connections to the target site: 20 Hz rTMS primarily engaged

corticocortical circuits, while 1 Hz rTMS altered corticolimbic circuits (Eldaief et al, 2011). It is

266

also worth mentioning that most of the studies that report on the neurobiological effects of rTMS

are restricted to motor cortex stimulation and plasticity in MEPs. Further research is needed to

better characterize the effects of rTMS over non-motor cortical regions, and to determine how

certain rTMS parameters - including stimulation intensity, intertrain interval, or the interval

between treatments - influence this plasticity.

A fifth limitation of the present set of studies is that the neuroimaging-based biomarkers

of treatment response are based on group-level differences between rTMS responders and non-

responders. In other words, these neuroimaging results assess differences in functional

connectivity between two groups (responders and non-responders) on average, instead of being

able to characterize more fully the differential distribution, overlap, or non-overlap of these

biomarkers across the full set of responder and non-responder individuals.

Significant average differences cannot always be translated into individual predictive

ability for a single metric, if the distributions have substantial overlap despite their difference in

means. Thus, the group-level neural predictors of the present study may or may not prove

suitable for individual-level prediction of treatment outcome in the clinical setting. For the

present set of studies, group-level differences were favoured over machine-learning classifiers

(which may achieve better individual-level classification by using large sets of metrics in a

multivariate rather than univariate predictive model), given that large training sets are needed to

develop robust machine-learning classifiers. The limited sample sizes in the present study would

negatively impact the accuracy, interpretability, and generalizability of automated prediction

models (for a review, see Patel et al, 2016). Furthermore, there are a number of methodological

challenges to training machine-learning classifiers on fMRI data that could influence a predictive

model’s accuracy and generalizability. For example, fMRI scans are composed of tens of

thousands of voxels, and the use of different, but equally valid, methods to parse these voxels

and select features to train a predictive model can lead to major differences in the model’s

predictive accuracy (for a systematic comparison of parcellation methods and their influence on

prediction model accuracy, see Arslan et al, 2018). Another example is cross-validation

methods: different cross-validation techniques can also substantially influence the accuracy and

generalizability of a prediction model (for a review, see Patel et al, 2016). As a result of these

limitations, substantial future work would be necessary to progress from the group-level

predictors identified in the present study to the eventual validation of robust individual-level

267

biomarkers of rTMS response, in large, multisite datasets, using machine-learning classifiers that

are uniquely designed to parse rs-fMRI data and that generalize beyond the initial training set.

Finally, regarding translational limitations, there are logistical constraints on the

scalability and accessibility of rTMS as a treatment for psychiatric disorders, and fMRI as a tool

to subtype patients or aid in treatment selection. First, rTMS treatments impose significant

logistical burdens on patients. As rTMS typically involves between 20-30 once daily sessions,

patients are required to come to the clinic every day, typically Monday to Friday, for between

four and six weeks. This is a considerable inconvenience for patients, especially those who are

able to work or must commute a long distance for daily treatment. It is also problematic for the

treatment capacity of rTMS clinics; long treatment courses mean that fewer patients can be

treated. Second, to date, rTMS clinics in Canada are largely found only in a handful of research

or clinical centres of excellence. This means that the current availability of rTMS in Canada is

limited to urban centres with teaching hospitals, requiring patients from underserved rural

communities to commute daily to the city or incur significant financial costs in short-term

housing downtown. As such, there is a need to decrease the overall logistical burden (i.e.,

number of required visits) on patients, perhaps via accelerated rTMS protocols using multiple

sessions per day (Holtzheimer et al, 2010). Even if rTMS clinics were to become widespread,

given the logistical burdens patients face in attending treatment sessions, treatment selection

techniques still need to be optimized so as to reliably identify patients for whom rTMS is likely

to be futile. Finally, the use of fMRI to identify biological markers of treatment response is not

readily scalable: MRI is not only expensive, but the number of MRI scanners per capita in

Canada significantly lags behind other developed countries like Japan and the United States

(Stein, 2005). It is therefore likely that even if a generalizable neuroimaging biomarker was

discovered, MRI wait times in Canada and abroad would drastically increase if it were

implemented in widespread practice. Again, 3 Tesla MRIs capable of scanning the brain at high

resolution are often localized to centres of excellence, which would require many patients to

commute from rural areas. Therefore, other techniques, such as EEG, or even a validated written

questionnaire, might be more cost effective, scalable, and appropriate tools to assess

neuropsychiatric endophenotypes, or aid in treatment selection in the future.

268

Conclusion In sum, this thesis provides novel evidence for the existence of transdiagnostic rsFC

predictors of, and mechanisms of response to, excitatory dmPFC-rTMS as a treatment for TRD,

ED, and OCD. Baseline SN frontostriatal rsFC significantly differed between treatment

responders and non-responders, such that TRD and ED rTMS responders had lower baseline

dACC frontostriatal rsFC and OCD rTMS responders had higher baseline dACC frontostriatal

rsFC relative to rTMS non-responders. Regarding mechanisms of response to excitatory dmPFC-

rTMS, treatment response in ED was significantly correlated with increased dACC frontostriatal

rsFC, while treatment response in OCD was significantly correlated with decreased dACC

frontostriatal rsFC. Furthermore, TRD patients who responded to active or placebo dmPFC-

rTMS showed a normalization of VMN and SN hypoconectivity, as well as improvements in

emotion regulation deficits following treatment. From a basic science perspective, these findings

shed light on a common underlying neurobiological etiology across multiple categories of mental

illness, and the neurobiological pathways through which individuals may recover from these

illnesses. From a clinical and translational perspective, the present study reveals new avenues for

the personalization of patient care strategies in psychiatric illness, and for the development of

novel treatment strategies that may one day prove effective in the most treatment-resistant

populations of patients.

269

Future Directions The findings from Study I-III point to several future clinical and neuroimaging questions

whose answers may further advance the field. These questions may be best addressed through a

series of randomized controlled trials, designed to clarify the clinical efficacy of dmPFC-rTMS

for treatment-refractory MDD, AN-BP, BN, and OCD. Such trials will also be well-positioned

to address the following questions for future study:

The first set of questions is related to the theme of demonstrating clinical efficacy under

randomized, placebo-controlled conditions, while employing an appropriate technique for sham

stimulation that does not directly modulate brain activity:

1) Does active dmPFC-rTMS achieve superior clinical outcomes to sham stimulation,

using a different sham technique? The results of Study III suggest that there is no

significant clinical difference between active and sham stimulation. However, this lack of

difference was driven by a substantially higher-than-expected sham rTMS response rate

rather than a lower-than-expected active rTMS response rate. As such, there are a number

of potential confounds in the sham stimulation design that warrant future study. For one, the

electrode montage over the frontal pole and DLPFC, used to mimic the somatosensory and

nociceptive qualities of dmPFC-rTMS, may have inadvertently induced an antidepressant

effect. Other recent sham-controlled studies of dmPFC-rTMS have used vertex stimulation

as an effective placebo that elicits no true antidepressant effect (Cho et al, 2015; Razza et al,

2018), and so a future study of dmPFC-rTMS should employ an alternative control

intervention involving either a shielded ‘placebo’ stimulation coil, or stimulation of a control

region not linked to the SN, to assess the efficacy of this dmPFC-rTMS for TRD.

2) What is the clinical efficacy of excitatory dmPFC-rTMS for treatment-refractory AN-

BP/BN, and OCD under placebo-controlled conditions? Given that the clinical results of

Studies I and II were under open-label conditions, a double-blind, sham-controlled trial is

necessary to establish the clinical efficacy of dmPFC-rTMS for AN-BP or BN, and OCD.

Given the potential confounding issues with the sham stimulation technique we used for our

MDD randomized controlled trial, it would be necessary to design the study to avoid any

inadvertent therapeutic effects from electromagnetic stimulation over the scalp aimed at

mimicking the somatosensation of rTMS. Given the recent advances in rTMS sham coil

270

design and increases in placebo response rates (Razza et al, 2018), it will be also important

for future studies to recruit a large sample size in order to achieve sufficient statistical power

to allow a demonstration of superiority at a clinically meaningful effect size. Such follow-up

studies may be conducted both for OCD and for ED patients – in the latter case, perhaps as a

useful adjunct intervention for patients who are completing an inpatient hospitalization or

enrolled in a day hospital program.

Another research theme of high importance for future investigation relates to the

optimization of rTMS parameters to maximize the clinical efficacy of this intervention.

3) Are there alternative rTMS targets that may be effective for patients who are non-

responsive to dmPFC- or DLPFC-rTMS? Observations from Study III suggest that rsFC

to the orbitofrontal cortex at baseline may be an important distinguishing feature of non-

responders versus responders to dmPFC-rTMS. If so, one potential alternative to dmPFC

stimulation is instead to stimulate the orbitofrontal cortex. OFC functional connectivity is

abnormal in MDD patients in general, compared to healthy controls (Fettes et al, 2018). The

OFC also has CSTC connections responsible for reward, non-reward, and affective

regulation; disturbances in this connectivity could lead to behavioural disturbances, and,

likewise, normalizing dysfunctional OFC CSTC connectivity could produce a therapeutic

effect in patients with abnormal network activity involving this region (for a review, see

Fettes et al, 2017b). Our group has recently published a case report where a patient who had

previously failed adequate courses of both dmPFC- and DLPFC-rTMS responded well to a

course of inhibitory OFC-rTMS (Fettes et al, 2017a). A subsequent case series on clinical

outcomes for 1 Hz right OFC-rTMS in N=42 patients who had already failed to respond to

dmPFC-rTMS found that 35.7% and 23.8% of TRD patients responded and remitted,

respectively (Feffer et al, 2018). Therefore, a randomized controlled trial of OFC-rTMS to

determine the efficacy of this intervention under placebo-controlled conditions is warranted.

4) Is there an alternative stimulatory protocol with more consistent neuroplastic effects

across individuals? ‘Conventional’ rTMS protocols are remarkably variable both in their

neurobiological effects on plasticity and their clinical efficacy. This heterogeneity is

apparent for both conventional protocols like 1 Hz and 10 Hz rTMS (Eldaief et al, 2011;

Maeda et al, 2000) and for newer protocols like iTBS and cTBS (Cárdenas-Morales et al,

2014; Hamada et al, 2013; Nettekoven et al, 2015). As such, even if the problem of reliable

271

individualized target selection could be reliably solved, the problem of failure to engage

plasticity as intended would remain, unless more consistent protocols can be developed.

QPS is one novel protocol that is reportedly more consistent in its neuroplastic effects than

previous stimulatory protocols; one recent study showed that the QPS-induced MEP effects

are consistent across 80% of subjects, for both its inhibitory and its excitatory forms

(Nakamura et al, 2016). While QPS appears to be safe and well-tolerated, clinical studies

involving QPS have yet to be performed in any psychiatric disorder (Nakatani-Enomoto et

al, 2011). Therefore, preliminary clinical studies of QPS versus conventional or theta-burst

protocols for the treatment of psychiatric disorders might be a worthwhile endeavour, to

maximize the intervention’s clinical efficacy and intended neurobiological effects.

5) Do the number of daily sessions and the interval between treatments impact treatment

efficacy to accelerated dmPFC-rTMS? While accelerated protocols of multiple daily

sessions of rTMS have seen increasing interest in the literature, the number of daily

treatments in a single course vary considerably between studies. For example, the first

accelerated rTMS study by Holtzheimer and colleagues administered 15 sessions of left

DLPFC-rTMS over 2 days (Holtzheimer et al, 2010), while others have successfully used

twice-daily (McGirr et al, 2015) or five times daily sessions (Baeken et al, 2013; Duprat et

al, 2016). Yet the optimal number and timing of rTMS sessions to balance treatment

efficacy and patient burden is still unclear. Are there any genetic or neurobiological factors

that could influence the optimal number of daily sessions for a particular patient? In

particular, does the interval between treatment sessions significantly influence clinical

efficacy or the direction of rTMS-induced neuronal plasticity? As the issues of optimal daily

session number and inter-session interval were not systematically explored in Study III, it

would be worthwhile to complete two randomized controlled trials: first, an rTMS study

evaluating the antidepressant effects and neurobiological predictors of once daily, twice

daily, and 5 times daily rTMS; and second, an rTMS study to establish the optimal interval

between multiple daily rTMS sessions. This work could help to reduce the logistical burden

on patients by reducing the number of clinic visits required to complete a course of rTMS

treatment.

6) Can rTMS be used in combination with other interventions to maximize its clinical

efficacy in psychiatric disorders? Most previous randomized controlled trials of rTMS for

psychiatric disorders like TRD and OCD have applied stimulation during resting conditions.

272

In other words, for most trials to date, during rTMS sessions the patient is in a seated

position, doing nothing in particular, without being instructed to engage in any particular

type of cognition or to keep any particular type of thought in mind. It could be the case that

the amplitude of rTMS effect over a cortical target, and therefore the clinical efficacy,

differs when the stimulation is delivered at rest compared to when that region is engaged in a

particular task. For example, Ruf and colleagues applied DLPFC-tDCS during working

memory training in a sample of healthy controls. They reported that individuals who

received task-relevant tDCS saw a steeper learning curve relative to those who received

placebo tDCS or task-irrelevant tDCS (Ruf et al, 2017). To the same point, the well-

established discrepancy between the ‘active’ and the ‘resting’ motor threshold during motor

cortex stimulation is a simple but clear illustration of the state-dependence of rTMS’ effects

on the brain. It would therefore be worthwhile to apply dmPFC-rTMS in the context of an

emotion regulation or impulsivity paradigm known to engage the SN, or during/before CBT

or mindfulness meditation, to see whether engaging the dmPFC with a task improves

clinical efficacy of this treatment.

Finally, there may be ways to improve the techniques with which we identify treatment

responders and distinguish them from non-responders prior to treatment.

7) Do transdiagnostic, biologically-derived subtypes of psychiatric disorders exist? Do

such biological subtypes differ in terms response to dmPFC-rTMS? The interpretation

of many previously-identified biomarkers of treatment response is limited, because many of

these studies have a modest sample size and offer only group-average predictive associations

rather than individual-patient predictive value. Many candidate biomarkers fail to replicate

in subsequent studies, and are therefore not generalizable to the population. Furthermore, as

this thesis has highlighted, psychiatric disorders are remarkably heterogeneous in terms of

their clinical presentation. As a result, many have pushed for new diagnostic approaches,

such as the Research Domain Criteria, to identify more biologically grounded ways to

categorize neurocognitive, affective, and behavioural abnormalities. Large, multi-site studies

involving hundreds or thousands of scans from individuals presenting with a variety of types

of abnormal cognition or behaviour, regardless of DSM-5 diagnosis, may be needed to

address this clinical heterogeneity, and to overcome the many challenges involved in

identifying generalizable biomarkers. In a well-cited example, Drysdale and colleagues

273

recently published such a study in a large dataset of individuals diagnosed with MDD, and

developed a novel method to define MDD subtypes by clustering subjects by distinct

patterns of abnormal rsFC in frontostriatal and limbic brain networks (Drysdale et al, 2017).

The four rsFC-based MDD subtypes predicted treatment responsiveness in individuals

receiving dmPFC-rTMS at superior accuracy rates compared to clinical or rsFC features

alone. Datasets that collect neuroimaging from patients afflicted with diverse psychiatric

conditions, such as the United Kingdom Biobank (Alfaro-Almagro et al, 2018), or the

Enhancing NeuroImaging Genetics through Meta-Analysis consortium (ENIGMA)

(Thompson et al, 2014b), will be necessary to classify transdiagnostic biological subtypes;

such markers will hopefully prove useful and generalizable for intervention selection and

prognosis in real-world clinical settings.

8) Are there more scalable biomarkers to predict dmPFC-rTMS treatment response or

monitor treatment mechanisms? fMRI has several major limitations for large-scale,

practical use in real-world clinical settings: it is costly, there are few scanners in Canada,

and it is technically challenging to simultaneously stimulate and image the brain, which

rules out real-time monitoring of brain activity during treatment sessions. However, TMS

coupled with EEG, if reliable, may be a more practical alternative to fMRI-based

biomarkers. EEG devices are relatively less expensive and plentiful, and some devices are

TMS compatible. Numerous studies have pointed to neurophysiologically-based biomarkers

of rTMS antidepressant response, including parietotemporal alpha power (Micoulaud-

Franchi et al, 2012), and frontal electrode power in multiple frequency bands (Arns et al,

2012). Changes in neurophysiological markers have also been identified as correlates of

rTMS treatment response, including changes in the P200 during an auditory oddball task

(Choi et al, 2014), and changes in gamma and theta at rest (Noda et al, 2017). Pellicciari and

colleagues found TMS-induced PFC asymmetry was restored following successful DLPFC-

iTBS in one case report (Pellicciari et al, 2017). Combining such techniques with dmPFC-

rTMS and transdiagnostic biomarker identification would potentially be a fruitful and

scalable alternative to rs-fMRI, if biomarkers are to be developed for use in real-world

settings in patients with MDD, OCD, ED, and other treatment-resistant psychiatric

disorders.

274

References Aaronson ST, Carpenter LL, Conway CR, Reimherr FW, Lisanby SH, Schwartz TL, et al

(2013). Vagus nerve stimulation therapy randomized to different amounts of electrical

charge for treatment-resistant depression: acute and chronic effects. Brain Stimul 6: 631–40.

Abelson JL, Curtis GC, Sagher O, Albucher RC, Harrigan M, Taylor SF, et al (2005). Deep

brain stimulation for refractory obsessive-compulsive disorder. Biol Psychiatry 57: 510–6.

Abou-Elseoud A, Starck T, Remes J, Nikkinen J, Tervonen O, Kiviniemi V (2010). The effect of

model order selection in group PICA. Hum Brain Mapp 31: 1207–16.

Abou Elseoud A, Littow H, Remes J, Starck T, Nikkinen J, Nissilä J, et al (2011). Group-ICA

Model Order Highlights Patterns of Functional Brain Connectivity. Front Syst Neurosci 5:

37.

Abramowitz JS, Taylor S, McKay D (2009). Obsessive-compulsive disorder. Lancet (London,

England) 374: 491–9.

Achard S, Bullmore E (2007). Efficiency and cost of economical brain functional networks.

PLoS Comput Biol 3: e17.

Addis DR, Wong AT, Schacter DL (2007). Remembering the past and imagining the future:

common and distinct neural substrates during event construction and elaboration.

Neuropsychologia 45: 1363–77.

Adey WR, Meyer M (1952). An experimental study of hippocampal afferent pathways from

prefrontal and cingulate areas in the monkey. J Anat 86: 58–74.

Adhikari BM, Goshorn ES, Lamichhane B, Dhamala M (2013). Temporal-order judgment of

audiovisual events involves network activity between parietal and prefrontal cortices. Brain

Connect 3: 536–45.

Admon R, Bleich-Cohen M, Weizmant R, Poyurovsky M, Faragian S, Hendler T (2012).

Functional and structural neural indices of risk aversion in obsessive-compulsive disorder

(OCD). Psychiatry Res 203: 207–13.

Admon R, Nickerson LD, Dillon DG, Holmes AJ, Bogdan R, Kumar P, et al (2015). Dissociable

cortico-striatal connectivity abnormalities in major depression in response to monetary

275

gains and penalties. Psychol Med 45: 121–31.

Agras WS, Walsh T, Fairburn CG, Wilson GT, Kraemer HC (2000). A multicenter comparison

of cognitive-behavioral therapy and interpersonal psychotherapy for bulimia nervosa. Arch

Gen Psychiatry 57: 459–66.

Agüera Z, Riesco N, Jiménez-Murcia S, Islam MA, Granero R, Vicente E, et al (2013).

Cognitive behaviour therapy response and dropout rate across purging and nonpurging

bulimia nervosa and binge eating disorder: DSM-5 implications. BMC Psychiatry 13: 285.

Aguirre I, Carretero B, Ibarra O, Kuhalainen J, Martínez J, Ferrer A, et al (2011). Age predicts

low-frequency transcranial magnetic stimulation efficacy in major depression. J Affect

Disord 130: 466–9.

Ahmari SE, Spellman T, Douglass NL, Kheirbek MA, Simpson HB, Deisseroth K, et al (2013).

Repeated cortico-striatal stimulation generates persistent OCD-like behavior. Science 340:

1234–9.

Aigner M, Treasure J, Kaye W, Kasper S, WFSBP Task Force On Eating Disorders (2011).

World Federation of Societies of Biological Psychiatry (WFSBP) guidelines for the

pharmacological treatment of eating disorders. World J Biol Psychiatry 12: 400–43.

Aigner M, Zitterl W, Prayer D, Demal U, Bach M, Prayer L, et al (2005). Magnetic resonance

imaging in patients with obsessive-compulsive disorder with good versus poor insight.

Psychiatry Res 140: 173–9.

Akerblad A-C, Bengtsson F, Knorring L von, Ekselius L (2006). Response, remission and

relapse in relation to adherence in primary care treatment of depression: a 2-year outcome

study. Int Clin Psychopharmacol 21: 117–24.

Alexander GE, DeLong MR, Strick PL (1986). Parallel organization of functionally segregated

circuits linking basal ganglia and cortex. Annu Rev Neurosci 9: 357–381.

Alexopoulos GS, Hoptman MJ, Kanellopoulos D, Murphy CF, Lim KO, Gunning FM (2012).

Functional connectivity in the cognitive control network and the default mode network in

late-life depression. J Affect Disord 139: 56–65.

Alfaro-Almagro F, Jenkinson M, Bangerter NK, Andersson JLR, Griffanti L, Douaud G, et al

(2018). Image processing and Quality Control for the first 10,000 brain imaging datasets

276

from UK Biobank. Neuroimage 166: 400–424.

Allen GI, Tsukahara N (1974). Cerebrocerebellar communication systems. Physiol Rev 54: 957–

1006.

Alonso J, Petukhova M, Vilagut G, Chatterji S, Heeringa S, Üstün TB, et al (2011). Days out of

role due to common physical and mental conditions: results from the WHO World Mental

Health surveys. Mol Psychiatry 16: 1234–46.

Alonso P, Pujol J, Cardoner N, Benlloch L, Deus J, Menchón JM, et al (2001). Right prefrontal

repetitive transcranial magnetic stimulation in obsessive-compulsive disorder: a double-

blind, placebo-controlled study. Am J Psychiatry 158: 1143–5.

Amassian VE, Cracco RQ, Maccabee PJ, Cracco JB, Rudell A, Eberle L (1989). Suppression of

visual perception by magnetic coil stimulation of human occipital cortex.

Electroencephalogr Clin Neurophysiol 74: 458–62.

American Psychiatric Association (2013). Diagnostic and Statistical Manual of Mental

Disorders, 5th Edition (DSM-5). Diagnostic Stat Man Ment Disord 4th Ed TR

doi:10.1176/appi.books.9780890425596.744053.

Amianto F, D’Agata F, Lavagnino L, Caroppo P, Abbate-Daga G, Righi D, et al (2013). Intrinsic

connectivity networks within cerebellum and beyond in eating disorders. Cerebellum 12:

623–31.

Anderson JS, Druzgal TJ, Lopez-Larson M, Jeong E-K, Desai K, Yurgelun-Todd D (2011).

Network anticorrelations, global regression, and phase-shifted soft tissue correction. Hum

Brain Mapp 32: 919–34.

Anderson RJ, Hoy KE, Daskalakis ZJ, Fitzgerald PB (2016). Repetitive transcranial magnetic

stimulation for treatment resistant depression: Re-establishing connections. Clin

Neurophysiol 127: 3394–3405.

Andrade C, Thyagarajan S (2007). The influence of name on the acceptability of ECT: the

importance of political correctness. J ECT 23: 75–7.

Andreescu C, Tudorascu DL, Butters MA, Tamburo E, Patel M, Price J, et al (2013). Resting

state functional connectivity and treatment response in late-life depression. Psychiatry Res

214: 313–21.

277

Andrews-Hanna JR, Reidler JS, Sepulcre J, Poulin R, Buckner RL (2010). Functional-anatomic

fractionation of the brain’s default network. Neuron 65: 550–62.

Anticevic A, Hu S, Zhang S, Savic A, Billingslea E, Wasylink S, et al (2014). Global resting-

state functional magnetic resonance imaging analysis identifies frontal cortex, striatal, and

cerebellar dysconnectivity in obsessive-compulsive disorder. Biol Psychiatry 75: 595–605.

Arcelus J, Mitchell AJ, Wales J, Nielsen S (2011). Mortality rates in patients with anorexia

nervosa and other eating disorders. A meta-analysis of 36 studies. Arch Gen Psychiatry 68:

724–31.

Argyelan M, Lencz T, Kaliora S, Sarpal DK, Weissman N, Kingsley PB, et al (2016). Subgenual

cingulate cortical activity predicts the efficacy of electroconvulsive therapy. Transl

Psychiatry 6: e789.

Arikuni T, Kubota K (1986). The organization of prefrontocaudate projections and their laminar

origin in the macaque monkey: a retrograde study using HRP-gel. J Comp Neurol 244: 492–

510.

Arns M, Drinkenburg WH, Fitzgerald PB, Kenemans JL (2012). Neurophysiological predictors

of non-response to rTMS in depression. Brain Stimul 5: 569–76.

Aron AR, Fletcher PC, Bullmore ET, Sahakian BJ, Robbins TW (2003). Stop-signal inhibition

disrupted by damage to right inferior frontal gyrus in humans. Nat Neurosci 6: 115–116.

Arslan S, Ktena SI, Makropoulos A, Robinson EC, Rueckert D, Parisot S (2018). Human brain

mapping: A systematic comparison of parcellation methods for the human cerebral cortex.

Neuroimage 170: 5–30.

Atlas LY, Wager TD (2014). A meta-analysis of brain mechanisms of placebo analgesia:

consistent findings and unanswered questions. Handb Exp Pharmacol 225: 37–69.

Atmaca M, Mermi O, Yildirim H, Gurok MG (2016). Orbito-frontal cortex and thalamus

volumes in obsessive-compulsive disorder before and after pharmacotherapy. Brain

Imaging Behav 10: 669–74.

Attwell D, Buchan AM, Charpak S, Lauritzen M, Macvicar BA, Newman EA (2010). Glial and

neuronal control of brain blood flow. Nature 468: 232–43.

Aurich NK, Alves Filho JO, Marques da Silva AM, Franco AR (2015). Evaluating the reliability

278

of different preprocessing steps to estimate graph theoretical measures in resting state fMRI

data. Front Neurosci 9: 48.

Averbeck BB, Lehman J, Jacobson M, Haber SN (2014). Estimates of projection overlap and

zones of convergence within frontal-striatal circuits. J Neurosci 34: 9497–505.

Avissar M, Powell F, Ilieva I, Respino M, Gunning FM, Liston C, et al (2017). Functional

connectivity of the left DLPFC to striatum predicts treatment response of depression to

TMS. Brain Stimul 10: 919–925.

Baek K, Shim WH, Jeong J, Radhakrishnan H, Rosen BR, Boas D, et al (2016). Layer-specific

interhemispheric functional connectivity in the somatosensory cortex of rats: resting state

electrophysiology and fMRI studies. Brain Struct Funct 221: 2801–15.

Baeken C, Marinazzo D, Everaert H, Wu G-R, Hove C Van, Audenaert K, et al (2015). The

Impact of Accelerated HF-rTMS on the Subgenual Anterior Cingulate Cortex in Refractory

Unipolar Major Depression: Insights From 18FDG PET Brain Imaging. Brain Stimul 8:

808–15.

Baeken C, Marinazzo D, Wu G-R, Schuerbeek P Van, Mey J De, Marchetti I, et al (2014).

Accelerated HF-rTMS in treatment-resistant unipolar depression: Insights from subgenual

anterior cingulate functional connectivity. World J Biol Psychiatry 15: 286–97.

Baeken C, Raedt R De, Hove C Van, Clerinx P, Mey J De, Bossuyt A (2009). HF-rTMS

treatment in medication-resistant melancholic depression: results from 18FDG-PET brain

imaging. CNS Spectr 14: 439–48.

Baeken C, Vanderhasselt M-A, Remue J, Herremans S, Vanderbruggen N, Zeeuws D, et al

(2013). Intensive HF-rTMS treatment in refractory medication-resistant unipolar depressed

patients. J Affect Disord 151: 625–31.

Bailer UF, Frank GK, Henry SE, Price JC, Meltzer CC, Mathis CA, et al (2007). Exaggerated 5-

HT1A but normal 5-HT2A receptor activity in individuals ill with anorexia nervosa. Biol

Psychiatry 61: 1090–9.

Bailer UF, Price JC, Meltzer CC, Mathis CA, Frank GK, Weissfeld L, et al (2004). Altered 5-

HT(2A) receptor binding after recovery from bulimia-type anorexia nervosa: relationships

to harm avoidance and drive for thinness. Neuropsychopharmacology 29: 1143–55.

279

Bailer UF, Price JC, Meltzer CC, Wagner A, Mathis CA, Gamst A, et al (2017). Dopaminergic

activity and altered reward modulation in anorexia nervosa-insight from multimodal

imaging. Int J Eat Disord 50: 593–596.

Bair MJ, Robinson RL, Katon W, Kroenke K (2003). Depression and pain comorbidity: a

literature review. Arch Intern Med 163: 2433–45.

Bajbouj M, Brakemeier E-L, Schubert F, Lang UE, Neu P, Schindowski C, et al (2005).

Repetitive transcranial magnetic stimulation of the dorsolateral prefrontal cortex and

cortical excitability in patients with major depressive disorder. Exp Neurol 196: 332–8.

Bakker N, Shahab S, Giacobbe P, Blumberger DM, Daskalakis ZJ, Kennedy SH, et al (2015).

rTMS of the dorsomedial prefrontal cortex for major depression: safety, tolerability,

effectiveness, and outcome predictors for 10 Hz versus intermittent theta-burst stimulation.

Brain Stimul 8: 208–15.

Balodis IM, Molina ND, Kober H, Worhunsky PD, White MA, Rajita Sinha, et al (2013).

Divergent neural substrates of inhibitory control in binge eating disorder relative to other

manifestations of obesity. Obesity (Silver Spring) 21: 367–77.

Bamford B, Sly R (2010). Exploring quality of life in the eating disorders. Eur Eat Disord Rev

18: 147–53.

Bandelow B, Zohar J, Hollander E, Kasper S, Möller H-J, WFSBP Task Force on Treatment

Guidelines for Anxiety O-C and P-TSD, et al (2008). World Federation of Societies of

Biological Psychiatry (WFSBP) guidelines for the pharmacological treatment of anxiety,

obsessive-compulsive and post-traumatic stress disorders - first revision. World J Biol


Bär K-J, la Cruz F de, Berger S, Schultz CC, Wagner G (2015). Structural and functional

differences in the cingulate cortex relate to disease severity in anorexia nervosa. J

Psychiatry Neurosci 40: 269–79.

Barbas H, Ghashghaei H, Dombrowski SM, Rempel-Clower NL (1999). Medial prefrontal

cortices are unified by common connections with superior temporal cortices and

distinguished by input from memory-related areas in the rhesus monkey. J Comp Neurol

410: 343–67.

280

Barbas H, Saha S, Rempel-Clower N, Ghashghaei T (2003). Serial pathways from primate

prefrontal cortex to autonomic areas may influence emotional expression. BMC Neurosci 4:

25.

Bartholdy S, McClelland J, Kekic M, O’Daly OG, Campbell IC, Werthmann J, et al (2015).

Clinical outcomes and neural correlates of 20 sessions of repetitive transcranial magnetic

stimulation in severe and enduring anorexia nervosa (the TIARA study): study protocol for

a randomised controlled feasibility trial. Trials 16: 548.

Bartova L, Meyer BM, Diers K, Rabl U, Scharinger C, Popovic A, et al (2015). Reduced default

mode network suppression during a working memory task in remitted major depression. J

Psychiatr Res 64: 9–18.

Bassett DS, Wymbs NF, Porter MA, Mucha PJ, Carlson JM, Grafton ST (2011). Dynamic

reconfiguration of human brain networks during learning. Proc Natl Acad Sci U S A 108:

7641–6.

Baumgartner R, Somorjai R, Summers R, Richter W (1999). Assessment of cluster homogeneity

in fMRI data using Kendall’s coefficient of concordance. Magn Reson Imaging 17: 1525–

32.

Baxter LR, Schwartz JM, Phelps ME, Mazziotta JC, Guze BH, Selin CE, et al (1989). Reduction

of prefrontal cortex glucose metabolism common to three types of depression. Arch Gen


Baxter MG, Parker A, Lindner CC, Izquierdo AD, Murray EA (2000). Control of response

selection by reinforcer value requires interaction of amygdala and orbital prefrontal cortex.

J Neurosci 20: 4311–9.

Beall EB, Lowe MJ (2010). The non-separability of physiologic noise in functional connectivity

MRI with spatial ICA at 3T. J Neurosci Methods 191: 263–76.

Beaty RE, Benedek M, Kaufman SB, Silvia PJ (2015). Default and Executive Network Coupling

Supports Creative Idea Production. Sci Rep 5: 10964.

Bechara A, Tranel D, Damasio H (2000). Characterization of the decision-making deficit of

patients with ventromedial prefrontal cortex lesions. Brain 123 ( Pt 1: 2189–202.

Beck AT, Epstein N, Brown G, Steer RA (1988). An inventory for measuring clinical anxiety:

281

Psychometric properties. J Consult Clin Psychol 56: 893–897.

Beck AT, Steer RA, Ball R, Ranieri W (1996). Comparison of Beck Depression Inventories -IA

and -II in psychiatric outpatients. J Pers Assess 67: 588–97.

Beck AT, Ward CH, Mendelson M, Mock J, Erbaugh J (1961). An inventory for measuring

depression. Arch Gen Psychiatry 4: 561–71.

Beckmann CF (2012). Modelling with independent components. Neuroimage 62: 891–901.

Beckmann CF, DeLuca M, Devlin JT, Smith SM (2005). Investigations into resting-state

connectivity using independent component analysis. Philos Trans R Soc Lond B Biol Sci

360: 1001–13.

Beckmann CF, Jenkinson M, Smith SM (2003). General multilevel linear modeling for group

analysis in FMRI. Neuroimage 20: 1052–63.

Behrens TEJ, Johansen-Berg H, Woolrich MW, Smith SM, Wheeler-Kingshott CAM, Boulby

PA, et al (2003). Non-invasive mapping of connections between human thalamus and

cortex using diffusion imaging. Nat Neurosci 6: 750–757.

Behzadi Y, Restom K, Liau J, Liu TT (2007a). A component based noise correction method

(CompCor) for BOLD and perfusion based fMRI. Neuroimage 37: 90–101.

Behzadi Y, Restom K, Liau J, Liu TT (2007b). A component based noise correction method

(CompCor) for BOLD and perfusion based fMRI. Neuroimage 37: 90–101.

Belleau EL, Taubitz LE, Larson CL (2015). Imbalance of default mode and regulatory networks

during externally focused processing in depression. Soc Cogn Affect Neurosci 10: 744–51.

Belliveau JW, Kennedy DN, McKinstry RC, Buchbinder BR, Weisskoff RM, Cohen MS, et al

(1991). Functional mapping of the human visual cortex by magnetic resonance imaging.

Science 254: 716–9.

Benali A, Trippe J, Weiler E, Mix A, Petrasch-Parwez E, Girzalsky W, et al (2011). Theta-Burst

Transcranial Magnetic Stimulation Alters Cortical Inhibition. J Neurosci 31: 1193–1203.

Benjamini Y, Hochberg Y (1995). Controlling the false discovery rate: a practical and powerful

approach to multiple testing. J R Stat Soc B 57: 289–300.

Bennett CM, Miller MB (2010). How reliable are the results from functional magnetic resonance

282

imaging? Ann N Y Acad Sci 1191: 133–55.

Berg KC, Peterson CB, Frazier P, Crow SJ (2012). Psychometric evaluation of the eating

disorder examination and eating disorder examination-questionnaire: a systematic review of

the literature. Int J Eat Disord 45: 428–38.

Berger H (1929). Über das Elektrenkephalogramm des Menschen. Arch Psychiatr Nervenkr 87:

527–570.

Berkman ND, Lohr KN, Bulik CM (2007). Outcomes of eating disorders: a systematic review of

the literature. Int J Eat Disord 40: 293–309.

Berlim MT, Eynde F van den, Tovar-Perdomo S, Daskalakis ZJ (2014). Response, remission and

drop-out rates following high-frequency repetitive transcranial magnetic stimulation (rTMS)

for treating major depression: a systematic review and meta-analysis of randomized,

double-blind and sham-controlled trials. Psychol Med 44: 225–39.

Berlim MT, McGirr A, Rodrigues Dos Santos N, Tremblay S, Martins R (2017). Efficacy of

theta burst stimulation (TBS) for major depression: An exploratory meta-analysis of

randomized and sham-controlled trials. J Psychiatr Res 90: 102–109.

Berlim MT, Neufeld NH, Eynde F Van den (2013). Repetitive transcranial magnetic stimulation

(rTMS) for obsessive-compulsive disorder (OCD): an exploratory meta-analysis of

randomized and sham-controlled trials. J Psychiatr Res 47: 999–1006.

Berlim MT, Turecki G (2007). Definition, assessment, and staging of treatment-resistant

refractory major depression: a review of current concepts and methods. Can J Psychiatry

52: 46–54.

Bermpohl F, Walter M, Sajonz B, Lücke C, Hägele C, Sterzer P, et al (2009). Attentional

modulation of emotional stimulus processing in patients with major depression--alterations

in prefrontal cortical regions. Neurosci Lett 463: 108–13.

Berner LA, Marsh R (2014). Frontostriatal circuits and the development of bulimia nervosa.

Front Behav Neurosci 8: 395.

Bernstein GA, Mueller BA, Schreiner MW, Campbell SM, Regan EK, Nelson PM, et al (2016).

Abnormal striatal resting-state functional connectivity in adolescents with obsessive-

compulsive disorder. Psychiatry Res 247: 49–56.

283

Berntson GG, Norman GJ, Bechara A, Bruss J, Tranel D, Cacioppo JT (2011). The insula and

evaluative processes. Psychol Sci 22: 80–6.

Beucke JC, Kaufmann C, Linnman C, Gruetzmann R, Endrass T, Deckersbach T, et al (2012).

Altered cingulostriatal coupling in obsessive-compulsive disorder. Brain Connect 2: 191–

202.

Beucke JC, Sepulcre J, Talukdar T, Linnman C, Zschenderlein K, Endrass T, et al (2013).

Abnormally high degree connectivity of the orbitofrontal cortex in obsessive-compulsive

disorder. JAMA psychiatry 70: 619–29.

Bhagalia R, Kim B (2008). Spin saturation artifact correction using slice-to-volume registration

motion estimates for fMRI time series. Med Phys 35: 424–34.

Biezonski D, Cha J, Steinglass J, Posner J (2016). Evidence for Thalamocortical Circuit

Abnormalities and Associated Cognitive Dysfunctions in Underweight Individuals with

Anorexia Nervosa. Neuropsychopharmacology 41: 1560–8.

Binder JR, Liebenthal E, Possing ET, Medler DA, Ward BD (2004). Neural correlates of sensory

and decision processes in auditory object identification. Nat Neurosci 7: 295–301.

Birbaumer N, Elbert T, Canavan AG, Rockstroh B (1990). Slow potentials of the cerebral cortex

and behavior. Physiol Rev 70: 1–41.

Birn RM (2012). The role of physiological noise in resting-state functional connectivity.


Birn RM, Diamond JB, Smith MA, Bandettini PA (2006). Separating respiratory-variation-

related fluctuations from neuronal-activity-related fluctuations in fMRI. Neuroimage 31:

1536–48.

Birnbaum HG, Kessler RC, Kelley D, Ben-Hamadi R, Joish VN, Greenberg PE (2010).

Employer burden of mild, moderate, and severe major depressive disorder: mental health

services utilization and costs, and work performance. Depress Anxiety 27: 78–89.

Bischoff-Grethe A, McCurdy D, Grenesko-Stevens E, Irvine LEZ, Wagner A, Yau W-YW, et al

(2013). Altered brain response to reward and punishment in adolescents with Anorexia

nervosa. Psychiatry Res 214: 331–40.

Bisley JW, Goldberg ME (2003). Neuronal activity in the lateral intraparietal area and spatial

284

attention. Science 299: 81–6.

Biswal B, DeYoe EA, Hyde JS (1996). Reduction of physiological fluctuations in fMRI using

digital filters. Magn Reson Med 35: 107–13.

Biswal B, Hudetz AG, Yetkin FZ, Haughton VM, Hyde JS (1997). Hypercapnia reversibly

suppresses low-frequency fluctuations in the human motor cortex during rest using echo-

planar MRI. J Cereb Blood Flow Metab 17: 301–8.

Biswal B, Yetkin FZ, Haughton VM, Hyde JS (1995). Functional connectivity in the motor

cortex of resting human brain using echo-planar MRI. Magn Reson Med 34: 537–41.

Bloch MH, McGuire J, Landeros-Weisenberger A, Leckman JF, Pittenger C (2010). Meta-

analysis of the dose-response relationship of SSRI in obsessive-compulsive disorder. Mol


Blumberger DM, Vila-Rodriguez F, Thorpe KE, Feffer K, Noda Y, Giacobbe P, et al (2018).

Effectiveness of theta burst versus high-frequency repetitive transcranial magnetic

stimulation in patients with depression (THREE-D): a randomised non-inferiority trial.

Lancet (London, England) 391: 1683–1692.

Boelema SR, Harakeh Z, Zandvoort MJE van, Reijneveld SA, Verhulst FC, Ormel J, et al

(2016). Executive functioning before and after onset of alcohol use disorder in adolescence.

A TRAILS study. J Psychiatr Res 78: 78–85.

Boes AD, Grafft AH, Joshi C, Chuang NA, Nopoulos P, Anderson SW (2011). Behavioral

effects of congenital ventromedial prefrontal cortex malformation. BMC Neurol 11: 151.

Bohon C, Stice E (2011). Reward abnormalities among women with full and subthreshold

bulimia nervosa: a functional magnetic resonance imaging study. Int J Eat Disord 44: 585–

95.

Boorman ED, O’Doherty JP, Adolphs R, Rangel A (2013). The behavioral and neural

mechanisms underlying the tracking of expertise. Neuron 80: 1558–71.

Bora E, Fornito A, Pantelis C, Yücel M (2012a). Gray matter abnormalities in Major Depressive

Disorder: a meta-analysis of voxel based morphometry studies. J Affect Disord 138: 9–18.

Bora E, Fornito A, Yücel M, Pantelis C (2012b). The effects of gender on grey matter

abnormalities in major psychoses: a comparative voxelwise meta-analysis of schizophrenia

285

and bipolar disorder. Psychol Med 42: 295–307.

Bora E, Harrison BJ, Davey CG, Yücel M, Pantelis C (2012c). Meta-analysis of volumetric

abnormalities in cortico-striatal-pallidal-thalamic circuits in major depressive disorder.

Psychol Med 42: 671–81.

Bora E, Murray RM (2014). Meta-analysis of cognitive deficits in ultra-high risk to psychosis

and first-episode psychosis: do the cognitive deficits progress over, or after, the onset of

psychosis? Schizophr Bull 40: 744–55.

Boubela RN, Kalcher K, Huf W, Kronnerwetter C, Filzmoser P, Moser E (2013). Beyond Noise:

Using Temporal ICA to Extract Meaningful Information from High-Frequency fMRI Signal

Fluctuations during Rest. Front Hum Neurosci 7: 168.

Bourne SK, Eckhardt CA, Sheth SA, Eskandar EN (2012). Mechanisms of deep brain

stimulation for obsessive compulsive disorder: effects upon cells and circuits. Front Integr

Neurosci 6: 29.

Boyacioglu R, Beckmann CF, Barth M (2013). An Investigation of RSN Frequency Spectra

Using Ultra-Fast Generalized Inverse Imaging. Front Hum Neurosci 7: 156.

Brakemeier E-L, Luborzewski A, Danker-Hopfe H, Kathmann N, Bajbouj M (2007). Positive

predictors for antidepressive response to prefrontal repetitive transcranial magnetic

stimulation (rTMS). J Psychiatr Res 41: 395–403.

Brand M, Franke-Sievert C, Jacoby GE, Markowitsch HJ, Tuschen-Caffier B (2007).

Neuropsychological correlates of decision making in patients with bulimia nervosa.

Neuropsychology 21: 742–50.

Brass M, Derrfuss J, Forstmann B, Cramon DY von (2005). The role of the inferior frontal

junction area in cognitive control. Trends Cogn Sci 9: 314–6.

Brass M, Haggard P (2007). To do or not to do: the neural signature of self-control. J Neurosci

27: 9141–5.

Brody AL, Saxena S, Mandelkern MA, Fairbanks LA, Ho ML, Baxter LR (2001). Brain

metabolic changes associated with symptom factor improvement in major depressive

disorder. Biol Psychiatry 50: 171–8.

Broft A, Shingleton R, Kaufman J, Liu F, Kumar D, Slifstein M, et al (2012). Striatal dopamine

286

in bulimia nervosa: a PET imaging study. Int J Eat Disord 45: 648–56.

Brookes MJ, Woolrich M, Luckhoo H, Price D, Hale JR, Stephenson MC, et al (2011).

Investigating the electrophysiological basis of resting state networks using

magnetoencephalography. Proc Natl Acad Sci U S A 108: 16783–8.

Brown GG, Perthen JE, Liu TT, Buxton RB (2007). A Primer on Functional Magnetic

Resonance Imaging. Neuropsychol Rev 17: 107–125.

Bruce KR, Steiger H (2005). Treatment implications of Axis-II comorbidity in eating disorders.

Eat Disord 13: 93–108.

Brunelin J, Jalenques I, Trojak B, Attal J, Szekely D, Gay A, et al (2014). The efficacy and

safety of low frequency repetitive transcranial magnetic stimulation for treatment-resistant

depression: the results from a large multicenter French RCT. Brain Stimul 7: 855–63.

Brunoni AR, Moffa AH, Fregni F, Palm U, Padberg F, Blumberger DM, et al (2016).

Transcranial direct current stimulation for acute major depressive episodes: meta-analysis of

individual patient data. Br J Psychiatry 208: 522–31.

Brunoni AR, Valiengo L, Baccaro A, Zanão TA, Oliveira JF de, Goulart A, et al (2013). The

sertraline vs. electrical current therapy for treating depression clinical study: results from a

factorial, randomized, controlled trial. JAMA psychiatry 70: 383–91.

Bryant-Waugh R, Hankins M, Shafran R, Lask B, Fosson A (1996). A prospective follow-up of

children with anorexia nervosa. J Youth Adolesc 25: 431–437.

Buckner RL (2012). The serendipitous discovery of the brain’s default network. Neuroimage 62:

1137–45.

Buckner RL, Andrews-Hanna JR, Schacter DL (2008). The brain’s default network: anatomy,

function, and relevance to disease. Ann N Y Acad Sci 1124: 1–38.

Buckner RL, Krienen FM, Castellanos A, Diaz JC, Yeo BTT (2011). The organization of the

human cerebellum estimated by intrinsic functional connectivity. J Neurophysiol 106:

2322–45.

Buckner RL, Sepulcre J, Talukdar T, Krienen FM, Liu H, Hedden T, et al (2009). Cortical hubs

revealed by intrinsic functional connectivity: mapping, assessment of stability, and relation

to Alzheimer’s disease. J Neurosci 29: 1860–73.

287

Bulik CM, Sullivan PF, Carter FA, McIntosh V V, Joyce PR (1999). Predictors of rapid and

sustained response to cognitive-behavioral therapy for bulimia nervosa. Int J Eat Disord 26:

137–44.

Bulik CM, Sullivan PF, Joyce PR, Carter FA, McIntosh V V (1998a). Predictors of 1-year

treatment outcome in bulimia nervosa. Compr Psychiatry 39: 206–14.

Bulik CM, Sullivan PF, Kendler KS (1998b). Heritability of binge-eating and broadly defined

bulimia nervosa. Biol Psychiatry 44: 1210–8.

Bulik CM, Sullivan PF, Tozzi F, Furberg H, Lichtenstein P, Pedersen NL (2006). Prevalence,

heritability, and prospective risk factors for anorexia nervosa. Arch Gen Psychiatry 63:

305–12.

Bullmore E, Sporns O (2009). Complex brain networks: graph theoretical analysis of structural

and functional systems. Nat Rev Neurosci 10: 186–98.

Bushara KO, Grafman J, Hallett M (2001). Neural correlates of auditory-visual stimulus onset

asynchrony detection. J Neurosci 21: 300–4.

Butts CT (2009). Revisiting the foundations of network analysis. Science 325: 414–6.

Buzsáki G, Draguhn A (2004). Neuronal oscillations in cortical networks. Science 304: 1926–9.

Caballero-Gaudes C, Reynolds RC (2017). Methods for cleaning the BOLD fMRI signal.


Cabral J, Hugues E, Sporns O, Deco G (2011). Role of local network oscillations in resting-state

functional connectivity. Neuroimage 57: 130–9.

Cai W, Chen T, Ryali S, Kochalka J, Li C-SR, Menon V (2016). Causal Interactions Within a

Frontal-Cingulate-Parietal Network During Cognitive Control: Convergent Evidence from a

Multisite-Multitask Investigation. Cereb Cortex 26: 2140–53.

Calhoun VD, Adali T, Pearlson GD, Pekar JJ (2001). Spatial and temporal independent

component analysis of functional MRI data containing a pair of task-related waveforms.

Hum Brain Mapp 13: 43–53.

Calhoun VD, Liu J, Adali T (2009). A review of group ICA for fMRI data and ICA for joint

inference of imaging, genetic, and ERP data. Neuroimage 45: S163-72.

288

Calvert GA (2001). Crossmodal processing in the human brain: insights from functional

neuroimaging studies. Cereb Cortex 11: 1110–23.

Calzavara R, Mailly P, Haber SN (2007). Relationship between the corticostriatal terminals from

areas 9 and 46, and those from area 8A, dorsal and rostral premotor cortex and area 24c: an

anatomical substrate for cognition to action. Eur J Neurosci 26: 2005–24.

Campbell-Meiklejohn DK, Woolrich MW, Passingham RE, Rogers RD (2008). Knowing when

to stop: the brain mechanisms of chasing losses. Biol Psychiatry 63: 293–300.

Canna A, Prinster A, Monteleone AM, Cantone E, Monteleone P, Volpe U, et al (2017).

Interhemispheric functional connectivity in anorexia and bulimia nervosa. Eur J Neurosci

45: 1129–1140.

Canolty RT, Knight RT (2010). The functional role of cross-frequency coupling. Trends Cogn

Sci 14: 506–15.

Cárdenas-Morales L, Nowak DA, Kammer T, Wolf RC, Schönfeldt-Lecuona C (2010).

Mechanisms and applications of theta-burst rTMS on the human motor cortex. Brain

Topogr 22: 294–306.

Cárdenas-Morales L, Volz LJ, Michely J, Rehme AK, Pool E-M, Nettekoven C, et al (2014).

Network connectivity and individual responses to brain stimulation in the human motor

system. Cereb Cortex 24: 1697–707.

Carew CL, Milne AM, Tatham EL, MacQueen GM, Hall GBC (2013). Neural systems

underlying thought suppression in young women with, and at-risk, for depression. Behav

Brain Res 257: 13–24.

Carpenter LL, Janicak PG, Aaronson ST, Boyadjis T, Brock DG, Cook IA, et al (2012).

Transcranial magnetic stimulation (TMS) for major depression: a multisite, naturalistic,

observational study of acute treatment outcomes in clinical practice. Depress Anxiety 29:

587–96.

Carpenter MB, Nakano K, Kim R (1976). Nigrothalamic projections in the monkey

demonstrated by autoradiographic technics. J Comp Neurol 165: 401–15.

Carragher N, Teesson M, Sunderland M, Newton NC, Krueger RF, Conrod PJ, et al (2016). The

structure of adolescent psychopathology: a symptom-level analysis. Psychol Med 46: 981–

289

94.

Carter JC, Blackmore E, Sutandar-Pinnock K, Woodside DB (2004). Relapse in anorexia

nervosa: a survival analysis. Psychol Med 34: 671–9.

Carter JC, Mercer-Lynn KB, Norwood SJ, Bewell-Weiss C V, Crosby RD, Woodside DB, et al

(2012). A prospective study of predictors of relapse in anorexia nervosa: implications for

relapse prevention. Psychiatry Res 200: 518–23.

Carver CS, White TL (1994). Behavioral inhibition, behavioral activation, and affective

responses to impending reward and punishment: The BIS/BAS Scales. J Pers Soc Psychol

67: 319–333.

Cauda F, Cavanna AE, D’agata F, Sacco K, Duca S, Geminiani GC (2011). Functional

connectivity and coactivation of the nucleus accumbens: a combined functional connectivity

and structure-based meta-analysis. J Cogn Neurosci 23: 2864–77.

Cauda F, Costa T, Torta DME, Sacco K, D’Agata F, Duca S, et al (2012). Meta-analytic

clustering of the insular cortex: characterizing the meta-analytic connectivity of the insula

when involved in active tasks. Neuroimage 62: 343–55.

Cavada C (2000). The Anatomical Connections of the Macaque Monkey Orbitofrontal Cortex. A

Review. Cereb Cortex 10: 220–242.

Cavada C, Goldman-Rakic PS (1989). Posterior parietal cortex in rhesus monkey: I. Parcellation

of areas based on distinctive limbic and sensory corticocortical connections. J Comp Neurol

287: 393–421.

Cavanna AE, Trimble MR (2006). The precuneus: a review of its functional anatomy and

behavioural correlates. Brain 129: 564–83.

Celone KA, Thompson-Brenner H, Ross RS, Pratt EM, Stern CE (2011). An fMRI investigation

of the fronto-striatal learning system in women who exhibit eating disorder behaviors.


Chai XJ, Castañón AN, Ongür D, Whitfield-Gabrieli S (2012). Anticorrelations in resting state

networks without global signal regression. Neuroimage 59: 1420–8.

Chan AW, Mohajerani MH, LeDue JM, Wang YT, Murphy TH (2015). Mesoscale infraslow

spontaneous membrane potential fluctuations recapitulate high-frequency activity cortical

290

motifs. Nat Commun 6: 7738.

Chan TWS, Ahn W-Y, Bates JE, Busemeyer JR, Guillaume S, Redgrave GW, et al (2014).

Differential impairments underlying decision making in anorexia nervosa and bulimia

nervosa: a cognitive modeling analysis. Int J Eat Disord 47: 157–67.

Chang C, Glover GH (2009). Effects of model-based physiological noise correction on default

mode network anti-correlations and correlations. Neuroimage 47: 1448–59.

Chang C, Glover GH (2010). Time-frequency dynamics of resting-state brain connectivity

measured with fMRI. Neuroimage 50: 81–98.

Chang LJ, Yarkoni T, Khaw MW, Sanfey AG (2013). Decoding the role of the insula in human

cognition: functional parcellation and large-scale reverse inference. Cereb Cortex 23: 739–

49.

Chang M, Womer FY, Edmiston EK, Bai C, Zhou Q, Jiang X, et al (2018). Neurobiological

Commonalities and Distinctions Among Three Major Psychiatric Diagnostic Categories: A

Structural MRI Study. Schizophr Bull 44: 65–74.

Chantiluke K, Halari R, Simic M, Pariante CM, Papadopoulos A, Giampietro V, et al (2012).

Fronto-striato-cerebellar dysregulation in adolescents with depression during motivated

attention. Biol Psychiatry 71: 59–67.

Charlson F, Siskind D, Doi SAR, McCallum E, Broome A, Lie DC (2012). ECT efficacy and

treatment course: a systematic review and meta-analysis of twice vs thrice weekly

schedules. J Affect Disord 138: 1–8.

Chavis DA, Pandya DN (1976). Further observations on corticofrontal connections in the rhesus

monkey. Brain Res 117: 369–386.

Chen JE, Glover GH (2015). BOLD fractional contribution to resting-state functional

connectivity above 0.1 Hz. Neuroimage 107: 207–218.

Chen JE, Jahanian H, Glover GH (2017). Nuisance Regression of High-Frequency Functional

Magnetic Resonance Imaging Data: Denoising Can Be Noisy. Brain Connect 7: 13–24.

Chen R, Classen J, Gerloff C, Celnik P, Wassermann EM, Hallett M, et al (1997). Depression of

motor cortex excitability by low-frequency transcranial magnetic stimulation. Neurology

48: 1398–403.

291

Chen S, Ross TJ, Zhan W, Myers CS, Chuang K-S, Heishman SJ, et al (2008). Group

independent component analysis reveals consistent resting-state networks across multiple

sessions. Brain Res 1239: 141–51.

Chen T, Michels L, Supekar K, Kochalka J, Ryali S, Menon V (2015). Role of the anterior

insular cortex in integrative causal signaling during multisensory auditory-visual attention.

Eur J Neurosci 41: 264–74.

Chen X, Pereira F, Lee W, Strother S, Mitchell T (2006). Exploring predictive and reproducible

modeling with the single-subject FIAC dataset. Hum Brain Mapp 27: 452–61.

Chen Y, Meng X, Hu Q, Cui H, Ding Y, Kang L, et al (2016). Altered resting-state functional

organization within the central executive network in obsessive-compulsive disorder.

Psychiatry Clin Neurosci 70: 448–456.

Cheng W, Rolls ET, Gu H, Zhang J, Feng J (2015). Autism: reduced connectivity between

cortical areas involved in face expression, theory of mind, and the sense of self. Brain 138:

1382–93.

Cheng W, Rolls ET, Qiu J, Liu W, Tang Y, Huang C-C, et al (2016). Medial reward and lateral

non-reward orbitofrontal cortex circuits change in opposite directions in depression. Brain

139: 3296–3309.

Cheng W, Rolls ET, Qiu J, Xie X, Wei D, Huang C-C, et al (2018). Increased functional

connectivity of the posterior cingulate cortex with the lateral orbitofrontal cortex in

depression. Transl Psychiatry 8: 90.

Cheng Y, Xu J, Nie B, Luo C, Yang T, Li H, et al (2013). Abnormal resting-state activities and

functional connectivities of the anterior and the posterior cortexes in medication-naïve

patients with obsessive-compulsive disorder. PLoS One 8: e67478.

Chikama M, McFarland NR, Amaral DG, Haber SN (1997). Insular cortical projections to

functional regions of the striatum correlate with cortical cytoarchitectonic organization in

the primate. J Neurosci 17: 9686–705.

Chistyakov A V, Kreinin B, Marmor S, Kaplan B, Khatib A, Darawsheh N, et al (2015).

Preliminary assessment of the therapeutic efficacy of continuous theta-burst magnetic

stimulation (cTBS) in major depression: a double-blind sham-controlled study. J Affect

292

Disord 170: 225–9.

Chistyakov A V, Rubicsek O, Kaplan B, Zaaroor M, Klein E (2010). Safety, tolerability and

preliminary evidence for antidepressant efficacy of theta-burst transcranial magnetic

stimulation in patients with major depression. Int J Neuropsychopharmacol 13: 387–393.

Cho SS, Koshimori Y, Aminian K, Obeso I, Rusjan P, Lang AE, et al (2015). Investing in the

future: stimulation of the medial prefrontal cortex reduces discounting of delayed rewards.

Neuropsychopharmacology 40: 546–53.

Cho SS, Pellecchia G, Aminian K, Ray N, Segura B, Obeso I, et al (2013). Morphometric

correlation of impulsivity in medial prefrontal cortex. Brain Topogr 26: 479–87.

Choi EY, Yeo BTT, Buckner RL (2012). The organization of the human striatum estimated by

intrinsic functional connectivity. J Neurophysiol 108: 2242–63.

Choi KM, Jang K-M, Jang KI, Um YH, Kim M-S, Kim D-W, et al (2014). The effects of 3

weeks of rTMS treatment on P200 amplitude in patients with depression. Neurosci Lett 577:

22–7.

Chouinard PA, Paus T (2010). What have We Learned from “Perturbing” the Human Cortical

Motor System with Transcranial Magnetic Stimulation? Front Hum Neurosci 4: 173.

Christakou A, Robbins TW, Everitt BJ (2004). Prefrontal cortical-ventral striatal interactions

involved in affective modulation of attentional performance: implications for corticostriatal

circuit function. J Neurosci 24: 773–80.

Ciobanu C, Girard M, Marin B, Labrunie A, Malauzat D (2013). rTMS for pharmacoresistant

major depression in the clinical setting of a psychiatric hospital: Effectiveness and effects of

age. J Affect Disord 150: 677–681.

Cirillo G, Pino G Di, Capone F, Ranieri F, Florio L, Todisco V, et al (2017). Neurobiological

after-effects of non-invasive brain stimulation. Brain Stimul 10: 1–18.

Claudino AM, Eynde F Van den, Stahl D, Dew T, Andiappan M, Kalthoff J, et al (2011).

Repetitive transcranial magnetic stimulation reduces cortisol concentrations in bulimic

disorders. Psychol Med 41: 1329–36.

Claudino AM, Oliveira IR de, Appolinario JC, Cordás TA, Duchesne M, Sichieri R, et al (2007).

Double-blind, randomized, placebo-controlled trial of topiramate plus cognitive-behavior

293

therapy in binge-eating disorder. J Clin Psychiatry 68: 1324–32.

Clausen L (2008). Time to remission for eating disorder patients: a 2(1/2)-year follow-up study

of outcome and predictors. Nord J Psychiatry 62: 151–9.

Cloutman LL, Binney RJ, Drakesmith M, Parker GJM, Lambon Ralph MA (2012). The variation

of function across the human insula mirrors its patterns of structural connectivity: evidence

from in vivo probabilistic tractography. Neuroimage 59: 3514–21.

Clower DM, West RA, Lynch JC, Strick PL (2001). The inferior parietal lobule is the target of

output from the superior colliculus, hippocampus, and cerebellum. J Neurosci 21: 6283–91.

Cocchi L, Harrison BJ, Pujol J, Harding IH, Fornito A, Pantelis C, et al (2012). Functional

alterations of large-scale brain networks related to cognitive control in obsessive-

compulsive disorder. Hum Brain Mapp 33: 1089–106.

Cohen LG, Roth BJ, Nilsson J, Dang N, Panizza M, Bandinelli S, et al (1990). Effects of coil

design on delivery of focal magnetic stimulation. Technical considerations.

Electroencephalogr Clin Neurophysiol 75: 350–7.

Cohen RB, Boggio PS, Fregni F (2009). Risk factors for relapse after remission with repetitive

transcranial magnetic stimulation for the treatment of depression. Depress Anxiety 26: 682–

8.

Cole DM, Smith SM, Beckmann CF (2010a). Advances and pitfalls in the analysis and

interpretation of resting-state FMRI data. Front Syst Neurosci 4: 8.

Cole MW, Bassett DS, Power JD, Braver TS, Petersen SE (2014). Intrinsic and task-evoked

network architectures of the human brain. Neuron 83: 238–51.

Cole MW, Pathak S, Schneider W (2010b). Identifying the brain’s most globally connected

regions. Neuroimage 49: 3132–48.

Cole MW, Reynolds JR, Power JD, Repovs G, Anticevic A, Braver TS (2013). Multi-task

connectivity reveals flexible hubs for adaptive task control. Nat Neurosci 16: 1348–55.

Cole MW, Schneider W (2007). The cognitive control network: Integrated cortical regions with

dissociable functions. Neuroimage 37: 343–60.

Connolly KR, Helmer A, Cristancho MA, Cristancho P, O’Reardon JP (2012). Effectiveness of

Transcranial Magnetic Stimulation in Clinical Practice Post-FDA Approval in the United

294

States. J Clin Psychiatry 73: e567–e573.

Cook IA, Schrader LM, Degiorgio CM, Miller PR, Maremont ER, Leuchter AF (2013).

Trigeminal nerve stimulation in major depressive disorder: acute outcomes in an open pilot

study. Epilepsy Behav 28: 221–6.

Cooney RE, Joormann J, Eugène F, Dennis EL, Gotlib IH (2010). Neural correlates of

rumination in depression. Cogn Affect Behav Neurosci 10: 470–8.

Cooper Z, Cooper PJ, Fairburn CG (1989). The validity of the eating disorder examination and

its subscales. Br J Psychiatry 154: 807–12.

Corbetta M, Patel G, Shulman GL (2008). The reorienting system of the human brain: from

environment to theory of mind. Neuron 58: 306–24.

Corbetta M, Shulman GL (2002). Control of goal-directed and stimulus-driven attention in the

brain. Nat Rev Neurosci 3: 201–15.

Cordes D, Haughton VM, Arfanakis K, Carew JD, Turski PA, Moritz CH, et al (2001).

Frequencies contributing to functional connectivity in the cerebral cortex in “resting-state”

data. AJNR Am J Neuroradiol 22: 1326–33.

Costa PT, R. M (Psychological Assessment Resources: 1992). Revised NEO personality

inventory (NEO PI-R) and NEO five-factor inventory (NEO-FFI): Professional manual. .

Coutinho J, Ramos AF, Maia L, Castro L, Conceição E, Geliebter A, et al (2015). Volumetric

alterations in the nucleus accumbens and caudate nucleus in bulimia nervosa: a structural

magnetic resonance imaging study. Int J Eat Disord 48: 206–14.

Cowdrey FA, Filippini N, Park RJ, Smith SM, McCabe C (2014). Increased resting state

functional connectivity in the default mode network in recovered anorexia nervosa. Hum

Brain Mapp 35: 483–91.

Cowdrey FA, Park RJ, Harmer CJ, McCabe C (2011). Increased neural processing of rewarding

and aversive food stimuli in recovered anorexia nervosa. Biol Psychiatry 70: 736–43.

Cox RW, Jesmanowicz A (1999). Real-time 3D image registration for functional MRI. Magn

Reson Med 42: 1014–8.

Craddock RC, Holtzheimer PE, Hu XP, Mayberg HS (2009). Disease state prediction from

resting state functional connectivity. Magn Reson Med 62: 1619–28.

295

Craddock RC, James GA, Holtzheimer PE, Hu XP, Mayberg HS (2012a). A whole brain fMRI

atlas generated via spatially constrained spectral clustering. Hum Brain Mapp 33: 1914–28.

Craddock RC, James GA, Holtzheimer PE, Hu XP, Mayberg HS (2012b). A whole brain fMRI

atlas generated via spatially constrained spectral clustering. Hum Brain Mapp 33: 1914–28.

Craddock RC, Jbabdi S, Yan C-G, Vogelstein JT, Castellanos FX, Martino A Di, et al (2013).

Imaging human connectomes at the macroscale. Nat Methods 10: 524–39.

Craig AD (2002). How do you feel? Interoception: the sense of the physiological condition of the

body. Nat Rev Neurosci 3: 655–66.

Craig AD (2010a). Once an island, now the focus of attention. Brain Struct Funct 214: 395–6.

Craig ADB (2009a). Emotional moments across time: a possible neural basis for time perception

in the anterior insula. Philos Trans R Soc Lond B Biol Sci 364: 1933–42.

Craig ADB (2009b). How do you feel--now? The anterior insula and human awareness. Nat Rev

Neurosci 10: 59–70.

Craig ADB (2010b). The sentient self. Brain Struct Funct 214: 563–77.

Crane AM, Roberts ME, Treasure J (2007). Are obsessive-compulsive personality traits

associated with a poor outcome in anorexia nervosa? A systematic review of randomized

controlled trials and naturalistic outcome studies. Int J Eat Disord 40: 581–8.

Crittenden BM, Mitchell DJ, Duncan J (2015). Recruitment of the default mode network during

a demanding act of executive control. Elife 4: e06481.

Croarkin PE, Nakonezny PA, Wall CA, Murphy LL, Sampson SM, Frye MA, et al (2016).

Transcranial magnetic stimulation potentiates glutamatergic neurotransmission in depressed

adolescents. Psychiatry Res Neuroimaging 247: 25–33.

Crofts JJ, Higham DJ, Bosnell R, Jbabdi S, Matthews PM, Behrens TEJ, et al (2011). Network

analysis detects changes in the contralesional hemisphere following stroke. Neuroimage 54:

161–9.

Crowther A, Smoski MJ, Minkel J, Moore T, Gibbs D, Petty C, et al (2015). Resting-state

connectivity predictors of response to psychotherapy in major depressive disorder.


296

Crutcher MD, DeLong MR (1984). Single cell studies of the primate putamen. II. Relations to

direction of movement and pattern of muscular activity. Exp brain Res 53: 244–58.

Cuijpers P (2015). Psychotherapies for adult depression: recent developments. Curr Opin


Cuijpers P, Sijbrandij M, Koole SL, Andersson G, Beekman AT, Reynolds CF (2013). The

efficacy of psychotherapy and pharmacotherapy in treating depressive and anxiety

disorders: a meta-analysis of direct comparisons. World Psychiatry 12: 137–48.

Cuijpers P, Weitz E, Twisk J, Kuehner C, Cristea I, David D, et al (2014). Gender as predictor

and moderator of outcome in cognitive behavior therapy and pharmacotherapy for adult

depression: an “individual patient data” meta-analysis. Depress Anxiety 31: 941–51.

Cullen KR, Klimes-Dougan B, Vu DP, Westlund Schreiner M, Mueller BA, Eberly LE, et al

(2016). Neural Correlates of Antidepressant Treatment Response in Adolescents with Major

Depressive Disorder. J Child Adolesc Psychopharmacol 26: 705–712.

Cummings JL (1993). Frontal-subcortical circuits and human behavior. Arch Neurol 50: 873–80.

Curtis CE, D’Esposito M (2003). Persistent activity in the prefrontal cortex during working

memory. Trends Cogn Sci 7: 415–423.

D’Urso G, Brunoni AR, Anastasia A, Micillo M, Bartolomeis A de, Mantovani A (2016).

Polarity-dependent effects of transcranial direct current stimulation in obsessive-compulsive

disorder. Neurocase 22: 60–4.

Daly JJ, Prudic J, Devanand DP, Nobler MS, Lisanby SH, Peyser S, et al (2001). ECT in bipolar

and unipolar depression: differences in speed of response. Bipolar Disord 3: 95–104.

Damoiseaux JS, Rombouts SARB, Barkhof F, Scheltens P, Stam CJ, Smith SM, et al (2006).

Consistent resting-state networks across healthy subjects. Proc Natl Acad Sci U S A 103:

13848–53.

Davidson RJ, Irwin W, Anderle MJ, Kalin NH (2003). The neural substrates of affective

processing in depressed patients treated with venlafaxine. Am J Psychiatry 160: 64–75.

Decker JH, Figner B, Steinglass JE (2015). On Weight and Waiting: Delay Discounting in

Anorexia Nervosa Pretreatment and Posttreatment. Biol Psychiatry 78: 606–14.

DeGuzman M, Shott ME, Yang TT, Riederer J, Frank GKW (2017). Association of Elevated

297

Reward Prediction Error Response With Weight Gain in Adolescent Anorexia Nervosa. Am

J Psychiatry 174: 557–565.

Delaveau P, Jabourian M, Lemogne C, Guionnet S, Bergouignan L, Fossati P (2011). Brain

effects of antidepressants in major depression: a meta-analysis of emotional processing

studies. J Affect Disord 130: 66–74.

Delgado MR, Nearing KI, Ledoux JE, Phelps EA (2008). Neural circuitry underlying the

regulation of conditioned fear and its relation to extinction. Neuron 59: 829–38.

Delva NJ, Graf P, Patry S, Gosselin C, Milev R, Gilron I, et al (2011). Access to

electroconvulsive therapy services in Canada. J ECT 27: 300–9.

Demyttenaere K, Mesters P, Boulanger B, Dewe W, Delsemme MH, Gregoire J, et al (2001).

Adherence to treatment regimen in depressed patients treated with amitriptyline or

fluoxetine. J Affect Disord 65: 243–52.

Denys D, Mantione M, Figee M, Munckhof P van den, Koerselman F, Westenberg H, et al

(2010). Deep brain stimulation of the nucleus accumbens for treatment-refractory

obsessive-compulsive disorder. Arch Gen Psychiatry 67: 1061–8.

Desmurget, Grafton (2000). Forward modeling allows feedback control for fast reaching

movements. Trends Cogn Sci 4: 423–431.

Devinsky O, Morrell MJ, Vogt BA (1995). Contributions of anterior cingulate cortex to

behaviour. Brain 118 ( Pt 1: 279–306.

Devlin B, Bacanu S-A, Klump KL, Bulik CM, Fichter MM, Halmi KA, et al (2002). Linkage

analysis of anorexia nervosa incorporating behavioral covariates. Hum Mol Genet 11: 689–

96.

DiFiglia M, Carey J (1986). Large neurons in the primate neostriatum examined with the

combined Golgi-electron microscopic method. J Comp Neurol 244: 36–52.

Dijk KRA Van, Hedden T, Venkataraman A, Evans KC, Lazar SW, Buckner RL (2010).

Intrinsic functional connectivity as a tool for human connectomics: theory, properties, and

optimization. J Neurophysiol 103: 297–321.

Dijk KRA Van, Sabuncu MR, Buckner RL (2012). The influence of head motion on intrinsic

functional connectivity MRI. Neuroimage 59: 431–8.

298

Ding J, Peterson JD, Surmeier DJ (2008). Corticostriatal and thalamostriatal synapses have

distinctive properties. J Neurosci 28: 6483–92.

Dixon ML, Fox KCR, Christoff K (2014). A framework for understanding the relationship

between externally and internally directed cognition. Neuropsychologia 62: 321–30.

Dong D, Lei X, Jackson T, Wang Y, Su Y, Chen H (2014). Altered regional homogeneity and

efficient response inhibition in restrained eaters. Neuroscience 266: 116–26.

Dosenbach NUF, Fair DA, Cohen AL, Schlaggar BL, Petersen SE (2008). A dual-networks

architecture of top-down control. Trends Cogn Sci 12: 99–105.

Dosenbach NUF, Visscher KM, Palmer ED, Miezin FM, Wenger KK, Kang HC, et al (2006). A

core system for the implementation of task sets. Neuron 50: 799–812.

Dougherty DD, Baer L, Cosgrove GR, Cassem EH, Price BH, Nierenberg AA, et al (2002).

Prospective long-term follow-up of 44 patients who received cingulotomy for treatment-

refractory obsessive-compulsive disorder. Am J Psychiatry 159: 269–75.

Dowdle LT, Brown TR, George MS, Hanlon CA (2018). Single pulse TMS to the DLPFC,

compared to a matched sham control, induces a direct, causal increase in caudate, cingulate,

and thalamic BOLD signal. Brain Stimul 11: 789–796.

Downar J, Crawley AP, Mikulis DJ, Davis KD (2000). A multimodal cortical network for the

detection of changes in the sensory environment. Nat Neurosci 3: 277–83.

Downar J, Crawley AP, Mikulis DJ, Davis KD (2001). The effect of task relevance on the

cortical response to changes in visual and auditory stimuli: an event-related fMRI study.


Downar J, Crawley AP, Mikulis DJ, Davis KD (2002). A cortical network sensitive to stimulus

salience in a neutral behavioral context across multiple sensory modalities. J Neurophysiol

87: 615–20.

Downar J, Daskalakis ZJ (2013). New targets for rTMS in depression: a review of convergent

evidence. Brain Stimul 6: 231–40.

Downar J, Geraci J, Salomons T V, Dunlop K, Wheeler S, McAndrews MP, et al (2014).

Anhedonia and reward-circuit connectivity distinguish nonresponders from responders to

dorsomedial prefrontal repetitive transcranial magnetic stimulation in major depression.

299

Biol Psychiatry 76: 176–85.

Downar J, Sankar A, Giacobbe P, Woodside B, Colton P (2012). Unanticipated Rapid Remission

of Refractory Bulimia Nervosa, during High-Dose Repetitive Transcranial Magnetic

Stimulation of the Dorsomedial Prefrontal Cortex: A Case Report. Front psychiatry 3: 30.

Drevets WC, Bogers W, Raichle ME (2002). Functional anatomical correlates of antidepressant

drug treatment assessed using PET measures of regional glucose metabolism. Eur

Neuropsychopharmacol 12: 527–44.

Drevets WC, Savitz J, Trimble M (2008). The subgenual anterior cingulate cortex in mood

disorders. CNS Spectr 13: 663–81.

Driessen E, Hollon SD (2010). Cognitive behavioral therapy for mood disorders: efficacy,

moderators and mediators. Psychiatr Clin North Am 33: 537–55.

Drysdale AT, Grosenick L, Downar J, Dunlop K, Mansouri F, Meng Y, et al (2017). Resting-

state connectivity biomarkers define neurophysiological subtypes of depression. Nat Med

23: 28–38.

Du M-Y, Wu Q-Z, Yue Q, Li J, Liao Y, Kuang W-H, et al (2012). Voxelwise meta-analysis of

gray matter reduction in major depressive disorder. Prog Neuropsychopharmacol Biol


Du W, Levin-Schwartz Y, Fu G-S, Ma S, Calhoun VD, Adalı T (2016). The role of diversity in

complex ICA algorithms for fMRI analysis. J Neurosci Methods 264: 129–35.

Dunlop BW, Rajendra JK, Craighead WE, Kelley ME, McGrath CL, Choi KS, et al (2017a).

Functional Connectivity of the Subcallosal Cingulate Cortex And Differential Outcomes to

Treatment With Cognitive-Behavioral Therapy or Antidepressant Medication for Major

Depressive Disorder. Am J Psychiatry 174: 533–545.

Dunlop K, Downar J (2017). Ensuring that novel resting-state fMRI metrics are physiologically

grounded, interpretable and meaningful (A commentary on Canna et al., 2017). Eur J

Neurosci 45: 1127–1128.

Dunlop K, Hanlon CA, Downar J (2017b). Noninvasive brain stimulation treatments for

addiction and major depression. Ann N Y Acad Sci 1394: 31–54.

Dunlop K, Woodside B, Lam E, Olmsted M, Colton P, Giacobbe P, et al (2015). Increases in

300

frontostriatal connectivity are associated with response to dorsomedial repetitive

transcranial magnetic stimulation in refractory binge/purge behaviors. NeuroImage Clin 8:

611–8.

Dunlop K, Woodside B, Olmsted M, Colton P, Giacobbe P, Downar J (2016a). Reductions in

Cortico-Striatal Hyperconnectivity Accompany Successful Treatment of Obsessive-

Compulsive Disorder with Dorsomedial Prefrontal rTMS. Neuropsychopharmacology 41:

1395–403.

Dunlop KA, Woodside B, Downar J (2016b). Targeting Neural Endophenotypes of Eating

Disorders with Non-invasive Brain Stimulation. Front Neurosci 10: 30.

Duprat R, Desmyter S, Rudi DR, Heeringen K van, Abbeele D Van den, Tandt H, et al (2016).

Accelerated intermittent theta burst stimulation treatment in medication-resistant major

depression: A fast road to remission? J Affect Disord 200: 6–14.

Duque J, Labruna L, Verset S, Olivier E, Ivry RB (2012). Dissociating the role of prefrontal and

premotor cortices in controlling inhibitory mechanisms during motor preparation. J


Ebert D, Speck O, König A, Berger M, Hennig J, Hohagen F (1997). 1H-magnetic resonance

spectroscopy in obsessive-compulsive disorder: evidence for neuronal loss in the cingulate

gyrus and the right striatum. Psychiatry Res 74: 173–6.

Ehrlich S, Geisler D, Ritschel F, King JA, Seidel M, Boehm I, et al (2015). Elevated cognitive

control over reward processing in recovered female patients with anorexia nervosa. J


Eippert F, Bingel U, Schoell ED, Yacubian J, Klinger R, Lorenz J, et al (2009). Activation of the

opioidergic descending pain control system underlies placebo analgesia. Neuron 63: 533–

43.

Eisen JL, Pinto A, Mancebo MC, Dyck IR, Orlando ME, Rasmussen SA (2010). A 2-year

prospective follow-up study of the course of obsessive-compulsive disorder. J Clin


Eklund A, Nichols TE, Knutsson H (2016). Cluster failure: Why fMRI inferences for spatial

extent have inflated false-positive rates. Proc Natl Acad Sci U S A 113: 7900–5.

301

El-Hage W, Leman S, Camus V, Belzung C (2013). Mechanisms of antidepressant resistance.

Front Pharmacol 4: 146.

Elbeh KAM, Elserogy YMB, Khalifa HE, Ahmed MA, Hafez MH, Khedr EM (2016). Repetitive

transcranial magnetic stimulation in the treatment of obsessive-compulsive disorders:

Double blind randomized clinical trial. Psychiatry Res 238: 264–269.

Eldaief MC, Halko MA, Buckner RL, Pascual-Leone A (2011). Transcranial magnetic

stimulation modulates the brain’s intrinsic activity in a frequency-dependent manner. Proc

Natl Acad Sci U S A 108: 21229–34.

Ellison Z, Foong J, Howard R, Bullmore E, Williams S, Treasure J (1998). Functional anatomy

of calorie fear in anorexia nervosa. Lancet (London, England) 352: 1192.

Enns MW, Larsen DK, Cox BJ (2000). Discrepancies between self and observer ratings of

depression. The relationship to demographic, clinical and personality variables. J Affect

Disord 60: 33–41.

Enomoto H, Terao Y, Kadowaki S, Nakamura K, Moriya A, Nakatani-Enomoto S, et al (2015).

Effects of L-Dopa and pramipexole on plasticity induced by QPS in human motor cortex. J

Neural Transm 122: 1253–61.

Epanechnikov V (1969). Non-parametric estimation of a multivariate probability density. Theory

Probab Its Appl 14: 153–158.

Evarts E V, Wise SP (1984). Basal ganglia outputs and motor control. Ciba Found Symp 107:

83–102.

Eynde F Van den, Claudino AM, Mogg A, Horrell L, Stahl D, Ribeiro W, et al (2010).

Repetitive transcranial magnetic stimulation reduces cue-induced food craving in bulimic

disorders. Biol Psychiatry 67: 793–5.

Eynde F Van den, Guillaume S, Broadbent H, Campbell IC, Schmidt U (2013). Repetitive

transcranial magnetic stimulation in anorexia nervosa: a pilot study. Eur Psychiatry 28: 98–

101.

Eynde F Van den, Suda M, Broadbent H, Guillaume S, Eynde M Van den, Steiger H, et al

(2012). Structural magnetic resonance imaging in eating disorders: a systematic review of

voxel-based morphometry studies. Eur Eat Disord Rev 20: 94–105.

302

Fairburn CG, Agras WS, Walsh BT, Wilson GT, Stice E (2004). Prediction of outcome in

bulimia nervosa by early change in treatment. Am J Psychiatry 161: 2322–4.

Fairburn CG, Beglin SJ (1994). Assessment of eating disorders: interview or self-report

questionnaire? Int J Eat Disord 16: 363–70.

Fairburn CG, Cooper Z (1993). The Eating Disorder Examination (12th edition). Binge Eat

Nature, assessment, Treat 317–360.

Fairburn CG, Peveler RC, Jones R, Hope RA, Doll HA (1993). Predictors of 12-month outcome

in bulimia nervosa and the influence of attitudes to shape and weight. J Consult Clin

Psychol 61: 696–8.

Fan J, Zhong M, Zhu X, Gan J, Liu W, Niu C, et al (2017). Resting-state functional connectivity

between right anterior insula and right orbital frontal cortex correlate with insight level in

obsessive-compulsive disorder. NeuroImage Clin 15: 1–7.

Fang J, Rong P, Hong Y, Fan Y, Liu J, Wang H, et al (2016). Transcutaneous Vagus Nerve

Stimulation Modulates Default Mode Network in Major Depressive Disorder. Biol


Farris SG, McLean CP, Meter PE Van, Simpson HB, Foa EB (2013). Treatment response,

symptom remission, and wellness in obsessive-compulsive disorder. J Clin Psychiatry 74:

685–90.

Fava M, Rush AJ, Alpert JE, Balasubramani GK, Wisniewski SR, Carmin CN, et al (2008).

Difference in treatment outcome in outpatients with anxious versus nonanxious depression:

a STAR*D report. Am J Psychiatry 165: 342–51.

Fayad SM, Guzick AG, Reid AM, Mason DM, Bertone A, Foote KD, et al (2016). Six-Nine

Year Follow-Up of Deep Brain Stimulation for Obsessive-Compulsive Disorder. PLoS One

11: e0167875.

Feffer K, Fettes P, Giacobbe P, Daskalakis ZJ, Blumberger DM, Downar J (2018). 1Hz rTMS of

the right orbitofrontal cortex for major depression: Safety, tolerability and clinical

outcomes. Eur Neuropsychopharmacol 28: 109–117.

Feis RA, Smith SM, Filippini N, Douaud G, Dopper EGP, Heise V, et al (2015). ICA-based

artifact removal diminishes scan site differences in multi-center resting-state fMRI. Front

303

Neurosci 9: 395.

Fellows LK (2007). The role of orbitofrontal cortex in decision making: a component process

account. Ann N Y Acad Sci 1121: 421–30.

Feng H, Yan L, Guan Y, Cui L (2005). Effects of transcranial magnetic stimulation on motor

cortical excitability and neurofunction after cerebral ischemia-reperfusion injury in rats.

Chinese Med Sci J 20: 226–30.

Ferrari AJ, Charlson FJ, Norman RE, Patten SB, Freedman G, Murray CJL, et al (2013). Burden

of depressive disorders by country, sex, age, and year: findings from the global burden of

disease study 2010. PLoS Med 10: e1001547.

Ferrazzi G, Kuklisova Murgasova M, Arichi T, Malamateniou C, Fox MJ, Makropoulos A, et al

(2014). Resting State fMRI in the moving fetus: a robust framework for motion, bias field

and spin history correction. Neuroimage 101: 555–68.

Fettes P, Peters S, Giacobbe P, Blumberger DM, Downar J (2017a). Neural correlates of

successful orbitofrontal 1 Hz rTMS following unsuccessful dorsolateral and dorsomedial

prefrontal rTMS in major depression: A case report. Brain Stimul 10: 165–167.

Fettes P, Schulze L, Downar J (2017b). Cortico-Striatal-Thalamic Loop Circuits of the

Orbitofrontal Cortex: Promising Therapeutic Targets in Psychiatric Illness. Front Syst

Neurosci 11: 25.

Fettes PW, Moayedi M, Dunlop K, Mansouri F, Vila-Rodriguez F, Giacobbe P, et al (2018).

Abnormal Functional Connectivity of Frontopolar Subregions in Treatment-Nonresponsive

Major Depressive Disorder. Biol psychiatry Cogn Neurosci neuroimaging 3: 337–347.

Field AE, Herzog DB, Keller MB, West J, Nussbaum K, Colditz GA (1997). Distinguishing

recovery from remission in a cohort of bulimic women: how should asymptomatic periods

be described? J Clin Epidemiol 50: 1339–45.

Figee M, Luigjes J, Smolders R, Valencia-Alfonso C-E, Wingen G van, Kwaasteniet B de, et al

(2013). Deep brain stimulation restores frontostriatal network activity in obsessive-

compulsive disorder. Nat Neurosci 16: 386–7.

First MB, Spitzer RL, Gibbon M, Williams JBW (2002). Structured Clinical Interview for DSM-

IV-TR Axis I Disorders. New York State Psychiatr Inst .

304

Fischer S, Smith GT, Anderson KG (2003). Clarifying the role of impulsivity in bulimia nervosa.

Int J Eat Disord 33: 406–11.

Fitzgerald PB, Hoy K, Gunewardene R, Slack C, Ibrahim S, Bailey M, et al (2011). A

randomized trial of unilateral and bilateral prefrontal cortex transcranial magnetic

stimulation in treatment-resistant major depression. Psychol Med 41: 1187–96.

Fitzgerald PB, Hoy K, McQueen S, Maller JJ, Herring S, Segrave R, et al (2009). A Randomized

Trial of rTMS Targeted with MRI Based Neuro-Navigation in Treatment-Resistant

Depression. Neuropsychopharmacology 34: 1255–1262.

Fitzgerald PB, Laird AR, Maller J, Daskalakis ZJ (2008). A meta-analytic study of changes in

brain activation in depression. Hum Brain Mapp 29: 683–95.

Fitzgerald PB, Sritharan A, Daskalakis ZJ, Castella AR de, Kulkarni J, Egan G (2007). A

functional magnetic resonance imaging study of the effects of low frequency right

prefrontal transcranial magnetic stimulation in depression. J Clin Psychopharmacol 27:

488–92.

Fiuzat EC, Rhodes SE V, Murray EA (2017). The Role of Orbitofrontal-Amygdala Interactions

in Updating Action-Outcome Valuations in Macaques. J Neurosci 37: 2463–2470.

Flaherty AW, Graybiel AM (1994). Input-output organization of the sensorimotor striatum in the

squirrel monkey. J Neurosci 14: 599–610.

Foa EB, Kozak MJ, Goodman WK, Hollander E, Jenike MA, Rasmussen SA (1995). DSM-IV

field trial: obsessive-compulsive disorder. Am J Psychiatry 152: 90–6.

Foa EB, Liebowitz MR, Kozak MJ, Davies S, Campeas R, Franklin ME, et al (2005).

Randomized, placebo-controlled trial of exposure and ritual prevention, clomipramine, and

their combination in the treatment of obsessive-compulsive disorder. Am J Psychiatry 162:

151–61.

Fontenelle IS, Fontenelle LF, Borges MC, Prazeres AM, Rangé BP, Mendlowicz M V, et al

(2010). Quality of life and symptom dimensions of patients with obsessive-compulsive

disorder. Psychiatry Res 179: 198–203.

Forbes EE, Olino TM, Ryan ND, Birmaher B, Axelson D, Moyles DL, et al (2010). Reward-

related brain function as a predictor of treatment response in adolescents with major

305

depressive disorder. Cogn Affect Behav Neurosci 10: 107–18.

Forman SD, Cohen JD, Fitzgerald M, Eddy WF, Mintun MA, Noll DC (1995). Improved

assessment of significant activation in functional magnetic resonance imaging (fMRI): use

of a cluster-size threshold. Magn Reson Med 33: 636–47.

Forstmann BU, Keuken MC, Jahfari S, Bazin PL, Neumann J, Schäfer A, et al (2012). Cortico-

subthalamic white matter tract strength predicts interindividual efficacy in stopping a motor

response. Neuroimage 60: 370–375.

Fox MD, Buckner RL, Liu H, Chakravarty MM, Lozano AM, Pascual-Leone A (2014). Resting-

state networks link invasive and noninvasive brain stimulation across diverse psychiatric

and neurological diseases. Proc Natl Acad Sci U S A 111: E4367-75.

Fox MD, Buckner RL, White MP, Greicius MD, Pascual-Leone A (2012a). Efficacy of

transcranial magnetic stimulation targets for depression is related to intrinsic functional

connectivity with the subgenual cingulate. Biol Psychiatry 72: 595–603.

Fox MD, Buckner RL, White MP, Greicius MD, Pascual-Leone A (2012b). Efficacy of

Transcranial Magnetic Stimulation Targets for Depression Is Related to Intrinsic Functional

Connectivity with the Subgenual Cingulate. Biol Psychiatry 72: 595–603.

Fox MD, Corbetta M, Snyder AZ, Vincent JL, Raichle ME (2006). Spontaneous neuronal

activity distinguishes human dorsal and ventral attention systems. Proc Natl Acad Sci U S A

103: 10046–51.

Fox MD, Liu H, Pascual-Leone A (2013a). Identification of reproducible individualized targets

for treatment of depression with TMS based on intrinsic connectivity. Neuroimage 66: 151–

60.

Fox MD, Liu H, Pascual-Leone A (2013b). Identification of reproducible individualized targets

for treatment of depression with TMS based on intrinsic connectivity. Neuroimage 66: 151–

160.

Fox MD, Raichle ME (2007). Spontaneous fluctuations in brain activity observed with functional

magnetic resonance imaging. Nat Rev Neurosci 8: 700–11.

Fox MD, Snyder AZ, Vincent JL, Corbetta M, Essen DC Van, Raichle ME (2005). The human

brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc Natl

306

Acad Sci U S A 102: 9673–8.

Fox MD, Zhang D, Snyder AZ, Raichle ME (2009). The global signal and observed

anticorrelated resting state brain networks. J Neurophysiol 101: 3270–83.

Franco AR, Mannell M V, Calhoun VD, Mayer AR (2013). Impact of analysis methods on the

reproducibility and reliability of resting-state networks. Brain Connect 3: 363–74.

Franco AR, Pritchard A, Calhoun VD, Mayer AR (2009). Interrater and intermethod reliability

of default mode network selection. Hum Brain Mapp 30: 2293–303.

Frank GK, Bailer UF, Henry SE, Drevets W, Meltzer CC, Price JC, et al (2005). Increased

dopamine D2/D3 receptor binding after recovery from anorexia nervosa measured by

positron emission tomography and [11c]raclopride. Biol Psychiatry 58: 908–12.

Frank GK, Kaye WH, Meltzer CC, Price JC, Greer P, McConaha C, et al (2002). Reduced 5-

HT2A receptor binding after recovery from anorexia nervosa. Biol Psychiatry 52: 896–906.

Frank GK, Shott ME, Hagman JO, Mittal VA (2013). Alterations in brain structures related to

taste reward circuitry in ill and recovered anorexia nervosa and in bulimia nervosa. Am J


Frank GK, Wagner A, Achenbach S, McConaha C, Skovira K, Aizenstein H, et al (2006).

Altered brain activity in women recovered from bulimic-type eating disorders after a

glucose challenge: a pilot study. Int J Eat Disord 39: 76–9.

Frank GKW, Reynolds JR, Shott ME, Jappe L, Yang TT, Tregellas JR, et al (2012a). Anorexia

nervosa and obesity are associated with opposite brain reward response.


Frank GKW, Reynolds JR, Shott ME, O’Reilly RC (2011). Altered temporal difference learning

in bulimia nervosa. Biol Psychiatry 70: 728–35.

Frank GKW, Roblek T, Shott ME, Jappe LM, Rollin MDH, Hagman JO, et al (2012b).

Heightened fear of uncertainty in anorexia and bulimia nervosa. Int J Eat Disord 45: 227–

32.

Frank GKW, Shott ME, Riederer J, Pryor TL (2016). Altered structural and effective

connectivity in anorexia and bulimia nervosa in circuits that regulate energy and reward

homeostasis. Transl Psychiatry 6: e932.

307

Fransson P, Marrelec G (2008). The precuneus/posterior cingulate cortex plays a pivotal role in

the default mode network: Evidence from a partial correlation network analysis.


Fraser LM, O’Carroll RE, Ebmeier KP (2008). The effect of electroconvulsive therapy on

autobiographical memory: a systematic review. J ECT 24: 10–7.

Fregni F, Marcolin MA, Myczkowski M, Amiaz R, Hasey G, Rumi DO, et al (2006). Predictors

of antidepressant response in clinical trials of transcranial magnetic stimulation. Int J


Fregni F, Orsati F, Pedrosa W, Fecteau S, Tome FAM, Nitsche MA, et al (2008). Transcranial

direct current stimulation of the prefrontal cortex modulates the desire for specific foods.

Appetite 51: 34–41.

Friederich H-C, Brooks S, Uher R, Campbell IC, Giampietro V, Brammer M, et al (2010).

Neural correlates of body dissatisfaction in anorexia nervosa. Neuropsychologia 48: 2878–

85.

Friederich H-C, Walther S, Bendszus M, Biller A, Thomann P, Zeigermann S, et al (2012). Grey

matter abnormalities within cortico-limbic-striatal circuits in acute and weight-restored

anorexia nervosa patients. Neuroimage 59: 1106–13.

Friedman DP, Aggleton JP, Saunders RC (2002). Comparison of hippocampal, amygdala, and

perirhinal projections to the nucleus accumbens: combined anterograde and retrograde

tracing study in the Macaque brain. J Comp Neurol 450: 345–65.

Frieling H, Fischer J, Wilhelm J, Engelhorn T, Bleich S, Hillemacher T, et al (2012).

Microstructural abnormalities of the posterior thalamic radiation and the mediodorsal

thalamic nuclei in females with anorexia nervosa--a voxel based diffusion tensor imaging

(DTI) study. J Psychiatr Res 46: 1237–42.

Friston KJ, Ashburner J, Frith CD, Poline J-B, Heather JD, Frackowiak RSJ (1995). Spatial

registration and normalization of images. Hum Brain Mapp 3: 165–189.

Friston KJ, Williams S, Howard R, Frackowiak RS, Turner R (1996). Movement-related effects

in fMRI time-series. Magn Reson Med 35: 346–55.

Friston KJ, Worsley KJ, Frackowiak RS, Mazziotta JC, Evans AC (1994). Assessing the

308

significance of focal activations using their spatial extent. Hum Brain Mapp 1: 210–20.

Frydman I, Salles Andrade JB de, Vigne P, Fontenelle LF (2016). Can Neuroimaging Provide

Reliable Biomarkers for Obsessive-Compulsive Disorder? A Narrative Review. Curr

Psychiatry Rep 18: 90.

Fu CHY, Steiner H, Costafreda SG (2013). Predictive neural biomarkers of clinical response in

depression: a meta-analysis of functional and structural neuroimaging studies of

pharmacological and psychological therapies. Neurobiol Dis 52: 75–83.

Fujiki M, Kobayashi H, Abe T, Kamida T (2003). Repetitive transcranial magnetic stimulation

for protection against delayed neuronal death induced by transient ischemia. J Neurosurg

99: 1063–9.

Funke K, Benali A (2011). Modulation of cortical inhibition by rTMS - findings obtained from

animal models. J Physiol 589: 4423–4435.

Furman DJ, Hamilton JP, Gotlib IH (2011). Frontostriatal functional connectivity in major

depressive disorder. Biol Mood Anxiety Disord 1: 11.

Furtado CP, Hoy KE, Maller JJ, Savage G, Daskalakis ZJ, Fitzgerald PB (2013). An

investigation of medial temporal lobe changes and cognition following antidepressant

response: a prospective rTMS study. Brain Stimul 6: 346–54.

Galusca B, Sigaud T, Costes N, Redoute J, Massoubre C, Estour B (2014). Wide impairment of

cerebral serotoninergic activity but inter-individual heterogeneity in bulimia nervosa

patients: a pilot [(18)F]MPPF/PET study. World J Biol Psychiatry 15: 599–608.

Gao F, Wang S, Guo Y, Wang J, Lou M, Wu J, et al (2010). Protective effects of repetitive

transcranial magnetic stimulation in a rat model of transient cerebral ischaemia: a microPET

study. Eur J Nucl Med Mol Imaging 37: 954–961.

Gatt L, Jan S, Mondraty N, Horsfield S, Hart S, Russell J, et al (2014). The household economic

burden of eating disorders and adherence to treatment in Australia. BMC Psychiatry 14:

338.

Gay A, Jaussent I, Sigaud T, Billard S, Attal J, Seneque M, et al (2016). A Lack of Clinical

Effect of High-frequency rTMS to Dorsolateral Prefrontal Cortex on Bulimic Symptoms: A

Randomised, Double-blind Trial. Eur Eat Disord Rev 24: 474–481.

309

Gedge L, Beaudoin A, Lazowski L, Toit R du, Jokic R, Milev R (2012). Effects of

Electroconvulsive Therapy and Repetitive Transcranial Magnetic Stimulation on Serum

Brain-Derived Neurotrophic Factor Levels in Patients with Depression. Front Psychiatry 3:

.

Gee DG, Biswal BB, Kelly C, Stark DE, Margulies DS, Shehzad Z, et al (2011). Low frequency

fluctuations reveal integrated and segregated processing among the cerebral hemispheres.


Geisler D, Borchardt V, Lord AR, Boehm I, Ritschel F, Zwipp J, et al (2016). Abnormal

functional global and local brain connectivity in female patients with anorexia nervosa. J


Genovese CR, Lazar NA, Nichols T (2002). Thresholding of Statistical Maps in Functional

Neuroimaging Using the False Discovery Rate. Neuroimage 15: 870–878.

George MS, Lisanby SH, Avery D, McDonald WM, Durkalski V, Pavlicova M, et al (2010).

Daily Left Prefrontal Transcranial Magnetic Stimulation Therapy for Major Depressive

Disorder. Arch Gen Psychiatry 67: 507.

George MS, Ward HE, Ninan PT, Pollack M, Nahas Z, Anderson B, et al (2008). A pilot study

of vagus nerve stimulation (VNS) for treatment-resistant anxiety disorders. Brain Stimul 1:

112–21.

George MS, Wassermann EM (1994). Rapid-rate transcranial magnetic stimulation and ECT.

Convuls Ther 10: 251–8.

George MS, Wassermann EM, Kimbrell TA, Little JT, Williams WE, Danielson AL, et al

(1997). Mood improvement following daily left prefrontal repetitive transcranial magnetic

stimulation in patients with depression: a placebo-controlled crossover trial. Am J


George MS, Wassermann EM, Williams WA, Callahan A, Ketter TA, Basser P, et al (1995).

Daily repetitive transcranial magnetic stimulation (rTMS) improves mood in depression.

Neuroreport 6: 1853–6.

George MS, Wassermann EM, Williams WA, Steppel J, Pascual-Leone A, Basser P, et al

(1996). Changes in mood and hormone levels after rapid-rate transcranial magnetic

310

stimulation (rTMS) of the prefrontal cortex. J Neuropsychiatry Clin Neurosci 8: 172–80.

Gersner R, Kravetz E, Feil J, Pell G, Zangen A (2011). Long-Term Effects of Repetitive

Transcranial Magnetic Stimulation on Markers for Neuroplasticity: Differential Outcomes

in Anesthetized and Awake Animals. J Neurosci 31: 7521–7526.

Ghashghaei HT, Barbas H (2002). Pathways for emotion: interactions of prefrontal and anterior

temporal pathways in the amygdala of the rhesus monkey. Neuroscience 115: 1261–79.

Ghiglieri V, Pendolino V, Sgobio C, Bagetta V, Picconi B, Calabresi P (2012). Theta-burst

stimulation and striatal plasticity in experimental parkinsonism. Exp Neurol 236: 395–398.

Giel KE, Kullmann S, Preißl H, Bischoff SC, Thiel A, Schmidt U, et al (2013). Understanding

the reward system functioning in anorexia nervosa: crucial role of physical activity. Biol

Psychol 94: 575–81.

Gijsman HJ, Geddes JR, Rendell JM, Nolen WA, Goodwin GM (2004). Antidepressants for

bipolar depression: a systematic review of randomized, controlled trials. Am J Psychiatry

161: 1537–47.

Gilbert AR, Keshavan MS, Diwadkar V, Nutche J, Macmaster F, Easter PC, et al (2008). Gray

matter differences between pediatric obsessive-compulsive disorder patients and high-risk

siblings: a preliminary voxel-based morphometry study. Neurosci Lett 435: 45–50.

Giménez M, Guinea-Izquierdo A, Villalta-Gil V, Martínez-Zalacaín I, Segalàs C, Subirà M, et al

(2017). Brain alterations in low-frequency fluctuations across multiple bands in obsessive

compulsive disorder. Brain Imaging Behav 11: 1690–1706.

Glover GH, Li TQ, Ress D (2000). Image-based method for retrospective correction of

physiological motion effects in fMRI: RETROICOR. Magn Reson Med 44: 162–7.

Godart NT, Flament MF, Curt F, Perdereau F, Lang F, Venisse JL, et al (2003). Anxiety

disorders in subjects seeking treatment for eating disorders: a DSM-IV controlled study.

Psychiatry Res 117: 245–58.

Gohel SR, Biswal BB (2015). Functional integration between brain regions at rest occurs in

multiple-frequency bands. Brain Connect 5: 23–34.

Golanov E V, Yamamoto S, Reis DJ (1994). Spontaneous waves of cerebral blood flow

associated with a pattern of electrocortical activity. Am J Physiol 266: R204-14.

311

Golden NH, Katzman DK, Kreipe RE, Stevens SL, Sawyer SM, Rees J, et al (2003). Eating

disorders in adolescents: position paper of the Society for Adolescent Medicine. J Adolesc

Health 33: 496–503.

Goldman-Rakic PS (1982). Cytoarchitectonic heterogeneity of the primate neostriatum:

subdivision into Island and Matrix cellular compartments. J Comp Neurol 205: 398–413.

Goldman RL, Borckardt JJ, Frohman HA, O’Neil PM, Madan A, Campbell LK, et al (2011).

Prefrontal cortex transcranial direct current stimulation (tDCS) temporarily reduces food

cravings and increases the self-reported ability to resist food in adults with frequent food

craving. Appetite 56: 741–6.

Golestani A-M, Goodyear BG (2011). Regions of interest for resting-state fMRI analysis

determined by inter-voxel cross-correlation. Neuroimage 56: 246–51.

Gomes PVO, Brasil-Neto JP, Allam N, Rodrigues de Souza E (2012). A randomized, double-

blind trial of repetitive transcranial magnetic stimulation in obsessive-compulsive disorder

with three-month follow-up. J Neuropsychiatry Clin Neurosci 24: 437–43.

Gong W, Wan L, Lu W, Ma L, Cheng F, Cheng W, et al (2018). Statistical testing and power

analysis for brain-wide association study. Med Image Anal 47: 15–30.

Goodkind M, Eickhoff SB, Oathes DJ, Jiang Y, Chang A, Jones-Hagata LB, et al (2015).

Identification of a common neurobiological substrate for mental illness. JAMA psychiatry

72: 305–15.

Goodman WK, Foote KD, Greenberg BD, Ricciuti N, Bauer R, Ward H, et al (2010). Deep brain

stimulation for intractable obsessive compulsive disorder: pilot study using a blinded,

staggered-onset design. Biol Psychiatry 67: 535–42.

Goodman WK, Price LH (1992). Assessment of severity and change in obsessive compulsive

disorder. Psychiatr Clin North Am 15: 861–9.

Goodman WK, Price LH, Rasmussen SA, Mazure C, Fleischmann RL, Hill CL, et al (1989). The

Yale-Brown Obsessive Compulsive Scale. I. Development, use, and reliability. Arch Gen


Götestam KG, Eriksen L, Heggestad T, Nielsen S (1998). Prevalence of eating disorders in

Norwegian general hospitals 1990-1994: admissions per year and seasonality. Int J Eat

312

Disord 23: 57–64.

Granger CWJ (1969). Investigating Causal Relations by Econometric Models and Cross-spectral

Methods. Econometrica 37: 424.

Gratz KL, Roemer L (2004). Multidimensional Assessment of Emotion Regulation and

Dysregulation: Development, Factor Structure, and Initial Validation of the Difficulties in

Emotion Regulation Scale. J Psychopathol Behav Assess 26: 41–54.

Graveland GA, DiFiglia M (1985). The frequency and distribution of medium-sized neurons

with indented nuclei in the primate and rodent neostriatum. Brain Res 327: 307–11.

Graybiel AM, Rauch SL (2000). Toward a neurobiology of obsessive-compulsive disorder.

Neuron 28: 343–7.

Greenberg BD, George MS, Martin JD, Benjamin J, Schlaepfer TE, Altemus M, et al (1997).

Effect of prefrontal repetitive transcranial magnetic stimulation in obsessive-compulsive

disorder: a preliminary study. Am J Psychiatry 154: 867–9.

Greenberg BD, Malone DA, Friehs GM, Rezai AR, Kubu CS, Malloy PF, et al (2006). Three-

year outcomes in deep brain stimulation for highly resistant obsessive-compulsive disorder.


Greenberg BD, Rauch SL, Haber SN (2010). Invasive circuitry-based neurotherapeutics:

stereotactic ablation and deep brain stimulation for OCD. Neuropsychopharmacology 35:

317–36.

Greenberg PE, Fournier A-A, Sisitsky T, Pike CT, Kessler RC (2015). The economic burden of

adults with major depressive disorder in the United States (2005 and 2010). J Clin


Grefkes C, Fink GR (2005). The functional organization of the intraparietal sulcus in humans

and monkeys. J Anat 207: 3–17.

Greicius MD, Flores BH, Menon V, Glover GH, Solvason HB, Kenna H, et al (2007). Resting-

state functional connectivity in major depression: abnormally increased contributions from

subgenual cingulate cortex and thalamus. Biol Psychiatry 62: 429–37.

Greicius MD, Krasnow B, Reiss AL, Menon V (2003). Functional connectivity in the resting

brain: a network analysis of the default mode hypothesis. Proc Natl Acad Sci U S A 100:

313

253–8.

Greicius MD, Supekar K, Menon V, Dougherty RF (2009). Resting-state functional connectivity

reflects structural connectivity in the default mode network. Cereb Cortex 19: 72–8.

Greist J, Chouinard G, DuBoff E, Halaris A, Kim SW, Koran L, et al (1995). Double-blind

parallel comparison of three dosages of sertraline and placebo in outpatients with obsessive-

compulsive disorder. Arch Gen Psychiatry 52: 289–95.

Grice DE, Halmi KA, Fichter MM, Strober M, Woodside DB, Treasure JT, et al (2002).

Evidence for a susceptibility gene for anorexia nervosa on chromosome 1. Am J Hum Genet

70: 787–92.

Grilo CM, Pagano ME, Stout RL, Markowitz JC, Ansell EB, Pinto A, et al (2012). Stressful life

events predict eating disorder relapse following remission: six-year prospective outcomes.


Grinband J, Hirsch J, Ferrera VP (2006). A neural representation of categorization uncertainty in

the human brain. Neuron 49: 757–63.

Grisham JR, Fullana MA, Mataix-Cols D, Moffitt TE, Caspi A, Poulton R (2011). Risk factors

prospectively associated with adult obsessive-compulsive symptom dimensions and

obsessive-compulsive disorder. Psychol Med 41: 2495–506.

Grosenick L, Greer S, Knutson B (2008). Interpretable classifiers for FMRI improve prediction

of purchases. IEEE Trans Neural Syst Rehabil Eng 16: 539–48.

Gueorguieva R, Mallinckrodt C, Krystal JH (2011). Trajectories of depression severity in clinical

trials of duloxetine: insights into antidepressant and placebo responses. Arch Gen


Guo W, Liu F, Zhang J, Zhang Z, Yu L, Liu J, et al (2013). Dissociation of regional activity in

the default mode network in first-episode, drug-naive major depressive disorder at rest. J

Affect Disord 151: 1097–101.

Gurden H (2013). Astrocytes: can they be the missing stars linking neuronal activity to

neurofunctional imaging signals? Front Cell Neurosci 7: 21.

Gusnard DA, Akbudak E, Shulman GL, Raichle ME (2001). Medial prefrontal cortex and self-

referential mental activity: relation to a default mode of brain function. Proc Natl Acad Sci

314

U S A 98: 4259–64.

Haber SN (2003). The primate basal ganglia: parallel and integrative networks. J Chem

Neuroanat 26: 317–30.

Haber SN (2014). The place of dopamine in the cortico-basal ganglia circuit. Neuroscience 282:

248–57.

Haber SN (2016). Corticostriatal circuitry. Dialogues Clin Neurosci 18: 7–21.

Haber SN, Kim K-S, Mailly P, Calzavara R (2006). Reward-related cortical inputs define a large

striatal region in primates that interface with associative cortical connections, providing a

substrate for incentive-based learning. J Neurosci 26: 8368–76.

Haber SN, Kunishio K, Mizobuchi M, Lynd-Balta E (1995). The orbital and medial prefrontal

circuit through the primate basal ganglia. J Neurosci 15: 4851–67.

Haber SN, Lynd-Balta E, Mitchell SJ (1993). The organization of the descending ventral pallidal

projections in the monkey. J Comp Neurol 329: 111–28.

Haber SN, Lynd E, Klein C, Groenewegen HJ (1990). Topographic organization of the ventral

striatal efferent projections in the rhesus monkey: an anterograde tracing study. J Comp

Neurol 293: 282–98.

Hale JR, Mayhew SD, Mullinger KJ, Wilson RS, Arvanitis TN, Francis ST, et al (2015).

Comparison of functional thalamic segmentation from seed-based analysis and ICA.


Hallett M (2000). Transcranial magnetic stimulation and the human brain. Nature 406: 147–50.

Hallett M (2007). Transcranial magnetic stimulation: a primer. Neuron 55: 187–99.

Halligan PW, Fink GR, Marshall JC, Vallar G (2003). Spatial cognition: evidence from visual

neglect. Trends Cogn Sci 7: 125–133.

Halligan PW, Marshall JC (1998). Neglect of awareness. Conscious Cogn 7: 356–80.

Hallquist MN, Hwang K, Luna B (2013). The nuisance of nuisance regression: spectral

misspecification in a common approach to resting-state fMRI preprocessing reintroduces

noise and obscures functional connectivity. Neuroimage 82: 208–25.

Halmi KA (2005). The multimodal treatment of eating disorders. World Psychiatry 4: 69–73.

315

Halmi KA, Agras WS, Mitchell J, Wilson GT, Crow S, Bryson SW, et al (2002). Relapse

predictors of patients with bulimia nervosa who achieved abstinence through cognitive

behavioral therapy. Arch Gen Psychiatry 59: 1105–9.

Ham T, Leff A, Boissezon X de, Joffe A, Sharp DJ (2013). Cognitive control and the salience

network: an investigation of error processing and effective connectivity. J Neurosci 33:

7091–8.

Hamada M, Hanajima R, Terao Y, Arai N, Furubayashi T, Inomata-Terada S, et al (2007).

Quadro-pulse stimulation is more effective than paired-pulse stimulation for plasticity

induction of the human motor cortex. Clin Neurophysiol 118: 2672–82.

Hamada M, Murase N, Hasan A, Balaratnam M, Rothwell JC (2013). The role of interneuron

networks in driving human motor cortical plasticity. Cereb Cortex 23: 1593–605.

Hamada M, Terao Y, Hanajima R, Shirota Y, Nakatani-Enomoto S, Furubayashi T, et al (2008).

Bidirectional long-term motor cortical plasticity and metaplasticity induced by quadripulse

transcranial magnetic stimulation. J Physiol 586: 3927–47.

Hamilton JP, Chen G, Thomason ME, Schwartz ME, Gotlib IH (2011a). Investigating neural

primacy in Major Depressive Disorder: multivariate Granger causality analysis of resting-

state fMRI time-series data. Mol Psychiatry 16: 763–72.

Hamilton JP, Furman DJ, Chang C, Thomason ME, Dennis E, Gotlib IH (2011b). Default-mode

and task-positive network activity in major depressive disorder: implications for adaptive

and maladaptive rumination. Biol Psychiatry 70: 327–33.

Hamilton MC (1960). A rating scale for depression. J Neurol Neurosurg Psychiatry 23: 56–62.

Hampton AN, Adolphs R, Tyszka MJ, O’Doherty JP (2007). Contributions of the amygdala to

reward expectancy and choice signals in human prefrontal cortex. Neuron 55: 545–55.

Hanakawa T, Goldfine AM, Hallett M (2017). A Common Function of Basal Ganglia-Cortical

Circuits Subserving Speed in Both Motor and Cognitive Domains. eNeuro 4:

ENEURO.0200-17.2017.

Hanlon CA, Canterberry M, Taylor JJ, DeVries W, Li X, Brown TR, et al (2013). Probing the

frontostriatal loops involved in executive and limbic processing via interleaved TMS and

functional MRI at two prefrontal locations: a pilot study. PLoS One 8: e67917.

316

Haq AU, Sitzmann AF, Goldman ML, Maixner DF, Mickey BJ (2015). Response of depression

to electroconvulsive therapy: a meta-analysis of clinical predictors. J Clin Psychiatry 76:

1374–84.

Harmer CJ, Thilo K V, Rothwell JC, Goodwin GM (2001). Transcranial magnetic stimulation of

medial-frontal cortex impairs the processing of angry facial expressions. Nat Neurosci 4:

17–8.

Harrison BJ, Pujol J, Cardoner N, Deus J, Alonso P, López-Solà M, et al (2013). Brain

corticostriatal systems and the major clinical symptom dimensions of obsessive-compulsive

disorder. Biol Psychiatry 73: 321–8.

Harrison BJ, Soriano-Mas C, Pujol J, Ortiz H, López-Solà M, Hernández-Ribas R, et al (2009).

Altered corticostriatal functional connectivity in obsessive-compulsive disorder. Arch Gen


Hart LM, Granillo MT, Jorm AF, Paxton SJ (2011). Unmet need for treatment in the eating

disorders: a systematic review of eating disorder specific treatment seeking among

community cases. Clin Psychol Rev 31: 727–35.

Harvey P-O, Fossati P, Pochon J-B, Levy R, Lebastard G, Lehéricy S, et al (2005). Cognitive

control and brain resources in major depression: an fMRI study using the n-back task.


Hasin DS, Goodwin RD, Stinson FS, Grant BF (2005). Epidemiology of major depressive

disorder: results from the National Epidemiologic Survey on Alcoholism and Related

Conditions. Arch Gen Psychiatry 62: 1097–106.

Hasin DS, Grant BF (2015). The National Epidemiologic Survey on Alcohol and Related

Conditions (NESARC) Waves 1 and 2: review and summary of findings. Soc Psychiatry

Psychiatr Epidemiol 50: 1609–40.

Hausmann A, Mangweth B, Walpoth M, Hoertnagel C, Kramer-Reinstadler K, Rupp CI, et al

(2004). Repetitive transcranial magnetic stimulation (rTMS) in the double-blind treatment

of a depressed patient suffering from bulimia nervosa: a case report. Int J


Hawken ER, Dilkov D, Kaludiev E, Simek S, Zhang F, Milev R (2016). Transcranial Magnetic

317

Stimulation of the Supplementary Motor Area in the Treatment of Obsessive-Compulsive

Disorder: A Multi-Site Study. Int J Mol Sci 17: 420.

Hay PJ, Touyz S, Sud R (2012). Treatment for severe and enduring anorexia nervosa: a review.

Aust N Z J Psychiatry 46: 1136–44.

Hay PP, Bacaltchuk J, Stefano S, Kashyap P (2009). Psychological treatments for bulimia

nervosa and binging. Cochrane database Syst Rev

CD000562doi:10.1002/14651858.CD000562.pub3.

Hayasaka S, Laurienti PJ (2010). Comparison of characteristics between region-and voxel-based

network analyses in resting-state fMRI data. Neuroimage 50: 499–508.

Hayasaka S, Nichols TE (2003). Validating cluster size inference: random field and permutation

methods. Neuroimage 20: 2343–56.

Hayden BY, Smith D V, Platt ML (2010). Cognitive control signals in posterior cingulate cortex.

Front Hum Neurosci 4: 223.

Hayes DJ, Lipsman N, Chen DQ, Woodside DB, Davis KD, Lozano AM, et al (2015).

Subcallosal Cingulate Connectivity in Anorexia Nervosa Patients Differs From Healthy

Controls: A Multi-tensor Tractography Study. Brain Stimul 8: 758–68.

Haynes J-D, Rees G (2006). Decoding mental states from brain activity in humans. Nat Rev


Haynes WIA, Haber SN (2013). The organization of prefrontal-subthalamic inputs in primates

provides an anatomical substrate for both functional specificity and integration: implications

for Basal Ganglia models and deep brain stimulation. J Neurosci 33: 4804–14.

Hayward G, Mehta MA, Harmer C, Spinks TJ, Grasby PM, Goodwin GM (2007). Exploring the

physiological effects of double-cone coil TMS over the medial frontal cortex on the anterior

cingulate cortex: an H2(15)O PET study. Eur J Neurosci 25: 2224–33.

Hazari N, Narayanaswamy JC, Arumugham SS (2016). Predictors of response to serotonin

reuptake inhibitors in obsessive-compulsive disorder. Expert Rev Neurother 16: 1175–91.

He BJ, Snyder AZ, Zempel JM, Smyth MD, Raichle ME (2008). Electrophysiological correlates

of the brain’s intrinsic large-scale functional architecture. Proc Natl Acad Sci U S A 105:

16039–44.

318

He Y, Wang J, Wang L, Chen ZJ, Yan C, Yang H, et al (2009). Uncovering intrinsic modular

organization of spontaneous brain activity in humans. PLoS One 4: e5226.

Hebb DO (John Wiley and Sons: 1949). The organization of behavior: A neuropsychological

theory. doi:10.1002/sce.37303405110.

Heeger DJ, Ress D (2002). What does fMRI tell us about neuronal activity? Nat Rev Neurosci 3:

142–51.

Hees MLJM van, Rotter T, Ellermann T, Evers SMAA (2013). The effectiveness of individual

interpersonal psychotherapy as a treatment for major depressive disorder in adult

outpatients: a systematic review. BMC Psychiatry 13: 22.

Hegde A, Ravi M, S S V, Arumugham SS, Thirthalli J, Janardhan Reddy YC (2016). Repetitive

Transcranial Magnetic Stimulation Over Presupplementary Motor Area May Not Be

Helpful in Treatment-Refractory Obsessive-Compulsive Disorder: A Case Series. J ECT 32:

139–42.

Henssen A, Zilles K, Palomero-Gallagher N, Schleicher A, Mohlberg H, Gerboga F, et al (2016).

Cytoarchitecture and probability maps of the human medial orbitofrontal cortex. Cortex 75:

87–112.

Herbsman T, Avery D, Ramsey D, Holtzheimer P, Wadjik C, Hardaway F, et al (2009). More

lateral and anterior prefrontal coil location is associated with better repetitive transcranial

magnetic stimulation antidepressant response. Biol Psychiatry 66: 509–15.

Hermes D, Miller KJ, Vansteensel MJ, Aarnoutse EJ, Leijten FSS, Ramsey NF (2012).

Neurophysiologic correlates of fMRI in human motor cortex. Hum Brain Mapp 33: 1689–

99.

Hermundstad AM, Bassett DS, Brown KS, Aminoff EM, Clewett D, Freeman S, et al (2013).

Structural foundations of resting-state and task-based functional connectivity in the human

brain. Proc Natl Acad Sci U S A 110: 6169–74.

Herpertz-Dahlmann BM, Wewetzer C, Schulz E, Remschmidt H (1996). Course and outcome in

adolescent anorexia nervosa. Int J Eat Disord 19: 335–45.

Herwig U, Padberg F, Unger J, Spitzer M, Schönfeldt-Lecuona C (2001a). Transcranial magnetic

stimulation in therapy studies: examination of the reliability of “standard” coil positioning

319

by neuronavigation. Biol Psychiatry 50: 58–61.

Herwig U, Schönfeldt-Lecuona C, Wunderlich AP, Tiesenhausen C von, Thielscher A, Walter H,

et al (2001b). The navigation of transcranial magnetic stimulation. Psychiatry Res 108:

123–31.

Herzog DB, Dorer DJ, Keel PK, Selwyn SE, Ekeblad ER, Flores AT, et al (1999). Recovery and

relapse in anorexia and bulimia nervosa: a 7.5-year follow-up study. J Am Acad Child

Adolesc Psychiatry 38: 829–37.

Heuvel MP van den, Mandl RCW, Kahn RS, Hulshoff Pol HE (2009). Functionally linked

resting-state networks reflect the underlying structural connectivity architecture of the

human brain. Hum Brain Mapp 30: 3127–41.

Heuvel OA van den, Werf YD van der, Verhoef KMW, Wit S de, Berendse HW, Wolters EC, et

al (2010). Frontal-striatal abnormalities underlying behaviours in the compulsive-impulsive

spectrum. J Neurol Sci 289: 55–9.

Hewitt PL, Flett GL, Turnbull-Donovan W, Mikail SF (1991). The Multidimensional

Perfectionism Scale: Reliability, validity, and psychometric properties in psychiatric

samples. Psychol Assess 3: 464–468.

Hieronymus F, Emilsson JF, Nilsson S, Eriksson E (2016). Consistent superiority of selective

serotonin reuptake inhibitors over placebo in reducing depressed mood in patients with

major depression. Mol Psychiatry 21: 523–30.

Hiltunen T, Kantola J, Abou Elseoud A, Lepola P, Suominen K, Starck T, et al (2014). Infra-

slow EEG fluctuations are correlated with resting-state network dynamics in fMRI. J


Hindriks R, Adhikari MH, Murayama Y, Ganzetti M, Mantini D, Logothetis NK, et al (2016).

Can sliding-window correlations reveal dynamic functional connectivity in resting-state

fMRI? Neuroimage 127: 242–256.

Hintiryan H, Foster NN, Bowman I, Bay M, Song MY, Gou L, et al (2016). The mouse cortico-

striatal projectome. Nat Neurosci 19: 1100–14.

Hipp JF, Hawellek DJ, Corbetta M, Siegel M, Engel AK (2012). Large-scale cortical correlation

structure of spontaneous oscillatory activity. Nat Neurosci 15: 884–90.

320

Hiser J, Koenigs M (2018). The Multifaceted Role of the Ventromedial Prefrontal Cortex in

Emotion, Decision Making, Social Cognition, and Psychopathology. Biol Psychiatry 83:

638–647.

Ho TC, Brown S, Serences JT (2009). Domain general mechanisms of perceptual decision

making in human cortex. J Neurosci 29: 8675–87.

Hoek HW (2006). Incidence, prevalence and mortality of anorexia nervosa and other eating

disorders. Curr Opin Psychiatry 19: 389–94.

Hoek HW, Harten PN van, Hermans KME, Katzman MA, Matroos GE, Susser ES (2005). The

incidence of anorexia nervosa on Curaçao. Am J Psychiatry 162: 748–52.

Hoexter MQ, Diniz JB, Lopes AC, Batistuzzo MC, Shavitt RG, Dougherty DD, et al (2015).

ORBITOFRONTAL THICKNESS AS A MEASURE FOR TREATMENT RESPONSE

PREDICTION IN OBSESSIVE-COMPULSIVE DISORDER. Depress Anxiety 32: 900–8.

Hoexter MQ, Souza Duran FL de, D’Alcante CC, Dougherty DD, Shavitt RG, Lopes AC, et al

(2012). Gray matter volumes in obsessive-compulsive disorder before and after fluoxetine

or cognitive-behavior therapy: a randomized clinical trial. Neuropsychopharmacology 37:

734–45.

Höflich G, Kasper S, Hufnagel A, Ruhrmann S, Möller H -J (1993). Application of transcranial

magnetic stimulation in treatment of drug-resistant major depression—a report of two cases.

Hum Psychopharmacol Clin Exp 8: 361–365.

Hofmann SG, Smits JAJ (2008). Cognitive-behavioral therapy for adult anxiety disorders: a

meta-analysis of randomized placebo-controlled trials. J Clin Psychiatry 69: 621–32.

Holmes AJ, Pizzagalli DA (2008). Response conflict and frontocingulate dysfunction in

unmedicated participants with major depression. Neuropsychologia 46: 2904–13.

Holtzheimer PE, Husain MM, Lisanby SH, Taylor SF, Whitworth LA, McClintock S, et al

(2017). Subcallosal cingulate deep brain stimulation for treatment-resistant depression: a

multisite, randomised, sham-controlled trial. The lancet Psychiatry 4: 839–849.

Holtzheimer PE, Kelley ME, Gross RE, Filkowski MM, Garlow SJ, Barrocas A, et al (2012).

Subcallosal cingulate deep brain stimulation for treatment-resistant unipolar and bipolar

depression. Arch Gen Psychiatry 69: 150–8.

321

Holtzheimer PE, McDonald WM, Mufti M, Kelley ME, Quinn S, Corso G, et al (2010).

Accelerated repetitive transcranial magnetic stimulation for treatment-resistant depression.

Depress Anxiety 27: 960–3.

Honey CJ, Sporns O, Cammoun L, Gigandet X, Thiran JP, Meuli R, et al (2009). Predicting

human resting-state functional connectivity from structural connectivity. Proc Natl Acad Sci

U S A 106: 2035–40.

Hornak J, O’Doherty J, Bramham J, Rolls ET, Morris RG, Bullock PR, et al (2004). Reward-

related reversal learning after surgical excisions in orbito-frontal or dorsolateral prefrontal

cortex in humans. J Cogn Neurosci 16: 463–78.

Hou J, Song L, Zhang W, Wu W, Wang J, Zhou D, et al (2013). Morphologic and functional

connectivity alterations of corticostriatal and default mode network in treatment-naïve

patients with obsessive-compulsive disorder. PLoS One 8: e83931.

Huang Y-Z, Chen R-S, Rothwell JC, Wen H-Y (2007). The after-effect of human theta burst

stimulation is NMDA receptor dependent. Clin Neurophysiol 118: 1028–1032.

Huang Y-Z, Edwards MJ, Rounis E, Bhatia KP, Rothwell JC (2005). Theta burst stimulation of

the human motor cortex. Neuron 45: 201–6.

Hubel DH (W. H. Freeman: 1988). Eye, Brain, and Vision. Color Res Appl 13: .

Hudson JI, Hiripi E, Pope HG, Kessler RC (2007). The prevalence and correlates of eating

disorders in the National Comorbidity Survey Replication. Biol Psychiatry 61: 348–58.

Huettel SA, McKeown MJ, Song AW, Hart S, Spencer DD, Allison T, et al (2004). Linking

hemodynamic and electrophysiological measures of brain activity: evidence from functional

MRI and intracranial field potentials. Cereb Cortex 14: 165–73.

Huettel SA, Stowe CJ, Gordon EM, Warner BT, Platt ML (2006). Neural signatures of economic

preferences for risk and ambiguity. Neuron 49: 765–75.

Hutchison RM, Womelsdorf T, Allen EA, Bandettini PA, Calhoun VD, Corbetta M, et al (2013).

Dynamic functional connectivity: promise, issues, and interpretations. Neuroimage 80: 360–

78.

Huyser C, Heuvel OA van den, Wolters LH, Haan E de, Boer F, Veltman DJ (2013). Increased

orbital frontal gray matter volume after cognitive behavioural therapy in paediatric

322

obsessive compulsive disorder. World J Biol Psychiatry 14: 319–31.

Hwang JM, Kim Y-H, Yoon KJ, Uhm KE, Chang WH (2015). Different responses to facilitatory

rTMS according to BDNF genotype. Clin Neurophysiol 126: 1348–53.

Hwang K, Bertolero MA, Liu WB, D’Esposito M (2017). The Human Thalamus Is an Integrative

Hub for Functional Brain Networks. J Neurosci 37: 5594–5607.

Insel TR (2014). The NIMH Research Domain Criteria (RDoC) Project: precision medicine for

psychiatry. Am J Psychiatry 171: 395–7.

Insel TR, Cuthbert BN (2015). Medicine. Brain disorders? Precisely. Science 348: 499–500.

Ishak WW, Balayan K, Bresee C, Greenberg JM, Fakhry H, Christensen S, et al (2013). A

descriptive analysis of quality of life using patient-reported measures in major depressive

disorder in a naturalistic outpatient setting. Qual Life Res 22: 585–96.

Israël M, Steiger H, Kolivakis T, McGregor L, Sadikot AF (2010). Deep brain stimulation in the

subgenual cingulate cortex for an intractable eating disorder. Biol Psychiatry 67: e53-4.

Isserles M, Shalev AY, Roth Y, Peri T, Kutz I, Zlotnick E, et al (2013). Effectiveness of deep

transcranial magnetic stimulation combined with a brief exposure procedure in post-

traumatic stress disorder--a pilot study. Brain Stimul 6: 377–83.

Iwabuchi SJ, Krishnadas R, Li C, Auer DP, Radua J, Palaniyappan L (2015). Localized

connectivity in depression: a meta-analysis of resting state functional imaging studies.

Neurosci Biobehav Rev 51: 77–86.

Jakab A, Molnár PP, Bogner P, Béres M, Berényi EL (2012). Connectivity-based parcellation

reveals interhemispheric differences in the insula. Brain Topogr 25: 264–71.

Jakobs O, Langner R, Caspers S, Roski C, Cieslik EC, Zilles K, et al (2012). Across-study and

within-subject functional connectivity of a right temporo-parietal junction subregion

involved in stimulus-context integration. Neuroimage 60: 2389–98.

Jakobsen JC, Hansen JL, Simonsen S, Simonsen E, Gluud C (2012). Effects of cognitive therapy

versus interpersonal psychotherapy in patients with major depressive disorder: a systematic

review of randomized clinical trials with meta-analyses and trial sequential analyses.


Jakubovski E, Diniz JB, Valerio C, Fossaluza V, Belotto-Silva C, Gorenstein C, et al (2013).

323

Clinical predictors of long-term outcome in obsessive-compulsive disorder. Depress Anxiety

30: 763–72.

Jenkinson M, Bannister P, Brady M, Smith S (2002). Improved optimization for the robust and

accurate linear registration and motion correction of brain images. Neuroimage 17: 825–41.

Jenkinson M, Beckmann CF, Behrens TEJ, Woolrich MW, Smith SM (2012). FSL. Neuroimage

62: 782–90.

Jiang A, Kennedy DN, Baker JR, Weisskoff RM, Tootell RBH, Woods RP, et al (1995). Motion

detection and correction in functional MR imaging. Hum Brain Mapp 3: 224–235.

Jiang J, Beck J, Heller K, Egner T (2015). An insula-frontostriatal network mediates flexible

cognitive control by adaptively predicting changing control demands. Nat Commun 6: 8165.

Jiménez F, Nicolini H, Lozano AM, Piedimonte F, Salín R, Velasco F (2013). Electrical

stimulation of the inferior thalamic peduncle in the treatment of major depression and

obsessive compulsive disorders. World Neurosurg 80: S30.e17-25.

Jo HJ, Gotts SJ, Reynolds RC, Bandettini PA, Martin A, Cox RW, et al (2013). Effective

Preprocessing Procedures Virtually Eliminate Distance-Dependent Motion Artifacts in

Resting State FMRI. J Appl Math 2013: .

Jo HJ, Saad ZS, Simmons WK, Milbury LA, Cox RW (2010). Mapping sources of correlation in

resting state FMRI, with artifact detection and removal. Neuroimage 52: 571–82.

Johnstone T, Reekum CM van, Urry HL, Kalin NH, Davidson RJ (2007). Failure to regulate:

counterproductive recruitment of top-down prefrontal-subcortical circuitry in major

depression. J Neurosci 27: 8877–8884.

Jones EG, Burton H (1976). Areal differences in the laminar distribution of thalamic afferents in

cortical fields of the insular, parietal and temporal regions of primates. J Comp Neurol 168:

197–247.

Jones TB, Bandettini PA, Kenworthy L, Case LK, Milleville SC, Martin A, et al (2010). Sources

of group differences in functional connectivity: an investigation applied to autism spectrum

disorder. Neuroimage 49: 401–14.

Joos AAB, Saum B, Zeeck A, Perlov E, Glauche V, Hartmann A, et al (2011). Frontocingular

dysfunction in bulimia nervosa when confronted with disease-specific stimuli. Eur Eat

324

Disord Rev 19: 447–53.

Jung Y-C, Schulte T, Müller-Oehring EM, Hawkes W, Namkoong K, Pfefferbaum A, et al

(2014). Synchrony of anterior cingulate cortex and insular-striatal activation predicts

ambiguity aversion in individuals with low impulsivity. Cereb Cortex 24: 1397–408.

Jutten C, Herault J (1991). Blind separation of sources, part I: An adaptive algorithm based on

neuromimetic architecture. Signal Processing 24: 1–10.

Kaczkurkin AN, Moore TM, Calkins ME, Ciric R, Detre JA, Elliott MA, et al (2017). Common

and dissociable regional cerebral blood flow differences associate with dimensions of

psychopathology across categorical diagnoses. Mol Psychiatry (in press): .

Kahn I, Desai M, Knoblich U, Bernstein J, Henninger M, Graybiel AM, et al (2011).

Characterization of the functional MRI response temporal linearity via optical control of

neocortical pyramidal neurons. J Neurosci 31: 15086–91.

Kahn I, Knoblich U, Desai M, Bernstein J, Graybiel AM, Boyden ES, et al (2013). Optogenetic

drive of neocortical pyramidal neurons generates fMRI signals that are correlated with

spiking activity. Brain Res 1511: 33–45.

Kahnt T, Tobler PN (2013). Salience signals in the right temporoparietal junction facilitate

value-based decisions. J Neurosci 33: 863–9.

Kaiser RH, Whitfield-Gabrieli S, Dillon DG, Goer F, Beltzer M, Minkel J, et al (2016). Dynamic

Resting-State Functional Connectivity in Major Depression. Neuropsychopharmacology 41:

1822–30.

Kalaska JF, Cisek P, Gosselin-Kessiby N (2003). Mechanisms of selection and guidance of

reaching movements in the parietal lobe. Adv Neurol 93: 97–119.

Kanno M, Matsumoto M, Togashi H, Yoshioka M, Mano Y (2004). Effects of acute repetitive

transcranial magnetic stimulation on dopamine release in the rat dorsolateral striatum. J

Neurol Sci 217: 73–81.

Kannurpatti SS, Biswal BB, Kim YR, Rosen BR (2008). Spatio-temporal characteristics of low-

frequency BOLD signal fluctuations in isoflurane-anesthetized rat brain. Neuroimage 40:

1738–47.

Kaplan AS, Walsh BT, Olmsted M, Attia E, Carter JC, Devlin MJ, et al (2009). The slippery

325

slope: prediction of successful weight maintenance in anorexia nervosa. Psychol Med 39:

1037–45.

Kaye WH, Nagata T, Weltzin TE, Hsu LK, Sokol MS, McConaha C, et al (2001). Double-blind

placebo-controlled administration of fluoxetine in restricting- and restricting-purging-type

anorexia nervosa. Biol Psychiatry 49: 644–52.

Kayser S, Bewernick BH, Grubert C, Hadrysiewicz BL, Axmacher N, Schlaepfer TE (2011).

Antidepressant effects, of magnetic seizure therapy and electroconvulsive therapy, in

treatment-resistant depression. J Psychiatr Res 45: 569–76.

Kayser S, Bewernick BH, Matusch A, Hurlemann R, Soehle M, Schlaepfer TE (2015). Magnetic

seizure therapy in treatment-resistant depression: clinical, neuropsychological and

metabolic effects. Psychol Med 45: 1073–92.

Kedzior KK, Azorina V, Reitz SK (2014). More female patients and fewer stimuli per session

are associated with the short-term antidepressant properties of repetitive transcranial

magnetic stimulation (rTMS): a meta-analysis of 54 sham-controlled studies published

between 1997-2013. Neuropsychiatr Dis Treat 10: 727–56.

Keel PK, Brown TA (2010). Update on course and outcome in eating disorders. Int J Eat Disord

43: 195–204.

Keel PK, Dorer DJ, Franko DL, Jackson SC, Herzog DB (2005). Postremission predictors of

relapse in women with eating disorders. Am J Psychiatry 162: 2263–8.

Keel PK, Klump KL (2003). Are eating disorders culture-bound syndromes? Implications for

conceptualizing their etiology. Psychol Bull 129: 747–69.

Kekic M, McClelland J, Bartholdy S, Boysen E, Musiat P, Dalton B, et al (2017). Single-Session

Transcranial Direct Current Stimulation Temporarily Improves Symptoms, Mood, and Self-

Regulatory Control in Bulimia Nervosa: A Randomised Controlled Trial. PLoS One 12:

e0167606.

Kekic M, McClelland J, Campbell I, Nestler S, Rubia K, David AS, et al (2014). The effects of

prefrontal cortex transcranial direct current stimulation (tDCS) on food craving and

temporal discounting in women with frequent food cravings. Appetite 78: 55–62.

Keller CJ, Bickel S, Honey CJ, Groppe DM, Entz L, Craddock RC, et al (2013).

326

Neurophysiological investigation of spontaneous correlated and anticorrelated fluctuations

of the BOLD signal. J Neurosci 33: 6333–42.

Keller MB, Herzog DB, Lavori PW, Bradburn IS, Mahoney ES (1992). The naturalistic history

of bulimia nervosa: Extraordinarily high rates of chronicity, relapse, recurrence, and

psychosocial morbidity. Int J Eat Disord 12: 1–9.

Kellner CH, Knapp R, Husain MM, Rasmussen K, Sampson S, Cullum M, et al (2010).

Bifrontal, bitemporal and right unilateral electrode placement in ECT: randomised trial. Br J


Kellner M (2010). Drug treatment of obsessive-compulsive disorder. Dialogues Clin Neurosci

12: 187–97.

Kelly AMC, Uddin LQ, Biswal BB, Castellanos FX, Milham MP (2008). Competition between

functional brain networks mediates behavioral variability. Neuroimage 39: 527–37.

Kemp JM, Powell TP (1971). The connexions of the striatum and globus pallidus: synthesis and

speculation. Philos Trans R Soc Lond B Biol Sci 262: 441–57.

Kendler KS, Aggen SH, Knudsen GP, Røysamb E, Neale MC, Reichborn-Kjennerud T (2011).

The structure of genetic and environmental risk factors for syndromal and subsyndromal

common DSM-IV axis I and all axis II disorders. Am J Psychiatry 168: 29–39.

Kendler KS, Prescott CA, Myers J, Neale MC (2003). The structure of genetic and

environmental risk factors for common psychiatric and substance use disorders in men and

women. Arch Gen Psychiatry 60: 929–37.

Kennedy SH, Giacobbe P, Rizvi SJ, Placenza FM, Nishikawa Y, Mayberg HS, et al (2011).

Deep brain stimulation for treatment-resistant depression: follow-up after 3 to 6 years. Am J


Kennedy SH, Lam RW, McIntyre RS, Tourjman SV, Bhat V, Blier P, et al (2016). Canadian

Network for Mood and Anxiety Treatments (CANMAT) 2016 Clinical Guidelines for the

Management of Adults with Major Depressive Disorder: Section 3. Pharmacological

Treatments. Can J Psychiatry 61: 540–60.

Kerestes R, Harrison BJ, Dandash O, Stephanou K, Whittle S, Pujol J, et al (2015). Specific

functional connectivity alterations of the dorsal striatum in young people with depression.

327

NeuroImage Clin 7: 266–72.

Keski-Rahkonen A, Hoek HW, Linna MS, Raevuori A, Sihvola E, Bulik CM, et al (2009).

Incidence and outcomes of bulimia nervosa: a nationwide population-based study. Psychol

Med 39: 823–31.

Kessler D, Angstadt M, Sripada CS (2017). Reevaluating “cluster failure” in fMRI using

nonparametric control of the false discovery rate. Proc Natl Acad Sci U S A 114: E3372–

E3373.

Kessler RC, Barber C, Birnbaum HG, Frank RG, Greenberg PE, Rose RM, et al (1999).

Depression in the workplace: effects on short-term disability. Health Aff (Millwood) 18:

163–71.

Kessler RC, Berglund P, Demler O, Jin R, Koretz D, Merikangas KR, et al (2003). The

epidemiology of major depressive disorder: results from the National Comorbidity Survey

Replication (NCS-R). JAMA 289: 3095–105.

Kessler RC, Chiu WT, Demler O, Merikangas KR, Walters EE (2005). Prevalence, severity, and

comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey

Replication. Arch Gen Psychiatry 62: 617–27.

Kety SS, Schmidt CF (1948). THE NITROUS OXIDE METHOD FOR THE QUANTITATIVE

DETERMINATION OF CEREBRAL BLOOD FLOW IN MAN: THEORY, PROCEDURE

AND NORMAL VALUES. J Clin Invest 27: 476–83.

Khedr EM, Elfetoh NA, Ali AM, Noamany M (2014). Anodal transcranial direct current

stimulation over the dorsolateral prefrontal cortex improves anorexia nervosa: A pilot study.

Restor Neurol Neurosci 32: 789–97.

Kim B, Boes JL, Bland PH, Chenevert TL, Meyer CR (1999). Motion correction in fMRI via

registration of individual slices into an anatomical volume. Magn Reson Med 41: 964–72.

Kim C, Choi HE, Jung H, Lee B-J, Lee KH, Lim Y-J (2014). Comparison of the Effects of 1 Hz

and 20 Hz rTMS on Motor Recovery in Subacute Stroke Patients. Ann Rehabil Med 38:

585–91.

Kim H (2012). A dual-subsystem model of the brain’s default network: self-referential

processing, memory retrieval processes, and autobiographical memory retrieval.

328


Kim HF, Hikosaka O (2015). Parallel basal ganglia circuits for voluntary and automatic

behaviour to reach rewards. Brain 138: 1776–800.

Kimbrell TA, Little JT, Dunn RT, Frye MA, Greenberg BD, Wassermann EM, et al (1999).

Frequency dependence of antidepressant response to left prefrontal repetitive transcranial

magnetic stimulation (rTMS) as a function of baseline cerebral glucose metabolism. Biol


Kimura M (1990). Behaviorally contingent property of movement-related activity of the primate

putamen. J Neurophysiol 63: 1277–96.

Kimura M, Kato M, Shimazaki H, Watanabe K, Matsumoto N (1996). Neural information

transferred from the putamen to the globus pallidus during learned movement in the

monkey. J Neurophysiol 76: 3771–86.

Kirby KN, Marakovic NN (1996). Delay-discounting probabilistic rewards: Rates decrease as

amounts increase. Psychon Bull Rev 3: 100–104.

Kito S, Hasegawa T, Koga Y (2012a). Cerebral blood flow in the ventromedial prefrontal cortex

correlates with treatment response to low-frequency right prefrontal repetitive transcranial

magnetic stimulation in the treatment of depression. Psychiatry Clin Neurosci 66: 138–45.

Kito S, Hasegawa T, Koga Y (2012b). Cerebral blood flow ratio of the dorsolateral prefrontal

cortex to the ventromedial prefrontal cortex as a potential predictor of treatment response to

transcranial magnetic stimulation in depression. Brain Stimul 5: 547–53.

Kiviniemi V, Kantola J-H, Jauhiainen J, Hyvärinen A, Tervonen O (2003). Independent

component analysis of nondeterministic fMRI signal sources. Neuroimage 19: 253–60.

Kiviniemi V, Vire T, Remes J, Elseoud AA, Starck T, Tervonen O, et al (2011). A sliding time-

window ICA reveals spatial variability of the default mode network in time. Brain Connect

1: 339–47.

Klein E, Kolsky Y, Puyerovsky M, Koren D, Chistyakov A, Feinsod M (1999). Right prefrontal

slow repetitive transcranial magnetic stimulation in schizophrenia: a double-blind sham-

controlled pilot study. Biol Psychiatry 46: 1451–4.

Klein TA, Endrass T, Kathmann N, Neumann J, Cramon DY von, Ullsperger M (2007). Neural

329

correlates of error awareness. Neuroimage 34: 1774–81.

Klump KL, Miller KB, Keel PK, McGue M, Iacono WG (2001). Genetic and environmental

influences on anorexia nervosa syndromes in a population-based twin sample. Psychol Med

31: 737–40.

Knopp J, Knowles S, Bee P, Lovell K, Bower P (2013). A systematic review of predictors and

moderators of response to psychological therapies in OCD: do we have enough empirical

evidence to target treatment? Clin Psychol Rev 33: 1067–81.

Knutson B, Bhanji JP, Cooney RE, Atlas LY, Gotlib IH (2008). Neural responses to monetary

incentives in major depression. Biol Psychiatry 63: 686–92.

Knutson B, Greer SM (2008). Anticipatory affect: neural correlates and consequences for choice.

Philos Trans R Soc Lond B Biol Sci 363: 3771–86.

Knutson B, Rick S, Wimmer GE, Prelec D, Loewenstein G (2007). Neural predictors of

purchases. Neuron 53: 147–56.

Kobayashi Y, Amaral DG (2003). Macaque monkey retrosplenial cortex: II. Cortical afferents. J

Comp Neurol 466: 48–79.

Kobayashi Y, Amaral DG (2007). Macaque monkey retrosplenial cortex: III. Cortical efferents. J


Koch K, Reeß TJ, Rus OG, Gürsel DA, Wagner G, Berberich G, et al (2018). Increased Default

Mode Network Connectivity in Obsessive-Compulsive Disorder During Reward

Processing. Front psychiatry 9: 254.

Koechlin E, Ody C, Kouneiher F (2003). The architecture of cognitive control in the human

prefrontal cortex. Science 302: 1181–5.

Kolbinger HM, Höflich G, Hufnagel A, Müller H-J, Kasper S (1995). Transcranial magnetic

stimulation (TMS) in the treatment of major depression: A pilot study. Hum

Psychopharmacol Clin Exp 10: 305–310.

Korgaonkar MS, Williams LM, Song YJ, Usherwood T, Grieve SM (2014). Diffusion tensor

imaging predictors of treatment outcomes in major depressive disorder. Br J Psychiatry

205: 321–8.

Kortegaard LS, Hoerder K, Joergensen J, Gillberg C, Kyvik KO (2001). A preliminary

330

population-based twin study of self-reported eating disorder. Psychol Med 31: 361–5.

Kosillo P, Smith AT (2010). The role of the human anterior insular cortex in time processing.

Brain Struct Funct 214: 623–8.

Koyama MS, Martino A Di, Zuo X-N, Kelly C, Mennes M, Jutagir DR, et al (2011). Resting-

state functional connectivity indexes reading competence in children and adults. J Neurosci

31: 8617–24.

Kozel FA, Johnson KA, Nahas Z, Nakonezny PA, Morgan PS, Anderson BS, et al (2011).

Fractional anisotropy changes after several weeks of daily left high-frequency repetitive

transcranial magnetic stimulation of the prefrontal cortex to treat major depression. J ECT

27: 5–10.

Krall SC, Rottschy C, Oberwelland E, Bzdok D, Fox PT, Eickhoff SB, et al (2015). The role of

the right temporoparietal junction in attention and social interaction as revealed by ALE

meta-analysis. Brain Struct Funct 220: 587–604.

Kreuzer PM, Schecklmann M, Lehner A, Wetter TC, Poeppl TB, Rupprecht R, et al (2015). The

ACDC pilot trial: targeting the anterior cingulate by double cone coil rTMS for the

treatment of depression. Brain Stimul 8: 240–6.

Krienen FM, Yeo BTT, Buckner RL (2014). Reconfigurable task-dependent functional coupling

modes cluster around a core functional architecture. Philos Trans R Soc Lond B Biol Sci

369: .

Kringelbach ML (2005). The human orbitofrontal cortex: linking reward to hedonic experience.

Nat Rev Neurosci 6: 691–702.

Kringelbach ML, Rolls ET (2004). The functional neuroanatomy of the human orbitofrontal

cortex: evidence from neuroimaging and neuropsychology. Prog Neurobiol 72: 341–72.

Krings T, Chiappa KH, Foltys H, Reinges MH, Cosgrove GR, Thron A (2001). Introducing

navigated transcranial magnetic stimulation as a refined brain mapping methodology.

Neurosurg Rev 24: 171–9.

Kroenke K, Spitzer RL, Williams JBW (2001). The PHQ-9: Validity of a brief depression

severity measure. J Gen Intern Med 16: 606–613.

Krueger RF (1999). The structure of common mental disorders. Arch Gen Psychiatry 56: 921–6.

331

Kruggel F, Cramon DY von (1999). Temporal properties of the hemodynamic response in

functional MRI. Hum Brain Mapp 8: 259–71.

Kucyi A, Davis KD (2014). Dynamic functional connectivity of the default mode network tracks

daydreaming. Neuroimage 100: 471–80.

Kucyi A, Schrouff J, Bickel S, Foster BL, Shine JM, Parvizi J (2018). Intracranial

Electrophysiology Reveals Reproducible Intrinsic Functional Connectivity within Human

Brain Networks. J Neurosci 38: 4230–4242.

Kühn S, Gallinat J (2013). Gray Matter Correlates of Posttraumatic Stress Disorder: A

Quantitative Meta-Analysis. Biol Psychiatry 73: 70–74.

Kühn S, Gallinat J, Brass M (2011). “Keep calm and carry on”: structural correlates of

expressive suppression of emotions. PLoS One 6: e16569.

Kühn S, Kaufmann C, Simon D, Endrass T, Gallinat J, Kathmann N (2013). Reduced thickness

of anterior cingulate cortex in obsessive-compulsive disorder. Cortex 49: 2178–85.

Kullmann S, Giel KE, Hu X, Bischoff SC, Teufel M, Thiel A, et al (2014). Impaired inhibitory

control in anorexia nervosa elicited by physical activity stimuli. Soc Cogn Affect Neurosci

9: 917–23.

Kumar N, Chadda RK (2011). Augmentation effect of repetitive transcranial magnetic

stimulation over the supplementary motor cortex in treatment refractory patients with

obsessive compulsive disorder. Indian J Psychiatry 53: 340–2.

Kunishio K, Haber SN (1994). Primate cingulostriatal projection: limbic striatal versus

sensorimotor striatal input. J Comp Neurol 350: 337–56.

Künzle H (1975). Bilateral projections from precentral motor cortex to the putamen and other

parts of the basal ganglia. An autoradiographic study in Macaca fascicularis. Brain Res 88:

195–209.

Kuperberg GR, Lakshmanan BM, Greve DN, West WC (2008). Task and semantic relationship

influence both the polarity and localization of hemodynamic modulation during lexico-

semantic processing. Hum Brain Mapp 29: 544–61.

Kuroda Y, Motohashi N, Ito H, Ito S, Takano A, Nishikawa T, et al (2006). Effects of repetitive

transcranial magnetic stimulation on [11C]raclopride binding and cognitive function in

332

patients with depression. J Affect Disord 95: 35–42.

Kuroda Y, Motohashi N, Ito H, Ito S, Takano A, Takahashi H, et al (2010). Chronic repetitive

transcranial magnetic stimulation failed to change dopamine synthesis rate: preliminary L-

[β-11C]DOPA positron emission tomography study in patients with depression. Psychiatry

Clin Neurosci 64: 659–62.

Kurth F, Zilles K, Fox PT, Laird AR, Eickhoff SB (2010). A link between the systems:

functional differentiation and integration within the human insula revealed by meta-

analysis. Brain Struct Funct 214: 519–34.

Kuyken W, Hayes R, Barrett B, Byng R, Dalgleish T, Kessler D, et al (2015). Effectiveness and

cost-effectiveness of mindfulness-based cognitive therapy compared with maintenance

antidepressant treatment in the prevention of depressive relapse or recurrence (PREVENT):

a randomised controlled trial. Lancet (London, England) 386: 63–73.

Kwong KK, Belliveau JW, Chesler DA, Goldberg IE, Weisskoff RM, Poncelet BP, et al (1992).

Dynamic magnetic resonance imaging of human brain activity during primary sensory

stimulation. Proc Natl Acad Sci U S A 89: 5675–9.

Labedi A, Benali A, Mix A, Neubacher U, Funke K (2014). Modulation of Inhibitory Activity

Markers by Intermittent Theta-burst Stimulation in Rat Cortex is NMDA-receptor

Dependent. Brain Stimul 7: 394–400.

LaConte S, Strother S, Cherkassky V, Anderson J, Hu X (2005). Support vector machines for

temporal classification of block design fMRI data. Neuroimage 26: 317–29.

Lahey BB, Applegate B, Hakes JK, Zald DH, Hariri AR, Rathouz PJ (2012). Is there a general

factor of prevalent psychopathology during adulthood? J Abnorm Psychol 121: 971–7.

Lai C-H, Wu Y-T (2015). The patterns of fractional amplitude of low-frequency fluctuations in

depression patients: the dissociation between temporal regions and fronto-parietal regions. J


Lamichhane B, Dhamala M (2015). The Salience Network and Its Functional Architecture in a

Perceptual Decision: An Effective Connectivity Study. Brain Connect 5: 362–70.

Lan MJ, Chhetry BT, Liston C, Mann JJ, Dubin M (2016). Transcranial Magnetic Stimulation of

Left Dorsolateral Prefrontal Cortex Induces Brain Morphological Changes in Regions

333

Associated with a Treatment Resistant Major Depressive Episode: An Exploratory

Analysis. Brain Stimul 9: 577–83.

Landmann C, Dehaene S, Pappata S, Jobert A, Bottlaender M, Roumenov D, et al (2007).

Dynamics of prefrontal and cingulate activity during a reward-based logical deduction task.

Cereb Cortex 17: 749–59.

Lao-Kaim NP, Fonville L, Giampietro VP, Williams SCR, Simmons A, Tchanturia K (2015).

Aberrant function of learning and cognitive control networks underlie inefficient cognitive

flexibility in anorexia nervosa: a cross-sectional fMRI study. PLoS One 10: e0124027.

Lapenta OM, Sierve K Di, Macedo EC de, Fregni F, Boggio PS (2014). Transcranial direct

current stimulation modulates ERP-indexed inhibitory control and reduces food

consumption. Appetite 83: 42–8.

Larson-Prior LJ, Zempel JM, Nolan TS, Prior FW, Snyder AZ, Raichle ME (2009). Cortical

network functional connectivity in the descent to sleep. Proc Natl Acad Sci U S A 106:

4489–94.

Lask B, Bryant-Waugh R (1992). Early-onset anorexia nervosa and related eating disorders. J

Child Psychol Psychiatry 33: 281–300.

Latora V, Marchiori M (2001). Efficient behavior of small-world networks. Phys Rev Lett 87:

198701.

Laumann TO, Gordon EM, Adeyemo B, Snyder AZ, Joo SJ, Chen M-Y, et al (2015). Functional

System and Areal Organization of a Highly Sampled Individual Human Brain. Neuron 87:

657–70.

Lavenex P, Suzuki WA, Amaral DG (2002). Perirhinal and parahippocampal cortices of the

macaque monkey: projections to the neocortex. J Comp Neurol 447: 394–420.

Lazzaro V Di, Dileone M, Pilato F, Capone F, Musumeci G, Ranieri F, et al (2011). Modulation

of motor cortex neuronal networks by rTMS: comparison of local and remote effects of six

different protocols of stimulation. J Neurophysiol 105: 2150–6.

Leckman JF, Denys D, Simpson HB, Mataix-Cols D, Hollander E, Saxena S, et al (2010).

Obsessive-compulsive disorder: a review of the diagnostic criteria and possible subtypes

and dimensional specifiers for DSM-V. Depress Anxiety 27: 507–27.

334

LeDoux J (2003). The emotional brain, fear, and the amygdala. Cell Mol Neurobiol 23: 727–38.

Lee MH, Smyser CD, Shimony JS (2013). Resting-state fMRI: a review of methods and clinical

applications. AJNR Am J Neuroradiol 34: 1866–72.

Lee S, Ran Kim K, Ku J, Lee J-H, Namkoong K, Jung Y-C (2014). Resting-state synchrony

between anterior cingulate cortex and precuneus relates to body shape concern in anorexia

nervosa and bulimia nervosa. Psychiatry Res 221: 43–8.

Lefaucheur J-P, André-Obadia N, Antal A, Ayache SS, Baeken C, Benninger DH, et al (2014).

Evidence-based guidelines on the therapeutic use of repetitive transcranial magnetic

stimulation (rTMS). Clin Neurophysiol 125: 2150–206.

Lehéricy S, Ducros M, Krainik A, Francois C, Moortele P-F Van de, Ugurbil K, et al (2004). 3-

D diffusion tensor axonal tracking shows distinct SMA and pre-SMA projections to the

human striatum. Cereb Cortex 14: 1302–9.

Lehner A, Schecklmann M, Greenlee MW, Rupprecht R, Langguth B (2016). Triple-site rTMS

for the treatment of chronic tinnitus: a randomized controlled trial. Sci Rep 6: 22302.

Leichnetz GR (2001). Connections of the medial posterior parietal cortex (area 7m) in the

monkey. Anat Rec 263: 215–36.

Lemieux L, Salek-Haddadi A, Lund TE, Laufs H, Carmichael D (2007). Modelling large motion

events in fMRI studies of patients with epilepsy. Magn Reson Imaging 25: 894–901.

Lemogne C, Delaveau P, Freton M, Guionnet S, Fossati P (2012). Medial prefrontal cortex and

the self in major depression. J Affect Disord 136: e1–e11.

Lenz M, Galanis C, Müller-Dahlhaus F, Opitz A, Wierenga CJ, Szabó G, et al (2016). Repetitive

magnetic stimulation induces plasticity of inhibitory synapses. Nat Commun 7: 10020.

Lenz M, Platschek S, Priesemann V, Becker D, Willems LM, Ziemann U, et al (2015).

Repetitive magnetic stimulation induces plasticity of excitatory postsynapses on proximal

dendrites of cultured mouse CA1 pyramidal neurons. Brain Struct Funct 220: 3323–3337.

Lenz M, Vlachos A (2016). Releasing the Cortical Brake by Non-Invasive Electromagnetic

Stimulation? rTMS Induces LTD of GABAergic Neurotransmission. Front Neural Circuits

10: 96.

Leonardi N, Ville D Van De (2015). On spurious and real fluctuations of dynamic functional

335

connectivity during rest. Neuroimage 104: 430–6.

Levkovitz Y, Marx J, Grisaru N, Segal M (1999). Long-term effects of transcranial magnetic

stimulation on hippocampal reactivity to afferent stimulation. J Neurosci 19: 3198–203.

Lewis JW, Beauchamp MS, DeYoe EA (2000). A comparison of visual and auditory motion

processing in human cerebral cortex. Cereb Cortex 10: 873–88.

Li C-T, Chen M-H, Juan C-H, Huang H-H, Chen L-F, Hsieh J-C, et al (2014a). Efficacy of

prefrontal theta-burst stimulation in refractory depression: a randomized sham-controlled

study. Brain 137: 2088–98.

Li C-T, Lin C-P, Chou K-H, Chen I-Y, Hsieh J-C, Wu C-L, et al (2010a). Structural and

cognitive deficits in remitting and non-remitting recurrent depression: a voxel-based

morphometric study. Neuroimage 50: 347–56.

Li C-T, Wang S-J, Hirvonen J, Hsieh J-C, Bai Y-M, Hong C-J, et al (2010b). Antidepressant

mechanism of add-on repetitive transcranial magnetic stimulation in medication-resistant

depression using cerebral glucose metabolism. J Affect Disord 127: 219–29.

Li H-J, Cao X-H, Zhu X-T, Zhang A-X, Hou X-H, Xu Y, et al (2014b). Surface-based regional

homogeneity in first-episode, drug-naïve major depression: a resting-state FMRI study.

Biomed Res Int 2014: 374828.

Liang X, Wang J, Yan C, Shu N, Xu K, Gong G, et al (2012). Effects of different correlation

metrics and preprocessing factors on small-world brain functional networks: a resting-state

functional MRI study. PLoS One 7: e32766.

Likhtik E, Pelletier JG, Paz R, Paré D (2005). Prefrontal control of the amygdala. J Neurosci 25:

7429–37.

Lilenfeld LR, Kaye WH, Greeno CG, Merikangas KR, Plotnicov K, Pollice C, et al (1998). A

controlled family study of anorexia nervosa and bulimia nervosa: psychiatric disorders in

first-degree relatives and effects of proband comorbidity. Arch Gen Psychiatry 55: 603–10.

Lim K-L, Jacobs P, Ohinmaa A, Schopflocher D, Dewa CS (2008). A new population-based

measure of the economic burden of mental illness in Canada. Chronic Dis Can 28: 92–8.

Lindquist MA, Xu Y, Nebel MB, Caffo BS (2014). Evaluating dynamic bivariate correlations in

resting-state fMRI: a comparison study and a new approach. Neuroimage 101: 531–46.

336

Lippitz BE, Mindus P, Meyerson BA, Kihlström L, Lindquist C (1999). Lesion topography and

outcome after thermocapsulotomy or gamma knife capsulotomy for obsessive-compulsive

disorder: relevance of the right hemisphere. Neurosurgery 44: 452-8; discussion 458-60.

Lipsman N, Giacobbe P, Lozano AM (2013a). Deep brain stimulation in obsessive-compulsive

disorder: neurocircuitry and clinical experience. Handb Clin Neurol 116: 245–50.

Lipsman N, Lam E, Volpini M, Sutandar K, Twose R, Giacobbe P, et al (2017). Deep brain

stimulation of the subcallosal cingulate for treatment-refractory anorexia nervosa: 1 year

follow-up of an open-label trial. The lancet Psychiatry 4: 285–294.

Lipsman N, Woodside DB, Giacobbe P, Hamani C, Carter JC, Norwood SJ, et al (2013b).

Subcallosal cingulate deep brain stimulation for treatment-refractory anorexia nervosa: a

phase 1 pilot trial. Lancet (London, England) 381: 1361–70.

Lipsman N, Woodside DB, Giacobbe P, Lozano AM (2013c). Neurosurgical treatment of

anorexia nervosa: review of the literature from leucotomy to deep brain stimulation. Eur Eat


Liston C, Chen AC, Zebley BD, Drysdale AT, Gordon R, Leuchter B, et al (2014). Default mode

network mechanisms of transcranial magnetic stimulation in depression. Biol Psychiatry 76:

517–26.

Liu F, Guo W, Liu L, Long Z, Ma C, Xue Z, et al (2013). Abnormal amplitude low-frequency

oscillations in medication-naive, first-episode patients with major depressive disorder: a

resting-state fMRI study. J Affect Disord 146: 401–6.

Liu X, Hairston J, Schrier M, Fan J (2011). Common and distinct networks underlying reward

valence and processing stages: a meta-analysis of functional neuroimaging studies.

Neurosci Biobehav Rev 35: 1219–36.

Lock J, Garrett A, Beenhakker J, Reiss AL (2011). Aberrant brain activation during a response

inhibition task in adolescent eating disorder subtypes. Am J Psychiatry 168: 55–64.

Logothetis NK (2003). The underpinnings of the BOLD functional magnetic resonance imaging

signal. J Neurosci 23: 3963–71.

Logothetis NK, Pauls J, Augath M, Trinath T, Oeltermann A (2001). Neurophysiological

investigation of the basis of the fMRI signal. Nature 412: 150–7.

337

Long X-Y, Zuo X-N, Kiviniemi V, Yang Y, Zou Q-H, Zhu C-Z, et al (2008). Default mode

network as revealed with multiple methods for resting-state functional MRI analysis. J

Neurosci Methods 171: 349–55.

Long Z, Li R, Wen X, Jin Z, Chen K, Yao L (2013). Separating 4D multi-task fMRI data of

multiple subjects by independent component analysis with projection. Magn Reson Imaging

31: 60–74.

Lopatina N, McDannald MA, Styer C V, Peterson JF, Sadacca BF, Cheer JF, et al (2016).

Medial Orbitofrontal Neurons Preferentially Signal Cues Predicting Changes in Reward

during Unblocking. J Neurosci 36: 8416–24.

Lopizzo N, Bocchio Chiavetto L, Cattane N, Plazzotta G, Tarazi FI, Pariante CM, et al (2015).

Gene-environment interaction in major depression: focus on experience-dependent

biological systems. Front psychiatry 6: 68.

Lozano AM, Lipsman N (2013). Probing and regulating dysfunctional circuits using deep brain

stimulation. Neuron 77: 406–24.

Lozano AM, Mayberg HS, Giacobbe P, Hamani C, Craddock RC, Kennedy SH (2008).

Subcallosal cingulate gyrus deep brain stimulation for treatment-resistant depression. Biol


Lu H, Wang L, Rea WW, Brynildsen JK, Jaime S, Zuo Y, et al (2016). Low- but Not High-

Frequency LFP Correlates with Spontaneous BOLD Fluctuations in Rat Whisker Barrel

Cortex. Cereb Cortex 26: 683–694.

Lu H, Zou Q, Gu H, Raichle ME, Stein EA, Yang Y (2012). Rat brains also have a default mode

network. Proc Natl Acad Sci U S A 109: 3979–84.

Luborzewski A, Schubert F, Seifert F, Danker-Hopfe H, Brakemeier E-L, Schlattmann P, et al

(2007). Metabolic alterations in the dorsolateral prefrontal cortex after treatment with high-

frequency repetitive transcranial magnetic stimulation in patients with unipolar major

depression. J Psychiatr Res 41: 606–15.

Lucas AR, Beard CM, O’Fallon WM, Kurland LT (1991). 50-year trends in the incidence of

anorexia nervosa in Rochester, Minn.: a population-based study. Am J Psychiatry 148: 917–

22.

338

Luo Y, Huang X, Yang Z, Li B, Liu J, Wei D (2014). Regional homogeneity of intrinsic brain

activity in happy and unhappy individuals. PLoS One 9: e85181.

Lynd-Balta E, Haber SN (1994). Primate striatonigral projections: a comparison of the

sensorimotor-related striatum and the ventral striatum. J Comp Neurol 345: 562–78.

Lyons D, Friedman DP, Nader MA, Porrino LJ (1996). Cocaine alters cerebral metabolism

within the ventral striatum and limbic cortex of monkeys. J Neurosci 16: 1230–8.

Ma J, Zhang Z, Su Y, Kang L, Geng D, Wang Y, et al (2013). Magnetic stimulation modulates

structural synaptic plasticity and regulates BDNF–TrkB signal pathway in cultured

hippocampal neurons. Neurochem Int 62: 84–91.

Maarouf M, Neudorfer C, Majdoub F El, Lenartz D, Kuhn J, Sturm V (2016). Deep Brain

Stimulation of Medial Dorsal and Ventral Anterior Nucleus of the Thalamus in OCD: A

Retrospective Case Series. PLoS One 11: e0160750.

Maat SM de, Dekker J, Schoevers RA, Jonghe F de (2007). Relative efficacy of psychotherapy

and combined therapy in the treatment of depression: a meta-analysis. Eur Psychiatry 22:

1–8.

MacDonald AW, Cohen JD, Stenger VA, Carter CS (2000). Dissociating the role of the

dorsolateral prefrontal and anterior cingulate cortex in cognitive control. Science 288:

1835–8.

Maddocks SE, Kaplan AS, Woodside DB, Langdon L, Piran N (1992). Two year follow-up of

bulimia nervosa: The importance of abstinence as the criterion of outcome. Int J Eat Disord

12: 133–141.

Madhyastha TM, Askren MK, Boord P, Grabowski TJ (2015). Dynamic connectivity at rest

predicts attention task performance. Brain Connect 5: 45–59.

Maeda F, Keenan JP, Tormos JM, Topka H, Pascual-Leone A (2000). Modulation of

corticospinal excitability by repetitive transcranial magnetic stimulation. Clin Neurophysiol

111: 800–5.

Maillet D, Rajah MN (2014). Dissociable roles of default-mode regions during episodic

encoding. Neuroimage 89: 244–55.

Maintz JB, Viergever MA (1998). A survey of medical image registration. Med Image Anal 2: 1–

339

36.

Mallet L, Polosan M, Jaafari N, Baup N, Welter M-L, Fontaine D, et al (2008). Subthalamic

nucleus stimulation in severe obsessive-compulsive disorder. N Engl J Med 359: 2121–34.

Manelis A, Almeida JRC, Stiffler R, Lockovich JC, Aslam HA, Phillips ML (2016).

Anticipation-related brain connectivity in bipolar and unipolar depression: a graph theory

approach. Brain 139: 2554–66.

Manning JR, Jacobs J, Fried I, Kahana MJ (2009). Broadband shifts in local field potential

power spectra are correlated with single-neuron spiking in humans. J Neurosci 29: 13613–

20.

Mansur CG, Myczkowki ML, Barros Cabral S de, Sartorelli M do CB, Bellini BB, Dias AM, et

al (2011). Placebo effect after prefrontal magnetic stimulation in the treatment of resistant

obsessive-compulsive disorder: a randomized controlled trial. Int J Neuropsychopharmacol

14: 1389–97.

Mantini D, Perrucci MG, Gratta C Del, Romani GL, Corbetta M (2007). Electrophysiological

signatures of resting state networks in the human brain. Proc Natl Acad Sci U S A 104:

13170–5.

Mantovani A, Lisanby SH, Pieraccini F, Ulivelli M, Castrogiovanni P, Rossi S (2006).

Repetitive transcranial magnetic stimulation (rTMS) in the treatment of obsessive-

compulsive disorder (OCD) and Tourette’s syndrome (TS). Int J Neuropsychopharmacol 9:

95–100.

Mantovani A, Simpson HB, Fallon BA, Rossi S, Lisanby SH (2010). Randomized sham-

controlled trial of repetitive transcranial magnetic stimulation in treatment-resistant

obsessive-compulsive disorder. Int J Neuropsychopharmacol 13: 217–27.

Manwaring JL, Green L, Myerson J, Strube MJ, Wilfley DE (2011). Discounting of Various

types of rewards by women with and without binge eating Disorder: Evidence for general

rather than specific Differences. Psychol Rec 61: 561–582.

Margulies DS, Böttger J, Long X, Lv Y, Kelly C, Schäfer A, et al (2010). Resting developments:

a review of fMRI post-processing methodologies for spontaneous brain activity. MAGMA

23: 289–307.

340

Marsh R, Horga G, Wang Z, Wang P, Klahr KW, Berner LA, et al (2011). An FMRI study of

self-regulatory control and conflict resolution in adolescents with bulimia nervosa. Am J


Marsh R, Maia T V, Peterson BS (2009a). Functional disturbances within frontostriatal circuits

across multiple childhood psychopathologies. Am J Psychiatry 166: 664–74.

Marsh R, Steinglass JE, Gerber AJ, Graziano O’Leary K, Wang Z, Murphy D, et al (2009b).

Deficient activity in the neural systems that mediate self-regulatory control in bulimia

nervosa. Arch Gen Psychiatry 66: 51–63.

Martin JLR, Martín-Sánchez E (2012). Systematic review and meta-analysis of vagus nerve

stimulation in the treatment of depression: variable results based on study designs. Eur


Martino A Di, Scheres A, Margulies DS, Kelly AMC, Uddin LQ, Shehzad Z, et al (2008).

Functional connectivity of human striatum: a resting state FMRI study. Cereb Cortex 18:

2735–47.

Martuzzi R, Ramani R, Qiu M, Shen X, Papademetris X, Constable RT (2011). A whole-brain

voxel based measure of intrinsic connectivity contrast reveals local changes in tissue

connectivity with anesthetic without a priori assumptions on thresholds or regions of

interest. Neuroimage 58: 1044–50.

Mason MF, Norton MI, Horn JD Van, Wegner DM, Grafton ST, Macrae CN (2007). Wandering

minds: the default network and stimulus-independent thought. Science 315: 393–5.

Matsunaga H, Nagata T, Hayashida K, Ohya K, Kiriike N, Stein DJ (2009). A long-term trial of

the effectiveness and safety of atypical antipsychotic agents in augmenting SSRI-refractory

obsessive-compulsive disorder. J Clin Psychiatry 70: 863–8.

May A (2011). Experience-dependent structural plasticity in the adult human brain. Trends Cogn

Sci 15: 475–482.

May A, Hajak G, Gänßbauer S, Steffens T, Langguth B, Kleinjung T, et al (2007). Structural

Brain Alterations following 5 Days of Intervention: Dynamic Aspects of Neuroplasticity.

Cereb Cortex 17: 205–210.

Mayberg HS (2003). Modulating dysfunctional limbic-cortical circuits in depression: towards

341

development of brain-based algorithms for diagnosis and optimised treatment. Br Med Bull

65: 193–207.

Mayberg HS, Brannan SK, Mahurin RK, Jerabek PA, Brickman JS, Tekell JL, et al (1997).

Cingulate function in depression: a potential predictor of treatment response. Neuroreport 8:

1057–61.

Mayberg HS, Liotti M, Brannan SK, McGinnis S, Mahurin RK, Jerabek PA, et al (1999).

Reciprocal limbic-cortical function and negative mood: converging PET findings in

depression and normal sadness. Am J Psychiatry 156: 675–82.

Mayberg HS, Lozano AM, Voon V, McNeely HE, Seminowicz D, Hamani C, et al (2005). Deep

brain stimulation for treatment-resistant depression. Neuron 45: 651–60.

McAdams CJ, Jeon-Slaughter H, Evans S, Lohrenz T, Montague PR, Krawczyk DC (2016).

Neural differences in self-perception during illness and after weight-recovery in anorexia

nervosa. Soc Cogn Affect Neurosci 11: 1823–1831.

McCabe C, Cowen PJ, Harmer CJ (2009). Neural representation of reward in recovered

depressed patients. Psychopharmacology (Berl) 205: 667–77.

McClelland J, Bozhilova N, Nestler S, Campbell IC, Jacob S, Johnson-Sabine E, et al (2013).

Improvements in symptoms following neuronavigated repetitive transcranial magnetic

stimulation (rTMS) in severe and enduring anorexia nervosa: findings from two case

studies. Eur Eat Disord Rev 21: 500–6.

McClelland J, Kekic M, Bozhilova N, Nestler S, Dew T, Eynde F Van den, et al (2016). A

Randomised Controlled Trial of Neuronavigated Repetitive Transcranial Magnetic

Stimulation (rTMS) in Anorexia Nervosa. PLoS One 11: e0148606.

McClintock SM, Tirmizi O, Chansard M, Husain MM (2011). A systematic review of the

neurocognitive effects of magnetic seizure therapy. Int Rev Psychiatry 23: 413–23.

McCulloch WS, Pitts W (1943). A logical calculus of the ideas immanent in nervous activity.

Bull Math Biol 52: 99-115; discussion 73–97.

Mcdonald AJ, Mascagni F, Guo L (1996). Projections of the medial and lateral prefrontal

cortices to the amygdala: a Phaseolus vulgaris leucoagglutinin study in the rat.

Neuroscience 71: 55–75.

342

McDonald WM, Durkalski V, Ball ER, Holtzheimer PE, Pavlicova M, Lisanby SH, et al (2011).

Improving the antidepressant efficacy of transcranial magnetic stimulation: maximizing the

number of stimulations and treatment location in treatment-resistant depression. Depress

Anxiety 28: 973–980.

McFarland NR, Haber SN (2000). Convergent inputs from thalamic motor nuclei and frontal

cortical areas to the dorsal striatum in the primate. J Neurosci 20: 3798–813.

McGirr A, Eynde F Van den, Tovar-Perdomo S, Fleck MPA, Berlim MT (2015). Effectiveness

and acceptability of accelerated repetitive transcranial magnetic stimulation (rTMS) for

treatment-resistant major depressive disorder: an open label trial. J Affect Disord 173: 216–

20.

McKiernan KA, Kaufman JN, Kucera-Thompson J, Binder JR (2003). A parametric

manipulation of factors affecting task-induced deactivation in functional neuroimaging. J

Cogn Neurosci 15: 394–408.

McLean PD, Whittal ML, Thordarson DS, Taylor S, Söchting I, Koch WJ, et al (2001).

Cognitive versus behavior therapy in the group treatment of obsessive-compulsive disorder.

J Consult Clin Psychol 69: 205–14.

McTeague LM, Goodkind MS, Etkin A (2016). Transdiagnostic impairment of cognitive control

in mental illness. J Psychiatr Res 83: 37–46.

McTeague LM, Huemer J, Carreon DM, Jiang Y, Eickhoff SB, Etkin A (2017). Identification of

Common Neural Circuit Disruptions in Cognitive Control Across Psychiatric Disorders. Am

J Psychiatry 174: 676–685.

Meindl T, Teipel S, Elmouden R, Mueller S, Koch W, Dietrich O, et al (2010). Test-retest

reproducibility of the default-mode network in healthy individuals. Hum Brain Mapp 31:

237–46.

Menon V (2011). Large-scale brain networks and psychopathology: a unifying triple network

model. Trends Cogn Sci 15: 483–506.

Menon V, Uddin LQ (2010). Saliency, switching, attention and control: a network model of

insula function. Brain Struct Funct 214: 655–67.

Menzies L, Chamberlain SR, Laird AR, Thelen SM, Sahakian BJ, Bullmore ET (2008).

343

Integrating evidence from neuroimaging and neuropsychological studies of obsessive-

compulsive disorder: the orbitofronto-striatal model revisited. Neurosci Biobehav Rev 32:

525–49.

Mesulam MM, Mufson EJ (1982). Insula of the old world monkey. I. Architectonics in the

insulo-orbito-temporal component of the paralimbic brain. J Comp Neurol 212: 1–22.

Micallef-Trigona B (2014). Comparing the effects of repetitive transcranial magnetic stimulation

and electroconvulsive therapy in the treatment of depression: a systematic review and meta-

analysis. Depress Res Treat 2014: 135049.

Michael N, Gösling M, Reutemann M, Kersting A, Heindel W, Arolt V, et al (2003). Metabolic

changes after repetitive transcranial magnetic stimulation (rTMS) of the left prefrontal

cortex: A sham-controlled proton magnetic resonance spectroscopy (1H MRS) study of

healthy brain. Eur J Neurosci 17: 2462–2468.

Micoulaud-Franchi J-A, Richieri R, Cermolacce M, Loundou A, Lancon C, Vion-Dury J (2012).

Parieto-temporal alpha EEG band power at baseline as a predictor of antidepressant

treatment response with repetitive Transcranial Magnetic Stimulation: a preliminary study.

J Affect Disord 137: 156–60.

Mikl M, Marecek R, Hlustík P, Pavlicová M, Drastich A, Chlebus P, et al (2008). Effects of

spatial smoothing on fMRI group inferences. Magn Reson Imaging 26: 490–503.

Milev R V, Giacobbe P, Kennedy SH, Blumberger DM, Daskalakis ZJ, Downar J, et al (2016).

Canadian Network for Mood and Anxiety Treatments (CANMAT) 2016 Clinical Guidelines

for the Management of Adults with Major Depressive Disorder: Section 4.

Neurostimulation Treatments. Can J Psychiatry 61: 561–75.

Milner B (1982). Some cognitive effects of frontal-lobe lesions in man. Philos Trans R Soc Lond

B Biol Sci 298: 211–26.

Miniussi C, Bonato C, Bignotti S, Gazzoli A, Gennarelli M, Pasqualetti P, et al (2005).

Repetitive transcranial magnetic stimulation (rTMS) at high and low frequency: an

efficacious therapy for major drug-resistant depression? Clin Neurophysiol 116: 1062–71.

Mir-Moghtadaei A, Caballero R, Fried P, Fox MD, Lee K, Giacobbe P, et al (2015).

Concordance Between BeamF3 and MRI-neuronavigated Target Sites for Repetitive

344

Transcranial Magnetic Stimulation of the Left Dorsolateral Prefrontal Cortex. Brain Stimul

8: 965–73.

Mir-Moghtadaei A, Giacobbe P, Daskalakis ZJ, Blumberger DM, Downar J (2016). Validation

of a 25% Nasion-Inion Heuristic for Locating the Dorsomedial Prefrontal Cortex for

Repetitive Transcranial Magnetic Stimulation. Brain Stimul 9: 793–5.

Mischoulon D, Eddy KT, Keshaviah A, Dinescu D, Ross SL, Kass AE, et al (2011). Depression

and eating disorders: treatment and course. J Affect Disord 130: 470–7.

Mitchell JE, Agras S, Wonderlich S (2007). Treatment of bulimia nervosa: where are we and

where are we going? Int J Eat Disord 40: 95–101.

Mitchell JE, Davis L, Goff G (1985). The process of relapse in patients with bulimia. Int J Eat

Disord 4: 457–463.

Mitra A, Kraft A, Wright P, Acland B, Snyder AZ, Rosenthal Z, et al (2018). Spontaneous Infra-

slow Brain Activity Has Unique Spatiotemporal Dynamics and Laminar Structure. Neuron

98: 297–305.e6.

Mitzdorf U (1985). Current source-density method and application in cat cerebral cortex:

investigation of evoked potentials and EEG phenomena. Physiol Rev 65: 37–100.

Mix A, Benali A, Eysel UT, Funke K (2010). Continuous and intermittent transcranial magnetic

theta burst stimulation modify tactile learning performance and cortical protein expression

in the rat differently. Eur J Neurosci 32: 1575–1586.

Mix A, Hoppenrath K, Funke K (2015). Reduction in cortical parvalbumin expression due to

intermittent theta-burst stimulation correlates with maturation of the perineuronal nets in

young rats. Dev Neurobiol 75: 1–11.

Modirrousta M, Shams E, Katz C, Mansouri B, Moussavi Z, Sareen J, et al (2015). The efficacy

of deep repetitive transcranial magnetic stimulation over the medial prefrontal cortex in

obsessive compulsive disorder: results from an open-label study. Depress Anxiety 32: 445–

50.

Mole TB, Irvine MA, Worbe Y, Collins P, Mitchell SP, Bolton S, et al (2015). Impulsivity in

disorders of food and drug misuse. Psychol Med 45: 771–82.

Molenberghs P, Mesulam MM, Peeters R, Vandenberghe RRC (2007). Remapping attentional

345

priorities: differential contribution of superior parietal lobule and intraparietal sulcus. Cereb

Cortex 17: 2703–12.

Mondino M, Haesebaert F, Poulet E, Saoud M, Brunelin J (2015). Efficacy of Cathodal

Transcranial Direct Current Stimulation Over the Left Orbitofrontal Cortex in a Patient

With Treatment-Resistant Obsessive-Compulsive Disorder. J ECT 31: 271–2.

Montague PR, Berns GS (2002). Neural economics and the biological substrates of valuation.

Neuron 36: 265–84.

Montgomery SA, Gommoll CP, Chen C, Greenberg WM (2015). Efficacy of levomilnacipran

extended-release in major depressive disorder: pooled analysis of 5 double-blind, placebo-

controlled trials. CNS Spectr 20: 148–56.

Morecraft RJ, Stilwell-Morecraft KS, Cipolloni PB, Ge J, McNeal DW, Pandya DN (2012).

Cytoarchitecture and cortical connections of the anterior cingulate and adjacent

somatomotor fields in the rhesus monkey. Brain Res Bull 87: 457–97.

Morel A, Gallay MN, Baechler A, Wyss M, Gallay DS (2013). The human insula: Architectonic

organization and postmortem MRI registration. Neuroscience 236: 117–35.

Moritz S, Rufer M, Fricke S, Karow A, Morfeld M, Jelinek L, et al (2005). Quality of life in

obsessive-compulsive disorder before and after treatment. Compr Psychiatry 46: 453–9.

Morris R, Paxinos G, Petrides M (2000). Architectonic analysis of the human retrosplenial

cortex. J Comp Neurol 421: 14–28.

Morris R, Petrides M, Pandya DN (1999). Architecture and connections of retrosplenial area 30

in the rhesus monkey (Macaca mulatta). Eur J Neurosci 11: 2506–18.

Moss J, Bolam JP (2008). A dopaminergic axon lattice in the striatum and its relationship with

cortical and thalamic terminals. J Neurosci 28: 11221–30.

Mountcastle VB (1957). Modality and topographic properties of single neurons of cat’s somatic

sensory cortex. J Neurophysiol 20: 408–34.

Mourão-Miranda J, Bokde ALW, Born C, Hampel H, Stetter M (2005). Classifying brain states

and determining the discriminating activation patterns: Support Vector Machine on

functional MRI data. Neuroimage 28: 980–95.

Muakkassa KF, Strick PL (1979). Frontal lobe inputs to primate motor cortex: evidence for four

346

somatotopically organized “premotor” areas. Brain Res 177: 176–82.

Mufson EJ, Mesulam MM (1982). Insula of the old world monkey. II: Afferent cortical input and

comments on the claustrum. J Comp Neurol 212: 23–37.

Mukamel R, Gelbard H, Arieli A, Hasson U, Fried I, Malach R (2005). Coupling between

neuronal firing, field potentials, and FMRI in human auditory cortex. Science 309: 951–4.

Mulders AEP, Plantinga BR, Schruers K, Duits A, Janssen MLF, Ackermans L, et al (2016).

Deep brain stimulation of the subthalamic nucleus in obsessive-compulsive disorder:

Neuroanatomical and pathophysiological considerations. Eur Neuropsychopharmacol 26:

1909–1919.

Müller M (2000). Long-Term Repetitive Transcranial Magnetic Stimulation Increases the

Expression of Brain-Derived Neurotrophic Factor and Cholecystokinin mRNA, but not

Neuropeptide Tyrosine mRNA in Specific Areas of Rat Brain. Neuropsychopharmacology

23: 205–215.

Müller VI, Langner R, Cieslik EC, Rottschy C, Eickhoff SB (2015). Interindividual differences

in cognitive flexibility: influence of gray matter volume, functional connectivity and trait

impulsivity. Brain Struct Funct 220: 2401–14.

Mullins N, Power RA, Fisher HL, Hanscombe KB, Euesden J, Iniesta R, et al (2016). Polygenic

interactions with environmental adversity in the aetiology of major depressive disorder.


Murphy K, Birn RM, Handwerker DA, Jones TB, Bandettini PA (2009). The impact of global

signal regression on resting state correlations: are anti-correlated networks introduced?


Murphy K, Fox MD (2017). Towards a consensus regarding global signal regression for resting

state functional connectivity MRI. Neuroimage 154: 169–173.

Murray CJ, Lopez AD (1997). Global mortality, disability, and the contribution of risk factors:

Global Burden of Disease Study. Lancet (London, England) 349: 1436–42.

Murrough JW, Abdallah CG, Anticevic A, Collins KA, Geha P, Averill LA, et al (2016).

Reduced global functional connectivity of the medial prefrontal cortex in major depressive

disorder. Hum Brain Mapp 37: 3214–23.

347

Mutschler I, Wieckhorst B, Kowalevski S, Derix J, Wentlandt J, Schulze-Bonhage A, et al

(2009). Functional organization of the human anterior insular cortex. Neurosci Lett 457: 66–

70.

Nadeau SE, McCoy KJM, Crucian GP, Greer RA, Rossi F, Bowers D, et al (2002). Cerebral

blood flow changes in depressed patients after treatment with repetitive transcranial

magnetic stimulation: evidence of individual variability. Neuropsychiatry Neuropsychol

Behav Neurol 15: 159–75.

Naeser MA, Martin PI, Treglia E, Ho M, Kaplan E, Bashir S, et al (2010). Research with rTMS

in the treatment of aphasia. Restor Neurol Neurosci 28: 511–29.

Nakamae T, Sakai Y, Abe Y, Nishida S, Fukui K, Yamada K, et al (2014). Altered fronto-striatal

fiber topography and connectivity in obsessive-compulsive disorder. PLoS One 9: e112075.

Nakamura K, Groiss SJ, Hamada M, Enomoto H, Kadowaki S, Abe M, et al (2016). Variability

in Response to Quadripulse Stimulation of the Motor Cortex. Brain Stimul 9: 859–866.

Nakatani-Enomoto S, Hanajima R, Hamada M, Mochizuki H, Kobayashi S, Enomoto H, et al

(2011). Some evidence supporting the safety of quadripulse stimulation (QPS). Brain Stimul

4: 303–5.

Nakatani E, Nakagawa A, Nakao T, Yoshizato C, Nabeyama M, Kudo A, et al (2005). A

randomized controlled trial of Japanese patients with obsessive-compulsive disorder--

effectiveness of behavior therapy and fluvoxamine. Psychother Psychosom 74: 269–76.

Nambu A (2011). Somatotopic organization of the primate Basal Ganglia. Front Neuroanat 5:

26.

Narayanaswamy JC, Jose D, Chhabra H, Agarwal SM, Shrinivasa B, Hegde A, et al (2015).

Successful Application of Add-on Transcranial Direct Current Stimulation (tDCS) for

Treatment of SSRI Resistant OCD. Brain Stimul 8: 655–7.

Nauczyciel C, Jeune F Le, Naudet F, Douabin S, Esquevin A, Vérin M, et al (2014). Repetitive

transcranial magnetic stimulation over the orbitofrontal cortex for obsessive-compulsive

disorder: a double-blind, crossover study. Transl Psychiatry 4: e436.

Nekovarova T, Fajnerova I, Horacek J, Spaniel F (2014). Bridging disparate symptoms of

schizophrenia: a triple network dysfunction theory. Front Behav Neurosci 8: 171.

348

Nemeroff CB (2007). Prevalence and management of treatment-resistant depression. J Clin

Psychiatry 68 Suppl 8: 17–25.

Nemeroff CB, Mayberg HS, Krahl SE, McNamara J, Frazer A, Henry TR, et al (2006). VNS

therapy in treatment-resistant depression: clinical evidence and putative neurobiological

mechanisms. Neuropsychopharmacology 31: 1345–55.

Nestler EJ, Carlezon WA (2006). The mesolimbic dopamine reward circuit in depression. Biol


Nettekoven C, Volz LJ, Leimbach M, Pool E-M, Rehme AK, Eickhoff SB, et al (2015). Inter-

individual variability in cortical excitability and motor network connectivity following

multiple blocks of rTMS. Neuroimage 118: 209–18.

Newton AT, Morgan VL, Rogers BP, Gore JC (2011). Modulation of steady state functional

connectivity in the default mode and working memory networks by cognitive load. Hum

Brain Mapp 32: 1649–59.

Niazy RK, Xie J, Miller K, Beckmann CF, Smith SM (2011). Spectral characteristics of resting

state networks. Prog Brain Res 193: 259–76.

Nichols T, Hayasaka S (2003). Controlling the familywise error rate in functional neuroimaging:

a comparative review. Stat Methods Med Res 12: 419–46.

Nichols TE, Holmes AP (2002). Nonparametric permutation tests for functional neuroimaging: a

primer with examples. Hum Brain Mapp 15: 1–25.

Niendam TA, Laird AR, Ray KL, Dean YM, Glahn DC, Carter CS (2012). Meta-analytic

evidence for a superordinate cognitive control network subserving diverse executive

functions. Cogn Affect Behav Neurosci 12: 241–68.

Niessing J, Ebisch B, Schmidt KE, Niessing M, Singer W, Galuske RAW (2005). Hemodynamic

signals correlate tightly with synchronized gamma oscillations. Science 309: 948–51.

Nilsson K, Hägglöf B (2005). Long-term follow-up of adolescent onset anorexia nervosa in

northern Sweden. Eur Eat Disord Rev 13: 89–100.

Nir Y, Fisch L, Mukamel R, Gelbard-Sagiv H, Arieli A, Fried I, et al (2007). Coupling between

neuronal firing rate, gamma LFP, and BOLD fMRI is related to interneuronal correlations.

Curr Biol 17: 1275–85.

349

Nir Y, Mukamel R, Dinstein I, Privman E, Harel M, Fisch L, et al (2008). Interhemispheric

correlations of slow spontaneous neuronal fluctuations revealed in human sensory cortex.

Nat Neurosci 11: 1100–8.

Niso G, Carrasco S, Gudín M, Maestú F, Del-Pozo F, Pereda E (2015). What graph theory

actually tells us about resting state interictal MEG epileptic activity. NeuroImage Clin 8:

503–15.

Noda Y, Silverstein WK, Barr MS, Vila-Rodriguez F, Downar J, Rajji TK, et al (2015).

Neurobiological mechanisms of repetitive transcranial magnetic stimulation of the

dorsolateral prefrontal cortex in depression: a systematic review. Psychol Med 45: 3411–32.

Noda Y, Zomorrodi R, Saeki T, Rajji TK, Blumberger DM, Daskalakis ZJ, et al (2017). Resting-

state EEG gamma power and theta-gamma coupling enhancement following high-frequency

left dorsolateral prefrontal rTMS in patients with depression. Clin Neurophysiol 128: 424–

432.

Norman KA, Polyn SM, Detre GJ, Haxby J V (2006). Beyond mind-reading: multi-voxel pattern

analysis of fMRI data. Trends Cogn Sci 10: 424–30.

Novick D, Montgomery W, Aguado J, Kadziola Z, Peng X, Brugnoli R, et al (2013). Which

somatic symptoms are associated with an unfavorable course in Asian patients with major

depressive disorder? J Affect Disord 149: 182–8.

Nuttin B, Cosyns P, Demeulemeester H, Gybels J, Meyerson B (1999). Electrical stimulation in

anterior limbs of internal capsules in patients with obsessive-compulsive disorder. Lancet

(London, England) 354: 1526.

Nuttin BJ, Gabriëls LA, Cosyns PR, Meyerson BA, Andréewitch S, Sunaert SG, et al (2003).

Long-term electrical capsular stimulation in patients with obsessive-compulsive disorder.

Neurosurgery 52: 1263-72; discussion 1272–4.

O’Doherty J, Kringelbach ML, Rolls ET, Hornak J, Andrews C (2001). Abstract reward and

punishment representations in the human orbitofrontal cortex. Nat Neurosci 4: 95–102.

O’Donovan C (2004). Achieving and sustaining remission in depression and anxiety disorders:

introduction. Can J Psychiatry 49: 5S–9S.

O’Neill J, Gorbis E, Feusner JD, Yip JC, Chang S, Maidment KM, et al (2013). Effects of

350

intensive cognitive-behavioral therapy on cingulate neurochemistry in obsessive-

compulsive disorder. J Psychiatr Res 47: 494–504.

O’Reardon JP, Solvason HB, Janicak PG, Sampson S, Isenberg KE, Nahas Z, et al (2007).

Efficacy and Safety of Transcranial Magnetic Stimulation in the Acute Treatment of Major

Depression: A Multisite Randomized Controlled Trial. Biol Psychiatry 62: 1208–1216.

Oberndorfer T, Simmons A, McCurdy D, Strigo I, Matthews S, Yang T, et al (2013a). Greater

anterior insula activation during anticipation of food images in women recovered from

anorexia nervosa versus controls. Psychiatry Res 214: 132–41.

Oberndorfer TA, Frank GKW, Simmons AN, Wagner A, McCurdy D, Fudge JL, et al (2013b).

Altered insula response to sweet taste processing after recovery from anorexia and bulimia

nervosa. Am J Psychiatry 170: 1143–51.

Oberndorfer TA, Kaye WH, Simmons AN, Strigo IA, Matthews SC (2011). Demand-specific

alteration of medial prefrontal cortex response during an inhibition task in recovered

anorexic women. Int J Eat Disord 44: 1–8.

Obeso I, Cho SS, Antonelli F, Houle S, Jahanshahi M, Ko JH, et al (2013). Stimulation of the

pre-SMA influences cerebral blood flow in frontal areas involved with inhibitory control of

action. Brain Stimul 6: 769–76.

Ogawa S, Lee TM, Kay AR, Tank DW (1990). Brain magnetic resonance imaging with contrast

dependent on blood oxygenation. Proc Natl Acad Sci U S A 87: 9868–72.

Ogiue-Ikeda M, Kawato S, Ueno S (2005). Acquisition of ischemic tolerance by repetitive

transcranial magnetic stimulation in the rat hippocampus. Brain Res 1037: 7–11.

Okada K, Nakao T, Sanematsu H, Murayama K, Honda S, Tomita M, et al (2015). Biological

heterogeneity of obsessive-compulsive disorder: A voxel-based morphometric study based

on dimensional assessment. Psychiatry Clin Neurosci 69: 411–21.

Olmsted MP, Kaplan AS, Rockert W (1994). Rate and prediction of relapse in bulimia nervosa.

Am J Psychiatry 151: 738–43.

Olmsted MP, Kaplan AS, Rockert W (2005). Defining remission and relapse in bulimia nervosa.


Olmsted MP, MacDonald DE, McFarlane T, Trottier K, Colton P (2015). Predictors of rapid

351

relapse in bulimia nervosa. Int J Eat Disord 48: 337–40.

Olmsted MP, McFarlane TL, Carter JC, Trottier K, Woodside DB, Dimitropoulos G (2010).

Inpatient and day hospital treatment for anorexia nervosa. Treat Eat Disord A Clin Handb

198–211.

Ongür D, Ferry AT, Price JL (2003). Architectonic subdivision of the human orbital and medial

prefrontal cortex. J Comp Neurol 460: 425–49.

Ooms P, Mantione M, Figee M, Schuurman PR, Munckhof P van den, Denys D (2014). Deep

brain stimulation for obsessive-compulsive disorders: long-term analysis of quality of life. J

Neurol Neurosurg Psychiatry 85: 153–8.

Oppen P van, Haan E de, Balkom AJ van, Spinhoven P, Hoogduin K, Dyck R van (1995).

Cognitive therapy and exposure in vivo in the treatment of obsessive compulsive disorder.

Behav Res Ther 33: 379–90.

Owen AM, Downes JJ, Sahakian BJ, Polkey CE, Robbins TW (1990). Planning and spatial

working memory following frontal lobe lesions in man. Neuropsychologia 28: 1021–34.

Ozekes S, Erguzel T, Sayar GH, Tarhan N (2014). Analysis of Brain Functional Changes in

High-Frequency Repetitive Transcranial Magnetic Stimulation in Treatment-Resistant

Depression. Clin EEG Neurosci 45: 257–261.

Padberg F, Zwanzger P, Thoma H, Kathmann N, Haag C, Greenberg BD, et al (1999). Repetitive

transcranial magnetic stimulation (rTMS) in pharmacotherapy-refractory major depression:

comparative study of fast, slow and sham rTMS. Psychiatry Res 88: 163–71.

Palaniyappan L, Marques TR, Taylor H, Mondelli V, Reinders AATS, Bonaccorso S, et al

(2016). Globally Efficient Brain Organization and Treatment Response in Psychosis: A

Connectomic Study of Gyrification. Schizophr Bull 42: 1446–1456.

Pallanti S, Cantisani A, Grassi G, Antonini S, Cecchelli C, Burian J, et al (2012). rTMS age-

dependent response in treatment-resistant depressed subjects: a mini-review. CNS Spectr

17: 24–30.

Pallanti S, Hollander E, Bienstock C, Koran L, Leckman J, Marazziti D, et al (2002). Treatment

non-response in OCD: methodological issues and operational definitions. Int J


352

Pan J, Zhan L, Hu C, Yang J, Wang C, Gu L, et al (2018). Emotion Regulation and Complex

Brain Networks: Association Between Expressive Suppression and Efficiency in the Fronto-

Parietal Network and Default-Mode Network. Front Hum Neurosci 12: 70.

Pan W-J, Thompson GJ, Magnuson ME, Jaeger D, Keilholz S (2013). Infraslow LFP correlates

to resting-state fMRI BOLD signals. Neuroimage 74: 288–97.

Pandya DN, Dye P, Butters N (1971). Efferent cortico-cortical projections of the prefrontal

cortex in the rhesus monkey. Brain Res 31: 35–46.

Pandya DN, Hoesen GW Van, Mesulam MM (1981). Efferent connections of the cingulate gyrus

in the rhesus monkey. Exp brain Res 42: 319–30.

Pandya DN, Kuypers HG (1969). Cortico-cortical connections in the rhesus monkey. Brain Res

13: 13–36.

Papakostas GI, Ionescu DF (2015). Towards new mechanisms: an update on therapeutics for

treatment-resistant major depressive disorder. Mol Psychiatry 20: 1142–50.

Parent A, Hazrati LN (1995). Functional anatomy of the basal ganglia. I. The cortico-basal

ganglia-thalamo-cortical loop. Brain Res Brain Res Rev 20: 91–127.

Parigi A Del, Gautier J-F, Chen K, Salbe AD, Ravussin E, Reiman E, et al (2002).

Neuroimaging and obesity: mapping the brain responses to hunger and satiation in humans

using positron emission tomography. Ann N Y Acad Sci 967: 389–97.

Parikh S V, Quilty LC, Ravitz P, Rosenbluth M, Pavlova B, Grigoriadis S, et al (2016).

Canadian Network for Mood and Anxiety Treatments (CANMAT) 2016 Clinical Guidelines

for the Management of Adults with Major Depressive Disorder: Section 2. Psychological

Treatments. Can J Psychiatry 61: 524–39.

Parvizi J, Hoesen GW Van, Buckwalter J, Damasio A (2006). Neural connections of the

posteromedial cortex in the macaque. Proc Natl Acad Sci U S A 103: 1563–8.

Parvizi J, Rangarajan V, Shirer WR, Desai N, Greicius MD (2013). The will to persevere

induced by electrical stimulation of the human cingulate gyrus. Neuron 80: 1359–67.

Pascual-Leone A, Catalá MD, Pascual-Leone Pascual A (1996a). Lateralized effect of rapid-rate

transcranial magnetic stimulation of the prefrontal cortex on mood. Neurology 46: 499–502.

Pascual-Leone A, Rubio B, Pallardó F, Catalá MD (1996b). Rapid-rate transcranial magnetic

353

stimulation of left dorsolateral prefrontal cortex in drug-resistant depression. Lancet

(London, England) 348: 233–7.

Pascual-Leone A, Valls-Solé J, Wassermann EM, Hallett M (1994). Responses to rapid-rate

transcranial magnetic stimulation of the human motor cortex. Brain 117 ( Pt 4: 847–58.

Pasquale F de, Penna S Della, Snyder AZ, Lewis C, Mantini D, Marzetti L, et al (2010).

Temporal dynamics of spontaneous MEG activity in brain networks. Proc Natl Acad Sci U

S A 107: 6040–5.

Pasquale F de, Penna S Della, Snyder AZ, Marzetti L, Pizzella V, Romani GL, et al (2012). A

cortical core for dynamic integration of functional networks in the resting human brain.

Neuron 74: 753–64.

Patel AX, Kundu P, Rubinov M, Jones PS, Vértes PE, Ersche KD, et al (2014). A wavelet

method for modeling and despiking motion artifacts from resting-state fMRI time series.


Patel MJ, Khalaf A, Aizenstein HJ (2016). Studying depression using imaging and machine

learning methods. NeuroImage Clin 10: 115–23.

Patel SR, Galfavy H, Kimeldorf MB, Dixon LB, Simpson HB (2017). Patient Preferences and

Acceptability of Evidence-Based and Novel Treatments for Obsessive-Compulsive

Disorder. Psychiatr Serv 68: 250–257.

Patriat R, Molloy EK, Meier TB, Kirk GR, Nair VA, Meyerand ME, et al (2013). The effect of

resting condition on resting-state fMRI reliability and consistency: a comparison between

resting with eyes open, closed, and fixated. Neuroimage 78: 463–73.

Patten SB, Williams JVA, Lavorato DH, Wang JL, McDonald K, Bulloch AGM (2015).

Descriptive epidemiology of major depressive disorder in Canada in 2012. Can J Psychiatry

60: 23–30.

Patton JH, Stanford MS, Barratt ES (1995). Factor structure of the barratt impulsiveness scale. J

Clin Psychol 51: 768–774.

Pauli WM, O’Reilly RC, Yarkoni T, Wager TD (2016). Regional specialization within the

human striatum for diverse psychological functions. Proc Natl Acad Sci U S A 113: 1907–

12.

354

Pauling L, Coryell CD (1936). The Magnetic Properties and Structure of Hemoglobin,

Oxyhemoglobin and Carbonmonoxyhemoglobin. Proc Natl Acad Sci U S A 22: 210–6.

Pauls DL (2010). The genetics of obsessive-compulsive disorder: a review. Dialogues Clin


Pawela CP, Biswal BB, Cho YR, Kao DS, Li R, Jones SR, et al (2008). Resting-state functional

connectivity of the rat brain. Magn Reson Med 59: 1021–9.

Pearson C, Janz T, Ali J (2013). Mental and substance use disorders in Canada. Heal a Glance

doi:82-624-X.

Peeters M, Vollebergh WAM, Wiers RW, Field M (2014). Psychological changes and cognitive

impairments in adolescent heavy drinkers. Alcohol Alcohol 49: 182–6.

Pellicciari MC, Ponzo V, Caltagirone C, Koch G (2017). Restored Asymmetry of Prefrontal

Cortical Oscillatory Activity after Bilateral Theta Burst Stimulation Treatment in a Patient

with Major Depressive Disorder: A TMS-EEG Study. Brain Stimul 10: 147–149.

Peng H, Zheng H, Li L, Liu J, Zhang Y, Shan B, et al (2012). High-frequency rTMS treatment

increases white matter FA in the left middle frontal gyrus in young patients with treatment-

resistant depression. J Affect Disord 136: 249–57.

Perani D, Colombo C, Bressi S, Bonfanti A, Grassi F, Scarone S, et al (1995). [18F]FDG PET

study in obsessive-compulsive disorder. A clinical/metabolic correlation study after

treatment. Br J Psychiatry 166: 244–50.

Perlbarg V, Bellec P, Anton J-L, Pélégrini-Issac M, Doyon J, Benali H (2007). CORSICA:

correction of structured noise in fMRI by automatic identification of ICA components.

Magn Reson Imaging 25: 35–46.

Petrides M (2005). Lateral prefrontal cortex: architectonic and functional organization. Philos

Trans R Soc Lond B Biol Sci 360: 781–95.

Pettersson A, Boström KB, Gustavsson P, Ekselius L (2015). Which instruments to support

diagnosis of depression have sufficient accuracy? A systematic review. Nord J Psychiatry

69: 497–508.

Petzold GC, Albeanu DF, Sato TF, Murthy VN (2008). Coupling of neural activity to blood flow

in olfactory glomeruli is mediated by astrocytic pathways. Neuron 58: 897–910.

355

Philip NS, Barredo J, ’t Wout-Frank M van, Tyrka AR, Price LH, Carpenter LL (2018). Network

Mechanisms of Clinical Response to Transcranial Magnetic Stimulation in Posttraumatic

Stress Disorder and Major Depressive Disorder. Biol Psychiatry 83: 263–272.

Philippi CL, Motzkin JC, Pujara MS, Koenigs M (2015). Subclinical depression severity is

associated with distinct patterns of functional connectivity for subregions of anterior

cingulate cortex. J Psychiatr Res 71: 103–11.

Picard N, Strick PL (1996). Motor areas of the medial wall: a review of their location and

functional activation. Cereb Cortex 6: 342–53.

Pike KM (1998). Long-term course of anorexia nervosa: response, relapse, remission, and

recovery. Clin Psychol Rev 18: 447–75.

Pinheiro AP, Root T, Bulik CM (2009). The Genetics of Anorexia Nervosa: Current Findings

and Future Perspectives. Int J Child Adolesc health 2: 153–164.

Piras F, Piras F, Chiapponi C, Girardi P, Caltagirone C, Spalletta G (2015). Widespread

structural brain changes in OCD: a systematic review of voxel-based morphometry studies.

Cortex 62: 89–108.

Pizzagalli D, Pascual-Marqui RD, Nitschke JB, Oakes TR, Larson CL, Abercrombie HC, et al

(2001). Anterior cingulate activity as a predictor of degree of treatment response in major

depression: evidence from brain electrical tomography analysis. Am J Psychiatry 158: 405–

15.

Pizzagalli DA, Holmes AJ, Dillon DG, Goetz EL, Birk JL, Bogdan R, et al (2009). Reduced

caudate and nucleus accumbens response to rewards in unmedicated individuals with major

depressive disorder. Am J Psychiatry 166: 702–10.

Pleger B, Ruff CC, Blankenburg F, Bestmann S, Wiech K, Stephan KE, et al (2006). Neural

coding of tactile decisions in the human prefrontal cortex. J Neurosci 26: 12596–601.

Plewnia C, Pasqualetti P, Große S, Schlipf S, Wasserka B, Zwissler B, et al (2014). Treatment of

major depression with bilateral theta burst stimulation: a randomized controlled pilot trial. J


Ploran EJ, Nelson SM, Velanova K, Donaldson DI, Petersen SE, Wheeler ME (2007). Evidence

accumulation and the moment of recognition: dissociating perceptual recognition processes

356

using fMRI. J Neurosci 27: 11912–24.

Pogarell O, Koch W, Pöpperl G, Tatsch K, Jakob F, Mulert C, et al (2007). Acute prefrontal

rTMS increases striatal dopamine to a similar degree as d-amphetamine. Psychiatry Res -

Neuroimaging 156: 251–255.

Pogarell O, Koch W, Pöpperl G, Tatsch K, Jakob F, Zwanzger P, et al (2006). Striatal dopamine

release after prefrontal repetitive transcranial magnetic stimulation in major depression:

Preliminary results of a dynamic [123I] IBZM SPECT study. J Psychiatr Res 40: 307–314.

Poldrack RA, Mumford JA, Nichols TE (2011). Handbook of functional MRI data analysis.

Cambridge Univ Press 4: .

Polnay A, James VAW, Hodges L, Murray GD, Munro C, Lawrie SM (2014). Group therapy for

people with bulimia nervosa: systematic review and meta-analysis. Psychol Med 44: 2241–

54.

Popa D, Popescu AT, Paré D (2009). Contrasting activity profile of two distributed cortical

networks as a function of attentional demands. J Neurosci 29: 1191–201.

Posner J, Hellerstein DJ, Gat I, Mechling A, Klahr K, Wang Z, et al (2013). Antidepressants

normalize the default mode network in patients with dysthymia. JAMA psychiatry 70: 373–

82.

Posner J, Marsh R, Maia T V, Peterson BS, Gruber A, Simpson HB (2014). Reduced functional

connectivity within the limbic cortico-striato-thalamo-cortical loop in unmedicated adults

with obsessive-compulsive disorder. Hum Brain Mapp 35: 2852–60.

Posner MI, Petersen SE (1990). The attention system of the human brain. Annu Rev Neurosci 13:

25–42.

Power JD, Barnes KA, Snyder AZ, Schlaggar BL, Petersen SE (2012). Spurious but systematic

correlations in functional connectivity MRI networks arise from subject motion.


Power JD, Cohen AL, Nelson SM, Wig GS, Barnes KA, Church JA, et al (2011). Functional

network organization of the human brain. Neuron 72: 665–78.

Power JD, Mitra A, Laumann TO, Snyder AZ, Schlaggar BL, Petersen SE (2014). Methods to

detect, characterize, and remove motion artifact in resting state fMRI. Neuroimage 84: 320–

357

41.

Power JD, Plitt M, Laumann TO, Martin A (2017). Sources and implications of whole-brain

fMRI signals in humans. Neuroimage 146: 609–625.

Prasko J, Pasková B, Záleský R, Novák T, Kopecek M, Bares M, et al (2006). The effect of

repetitive transcranial magnetic stimulation (rTMS) on symptoms in obsessive compulsive

disorder. A randomized, double blind, sham controlled study. Neuro Endocrinol Lett 27:

327–32.

Prasser J, Schecklmann M, Poeppl TB, Frank E, Kreuzer PM, Hajak G, et al (2015). Bilateral

prefrontal rTMS and theta burst TMS as an add-on treatment for depression: a randomized

placebo controlled trial. World J Biol Psychiatry 16: 57–65.

Preti MG, Bolton TA, Ville D Van De (2017). The dynamic functional connectome: State-of-the-

art and perspectives. Neuroimage 160: 41–54.

Preuschoff K, Bossaerts P, Quartz SR (2006). Neural differentiation of expected reward and risk

in human subcortical structures. Neuron 51: 381–90.

Preuschoff K, Quartz SR, Bossaerts P (2008). Human insula activation reflects risk prediction

errors as well as risk. J Neurosci 28: 2745–52.

Preuss TM, Goldman-Rakic PS (1989). Connections of the ventral granular frontal cortex of

macaques with perisylvian premotor and somatosensory areas: anatomical evidence for

somatic representation in primate frontal association cortex. J Comp Neurol 282: 293–316.

Price JL (2007). Definition of the orbital cortex in relation to specific connections with limbic

and visceral structures and other cortical regions. Ann N Y Acad Sci 1121: 54–71.

Pruim RHR, Mennes M, Rooij D van, Llera A, Buitelaar JK, Beckmann CF (2015). ICA-

AROMA: A robust ICA-based strategy for removing motion artifacts from fMRI data.


Pyle RL, Mitchell JE, Eckert ED, Hatsukami D, Pomeroy C, Zimmerman R (1990). Maintenance

treatment and 6-month outcome for bulimic patients who respond to initial treatment. Am J


Qiu L, Fu X, Wang S, Tang Q, Chen X, Cheng L, et al (2017). Abnormal regional spontaneous

neuronal activity associated with symptom severity in treatment-naive patients with

358

obsessive-compulsive disorder revealed by resting-state functional MRI. Neurosci Lett 640:

99–104.

Quirk GJ, Likhtik E, Pelletier JG, Paré D (2003). Stimulation of medial prefrontal cortex

decreases the responsiveness of central amygdala output neurons. J Neurosci 23: 8800–7.

Racine SE, Martin SJ (2017). Integrating eating disorder-specific risk factors into the acquired

preparedness model of dysregulated eating: A moderated mediation analysis. Eat Behav 24:

54–60.

Radeloff D, Willmann K, Otto L, Lindner M, Putnam K, Leeuwen S Van, et al (2014). High-fat

taste challenge reveals altered striatal response in women recovered from bulimia nervosa:

A pilot study. World J Biol Psychiatry 15: 307–16.

Radua J, Heuvel OA van den, Surguladze S, Mataix-Cols D (2010). Meta-analytical comparison

of voxel-based morphometry studies in obsessive-compulsive disorder vs other anxiety

disorders. Arch Gen Psychiatry 67: 701–11.

Raichle ME (2009a). A paradigm shift in functional brain imaging. J Neurosci 29: 12729–34.

Raichle ME (2009b). A brief history of human brain mapping. Trends Neurosci 32: 118–26.

Raichle ME (2010). Two views of brain function. Trends Cogn Sci 14: 180–90.

Raichle ME, Gusnard DA (2002). Appraising the brain’s energy budget. Proc Natl Acad Sci U S

A 99: 10237–9.

Raichle ME, MacLeod AM, Snyder AZ, Powers WJ, Gusnard DA, Shulman GL (2001). A

default mode of brain function. Proc Natl Acad Sci U S A 98: 676–82.

Raichle ME, Mintun MA (2006). Brain work and brain imaging. Annu Rev Neurosci 29: 449–76.

Ramirez-Mahaluf JP, Perramon J, Otal B, Villoslada P, Compte A (2018). Subgenual anterior

cingulate cortex controls sadness-induced modulations of cognitive and emotional network

hubs. Sci Rep 8: 8566.

Ratnasingham S, Cairney J, Manson H, Rehm J, Lin E, Kurdyak P (2013). The burden of mental

illness and addiction in ontario. Can J Psychiatry 58: 529–37.

Rauch SL, Dougherty DD, Malone D, Rezai A, Friehs G, Fischman AJ, et al (2006). A

functional neuroimaging investigation of deep brain stimulation in patients with obsessive-

359

compulsive disorder. J Neurosurg 104: 558–65.

Rav-Acha M, Sagiv N, Segev I, Bergman H, Yarom Y (2005). Dynamic and spatial features of

the inhibitory pallidal GABAergic synapses. Neuroscience 135: 791–802.

Ravizza L, Maina G, Bogetto F (1997). Episodic and chronic obsessive-compulsive disorder.

Depress Anxiety 6: 154–8.

Ray S, Maunsell JHR (2011). Different origins of gamma rhythm and high-gamma activity in

macaque visual cortex. PLoS Biol 9: e1000610.

Razza LB, Moffa AH, Moreno ML, Carvalho AF, Padberg F, Fregni F, et al (2018). A

systematic review and meta-analysis on placebo response to repetitive transcranial magnetic

stimulation for depression trials. Prog Neuropsychopharmacol Biol Psychiatry 81: 105–

113.

Reess TJ, Rus OG, Schmidt R, Reus MA de, Zaudig M, Wagner G, et al (2016). Connectomics-

based structural network alterations in obsessive-compulsive disorder. Transl Psychiatry 6:

e882.

Reijneveld JC, Ponten SC, Berendse HW, Stam CJ (2007). The application of graph theoretical

analysis to complex networks in the brain. Clin Neurophysiol 118: 2317–31.

Rhee SH, Lahey BB, Waldman ID (2015). Comorbidity Among Dimensions of Childhood

Psychopathology: Converging Evidence from Behavior Genetics. Child Dev Perspect 9:

26–31.

Richard JM, Berridge KC (2013). Prefrontal cortex modulates desire and dread generated by

nucleus accumbens glutamate disruption. Biol Psychiatry 73: 360–70.

Richieri R, Boyer L, Padovani R, Adida M, Colavolpe C, Mundler O, et al (2012). Equivalent

brain SPECT perfusion changes underlying therapeutic efficiency in pharmacoresistant

depression using either high-frequency left or low-frequency right prefrontal rTMS. Prog

Neuropsychopharmacol Biol Psychiatry 39: 364–70.

Ridderinkhof KR, Ullsperger M, Crone EA, Nieuwenhuis S (2004a). The role of the medial

frontal cortex in cognitive control. Science 306: 443–7.

Ridderinkhof KR, Wildenberg WPM van den, Segalowitz SJ, Carter CS (2004b). Neurocognitive

mechanisms of cognitive control: the role of prefrontal cortex in action selection, response

360

inhibition, performance monitoring, and reward-based learning. Brain Cogn 56: 129–40.

Riedel M, Möller H-J, Obermeier M, Schennach-Wolff R, Bauer M, Adli M, et al (2010).

Response and remission criteria in major depression--a validation of current practice. J

Psychiatr Res 44: 1063–8.

Rigaud D, Pennacchio H, Bizeul C, Reveillard V, Vergès B (2011). Outcome in AN adult

patients: a 13-year follow-up in 484 patients. Diabetes Metab 37: 305–11.

Rilling JK, Barks SK, Parr LA, Preuss TM, Faber TL, Pagnoni G, et al (2007). A comparison of

resting-state brain activity in humans and chimpanzees. Proc Natl Acad Sci U S A 104:

17146–51.

Ritchey M, Dolcos F, Eddington KM, Strauman TJ, Cabeza R (2011). Neural correlates of

emotional processing in depression: changes with cognitive behavioral therapy and

predictors of treatment response. J Psychiatr Res 45: 577–87.

Rizvi SL, Peterson CB, Crow SJ, Agras WS (2000). Test-retest reliability of the eating disorder

examination. Int J Eat Disord 28: 311–6.

Roberson AM, Castro VM, Cagan A, Perlis RH (2016). Antidepressant nonadherence in routine

clinical settings determined from discarded blood samples. J Clin Psychiatry 77: 359–62.

Robinson OJ, Cools R, Carlisi CO, Sahakian BJ, Drevets WC (2012). Ventral striatum response

during reward and punishment reversal learning in unmedicated major depressive disorder.

Am J Psychiatry 169: 152–9.

Robinson RG, Kubos KL, Starr LB, Rao K, Price TR (1984). Mood disorders in stroke patients.

Importance of location of lesion. Brain 107 ( Pt 1: 81–93.

Rosen JC, Vara L, Wendt S, Leitenberg H (1990). Validity studies of the eating disorder

examination. Int J Eat Disord 9: 519–528.

Rosenblat JD, Kakar R, McIntyre RS (2015). The Cognitive Effects of Antidepressants in Major

Depressive Disorder: A Systematic Review and Meta-Analysis of Randomized Clinical

Trials. Int J Neuropsychopharmacol 19: .

Rosenkranz JA, Moore H, Grace AA (2003). The prefrontal cortex regulates lateral amygdala

neuronal plasticity and responses to previously conditioned stimuli. J Neurosci 23: 11054–

64.

361

Rotge J-Y, Langbour N, Guehl D, Bioulac B, Jaafari N, Allard M, et al (2010). Gray matter

alterations in obsessive-compulsive disorder: an anatomic likelihood estimation meta-

analysis. Neuropsychopharmacology 35: 686–91.

Royce GJ (1982). Laminar origin of cortical neurons which project upon the caudate nucleus: a

horseradish peroxidase investigation in the cat. J Comp Neurol 205: 8–29.

Rubio JM, Markowitz JC, Alegría A, Pérez-Fuentes G, Liu S-M, Lin K-H, et al (2011).

Epidemiology of chronic and nonchronic major depressive disorder: results from the

national epidemiologic survey on alcohol and related conditions. Depress Anxiety 28: 622–

31.

Rudebeck PH, Mitz AR, Chacko R V, Murray EA (2013). Effects of amygdala lesions on

reward-value coding in orbital and medial prefrontal cortex. Neuron 80: 1519–31.

Ruf SP, Fallgatter AJ, Plewnia C (2017). Augmentation of working memory training by

transcranial direct current stimulation (tDCS). Sci Rep 7: 876.

Rufer M, Hand I, Alsleben H, Braatz A, Ortmann J, Katenkamp B, et al (2005). Long-term

course and outcome of obsessive-compulsive patients after cognitive-behavioral therapy in

combination with either fluvoxamine or placebo: a 7-year follow-up of a randomized

double-blind trial. Eur Arch Psychiatry Clin Neurosci 255: 121–8.

Ruff CC, Driver J, Bestmann S (2009). Combining TMS and fMRI: from “virtual lesions” to

functional-network accounts of cognition. Cortex 45: 1043–9.

Ruffini C, Locatelli M, Lucca A, Benedetti F, Insacco C, Smeraldi E (2009). Augmentation

effect of repetitive transcranial magnetic stimulation over the orbitofrontal cortex in drug-

resistant obsessive-compulsive disorder patients: a controlled investigation. Prim Care

Companion J Clin Psychiatry 11: 226–30.

Ruscio AM, Stein DJ, Chiu WT, Kessler RC (2010). The epidemiology of obsessive-compulsive

disorder in the National Comorbidity Survey Replication. Mol Psychiatry 15: 53–63.

Rush AJ, Marangell LB, Sackeim HA, George MS, Brannan SK, Davis SM, et al (2005). Vagus

nerve stimulation for treatment-resistant depression: a randomized, controlled acute phase

trial. Biol Psychiatry 58: 347–54.

Rush AJ, Trivedi MH, Ibrahim HM, Carmody TJ, Arnow B, Klein DN, et al (2003). The 16-Item

362

Quick Inventory of Depressive Symptomatology (QIDS), clinician rating (QIDS-C), and

self-report (QIDS-SR): a psychometric evaluation in patients with chronic major

depression. Biol Psychiatry 54: 573–83.

Rush AJ, Trivedi MH, Wisniewski SR, Nierenberg AA, Stewart JW, Warden D, et al (2006).

Acute and longer-term outcomes in depressed outpatients requiring one or several treatment

steps: a STAR*D report. Am J Psychiatry 163: 1905–17.

Rushworth MFS, Behrens TEJ, Rudebeck PH, Walton ME (2007). Contrasting roles for

cingulate and orbitofrontal cortex in decisions and social behaviour. Trends Cogn Sci 11:

168–76.

Rushworth MFS, Walton ME, Kennerley SW, Bannerman DM (2004). Action sets and decisions

in the medial frontal cortex. Trends Cogn Sci 8: 410–7.

Rutherford G, Lithgow B, Moussavi Z (2015). Short and Long-term Effects of rTMS Treatment

on Alzheimer’s Disease at Different Stages: A Pilot Study. J Exp Neurosci 9: 43–51.

Rzepa E, McCabe C (2016). Decreased anticipated pleasure correlates with increased salience

network resting state functional connectivity in adolescents with depressive

symptomatology. J Psychiatr Res 82: 40–7.

Saad ZS, Gotts SJ, Murphy K, Chen G, Jo HJ, Martin A, et al (2012). Trouble at rest: how

correlation patterns and group differences become distorted after global signal regression.

Brain Connect 2: 25–32.

Saad ZS, Reynolds RC, Jo HJ, Gotts SJ, Chen G, Martin A, et al (2013). Correcting brain-wide

correlation differences in resting-state FMRI. Brain Connect 3: 339–52.

Sachdev PS, Loo CK, Mitchell PB, McFarquhar TF, Malhi GS (2007). Repetitive transcranial

magnetic stimulation for the treatment of obsessive compulsive disorder: a double-blind

controlled investigation. Psychol Med 37: 1645–9.

Sachdev PS, McBride R, Loo CK, Mitchell PB, Malhi GS, Croker VM (2001). Right versus left

prefrontal transcranial magnetic stimulation for obsessive-compulsive disorder: a

preliminary investigation. J Clin Psychiatry 62: 981–4.

Sadaghiani S, Scheeringa R, Lehongre K, Morillon B, Giraud A-L, D’Esposito M, et al (2012).

α-band phase synchrony is related to activity in the fronto-parietal adaptive control network.

363

J Neurosci 32: 14305–10.

Saggar M, Sporns O, Gonzalez-Castillo J, Bandettini PA, Carlsson G, Glover G, et al (2018).

Towards a new approach to reveal dynamical organization of the brain using topological

data analysis. Nat Commun 9: 1399.

Saito M, Iwata N, Kawakami N, Matsuyama Y, World Mental Health Japan 2002–2003

Collaborators, Ono Y, et al (2010). Evaluation of the DSM-IV and ICD-10 criteria for

depressive disorders in a community population in Japan using item response theory. Int J

Methods Psychiatr Res 19: 211–22.

Sakai Y, Narumoto J, Nishida S, Nakamae T, Yamada K, Nishimura T, et al (2011).

Corticostriatal functional connectivity in non-medicated patients with obsessive-compulsive

disorder. Eur Psychiatry 26: 463–9.

Salomons T V, Dunlop K, Kennedy SH, Flint A, Geraci J, Giacobbe P, et al (2014). Resting-

state cortico-thalamic-striatal connectivity predicts response to dorsomedial prefrontal

rTMS in major depressive disorder. Neuropsychopharmacology 39: 488–98.

Sanders N, Smeets PAM, Elburg AA van, Danner UN, Meer F van, Hoek HW, et al (2015).

Altered food-cue processing in chronically ill and recovered women with anorexia nervosa.

Front Behav Neurosci 9: 46.

Sarkhel S, Sinha VK, Praharaj SK (2010). Adjunctive high-frequency right prefrontal repetitive

transcranial magnetic stimulation (rTMS) was not effective in obsessive-compulsive

disorder but improved secondary depression. J Anxiety Disord 24: 535–9.

Satterthwaite TD, Elliott MA, Gerraty RT, Ruparel K, Loughead J, Calkins ME, et al (2013). An

improved framework for confound regression and filtering for control of motion artifact in

the preprocessing of resting-state functional connectivity data. Neuroimage 64: 240–56.

Satterthwaite TD, Wolf DH, Loughead J, Ruparel K, Elliott MA, Hakonarson H, et al (2012).

Impact of in-scanner head motion on multiple measures of functional connectivity:

relevance for studies of neurodevelopment in youth. Neuroimage 60: 623–32.

Saveanu R V, Nemeroff CB (2012). Etiology of depression: genetic and environmental factors.

Psychiatr Clin North Am 35: 51–71.

Saxena S, Brody AL, Maidment KM, Smith EC, Zohrabi N, Katz E, et al (2004). Cerebral

364

glucose metabolism in obsessive-compulsive hoarding. Am J Psychiatry 161: 1038–48.

Saxena S, Gorbis E, O’Neill J, Baker SK, Mandelkern MA, Maidment KM, et al (2009). Rapid

effects of brief intensive cognitive-behavioral therapy on brain glucose metabolism in

obsessive-compulsive disorder. Mol Psychiatry 14: 197–205.

Schäfer A, Vaitl D, Schienle A (2010). Regional grey matter volume abnormalities in bulimia

nervosa and binge-eating disorder. Neuroimage 50: 639–43.

Schebendach J, Broft A, Foltin RW, Walsh BT (2013). Can the reinforcing value of food be

measured in bulimia nervosa? Appetite 62: 70–5.

Scheperjans F, Hermann K, Eickhoff SB, Amunts K, Schleicher A, Zilles K (2008). Observer-

independent cytoarchitectonic mapping of the human superior parietal cortex. Cereb Cortex

18: 846–67.

Schienle A, Schäfer A, Hermann A, Vaitl D (2009). Binge-eating disorder: reward sensitivity

and brain activation to images of food. Biol Psychiatry 65: 654–61.

Schiller CE, Minkel J, Smoski MJ, Dichter GS (2013). Remitted major depression is

characterized by reduced prefrontal cortex reactivity to reward loss. J Affect Disord 151:

756–62.

Schmidt H, Petkov G, Richardson MP, Terry JR (2014). Dynamics on networks: the role of local

dynamics and global networks on the emergence of hypersynchronous neural activity. PLoS

Comput Biol 10: e1003947.

Schneibel R, Brakemeier E-L, Wilbertz G, Dykierek P, Zobel I, Schramm E (2012). Sensitivity

to detect change and the correlation of clinical factors with the Hamilton Depression Rating

Scale and the Beck Depression Inventory in depressed inpatients. Psychiatry Res 198: 62–7.

Schneider B, Koenigs M (2017). Human lesion studies of ventromedial prefrontal cortex.

Neuropsychologia 107: 84–93.

Schölvinck ML, Leopold DA, Brookes MJ, Khader PH (2013). The contribution of

electrophysiology to functional connectivity mapping. Neuroimage 80: 297–306.

Schölvinck ML, Maier A, Ye FQ, Duyn JH, Leopold DA (2010). Neural basis of global resting-

state fMRI activity. Proc Natl Acad Sci U S A 107: 10238–43.

Schönfeldt-Lecuona C, Thielscher A, Freudenmann RW, Kron M, Spitzer M, Herwig U (2005).

365

Accuracy of stereotaxic positioning of transcranial magnetic stimulation. Brain Topogr 17:

253–9.

Schöpf V, Windischberger C, Kasess CH, Lanzenberger R, Moser E (2010). Group ICA of

resting-state data: a comparison. MAGMA 23: 317–25.

Schrader LM, Cook IA, Miller PR, Maremont ER, DeGiorgio CM (2011). Trigeminal nerve

stimulation in major depressive disorder: first proof of concept in an open pilot trial.

Epilepsy Behav 22: 475–8.

Schultz W (1998). Predictive reward signal of dopamine neurons. J Neurophysiol 80: 1–27.

Schulz K, Sydekum E, Krueppel R, Engelbrecht CJ, Schlegel F, Schröter A, et al (2012).

Simultaneous BOLD fMRI and fiber-optic calcium recording in rat neocortex. Nat Methods

9: 597–602.

Schulze L, Feffer K, Lozano C, Giacobbe P, Daskalakis ZJ, Blumberger DM, et al (2018).

Number of pulses or number of sessions? An open-label study of trajectories of

improvement for once-vs. twice-daily dorsomedial prefrontal rTMS in major depression.

Brain Stimul 11: 327–336.

Schulze L, Wheeler S, McAndrews MP, Solomon CJE, Giacobbe P, Downar J (2016). Cognitive

safety of dorsomedial prefrontal repetitive transcranial magnetic stimulation in major

depression. Eur Neuropsychopharmacol 26: 1213–26.

Schummers J, Yu H, Sur M (2008). Tuned responses of astrocytes and their influence on

hemodynamic signals in the visual cortex. Science 320: 1638–43.

Schutter DJLG, Honk J van (2006). A Standardized Motor Threshold Estimation Procedure for

Transcranial Magnetic Stimulation Research. J ECT 22: 176–178.

Seeger G, Braus DF, Ruf M, Goldberger U, Schmidt MH (2002). Body image distortion reveals

amygdala activation in patients with anorexia nervosa -- a functional magnetic resonance

imaging study. Neurosci Lett 326: 25–8.

Seeley WW, Menon V, Schatzberg AF, Keller J, Glover GH, Kenna H, et al (2007). Dissociable

intrinsic connectivity networks for salience processing and executive control. J Neurosci 27:

2349–56.

Segarra N, Metastasio A, Ziauddeen H, Spencer J, Reinders NR, Dudas RB, et al (2016).

366

Abnormal Frontostriatal Activity During Unexpected Reward Receipt in Depression and

Schizophrenia: Relationship to Anhedonia. Neuropsychopharmacology 41: 2001–10.

Selemon LD, Goldman-Rakic PS (1985). Longitudinal topography and interdigitation of

corticostriatal projections in the rhesus monkey. J Neurosci 5: 776–94.

Semendeferi K, Armstrong E, Schleicher A, Zilles K, Hoesen GW Van (2001). Prefrontal cortex

in humans and apes: a comparative study of area 10. Am J Phys Anthropol 114: 224–41.

Seo D-O, Funderburk SC, Bhatti DL, Motard LE, Newbold D, Girven KS, et al (2016a). A

GABAergic Projection from the Centromedial Nuclei of the Amygdala to Ventromedial

Prefrontal Cortex Modulates Reward Behavior. J Neurosci 36: 10831–10842.

Seo H-J, Jung Y-E, Kim T-S, Kim J-B, Lee M-S, Kim J-M, et al (2011). Distinctive clinical

characteristics and suicidal tendencies of patients with anxious depression. J Nerv Ment Dis

199: 42–8.

Seo H-J, Jung Y-E, Lim HK, Um Y-H, Lee CU, Chae J-H (2016b). Adjunctive Low-frequency

Repetitive Transcranial Magnetic Stimulation over the Right Dorsolateral Prefrontal Cortex

in Patients with Treatment-resistant Obsessive-compulsive Disorder: A Randomized

Controlled Trial. Clin Psychopharmacol Neurosci 14: 153–60.

Sesack SR, Deutch AY, Roth RH, Bunney BS (1989). Topographical organization of the efferent

projections of the medial prefrontal cortex in the rat: an anterograde tract-tracing study with

Phaseolus vulgaris leucoagglutinin. J Comp Neurol 290: 213–42.

Shackman AJ, Salomons T V, Slagter HA, Fox AS, Winter JJ, Davidson RJ (2011). The

integration of negative affect, pain and cognitive control in the cingulate cortex. Nat Rev


Shajahan PM, Glabus MF, Steele JD, Doris AB, Anderson K, Jenkins JA, et al (2002). Left

dorso-lateral repetitive transcranial magnetic stimulation affects cortical excitability and

functional connectivity, but does not impair cognition in major depression. Prog

Neuropsychopharmacol Biol Psychiatry 26: 945–54.

Shanmugan S, Wolf DH, Calkins ME, Moore TM, Ruparel K, Hopson RD, et al (2016).

Common and Dissociable Mechanisms of Executive System Dysfunction Across

Psychiatric Disorders in Youth. Am J Psychiatry 173: 517–26.

367

Shannon BJ, Raichle ME, Snyder AZ, Fair DA, Mills KL, Zhang D, et al (2011). Premotor

functional connectivity predicts impulsivity in juvenile offenders. Proc Natl Acad Sci U S A

108: 11241–5.

Shapira-Lichter I, Oren N, Jacob Y, Gruberger M, Hendler T (2013). Portraying the unique

contribution of the default mode network to internally driven mnemonic processes. Proc


Shapiro JR, Berkman ND, Brownley KA, Sedway JA, Lohr KN, Bulik CM (2007). Bulimia

nervosa treatment: a systematic review of randomized controlled trials. Int J Eat Disord 40:

321–36.

Sharma A, Wolf DH, Ciric R, Kable JW, Moore TM, Vandekar SN, et al (2017). Common

Dimensional Reward Deficits Across Mood and Psychotic Disorders: A Connectome-Wide

Association Study. Am J Psychiatry 174: 657–666.

Sheehan D V, Lecrubier Y, Sheehan KH, Amorim P, Janavs J, Weiller E, et al (1998). The Mini-

International Neuropsychiatric Interview (M.I.N.I.): the development and validation of a

structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry 59

Suppl 2: 22-33;quiz 34-57.

Sheffield JM, Kandala S, Tamminga CA, Pearlson GD, Keshavan MS, Sweeney JA, et al (2017).

Transdiagnostic Associations Between Functional Brain Network Integrity and Cognition.

JAMA psychiatry 74: 605–613.

Shehzad Z, Kelly AMC, Reiss PT, Gee DG, Gotimer K, Uddin LQ, et al (2009). The resting

brain: unconstrained yet reliable. Cereb Cortex 19: 2209–29.

Sheline YI, Barch DM, Price JL, Rundle MM, Vaishnavi SN, Snyder AZ, et al (2009). The

default mode network and self-referential processes in depression. Proc Natl Acad Sci 106:

1942–1947.

Shen T, Qiu M, Li C, Zhang J, Wu Z, Wang B, et al (2014). Altered spontaneous neural activity

in first-episode, unmedicated patients with major depressive disorder. Neuroreport 25:

1302–7.

Shepherd GMG (2013). Corticostriatal connectivity and its role in disease. Nat Rev Neurosci 14:

278–91.

368

Shilyansky C, Williams LM, Gyurak A, Harris A, Usherwood T, Etkin A (2016). Effect of

antidepressant treatment on cognitive impairments associated with depression: a

randomised longitudinal study. The lancet Psychiatry 3: 425–35.

Shiozawa P, Duailibi MS, Silva ME da, Cordeiro Q (2014). Trigeminal nerve stimulation (TNS)

protocol for treating major depression: an open-label proof-of-concept trial. Epilepsy Behav

39: 6–9.

Shiozawa P, Silva ME da, Netto GTM, Taiar I, Cordeiro Q (2015). Effect of a 10-day trigeminal

nerve stimulation (TNS) protocol for treating major depressive disorder: a phase II, sham-

controlled, randomized clinical trial. Epilepsy Behav 44: 23–6.

Shmuel A, Leopold DA (2008). Neuronal correlates of spontaneous fluctuations in fMRI signals

in monkey visual cortex: Implications for functional connectivity at rest. Hum Brain Mapp

29: 751–61.

Shmueli K, Gelderen P van, Zwart JA de, Horovitz SG, Fukunaga M, Jansma JM, et al (2007).

Low-frequency fluctuations in the cardiac rate as a source of variance in the resting-state

fMRI BOLD signal. Neuroimage 38: 306–20.

Shott ME, Pryor TL, Yang TT, Frank GKW (2016). Greater Insula White Matter Fiber

Connectivity in Women Recovered from Anorexia Nervosa. Neuropsychopharmacology 41:

498–507.

Shulman GL, Fiez JA, Corbetta M, Buckner RL, Miezin FM, Raichle ME, et al (1997). Common

Blood Flow Changes across Visual Tasks: II. Decreases in Cerebral Cortex. J Cogn


Shulman RG, Rothman DL, Behar KL, Hyder F (2004). Energetic basis of brain activity:

implications for neuroimaging. Trends Neurosci 27: 489–95.

Siegel M, Donner TH, Engel AK (2012). Spectral fingerprints of large-scale neuronal

interactions. Nat Rev Neurosci 13: 121–34.

Siegle GJ, Carter CS, Thase ME (2006). Use of FMRI to predict recovery from unipolar

depression with cognitive behavior therapy. Am J Psychiatry 163: 735–8.

Siero JCW, Hendrikse J, Hoogduin H, Petridou N, Luijten P, Donahue MJ (2015). Cortical depth

dependence of the BOLD initial dip and poststimulus undershoot in human visual cortex at

369

7 Tesla. Magn Reson Med 73: 2283–95.

Sikora M, Heffernan J, Avery ET, Mickey BJ, Zubieta J-K, Peciña M (2016). Salience Network

Functional Connectivity Predicts Placebo Effects in Major Depression. Biol psychiatry

Cogn Neurosci neuroimaging 1: 68–76.

Simon GE, Perlis RH (2010). Personalized medicine for depression: can we match patients with

treatments? Am J Psychiatry 167: 1445–55.

Simpson HB, Franklin ME, Cheng J, Foa EB, Liebowitz MR (2005). Standard criteria for relapse

are needed in obsessive-compulsive disorder. Depress Anxiety 21: 1–8.

Simpson HB, Huppert JD, Petkova E, Foa EB, Liebowitz MR (2006). Response versus remission

in obsessive-compulsive disorder. J Clin Psychiatry 67: 269–76.

Simpson JR, Drevets WC, Snyder AZ, Gusnard DA, Raichle ME (2001a). Emotion-induced

changes in human medial prefrontal cortex: II. During anticipatory anxiety. Proc Natl Acad

Sci U S A 98: 688–93.

Simpson JR, Snyder AZ, Gusnard DA, Raichle ME (2001b). Emotion-induced changes in human

medial prefrontal cortex: I. During cognitive task performance. Proc Natl Acad Sci U S A

98: 683–7.

Skunde M, Walther S, Simon JJ, Wu M, Bendszus M, Herzog W, et al (2016). Neural signature

of behavioural inhibition in women with bulimia nervosa. J Psychiatry Neurosci 41: E69-

78.

Slotnick SD (2017). Resting-state fMRI data reflects default network activity rather than null

data: A defense of commonly employed methods to correct for multiple comparisons. Cogn


Small DM, Gregory MD, Mak YE, Gitelman D, Mesulam MM, Parrish T (2003). Dissociation of

neural representation of intensity and affective valuation in human gustation. Neuron 39:

701–11.

Small DM, Zatorre RJ, Dagher A, Evans AC, Jones-Gotman M (2001). Changes in brain activity

related to eating chocolate: from pleasure to aversion. Brain 124: 1720–33.

Smink FRE, Hoeken D van, Oldehinkel AJ, Hoek HW (2014). Prevalence and severity of DSM-

5 eating disorders in a community cohort of adolescents. Int J Eat Disord 47: 610–9.

370

Smith D V, Rigney AE, Delgado MR (2016). Distinct Reward Properties are Encoded via

Corticostriatal Interactions. Sci Rep 6: 20093.

Smith D V, Utevsky A V, Bland AR, Clement N, Clithero JA, Harsch AEW, et al (2014).

Characterizing individual differences in functional connectivity using dual-regression and

seed-based approaches. Neuroimage 95: 1–12.

Smith KS, Graybiel AM (2013). A dual operator view of habitual behavior reflecting cortical and

striatal dynamics. Neuron 79: 361–74.

Smith SM (2002). Fast robust automated brain extraction. Hum Brain Mapp 17: 143–55.

Smith WK (1945). THE FUNCTIONAL SIGNIFICANCE OF THE ROSTRAL CINGULAR

CORTEX AS REVEALED BY ITS RESPONSES TO ELECTRICAL EXCITATION. J


Smith Y, Raju D V, Pare J-F, Sidibe M (2004). The thalamostriatal system: a highly specific

network of the basal ganglia circuitry. Trends Neurosci 27: 520–7.

Smith Y, Shink E, Sidibé M (1998). Neuronal circuitry and synaptic connectivity of the basal

ganglia. Neurosurg Clin N Am 9: 203–22.

Smittenaar P, Kurth-Nelson Z, Mohammadi S, Weiskopf N, Dolan RJ (2017). Local striatal

reward signals can be predicted from corticostriatal connectivity. Neuroimage 159: 9–17.

Smoski MJ, Rittenberg A, Dichter GS (2011). Major depressive disorder is characterized by

greater reward network activation to monetary than pleasant image rewards. Psychiatry Res

194: 263–270.

Snyder AZ, Raichle ME (2012). A brief history of the resting state: the Washington University

perspective. Neuroimage 62: 902–10.

Snyder HR, Miyake A, Hankin BL (2015). Advancing understanding of executive function

impairments and psychopathology: bridging the gap between clinical and cognitive

approaches. Front Psychol 6: 328.

SOKOLOFF L, MANGOLD R, WECHSLER RL, KENNEY C, KETY SS (1955). The effect of

mental arithmetic on cerebral circulation and metabolism. J Clin Invest 34: 1101–8.

Somogyi P, Bolam JP, Smith AD (1981). Monosynaptic cortical input and local axon collaterals

of identified striatonigral neurons. A light and electron microscopic study using the Golgi-

371

peroxidase transport-degeneration procedure. J Comp Neurol 195: 567–84.

Song X, Zhang Y, Liu Y (2014). Frequency specificity of regional homogeneity in the resting-

state human brain. PLoS One 9: e86818.

Souery D, Papakostas GI, Trivedi MH (2006). Treatment-resistant depression. J Clin Psychiatry

67 Suppl 6: 16–22.

Speer AM, Benson BE, Kimbrell TK, Wassermann EM, Willis MW, Herscovitch P, et al (2009).

Opposite effects of high and low frequency rTMS on mood in depressed patients:

relationship to baseline cerebral activity on PET. J Affect Disord 115: 386–94.

Speer AM, Kimbrell TA, Wassermann EM, D Repella J, Willis MW, Herscovitch P, et al (2000).

Opposite effects of high and low frequency rTMS on regional brain activity in depressed

patients. Biol Psychiatry 48: 1133–41.

Spies M, Kraus C, Geissberger N, Auer B, Klöbl M, Tik M, et al (2017). Default mode network

deactivation during emotion processing predicts early antidepressant response. Transl

Psychiatry 7: e1008.

Sporns O (2011). The human connectome: a complex network. Ann N Y Acad Sci 1224: 109–

125.

Spreng RN, Grady CL (2010). Patterns of brain activity supporting autobiographical memory,

prospection, and theory of mind, and their relationship to the default mode network. J Cogn

Neurosci 22: 1112–23.

Spreng RN, Mar RA, Kim ASN (2009). The common neural basis of autobiographical memory,

prospection, navigation, theory of mind, and the default mode: a quantitative meta-analysis.

J Cogn Neurosci 21: 489–510.

Sridharan D, Levitin DJ, Menon V (2008). A critical role for the right fronto-insular cortex in

switching between central-executive and default-mode networks. Proc Natl Acad Sci U S A

105: 12569–74.

Stafford JM, Jarrett BR, Miranda-Dominguez O, Mills BD, Cain N, Mihalas S, et al (2014).

Large-scale topology and the default mode network in the mouse connectome. Proc Natl

Acad Sci U S A 111: 18745–50.

Stagg CJ, Nitsche MA (2011). Physiological basis of transcranial direct current stimulation.

372

Neuroscientist 17: 37–53.

Stam CJ, Reijneveld JC (2007). Graph theoretical analysis of complex networks in the brain.

Nonlinear Biomed Phys 1: 3.

Stein DJ, Andersen EW, Tonnoir B, Fineberg N (2007). Escitalopram in obsessive-compulsive

disorder: a randomized, placebo-controlled, paroxetine-referenced, fixed-dose, 24-week

study. Curr Med Res Opin 23: 701–11.

Stein LA (2005). Making the best use of radiological resources in Canada. Healthc Pap 6: 18–

23.

Steinglass JE, Figner B, Berkowitz S, Simpson HB, Weber EU, Walsh BT (2012). Increased

capacity to delay reward in anorexia nervosa. J Int Neuropsychol Soc 18: 773–80.

Steinhausen H-C (2002). The outcome of anorexia nervosa in the 20th century. Am J Psychiatry

159: 1284–93.

Steinhausen HC (1997). Outcome of anorexia nervosa in the younger patient. J Child Psychol


Steketee G (1997). Disability and family burden in obsessive-compulsive disorder. Can J


Stevens MC, Pearlson GD, Calhoun VD (2009). Changes in the interaction of resting-state neural

networks from adolescence to adulthood. Hum Brain Mapp 30: 2356–66.

Storch EA, Rasmussen SA, Price LH, Larson MJ, Murphy TK, Goodman WK (2010).

Development and psychometric evaluation of the Yale-Brown Obsessive-Compulsive

Scale--Second Edition. Psychol Assess 22: 223–32.

Stoy M, Schlagenhauf F, Sterzer P, Bermpohl F, Hägele C, Suchotzki K, et al (2012).

Hyporeactivity of ventral striatum towards incentive stimuli in unmedicated depressed

patients normalizes after treatment with escitalopram. J Psychopharmacol 26: 677–88.

Strafella AP, Paus T, Barrett J, Dagher A (2001). Repetitive transcranial magnetic stimulation of

the human prefrontal cortex induces dopamine release in the caudate nucleus. J Neurosci

21: RC157.

Striegel-Moore RH, Bulik CM (2007). Risk factors for eating disorders. Am Psychol 62: 181–98.

373

Strigo IA, Matthews SC, Simmons AN, Oberndorfer T, Klabunde M, Reinhardt LE, et al (2013).

Altered insula activation during pain anticipation in individuals recovered from anorexia

nervosa: evidence of interoceptive dysregulation. Int J Eat Disord 46: 23–33.

Strober M, Freeman R, Lampert C, Diamond J, Kaye W (2000). Controlled family study of

anorexia nervosa and bulimia nervosa: evidence of shared liability and transmission of

partial syndromes. Am J Psychiatry 157: 393–401.

Strober M, Freeman R, Lampert C, Diamond J, Kaye W (2001). Males with anorexia nervosa: a

controlled study of eating disorders in first-degree relatives. Int J Eat Disord 29: 263–9.

Strober M, Freeman R, Morrell W (1997). The long-term course of severe anorexia nervosa in

adolescents: survival analysis of recovery, relapse, and outcome predictors over 10-15 years

in a prospective study. Int J Eat Disord 22: 339–60.

Subirà M, Alonso P, Segalàs C, Real E, López-Solà C, Pujol J, et al (2013). Brain structural

alterations in obsessive-compulsive disorder patients with autogenous and reactive

obsessions. PLoS One 8: e75273.

Sullivan PF (1995). Mortality in anorexia nervosa. Am J Psychiatry 152: 1073–4.

Sullivan PF, Neale MC, Kendler KS (2000). Genetic epidemiology of major depression: review

and meta-analysis. Am J Psychiatry 157: 1552–62.

Sutoh C, Koga Y, Kimura H, Kanahara N, Numata N, Hirano Y, et al (2016). Repetitive

Transcranial Magnetic Stimulation Changes Cerebral Oxygenation on the Left Dorsolateral

Prefrontal Cortex in Bulimia Nervosa: A Near-Infrared Spectroscopy Pilot Study. Eur Eat


Suzuki WA, Amaral DG (1994). Perirhinal and parahippocampal cortices of the macaque

monkey: cortical afferents. J Comp Neurol 350: 497–533.

Svoboda E, McKinnon MC, Levine B (2006). The functional neuroanatomy of autobiographical

memory: a meta-analysis. Neuropsychologia 44: 2189–208.

Swanson SA, Crow SJ, Grange D Le, Swendsen J, Merikangas KR (2011). Prevalence and

correlates of eating disorders in adolescents. Results from the national comorbidity survey

replication adolescent supplement. Arch Gen Psychiatry 68: 714–23.

Sykes M, Makowiecki K, Rodger J (2013). Long term delivery of pulsed magnetic fields does

374

not alter visual discrimination learning or dendritic spine density in the mouse CA1

pyramidal or dentate gyrus neurons. F1000Research 2: 180.

Tadayonnejad R, Yang S, Kumar A, Ajilore O (2015). Clinical, cognitive, and functional

connectivity correlations of resting-state intrinsic brain activity alterations in unmedicated

depression. J Affect Disord 172: 241–50.

Tagliazucchi E, Balenzuela P, Fraiman D, Chialvo DR (2012). Criticality in large-scale brain

FMRI dynamics unveiled by a novel point process analysis. Front Physiol 3: 15.

Tagliazucchi E, Siniatchkin M, Laufs H, Chialvo DR (2016). The Voxel-Wise Functional

Connectome Can Be Efficiently Derived from Co-activations in a Sparse Spatio-Temporal

Point-Process. Front Neurosci 10: 381.

Takahashi S, Ukai S, Tsuji T, Kose A, Shoyama M, Yamamoto M, et al (2013). Cerebral blood

flow in the subgenual anterior cingulate cortex and modulation of the mood-regulatory

networks in a successful rTMS treatment for major depressive disorder. Neurocase 19: 262–

7.

Talairach J, Tournoux P (1988). Co-planar stereotaxic atlas of the human brain: 3-dimensional

proportional system: an approach to cerebral imaging. Neuropsychologia 39: 145.

Tang A, Thickbroom G, Rodger J (2017). Repetitive Transcranial Magnetic Stimulation of the

Brain. Neurosci 23: 82–94.

Tapajóz P de Sampaio F, Soneira S, Aulicino A, Harris P, Allegri RF (2015). Emotional

reactivity to social stimuli in patients with eating disorders. Psychiatry Res 229: 887–94.

Tataranni PA, Gautier JF, Chen K, Uecker A, Bandy D, Salbe AD, et al (1999).

Neuroanatomical correlates of hunger and satiation in humans using positron emission

tomography. Proc Natl Acad Sci U S A 96: 4569–74.

Taylor PCJ, Nobre AC, Rushworth MFS (2007). Subsecond changes in top down control exerted

by human medial frontal cortex during conflict and action selection: a combined

transcranial magnetic stimulation electroencephalography study. J Neurosci 27: 11343–53.

Taylor S (2016). Disorder-specific genetic factors in obsessive-compulsive disorder: A

comprehensive meta-analysis. Am J Med Genet B Neuropsychiatr Genet 171B: 325–32.

Taylor S, Jang KL (2011). Biopsychosocial etiology of obsessions and compulsions: an

375

integrated behavioral-genetic and cognitive-behavioral analysis. J Abnorm Psychol 120:

174–86.

Taylor SF, Ho SS, Abagis T, Angstadt M, Maixner DF, Welsh RC, et al (2018). Changes in

brain connectivity during a sham-controlled, transcranial magnetic stimulation trial for

depression. J Affect Disord 232: 143–151.

Taylor WD, Kudra K, Zhao Z, Steffens DC, MacFall JR (2014). Cingulum bundle white matter

lesions influence antidepressant response in late-life depression: a pilot study. J Affect

Disord 162: 8–11.

Tedeschini E, Levkovitz Y, Iovieno N, Ameral VE, Nelson JC, Papakostas GI (2011). Efficacy

of antidepressants for late-life depression: a meta-analysis and meta-regression of placebo-

controlled randomized trials. J Clin Psychiatry 72: 1660–8.

Teneback CC, Nahas Z, Speer AM, Molloy M, Stallings LE, Spicer KM, et al (1999). Changes

in prefrontal cortex and paralimbic activity in depression following two weeks of daily left

prefrontal TMS. J Neuropsychiatry Clin Neurosci 11: 426–35.

Teng S, Guo Z, Peng H, Xing G, Chen H, He B, et al (2017). High-frequency repetitive

transcranial magnetic stimulation over the left DLPFC for major depression: Session-

dependent efficacy: A meta-analysis. Eur Psychiatry 41: 75–84.

Terao Y, Ugawa Y, Enomoto H, Furubayashi T, Shiio Y, Machii K, et al (2001). Hemispheric

lateralization in the cortical motor preparation for human vocalization. J Neurosci 21:

1600–9.

Thase ME, Larsen KG, Kennedy SH (2011). Assessing the “true” effect of active antidepressant

therapy v. placebo in major depressive disorder: use of a mixture model. Br J Psychiatry

199: 501–7.

Thielscher A, Pessoa L (2007). Neural correlates of perceptual choice and decision making

during fear-disgust discrimination. J Neurosci 27: 2908–17.

Thomas JJ, Roberto CA, Brownell KD (2009). Eighty-five per cent of what? Discrepancies in

the weight cut-off for anorexia nervosa substantially affect the prevalence of underweight.


Thompson-Brenner H, Shingleton RM, Sauer-Zavala S, Richards LK, Pratt EM (2015). Multiple

376

measures of rapid response as predictors of remission in cognitive behavior therapy for

bulimia nervosa. Behav Res Ther 64: 9–14.

Thompson GJ, Pan W-J, Billings JCW, Grooms JK, Shakil S, Jaeger D, et al (2014a). Phase-

amplitude coupling and infraslow (<1 Hz) frequencies in the rat brain: relationship to

resting state fMRI. Front Integr Neurosci 8: 41.

Thompson PM, Stein JL, Medland SE, Hibar DP, Vasquez AA, Renteria ME, et al (2014b). The

ENIGMA Consortium: large-scale collaborative analyses of neuroimaging and genetic data.

Brain Imaging Behav 8: 153–82.

Thornton LM, Mazzeo SE, Bulik CM (2011). The heritability of eating disorders: methods and

current findings. Curr Top Behav Neurosci 6: 141–56.

Titova OE, Hjorth OC, Schiöth HB, Brooks SJ (2013). Anorexia nervosa is linked to reduced

brain structure in reward and somatosensory regions: a meta-analysis of VBM studies. BMC

Psychiatry 13: 110.

Tokay T, Holl N, Kirschstein T, Zschorlich V, Köhling R (2009). High-frequency magnetic

stimulation induces long-term potentiation in rat hippocampal slices. Neurosci Lett 461:

150–154.

Tollefson GD, Rampey AH, Potvin JH, Jenike MA, Rush AJ, Kominguez RA, et al (1994). A

multicenter investigation of fixed-dose fluoxetine in the treatment of obsessive-compulsive

disorder. Arch Gen Psychiatry 51: 559–67.

Tor P-C, Gálvez V, Goldstein J, George D, Loo CK (2016). Pilot Study of Accelerated Low-

Frequency Right-Sided Transcranial Magnetic Stimulation for Treatment-Resistant

Depression. J ECT 32: 180–2.

Tremblay L, Schultz W (1999). Relative reward preference in primate orbitofrontal cortex.

Nature 398: 704–8.

Tremblay LK, Naranjo CA, Graham SJ, Herrmann N, Mayberg HS, Hevenor S, et al (2005).

Functional neuroanatomical substrates of altered reward processing in major depressive

disorder revealed by a dopaminergic probe. Arch Gen Psychiatry 62: 1228–36.

Treynor W, Gonzalez R, Nolen-Hoeksema S (2003). Rumination reconsidered: A psychometric

analysis. Cognit Ther Res 27: 247–259.

377

Triggs WJ, Ricciuti N, Ward HE, Cheng J, Bowers D, Goodman WK, et al (2010). Right and left

dorsolateral pre-frontal rTMS treatment of refractory depression: a randomized, sham-

controlled trial. Psychiatry Res 178: 467–74.

Trippe J, Mix A, Aydin-Abidin S, Funke K, Benali A (2009). Theta burst and conventional low-

frequency rTMS differentially affect GABAergic neurotransmission in the rat cortex. Exp

Brain Res 199: 411–421.

Trivedi MH, Rush AJ, Wisniewski SR, Nierenberg AA, Warden D, Ritz L, et al (2006).

Evaluation of outcomes with citalopram for depression using measurement-based care in

STAR*D: implications for clinical practice. Am J Psychiatry 163: 28–40.

Tziortzi AC, Haber SN, Searle GE, Tsoumpas C, Long CJ, Shotbolt P, et al (2014).

Connectivity-based functional analysis of dopamine release in the striatum using diffusion-

weighted MRI and positron emission tomography. Cereb Cortex 24: 1165–77.

Uddin LQ, Kelly AM, Biswal BB, Castellanos FX, Milham MP (2009). Functional connectivity

of default mode network components: correlation, anticorrelation, and causality. Hum Brain

Mapp 30: 625–37.

Uher R, Murphy T, Brammer MJ, Dalgleish T, Phillips ML, Ng VW, et al (2004). Medial

prefrontal cortex activity associated with symptom provocation in eating disorders. Am J


Uher R, Payne JL, Pavlova B, Perlis RH (2014). Major depressive disorder in DSM-5:

implications for clinical practice and research of changes from DSM-IV. Depress Anxiety

31: 459–71.

Uher R, Yoganathan D, Mogg A, Eranti S V, Treasure J, Campbell IC, et al (2005). Effect of left

prefrontal repetitive transcranial magnetic stimulation on food craving. Biol Psychiatry 58:

840–2.

Urry HL, Reekum CM van, Johnstone T, Kalin NH, Thurow ME, Schaefer HS, et al (2006).

Amygdala and ventromedial prefrontal cortex are inversely coupled during regulation of

negative affect and predict the diurnal pattern of cortisol secretion among older adults. J

Neurosci 26: 4415–25.

Utevsky A V, Smith D V, Huettel SA (2014). Precuneus is a functional core of the default-mode

378

network. J Neurosci 34: 932–40.

Uylings HBM, -Arigita EJS, Vos K de, Pool CW, Evers P, Rajkowska G (2010). 3-D

Cytoarchitectonic parcellation of human orbitofrontal cortex. Psychiatry Res Neuroimaging

183: 1–20.

Vaghi MM, Vértes PE, Kitzbichler MG, Apergis-Schoute AM, Flier FE van der, Fineberg NA, et

al (2017). Specific Frontostriatal Circuits for Impaired Cognitive Flexibility and Goal-

Directed Planning in Obsessive-Compulsive Disorder: Evidence From Resting-State

Functional Connectivity. Biol Psychiatry 81: 708–717.

Vaishnavi SN, Vlassenko AG, Rundle MM, Snyder AZ, Mintun MA, Raichle ME (2010).

Regional aerobic glycolysis in the human brain. Proc Natl Acad Sci U S A 107: 17757–62.

Val-Laillet D, Aarts E, Weber B, Ferrari M, Quaresima V, Stoeckel LE, et al (2015).

Neuroimaging and neuromodulation approaches to study eating behavior and prevent and

treat eating disorders and obesity. NeuroImage Clin 8: 1–31.

Vallar G (1998). Spatial hemineglect in humans. Trends Cogn Sci 2: 87–97.

Vanhatalo S, Palva JM, Holmes MD, Miller JW, Voipio J, Kaila K (2004). Infraslow oscillations

modulate excitability and interictal epileptic activity in the human cortex during sleep. Proc


Vanhaudenhuyse A, Noirhomme Q, Tshibanda LJ-F, Bruno M-A, Boveroux P, Schnakers C, et

al (2010). Default network connectivity reflects the level of consciousness in non-

communicative brain-damaged patients. Brain 133: 161–71.

Vasiliadis H-M, Lesage A, Adair C, Wang PS, Kessler RC (2007). Do Canada and the United

States differ in prevalence of depression and utilization of services? Psychiatr Serv 58: 63–

71.

Velden AM van der, Kuyken W, Wattar U, Crane C, Pallesen KJ, Dahlgaard J, et al (2015). A

systematic review of mechanisms of change in mindfulness-based cognitive therapy in the

treatment of recurrent major depressive disorder. Clin Psychol Rev 37: 26–39.

Ven VG van de, Formisano E, Prvulovic D, Roeder CH, Linden DEJ (2004). Functional

connectivity as revealed by spatial independent component analysis of fMRI measurements

during rest. Hum Brain Mapp 22: 165–78.

379

Vermeiden M, Broek WW van den, Mulder PGH, Birkenhäger TK (2010). Influence of gender

and menopausal status on antidepressant treatment response in depressed inpatients. J

Psychopharmacol 24: 497–502.

Vincent JL, Kahn I, Snyder AZ, Raichle ME, Buckner RL (2008). Evidence for a frontoparietal

control system revealed by intrinsic functional connectivity. J Neurophysiol 100: 3328–42.

Vincent JL, Patel GH, Fox MD, Snyder AZ, Baker JT, Essen DC Van, et al (2007). Intrinsic

functional architecture in the anaesthetized monkey brain. Nature 447: 83–6.

Vincent JL, Snyder AZ, Fox MD, Shannon BJ, Andrews JR, Raichle ME, et al (2006). Coherent

spontaneous activity identifies a hippocampal-parietal memory network. J Neurophysiol 96:

3517–31.

Vlachos A, Muller-Dahlhaus F, Rosskopp J, Lenz M, Ziemann U, Deller T (2012). Repetitive

Magnetic Stimulation Induces Functional and Structural Plasticity of Excitatory

Postsynapses in Mouse Organotypic Hippocampal Slice Cultures. J Neurosci 32: 17514–

17523.

Vocks S, Busch M, Grönemeyer D, Schulte D, Herpertz S, Suchan B (2010). Neural correlates of

viewing photographs of one’s own body and another woman’s body in anorexia and bulimia

nervosa: an fMRI study. J Psychiatry Neurosci 35: 163–76.

Vogeley K, May M, Ritzl A, Falkai P, Zilles K, Fink GR (2004). Neural correlates of first-person

perspective as one constituent of human self-consciousness. J Cogn Neurosci 16: 817–27.

Vogt BA, Finch DM, Olson CR (1992). Functional heterogeneity in cingulate cortex: the anterior

executive and posterior evaluative regions. Cereb Cortex 2: 435–43.

Vogt BA, Pandya DN (1987). Cingulate cortex of the rhesus monkey: II. Cortical afferents. J


Vogt BA, Pandya DN, Rosene DL (1987). Cingulate cortex of the rhesus monkey: I.

Cytoarchitecture and thalamic afferents. J Comp Neurol 262: 256–70.

Vogt BA, Rosene DL, Pandya DN (1979). Thalamic and cortical afferents differentiate anterior

from posterior cingulate cortex in the monkey. Science 204: 205–7.

Vogt BA, Vogt LJ, Perl DP, Hof PR (2001). Cytology of human caudomedial cingulate,

retrosplenial, and caudal parahippocampal cortices. J Comp Neurol 438: 353–76.

380

Volz LJ, Benali A, Mix A, Neubacher U, Funke K (2013). Dose-Dependence of Changes in

Cortical Protein Expression Induced with Repeated Transcranial Magnetic Theta-Burst

Stimulation in the Rat. Brain Stimul 6: 598–606.

Voorn P, Vanderschuren LJMJ, Groenewegen HJ, Robbins TW, Pennartz CMA (2004). Putting

a spin on the dorsal-ventral divide of the striatum. Trends Neurosci 27: 468–74.

Waarde JA van, Scholte HS, Oudheusden LJB van, Verwey B, Denys D, Wingen GA van

(2015). A functional MRI marker may predict the outcome of electroconvulsive therapy in

severe and treatment-resistant depression. Mol Psychiatry 20: 609–14.

Wade TD, Bulik CM, Neale M, Kendler KS (2000). Anorexia nervosa and major depression:

shared genetic and environmental risk factors. Am J Psychiatry 157: 469–71.

Wager TD, Davidson ML, Hughes BL, Lindquist MA, Ochsner KN (2008). Prefrontal-

Subcortical Pathways Mediating Successful Emotion Regulation. Neuron 59: 1037–1050.

Wager TD, Rilling JK, Smith EE, Sokolik A, Casey KL, Davidson RJ, et al (2004). Placebo-

induced changes in FMRI in the anticipation and experience of pain. Science 303: 1162–7.

Wagner A, Aizenstein H, Venkatraman VK, Bischoff-Grethe A, Fudge J, May JC, et al (2010).

Altered striatal response to reward in bulimia nervosa after recovery. Int J Eat Disord 43:

289–94.

Wagner A, Aizenstein H, Venkatraman VK, Fudge J, May JC, Mazurkewicz L, et al (2007).

Altered reward processing in women recovered from anorexia nervosa. Am J Psychiatry

164: 1842–9.

Wagner G, Koch K, Schachtzabel C, Peikert G, Schultz CC, Reichenbach JR, et al (2013). Self-

referential processing influences functional activation during cognitive control: an fMRI

study. Soc Cogn Affect Neurosci 8: 828–37.

Wagner G, Schachtzabel C, Peikert G, Bär K-J (2015). The neural basis of the abnormal self-

referential processing and its impact on cognitive control in depressed patients. Hum Brain

Mapp 36: 2781–94.

Wagner G, Sinsel E, Sobanski T, Köhler S, Marinou V, Mentzel H-J, et al (2006). Cortical

inefficiency in patients with unipolar depression: an event-related FMRI study with the

Stroop task. Biol Psychiatry 59: 958–65.

381

Walker AE (1940). A cytoarchitectural study of the prefrontal area of the macaque monkey. J


Walpoth M, Hoertnagl C, Mangweth-Matzek B, Kemmler G, Hinterhölzl J, Conca A, et al

(2008). Repetitive transcranial magnetic stimulation in bulimia nervosa: preliminary results

of a single-centre, randomised, double-blind, sham-controlled trial in female outpatients.

Psychother Psychosom 77: 57–60.

Walsh BT, Kaplan AS, Attia E, Olmsted M, Parides M, Carter JC, et al (2006). Fluoxetine after

weight restoration in anorexia nervosa: a randomized controlled trial. JAMA 295: 2605–12.

Walsh BT, Wilson GT, Loeb KL, Devlin MJ, Pike KM, Roose SP, et al (1997). Medication and

psychotherapy in the treatment of bulimia nervosa. Am J Psychiatry 154: 523–31.

Walsh E, Carl H, Eisenlohr-Moul T, Minkel J, Crowther A, Moore T, et al (2017). Attenuation

of Frontostriatal Connectivity During Reward Processing Predicts Response to

Psychotherapy in Major Depressive Disorder. Neuropsychopharmacology 42: 831–843.

Wang H-Y, Crupi D, Liu J, Stucky A, Cruciata G, Rocco A Di, et al (2011). Repetitive

Transcranial Magnetic Stimulation Enhances BDNF-TrkB Signaling in Both Brain and

Lymphocyte. J Neurosci 31: 11044–11054.

Wang J, Wang L, Zang Y, Yang H, Tang H, Gong Q, et al (2009). Parcellation-dependent small-

world brain functional networks: a resting-state fMRI study. Hum Brain Mapp 30: 1511–23.

Wang J, Zuo X, He Y (2010). Graph-based network analysis of resting-state functional MRI.

Front Syst Neurosci 4: 16.

Wang L, Dai W, Su Y, Wang G, Tan Y, Jin Z, et al (2012a). Amplitude of low-frequency

oscillations in first-episode, treatment-naive patients with major depressive disorder: a

resting-state functional MRI study. PLoS One 7: e48658.

Wang L, Kong Q, Li K, Su Y, Zeng Y, Zhang Q, et al (2016). Frequency-dependent changes in

amplitude of low-frequency oscillations in depression: A resting-state fMRI study. Neurosci

Lett 614: 105–11.

Wang L, Li X, Li K, Su Y, Zeng Y, Zhang Q, et al (2017). Mapping the effect of escitalopram

treatment on amplitude of low-frequency fluctuations in patients with depression: a resting-

state fMRI study. Metab Brain Dis 32: 147–154.

382

Wang L, Saalmann YB, Pinsk MA, Arcaro MJ, Kastner S (2012b). Electrophysiological low-

frequency coherence and cross-frequency coupling contribute to BOLD connectivity.

Neuron 76: 1010–20.

Wang L, Xia M, Li K, Zeng Y, Su Y, Dai W, et al (2015). The effects of antidepressant

treatment on resting-state functional brain networks in patients with major depressive

disorder. Hum Brain Mapp 36: 768–78.

Wang Y, Li T-Q (2015). Dimensionality of ICA in resting-state fMRI investigated by feature

optimized classification of independent components with SVM. Front Hum Neurosci 9:

259.

Wassermann EM, McShane LM, Hallett M, Cohen LG (1992). Noninvasive mapping of muscle

representations in human motor cortex. Electroencephalogr Clin Neurophysiol 85: 1–8.

Watanabe M, Bartels A, Macke JH, Murayama Y, Logothetis NK (2013). Temporal jitter of the

BOLD signal reveals a reliable initial dip and improved spatial resolution. Curr Biol 23:

2146–50.

Webb CA, Dillon DG, Pechtel P, Goer FK, Murray L, Huys QJM, et al (2016). Neural

Correlates of Three Promising Endophenotypes of Depression: Evidence from the

EMBARC Study. Neuropsychopharmacology 41: 454–63.

Webb CA, Weber M, Mundy EA, Killgore WDS (2014). Reduced gray matter volume in the

anterior cingulate, orbitofrontal cortex and thalamus as a function of mild depressive

symptoms: a voxel-based morphometric analysis. Psychol Med 44: 2833–43.

Weissenbacher A, Kasess C, Gerstl F, Lanzenberger R, Moser E, Windischberger C (2009).

Correlations and anticorrelations in resting-state functional connectivity MRI: a quantitative

comparison of preprocessing strategies. Neuroimage 47: 1408–16.

Weissman MM, Bland RC, Canino GJ, Greenwald S, Hwu HG, Lee CK, et al (1994). The cross

national epidemiology of obsessive compulsive disorder. The Cross National Collaborative

Group. J Clin Psychiatry 55 Suppl: 5–10.

Weller JA, Levin IP, Shiv B, Bechara A (2009). The effects of insula damage on decision-

making for risky gains and losses. Soc Neurosci 4: 347–58.

Welton T, Kent DA, Auer DP, Dineen RA (2015). Reproducibility of graph-theoretic brain

383

network metrics: a systematic review. Brain Connect 5: 193–202.

Wens V, Bourguignon M, Goldman S, Marty B, Beeck M Op de, Clumeck C, et al (2014). Inter-

and intra-subject variability of neuromagnetic resting state networks. Brain Topogr 27:

620–34.

Weygandt M, Schaefer A, Schienle A, Haynes J-D (2012). Diagnosing different binge-eating

disorders based on reward-related brain activation patterns. Hum Brain Mapp 33: 2135–46.

Whiteside SP, Lynam DR (2001). The Five Factor Model and impulsivity: using a structural

model of personality to understand impulsivity. Pers Individ Dif 30: 669–689.

Whitfield-Gabrieli S, Nieto-Castanon A (2012). Conn: A Functional Connectivity Toolbox for

Correlated and Anticorrelated Brain Networks. Brain Connect 2: 125–141.

Whittal ML, Thordarson DS, McLean PD (2005). Treatment of obsessive-compulsive disorder:

cognitive behavior therapy vs. exposure and response prevention. Behav Res Ther 43:

1559–76.

Wierenga CE, Ely A, Bischoff-Grethe A, Bailer UF, Simmons AN, Kaye WH (2014). Are

Extremes of Consumption in Eating Disorders Related to an Altered Balance between

Reward and Inhibition? Front Behav Neurosci 8: 410.

Wiles NJ, Thomas L, Turner N, Garfield K, Kounali D, Campbell J, et al (2016). Long-term

effectiveness and cost-effectiveness of cognitive behavioural therapy as an adjunct to

pharmacotherapy for treatment-resistant depression in primary care: follow-up of the

CoBalT randomised controlled trial. The lancet Psychiatry 3: 137–44.

Williams ZM, Bush G, Rauch SL, Cosgrove GR, Eskandar EN (2004). Human anterior cingulate

neurons and the integration of monetary reward with motor responses. Nat Neurosci 7:

1370–5.

Wilson GT, Fairburn CC, Agras WS, Walsh BT, Kraemer H (2002). Cognitive-behavioral

therapy for bulimia nervosa: time course and mechanisms of change. J Consult Clin Psychol

70: 267–74.

Wise RA, Rompre PP (1989). Brain dopamine and reward. Annu Rev Psychol 40: 191–225.

Wise RG, Ide K, Poulin MJ, Tracey I (2004). Resting fluctuations in arterial carbon dioxide

induce significant low frequency variations in BOLD signal. Neuroimage 21: 1652–64.

384

Wise T, Radua J, Via E, Cardoner N, Abe O, Adams TM, et al (2017). Common and distinct

patterns of grey-matter volume alteration in major depression and bipolar disorder: evidence

from voxel-based meta-analysis. Mol Psychiatry 22: 1455–1463.

Woo C-W, Krishnan A, Wager TD (2014). Cluster-extent based thresholding in fMRI analyses:

pitfalls and recommendations. Neuroimage 91: 412–9.

Woody SR, Steketee G, Chambless DL (1995). Reliability and validity of the Yale-Brown

Obsessive-Compulsive Scale. Behav Res Ther 33: 597–605.

Woolgar A, Hampshire A, Thompson R, Duncan J (2011). Adaptive coding of task-relevant

information in human frontoparietal cortex. J Neurosci 31: 14592–9.

Workman CI, Lythe KE, McKie S, Moll J, Gethin JA, Deakin JFW, et al (2017). A novel

resting-state functional magnetic resonance imaging signature of resilience to recurrent

depression. Psychol Med 47: 597–607.

World Health Organization (1992). The ICD-10 Classification of Mental and Behavioural

Disorders. Int Classif 10: 1–267.

Worsley KJ, Marrett S, Neelin P, Evans AC (1996). Searching scale space for activation in PET

images. Hum Brain Mapp 4: 74–90.

Wu CC, Sacchet MD, Knutson B (2012). Toward an affective neuroscience account of financial

risk taking. Front Neurosci 6: 159.

Wu CW, Chen C-L, Liu P-Y, Chao Y-P, Biswal BB, Lin C-P (2011). Empirical evaluations of

slice-timing, smoothing, and normalization effects in seed-based, resting-state functional

magnetic resonance imaging analyses. Brain Connect 1: 401–10.

Wu G-R, Baeken C (2017). Longer depressive episode duration negatively influences HF-rTMS

treatment response: a cerebellar metabolic deficiency? Brain Imaging Behav 11: 8–16.

Wu H, Dyck-Lippens PJ Van, Santegoeds R, Kuyck K van, Gabriëls L, Lin G, et al (2013).

Deep-brain stimulation for anorexia nervosa. World Neurosurg 80: S29.e1-10.

Xie J, Chen J, Wei Q (2013). Repetitive transcranial magnetic stimulation versus

electroconvulsive therapy for major depression: a meta-analysis of stimulus parameter

effects. Neurol Res 35: 1084–91.

Xue S, Wang X, Wang W, Liu J, Qiu J (2016). Frequency-dependent alterations in regional

385

homogeneity in major depression. Behav Brain Res 306: 13–9.

Yan C-G, Cheung B, Kelly C, Colcombe S, Craddock RC, Martino A Di, et al (2013). A

comprehensive assessment of regional variation in the impact of head micromovements on

functional connectomics. Neuroimage 76: 183–201.

Yang T, Cheng Y, Li H, Jiang H, Luo C, Shan B, et al (2010). Abnormal regional homogeneity

of drug-naïve obsessive-compulsive patients. Neuroreport 21: 786–90.

Yang X-R, Kirton A, Wilkes TC, Pradhan S, Liu I, Jaworska N, et al (2014). Glutamate

Alterations Associated With Transcranial Magnetic Stimulation in Youth Depression: A

Case Series. J ECT 00: 1–6.

Yang X-Y, Sun J, Luo J, Zhong Z-X, Li P, Yao S-M, et al (2015). Regional homogeneity of

spontaneous brain activity in adult patients with obsessive-compulsive disorder before and

after cognitive behavioural therapy. J Affect Disord 188: 243–51.

Yeo BTT, Krienen FM, Sepulcre J, Sabuncu MR, Lashkari D, Hollinshead M, et al (2011). The

organization of the human cerebral cortex estimated by intrinsic functional connectivity. J


Yeterian EH, Hoesen GW Van (1978). Cortico-striate projections in the rhesus monkey: the

organization of certain cortico-caudate connections. Brain Res 139: 43–63.

Yin HH, Mulcare SP, Hilário MRF, Clouse E, Holloway T, Davis MI, et al (2009). Dynamic

reorganization of striatal circuits during the acquisition and consolidation of a skill. Nat


Yoon KJ, Lee Y-T, Han TR (2011). Mechanism of functional recovery after repetitive

transcranial magnetic stimulation (rTMS) in the subacute cerebral ischemic rat model:

neural plasticity or anti-apoptosis? Exp Brain Res 214: 549–556.

Young L, Camprodon JA, Hauser M, Pascual-Leone A, Saxe R (2010). Disruption of the right

temporoparietal junction with transcranial magnetic stimulation reduces the role of beliefs

in moral judgments. Proc Natl Acad Sci U S A 107: 6753–8.

Yukimasa T, Yoshimura R, Tamagawa A, Uozumi T, Shinkai K, Ueda N, et al (2006). High-

Frequency Repetitive Transcranial Magnetic Stimulation Improves Refractory Depression

by Influencing Catecholamine and Brain-Derived Neurotrophic Factors.

386

Pharmacopsychiatry 39: 52–59.

Yuste R (2015). From the neuron doctrine to neural networks. Nat Rev Neurosci 16: 487–97.

Záborszky L, Alheid GF, Beinfeld MC, Eiden LE, Heimer L, Palkovits M (1985).

Cholecystokinin innervation of the ventral striatum: a morphological and

radioimmunological study. Neuroscience 14: 427–53.

Zalesky A, Fornito A, Harding IH, Cocchi L, Yücel M, Pantelis C, et al (2010). Whole-brain

anatomical networks: does the choice of nodes matter? Neuroimage 50: 970–83.

Zanardini R, Gazzoli A, Ventriglia M, Perez J, Bignotti S, Mariarossini P, et al (2006). Effect of

repetitive transcranial magnetic stimulation on serum brain derived neurotrophic factor in

drug resistant depressed patients. J Affect Disord 91: 83–86.

Zang Y-F, He Y, Zhu C-Z, Cao Q-J, Sui M-Q, Liang M, et al (2007). Altered baseline brain

activity in children with ADHD revealed by resting-state functional MRI. Brain Dev 29:

83–91.

Zang Y, Jiang T, Lu Y, He Y, Tian L (2004). Regional homogeneity approach to fMRI data

analysis. Neuroimage 22: 394–400.

Zangen A, Hyodo K (2002). Transcranial magnetic stimulation induces increases in extracellular

levels of dopamine and glutamate in the nucleus accumbens. Neuroreport 13: 2401–5.

Zastrow A, Kaiser S, Stippich C, Walther S, Herzog W, Tchanturia K, et al (2009). Neural

correlates of impaired cognitive-behavioral flexibility in anorexia nervosa. Am J Psychiatry

166: 608–16.

Zerwas S, Lund BC, Holle A Von, Thornton LM, Berrettini WH, Brandt H, et al (2013). Factors

associated with recovery from anorexia nervosa. J Psychiatr Res 47: 972–9.

Zhang K, Liu Z, Cao X, Yang C, Xu Y, Xu T, et al (2017). Amplitude of low-frequency

fluctuations in first-episode, drug-naïve depressive patients: A 5-year retrospective study.

PLoS One 12: e0174564.

Zhang W-N, Chang S-H, Guo L-Y, Zhang K-L, Wang J (2013). The neural correlates of reward-

related processing in major depressive disorder: a meta-analysis of functional magnetic

resonance imaging studies. J Affect Disord 151: 531–9.

Zhang Y, Brady M, Smith S (2001). Segmentation of brain MR images through a hidden Markov

387

random field model and the expectation-maximization algorithm. IEEE Trans Med Imaging

20: 45–57.

Zhao Y-J, Du M-Y, Huang X-Q, Lui S, Chen Z-Q, Liu J, et al (2014). Brain grey matter

abnormalities in medication-free patients with major depressive disorder: a meta-analysis.


Zhu X, Zhu Q, Shen H, Liao W, Yuan F (2017). Rumination and Default Mode Network

Subsystems Connectivity in First-episode, Drug-Naive Young Patients with Major

Depressive Disorder. Sci Rep 7: 43105.

Zimmerman M, Martinez JH, Young D, Chelminski I, Dalrymple K (2013). Severity

classification on the Hamilton Depression Rating Scale. J Affect Disord 150: 384–8.

Zohar J, Judge R (1996). Paroxetine versus clomipramine in the treatment of obsessive-

compulsive disorder. OCD Paroxetine Study Investigators. Br J Psychiatry 169: 468–74.

Zotev V, Yuan H, Phillips R, Bodurka J (2012). EEG-assisted retrospective motion correction

for fMRI: E-REMCOR. Neuroimage 63: 698–712.

Zou K, Deng W, Li T, Zhang B, Jiang L, Huang C, et al (2010). Changes of brain morphometry

in first-episode, drug-naïve, non-late-life adult patients with major depression: an optimized

voxel-based morphometry study. Biol Psychiatry 67: 186–8.

Zou Q-H, Zhu C-Z, Yang Y, Zuo X-N, Long X-Y, Cao Q-J, et al (2008). An improved approach

to detection of amplitude of low-frequency fluctuation (ALFF) for resting-state fMRI:

fractional ALFF. J Neurosci Methods 172: 137–41.

Zuo C, Ma Y, Sun B, Peng S, Zhang H, Eidelberg D, et al (2013). Metabolic imaging of bilateral

anterior capsulotomy in refractory obsessive compulsive disorder: an FDG PET study. J

Cereb Blood Flow Metab 33: 880–7.

Zuo X-N, Kelly C, Adelstein JS, Klein DF, Castellanos FX, Milham MP (2010a). Reliable

intrinsic connectivity networks: test-retest evaluation using ICA and dual regression

approach. Neuroimage 49: 2163–77.

Zuo X-N, Kelly C, Martino A Di, Mennes M, Margulies DS, Bangaru S, et al (2010b). Growing

together and growing apart: regional and sex differences in the lifespan developmental

trajectories of functional homotopy. J Neurosci 30: 15034–43.

388

Zuo X-N, Xing X-X (2014). Test-retest reliabilities of resting-state FMRI measurements in

human brain functional connectomics: a systems neuroscience perspective. Neurosci

Biobehav Rev 45: 100–18.

(1991). Clomipramine in the treatment of patients with obsessive-compulsive disorder. The

Clomipramine Collaborative Study. Arch Gen Psychiatry 48: 730–8.

390

Appendices

AppendixI.This study was published in the Journal of Visualized Experiments:

Dunlop, K., Gaprielian, P., Blumberger, D., Daskalakis, Z.J., Kennedy, S.H., Giacobbe, P., Downar, J. (2015). MRI-guided dmPFC-rTMS as a Treatment for Treatment-resistant Major Depressive Disorder. Journal of Visualized Experiments, (102): 53129.

Abstract

Abstract

Here we outline the protocol for magnetic resonance imaging (MRI) guided repetitive transcranial magnetic stimulation (rTMS) to the dorsal medial prefrontal cortex (dmPFC) in patients with major depressive disorder (MDD). Technicians used a neuronavigation system to process patient MRIs to generate a 3-dimensional head model. The head model was subsequently used to identify patient-specific stimulatory targets. The dmPFC was stimulated daily for 20 sessions. Stimulation intensity was titrated to address scalp pain associated with rTMS. Weekly assessments were conducted on the patients using the Hamilton Rating Scale for Depression (HamD17) and Beck Depression Index II (BDI-II). Treatment-resistant MDD patients achieved significant improvements on both HAMD and BDI-II. Of note, angled, double-cone coil rTMS at 120% resting motor threshold allows for optimal stimulation of deeper midline prefrontal regions, which results in a possible therapeutic application for MDD. One major limitation of the rTMS field is the heterogeneity of treatment parameters across studies, including duty cycle, number of pulses per session and intensity. Further work should be done to clarify the effect of stimulation parameters on outcome. Future dmPFC-rTMS work should include sham-controlled studies to confirm its clinical efficacy in MDD.

Introduction

Repetitive transcranial magnetic stimulation (rTMS) is a form of indirect focal cortical stimulation. rTMS employs brief, focal electromagnetic field pulses that penetrate the skull to stimulate target brain regions. rTMS is thought to engage the mechanisms of synaptic long-term potentiation and long-term depression, thereby increasing or decreasing the cortical excitability of the region stimulated1. Generally, the rTMS pulse frequency determines its effects: higher frequency stimulation tends to be excitatory, while lower frequency is inhibitory. Non-invasive stimulatory procedures are also widely used as a causal probe to induce temporary ‘cortical lesions’, and establish neural-behavior relationships or functional regions by temporarily disabling the function of a desired cortical region2–4.

Therapeutic rTMS involves multiple stimulation sessions, usually applied once daily over several weeks, to treat a variety of disorders, including major depressive disorder (MDD)5, eating disorders6, and obsessive-compulsive disorder7. rTMS for MDD is a potential option for medically refractory patients, and allows the clinician to noninvasively target and alter the

391

excitability of a cortical region directly involved with depressive etiology or pathophysiology. The conventional cortical target for MDD-rTMS is the dorsolateral prefrontal cortex (DLPFC)8. However, convergent evidence from neuroimaging, lesion, and stimulation studies identifies the dorsomedial prefrontal cortex (dmPFC) as a potentially important therapeutic target for MDD9 and a variety of other psychiatric disorders characterized by deficits in self-regulation of thoughts, behaviors, and emotional states10. The dmPFC is a region of consistent activation in emotional regulation11, behavioral regulation12,13. The dmPFC is also associated with neurochemical14, structural15, and functional16 abnormalities in MDD.

Described here is the procedure for 20 sessions (4 weeks) of magnetic resonance imaging (MRI) guided rTMS to the dmPFC bilaterally, as a treatment for major depressive disorder. In addition to a conventional 10 Hz protocol applied over 30 min, an intermittent theta burst stimulation protocol (TBS) is discussed, which applies 50 Hz triplet bursts at 5 Hz over a 6 min session17. Both protocols are thought to be excitatory, with the TBS protocol having the potential to achieve comparable effects using a much shorter session18. In both protocols, anatomical MRIs as well as clinical assessments are acquired prior to rTMS. Neuronavigation uses the anatomical scans to account for anatomical variability of dmPFC and optimize the location of rTMS. A relatively new 120°-angled fluid-cooled rTMS coil was also used in order to stimulate deeper midline cortical structures. Finally, rTMS intensity titration was used over the first week of rTMS sessions to ensure that patients could habituate to the higher pain levels associated with dmPFC stimulation as compared to conventional DLPFC stimulation.

ProtocolSubjectSelection

1. Conduct an initial assessment on a prospective patient. The inclusion criteria included the presence of a current depressive episode that is resistant to at least 1 adequate trial of medication, and a Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition, (DSM-5) diagnosis of MDD as established by the assessing psychiatrist. Confirm the diagnosis with a standardized Mini Mental State Examination (MINI).

2. Ensure that patients are on a stable medication or are washed out of their medication routine for at least 4 weeks prior to their first rTMS treatment session. Do not alter this medication regiment throughout rTMS treatment to help disambiguate the cause of any observed clinical improvement or deterioration.

Exclude patients who may have a potential contraindication to rTMS or MRI, including seizure history, cardiac arrhythmia, implanted or foreign devices/metal particulates, unstable medical conditions, or pregnancy. Patients with comorbid post-traumatic stress disorder, obsessive-compulsive disorder, other anxiety disorders, attention deficit hyperactivity disorder, bulimia nervosa or binge eating disorder, or moderate Cluster B personality features are also suitable for this treatment and need not be excluded. Patients with bipolar disorder rather than MDD may also be suitable for this treatment. Patients with psychotic disorders, active substance use, a primary diagnosis of borderline or antisocial personality disorder, or persistent depressive disorder (dysthymia) may be less suitable for treatment and may require exclusion.

AcquiringMagneticResonanceImages

392

1. Acquire patients’ MRIs at any time prior to treatment. Here, use a 3 Tesla scanner with an 8-channel phased-array head coil (refer to Table of Materials), or any scanner capable of created a 3D representation of a patient’s brain.

2. Adhering to local site protocol, acquire a T1-weighted fast spoiled gradient-echo anatomical scan. Use the following parameters: TE = 12 msec, TI = 300 msec, flip angle = 20°, 116 sagittal slices, thickness = 1.5 mm, no gap, 256 x 256 matrix, FOV 240 mm. This scan will be used for real-time rTMS neuronavigation during motor thresholding and treatment sessions.

Preprocessing Anatomical Scans for Real-time Neuronavigation 1. Prepare for MRI guidance using a neuronavigation system. Note: The following steps

employ the Visor 2.0 neuronavigation system (refer to Table of Materials), but other

navigation systems such as the Brainsight TMS Navigation, StealthStation, Aimnav, and

NBS System 4 use similar procedures.

2. Segment anatomical MRIs into its scalp and brain components. Register the two segments into standard stereotactic space, such as Talairach and Tournoux space19.

3. Place target markers by selecting the following points on the MRI: Nasion; Left and right ear, targeting the tragus; Anterior commissure; Posterior commissure; Interhemispheric point (point between the two hemispheres); the anterior most point of the brain; the posterior most point of the brain; the superior most point of the brain; and the left and right most point of the brain.

4. Reconstruct the surfaces of the patient’s scalp and brain in standard space to create a three-dimensional surface-based head model – this image will be used to identify stereotactic scalp coordinates overlying the dmPFC (Talairach and Tournoux coordinate X0, Y+60, Z+60) for optimal coil vertex placement during treatment. Note: This method uses population coordinates to identify the stimulation target. Other methods to identify a stimulation target, outlined in the Discussion, include single-subject anatomy or fMRI activation maps.

5. Register brain and scalp coordinates from stereotactic space to patient space for individualized coil placement.

MotorThresholdAssessment

1. Seat patient in the treatment chair, adjusting the camera for an unobstructed view of the patient.

2. Place the headband with the marker clip attached to it around the patient’s head. The marker clip should sit above bridge of nose.

3. Preprocess the anatomical scan for the patient as described above in Step 3. 4. Load the preprocessed anatomical scans to the neuronavigation program and turn on the

camera. 5. Using a neuronavigation pen, highlight each scalp target point on the patient. The

movements made with the neuronavigation pen will be projected on the television screen in the form of red lines.

393

6. Assess patients’ motor thresholds, the minimum intensity needed to globally excite the motor pathway, prior to rTMS treatment. For this step, begin by having the patient’s lower limbs extended and supported from below, using a stool or a chair equipped with an extensible leg rest.

7. For motor threshold determination, under neuronavigation, target the medial primary motor cortex. Place the coil vertex over the sagittal fissure, 0.5-1.0 cm anterior to the central sulcus. Use an angled or double-cone coil for deeper pulse penetration into medial areas. Use stimulator equipped with a fluid-cooled coil, whose windings are angled at 120° to allow deeper penetration of the pulses (refer to Table of Materials).

8. Perform motor thresholding separately for the left and right hemispheres. Orient the coil laterally to direct rTMS-evoked current flow to the desired hemisphere20. For example, to stimulate the left hemisphere, orient the coil with the handle pointing rightwards and the direction of current flow toward the left hemisphere. Observe the contralateral (right) lower limb for movements during this procedure.

9. Determine threshold and elicited motor movement visually by the halluces longus muscle of the big toe. Note: Unlike conventional motor threshold testing that targets the hand muscle, stimulating the medial wall of the motor cortex will target the toe muscle. Motor evoked potentials (MEPs) may also be used as a more accurate determination of motor threshold, however it is a much lengthier approach.

1. Begin by stimulating at 55% of maximum machine intensity, then adjust upwards or downwards in increments of ~5% depending on whether a response is observed. Reduce the increment size steadily to ~1% as the motor threshold is approached, as previously described21. Stimulate no more frequently than 0.2 Hz (once per 5 sec) to avoid inhibitory or excitatory effects over time.

2. Once a motor threshold is established, move the vertex 1-2 cm anteriorly and posteriorly, in exploratory increments of 2-3 mm, to determine whether any alternative site offers a lower motor threshold. Use the lowest threshold achieved along this arc for each side.

rTMSTreatment&AdaptiveTitration

1. Perform a course of neuronavigated dmPFC-rTMS, using a total of 20-30 daily sessions over 4-6 weeks. For treatments, use the 120° angled, fluid-cooled coil and the parameters listed below for dmPFC stimulation in each treatment session (refer to Table of Materials).

2. Seat the patient in the treatment chair, adjusting the camera for an unobstructed view of the patient.

3. Place a headband with a marker clip attached to it around the patient’s head (placed laterally so as not to block the rTMS coil placement over the medial target site) as described above. Using a camera, the neuronavigation system, will detect the marker clip and will allow for preprocessing and neuronavigation.

4. Load the preprocessed anatomical scans to the neuronavigation program and turn on the camera.

5. Using a neuronavigation pen, highlight each scalp target point on the patient. The movements made with the neuronavigation pen will be projected on the television screen in the form of red lines.

394

6. Place the coil over the dmPFC target under MRI guidance using the neuronavigation system. For verification purposes, this point should lie close to 25% of the distance from nasion to inion. Laterally. Orient the coil laterally, with the handle pointing away from the hemisphere to be stimulated. Stimulate the left hemisphere, then re-orient the coil by 180° to stimulate the right hemisphere, maintaining the vertex in the same location over the dmPFC scalp site.

7. Ensure that the scalp site for dmFPC remains in close contact with the coil itself throughout treatment. Ensure that the patient and operator wear earplugs or other hearing protection during treatment.

8. For 10 Hz stimulation, use a duty cycle of 5 sec on, 10 seconds off for a total of 60 trains (3,000 pulses) per hemisphere per session. Perform this protocol of the left then right hemisphere by orienting the coil laterally, as previously described20. Note: The described protocol for 10 Hz rTMS is outside international safety guidelines (Rossi et al., 2009). There is evidence for its safety18,22.

9. For TBS stimulation, use a duty cycle of 2 sec on, 8 sec off for a total of 600 pulses per hemisphere per session. Perform this protocol of the left then right hemisphere by orienting the coil laterally, as previously described20.

10. Adaptively titrate the rTMS stimulus intensity upwards from an initial value of 20% maximum stimulator intensity, to allow the patient to habituate to the pain and scalp discomfort associated with rTMS during the initial sessions23. Increment the stimulation intensity by 2-5% on each train of stimulation, as tolerated.

1. To assess tolerability, have the patient rate pain on a verbal analogue scale (VAS) from 0 to 10 (0 = no pain, 10 = limit of tolerability without emotional distress) after each train of stimulation is delivered.

11. Begin with a higher stimulation intensity on each session, using a level associated with moderate tolerability (VAS 5-6) from the previous session, until the patient is starting at the target intensity of 120% of resting motor threshold on each hemisphere. Maintain a verbal analogue scale of less than 9 throughout treatments during this titration process. Titration is typically completed in 2-5 days.

12. Monitor the patient for other adverse effects during treatment. Note: The most common treatment-interrupting adverse effect is a syncopal episode, arising during the first or second session of treatment in ~1% of patients. The patient may recount feeling dizzy, faint, or disoriented, and may transiently (~10 sec) lose consciousness. Regular, repeated convulsive movements or post-episode confusion lasting more than a few seconds should be absent, however. In the event of a syncopal episode, lower the headrest on the chair if possible and encourage the patient to remain still until recovered. The session may proceed if the patient is recovered and willing to go on after a few min.

13. Monitor the patient for a generalized tonic-clonic seizure during treatment. Note: These events are rare, and we have not observed a seizure in ~8,000 sessions of dmPFC-rTMS across >200 individual patients to date. Regular, rhythmical, vigorous convulsive movements lasting 10-40 sec, initially around 3 Hz and becoming progressively less rapid, accompanied by unresponsiveness, are suggestive of seizure rather than syncope. However, the two may be difficult to distinguish for an untrained observer.

1. Use video monitoring during all treatments so that the episode can be reviewed by a neurologist at subsequent assessment, if necessary. In the event of such an

395

episode, apply standard seizure first aid steps, including clearing the area of objects with the potential to cause injury, placing the patient on the ground if possible or lowering the treatment chair to the horizontal position if not, laying the patient on the left side if possible, ensuring a clear airway, and ensuring that someone remains with the patient until the seizure terminates and the person regains full alertness.

2. Call emergency services if the seizure does not self-terminate after ~60 sec.

ClinicalDataCollection

1. Collect standardized self-reported questionnaires at baseline, weekly throughout treatment and at follow-up (e.g., 2, 4, 6, 12, and 26 weeks post-treatment). Collect the following self-report data: Beck Depression Inventory (BDI-II)24, and Beck Anxiety Inventory25 on a daily basis throughout treatment.

2. Collect depression severity scores via the clinician-rated 17-item Hamilton Rating Scale for Depression score26 (HamD17) at baseline, weekly during treatment, and at 2, 4, 6, 12 and 26 weeks post-treatment in follow-up.

RepresentativeResultsIn previous work, HamD17 was used as a measure of treatment response for 10 Hz

dmPFC-rTMS. Table 1 displays the pre- and post-treatment HamD17 scores in a previously published case series27. Among all subjects, pre-treatment HamD17 score was 21.66.9 that significantly decreased by 4,331% to 12.58.2 post-rTMS (t22 = 6.54, p <0.0001)27. Using a remission criterion of HamD17 ≤7, 8 of 23 subjects remitted following treatment. Table 2 displays the pre- and post-treatment BDI-II scores in the same case series27. Pre-treatment BDI-II was 32.59.9 and significantly decreased by 34.231.7% to 22.012.8 post-rTMS (t22 = 5.11, p <0.001). HamD17 and BDI-II percent improvement was correlated to determine whether the same subjects responded on both measures (r = 0.72, p = 0.0001).

Adaptive titration was reported in a larger subset of 47 MDD patients undergoing 10 Hz dmPFC-rTMS23. In a case series that included this subset of patients, subjects achieved the target stimulus intensity in 0.91.8 sessions and were able to complete an entire rTMS session at the intended intensity at 4.53.7 sessions23. Adaptive titration was not correlated to treatment improvement.

A comparison of TBS to 10 Hz dmPFC stimulation was recently performed in a recent 185-subject chart review18. Outcomes did not differ significantly between groups. On the HamD17, 10 Hz patients had a 50.6% response and 38.5% remission rate, while TBS patients achieved a 48.5% response and 27.9% remission rate. On the BDI-II, 10 Hz patients had a 40.6% response an 29.2% remission rate, while TBS patients achieved a 43.0% response and 31.0% remission rate18.

396

Table 1: Individual subject HamD17 improvement, using baseline and post-treatment HamD17 scores.

Subject # Pre-Treatment HAMD Post-Treatment HAMD % Improvement

11 21 1 95.24

6 18 2 88.89

4 28 4 85.71

2 12 2 83.33

9 22 4 81.82 25 19 4 78.95

12 20 5 75.00

10 20 5 75.00

14 14 4 71.43

16 26 10 61.54

7 19 8 57.89

24 17 9 47.06

3 19 11 42.11

8 21 14 33.33

5 36 24 33.33

17 23 16 30.43 15 37 27 27.03

23 12 9 25.00

19 28 21 25.00

13 29 22 24.14

1 12 10 16.67

21 13 12 7.69

18 23 22 4.35

22 21 22 -4.76

20 22 24 -9.09

Mean 21.28 11.68 46.28

Standard Dev. 6.68 8.24 31.81

397

Table 2: Individual subject BDI-II improvement, using baseline and post-treatment BDI-II scores.

Subject # Pre-rTMS BDI Post-rTMS BDI % Improvement

11 26 3 88.46

6 21 4 80.95

4 45 9 80.00

2 17 4 76.47

16 36 13 63.89 5 35 17 51.43

3 30 15 50.00

12 26 14 46.15

14 22 12 45.45

1 33 19 42.42

10 34 20 41.18

23 32 19 40.63

9 22 15 31.82

15 57 40 29.82

19 38 28 26.32

7 25 22 12.00 18 45 41 8.89

20 45 43 4.44

17 25 24 4.00

13 44 44 0.00

22 36 37 -2.78

21 30 32 -6.67

8 24 31 -29.17

Mean 32.52 22.00 34.16

Standard Dev. 9.86 12.83 31.70

DiscussionHere, MRI-guided dmPFC-rTMS was applied for treatment-resistant MDD. In general,

rTMS at this site was well tolerated, with mild scalp discomfort and pain at the site of stimulation that was adequately managed using adaptive titration. In open-label trials and a chart review, both 10 Hz and theta burst stimulation resulted in significant improvements in depressive severity as measured by the HamD17 and BDI-II.

There are two critical steps worth noting in the rTMS treatment procedure for optimal dmPFC stimulation. First, an angled, double-cone coil allows for optimal stimulation of deeper structures within the medial aspect of the prefrontal cortex28. Second, a treatment stimulation intensity of 120% resting motor threshold at this medial site is well-tolerated and without serious adverse events, despite the relatively high intensity of the applied stimulation in absolute terms when compared to the lower absolute intensities required for conventional DLPFC-rTMS. This

398

same intensity also appears to be safe and tolerable for TBS protocols with dmPFC-rTMS, notwithstanding the significantly lower values of 80% active motor threshold more commonly used with TBS18. As previously mentioned, significant pain and discomfort is associated with anterior medial prefrontal stimulation at higher intensities29. Adaptive titration was quickly and successfully used to aide in rTMS-related discomfort adaptation. In sum, the use of an angled rTMS coil and relatively high stimulation intensity (with adaptive titration) may allow for deeper penetration of stimulation to the medial prefrontal and underlying cingulate cortices28, without incurring higher risks of seizure of intolerable scalp pain.

Neuronavigation is often used for precise individualized anatomical landmarking for coil vertex placement. However, one problem with MRI-guided neuronavigation is that it potentially omits the functional relationships of the desired stimulation target to other brain regions in favor of anatomical specificity across subjects. Indeed, there is significant functional connectivity variability found in association cortices, including regions of prefrontal cortex, which may impede treatment efficacy30. For example, a recent study used resting-state functional connectivity to show that left DLPFC-rTMS treatment efficacy in MDD was dependent on left DLPFC connectivity to the subgenual cingulate cortex31. Patients that improved with left DLPFC-rTMS tended to have anticorrelated functional connectivity between the DLPFC and the subgenual cingulate cortex at baseline. Therefore, resting-state functional connectivity could be harnessed to further optimize target placement and identify potential biomarkers once the functional characteristics of response are identified32.

One major limitation of rTMS as a treatment is that it is unclear how certain stimulation parameters influence its treatment efficacy. There is substantial variability in the parameters of conventional left DLPFC stimulation for MDD across studies, and there is also increasing evidence of substantial inter-individual variability in how some rTMS parameters affect cortical excitation and inhibition or treatment efficacy33,34. For example, the effects of 10 Hz stimulation on motor evoked potential (MEP) was recently shown to vary considerably across subjects, with some showing decreasesrather than increases in MEP strength after stimulation35. Other rTMS treatment parameters that potentially require further optimization (or individualization) to maximize treatment efficacy include the number of pulses per session, the number of sessions per day, stimulation intensity and the duty cycle (how many seconds stimulation is on and off per cycle).

There are also general limitations to rTMS as a treatment. These include the logistical requirements for patients to make multiple visits to hospital for treatment, limited access to treatment for patients in remote areas, the high cost of treatment (>$250 per session) with conventional parameters, and the low volumes of patients who can be treated per device using conventional parameters (1-2 per hr at most). Parameter optimization may help to address some of these problems in future. Other forms of non-invasive stimulation, such as transcranial direct current stimulation (tDCS), may also come to serve as a less expensive alternative to rTMS, suitable for use at home rather than in the clinic36.

Despite its technical limitations, dmPFC-rTMS is clinically promising for treatment-resistant MDD. rTMS, and dmPFC-rTMS in particular, may also probe to be a promising option in other medication-resistant psychiatric illnesses including eating disorders10, obsessive-compulsive disorder37, and post-traumatic stress disorder38. Identifying good treatment candidates for these disorders may require additional tools other than traditional symptom-based diagnostic classification schemas – in particular, neuroimaging. Acquiring patient neuroimaging data before and after treatment allows for the identification of potential biological pre-treatment

399

predictors and mechanisms of treatment response. Dorsomedial and subgenual cingulate resting-state functional connectivity have been identified as potential predictors to treatment response27. Additionally, graph theoretical measures such as betweenness centrality have been shown to differentiate dmPFC-rTMS responders and non-responders at baseline based on subscales for hedonic responses23. Neuroimaging also points to anterior mid-cingulate cortex and dorsomedial thalamic resting state functional connectivity change that correlates to treatment response27. In sum, functional neuroimaging may become a useful clinical tool as potential predictors and mechanisms of treatment response are identified.

Since current dmPFC-rTMS studies have used an open-label design, future directions should include the creation of a sham-controlled trial to assess its therapeutic efficacy in MDD versus sham and conventional stimulation. However, creating a convincing sham-control arm is technically challenging, particularly for simulating somatosensory or nociceptive sensations, as well as convincingly blinding the rTMS technician39. In a recent meta-analysis, over half of patients were able to correctly guess their treatment arm39. In another meta-analysis, placebo effects were large, but comparable to escitalopram trials40. Future studies involving a rTMS sham arm should consider a design that addresses all sensory aspects of rTMS for both the patient and the technician. Nonetheless, augmenting magnetic stimulation techniques through TBS41, priming stimulation42 or adjunctive cognitive behavioral therapy43 or pharmacotherapy44 may also help to optimize the therapeutic effects of rTMS. TBS in particular has the potential to achieve significant improvements in treatment duration and thus in patient volumes, access times, and treatment cost, while achieving equivalent outcomes to much longer conventional protocols18,45.

In summary, rTMS of the dmPFC is a promising novel approach to therapeutic brain stimulation for treatment-resistant MDD. By incorporating the use of a MRI-guided neuronavigation system, a fluid-cooled, 120° angled stimulation coil, a high stimulation intensity and an adaptive titration schedule, dmPFC-rTMS can be safely and accurately delivered to deep targets in the medial prefrontal cortex. As these regions are central to the pathophysiology of many neuropsychiatric disorders, this approach may have promising applications not only for MDD, but also for a variety of other psychiatric conditions that are resistant to standard treatments.

References1. Fitzgerald PB, Fountain S, Daskalakis ZJ. A comprehensive review of the effects of rTMS on motor

cortical excitability and inhibition. Clinical Neurophysiology. 2006;117:2584–2596.

2. Pascual-Leone A, Gates JR, Dhuna A. Induction of speech arrest and counting errors with rapid-rate transcranial magnetic stimulation. Neurology. 1991;41:697–702.

3. Young L, Camprodon JA, Hauser M, Pascual-Leone A, Saxe R. Disruption of the right temporoparietal junction with transcranial magnetic stimulation reduces the role of beliefs in moral judgments. Proceedings of the National Academy of Sciences of the United States of America. 2010;107:6753–6758.

4. Hilgetag CC, Théoret H, Pascual-Leone A. Enhanced visual spatial attention ipsilateral to rTMS-induced “virtual lesions” of human parietal cortex. Nature neuroscience. 2001;4:953–957.

5. Berman RM, et al. A randomized clinical trial of repetitive transcranial magnetic stimulation in the treatment of major depression. Biological psychiatry. 2000;47:332–337.

6. Van den Eynde F, et al. Repetitive transcranial magnetic stimulation reduces cue-induced food craving in bulimic disorders. Biological psychiatry. 2010;67(8):793–795.

400

7. Berlim MT, Neufeld NH, Vanden Eynde F. Repetitive transcranial magnetic stimulation (rTMS) for obsessive-compulsive disorder (OCD): an exploratory meta-analysis of randomized and sham-controlled trials. Journal of psychiatric research. 2013;47(8):999–1006.

8. Fitzgerald PB, et al. A randomized trial of unilateral and bilateral prefrontal cortex transcranial magnetic stimulation in treatment-resistant major depression. Psychological Medicine. 2011;41:1187–1196.

9. Downar J, Daskalakis ZJ. New targets for rTMS in depression: A review of convergent evidence. Brain Stimulation. 2013;6:231–240.

10. Downar J, Sankar A, Giacobbe P, Woodside B, Colton P. Unanticipated Rapid Remission of Refractory Bulimia Nervosa, during High-Dose Repetitive Transcranial Magnetic Stimulation of the Dorsomedial Prefrontal Cortex: A Case Report. Frontiers in psychiatry. 2012;3(30):1–5.

11. Gallinat J, Brass M. Keep Calm and Carry On”: Structural Correlates of expressive suppression of emotions. PLoS ONE. 2011;6:e1–e4.

12. Langner R, Cieslik EC, Rottschy C, Eickhoff SB. Interindividual differences in cognitive flexibility: influence of gray matter volume, functional connectivity and trait impulsivity. Brain structure, & function. 2014.

13. Jung Y-C, et al. Synchrony of anterior cingulate cortex and insular-striatal activation predicts ambiguity aversion in individuals with low impulsivity. Cerebral cortex. 2014;24(5):1397–1408.

14. Auer DP, Pütz B, Kraft E, Lipinski B, Schill J, Holsboer F. Reduced glutamate in the anterior cingulate cortex in depression: An in vivo proton magnetic resonance spectroscopy study. Biological Psychiatry. 2000;47:305–313.

15. Bora E, Fornito A, Pantelis C, Yucel M. Gray matter volume in major depressive disorder: a meta-analysis of voxel-based morphometry studies. Psychiatry research. 2013;211(1):37–46.

16. Sheline YI, Price JL, Yan Z, Mintun MA. Resting-state functional MRI in depression unmasks increased connectivity between networks via the dorsal nexus. Proceedings of the National Academy of Sciences of the United States of America. 2010;107:11020–11025.

17. Huang Y-Z, Edwards MJ, Rounis E, Bhatia KP, Rothwell JC. Theta burst stimulation of the human motor cortex. Neuron. 2005;45:201–206.

18. Bakker N, et al. rTMS of the dorsomedial prefrontal cortex for major depression: safety, tolerability, effectiveness, and outcome predictors for 10 Hz versus intermittent theta-burst stimulation. Brain Stimulation. 2014;In Press:1–22.

19. Talairach J, Tournoux P. Co-planar stereotaxic atlas of the human brain: 3-dimensional proportional system: an approach to cerebral imaging. Neuropsychologia. 1988;39:145.

20. Terao Y, et al. A single motor unit recording technique for studying the differential activation of corticospinal volleys by transcranial magnetic stimulation. Brain Research Protocols. 2001;7:61–67.

21. Schutter DJLG, van Honk J. A standardized motor threshold estimation procedure for transcranial magnetic stimulation research. The journal of ECT. 2006;22:176–178.

22. Downar J, Geraci J, et al. Anhedonia and Reward-Circuit Connectivity Distinguish Nonresponders from Responders to Dorsomedial Prefrontal Repetitive Transcranial Magnetic Stimulation in Major Depression. Biological psychiatry. 2013. pp. 1–26.

23. Downar J, Geraci J, et al. Anhedonia and Reward-Circuit Connectivity Distinguish Nonresponders from Responders to Dorsomedial Prefrontal Repetitive Transcranial Magnetic Stimulation in Major Depression. Biological Psychiatry. 2014;76(3):176–185.

24. Beck AT, Steer RA, Brown GK. Manual for the Beck depression inventory-II. San Antonio, TX: Psychological Corporation; 1996. pp. 1–82.

25. Beck AT, Epstein N, Brown G, Steer RA. An inventory for measuring clinical anxiety: psychometric properties. Journal of consulting and clinical psychology. 1988;56:893–897.

26. Hamilton MC. Hamilton Depression Rating Scale (HAM-D) REDLOC. 1960;23:56–62.

401

27. Salomons TV, et al. Resting-State Cortico-Thalamic-Striatal Connectivity Predicts Response to Dorsomedial Prefrontal rTMS in Major Depressive Disorder. Neuropsychopharmacology. 2014;39:488–498.

28. Hayward G, et al. Exploring the physiological effects of double-cone coil TMS over the medial frontal cortex on the anterior cingulate cortex: an H2(15)O PET study. The European journal of neuroscience. 2007;25:2224–2233.

29. Vanneste S, Ost J, Langguth B, De Ridder D. TMS by double-cone coil prefrontal stimulation for medication resistant chronic depression: a case report. Neurocase. 2014;20(1):61–68.

30. Mueller S, et al. Individual Variability in Functional Connectivity Architecture of the Human Brain. Neuron. 2013;77:586–595.

31. Fox MD, Buckner RL, White MP, Greicius MD, Pascual-Leone A. Efficacy of transcranial magnetic stimulation targets for depression is related to intrinsic functional connectivity with the subgenual cingulate. Biological Psychiatry. 2012;72:595–603.

32. Fox MD, Liu H, Pascual-Leone A. Identification of reproducible individualized targets for treatment of depression with TMS based on intrinsic connectivity. NeuroImage. 2013;66:151–160.

33. Kedzior K, Azorina V, Reitz S. More female patients and fewer stimuli per session are associated with the short-term antidepressant properties of repetitive transcranial magnetic stimulation (rTMS): a meta-analysis of 54 sham-controlled studies published between 1997-2013. Neuropsychiatric disease and treatment. 2014;10:727–756.

34. Lee JC, Blumberger DM, Fitzgerald PB, Daskalakis ZJ, Levinson AJ. The Role of Transcranial Magnetic Stimulation in Treatment-Resistant Depression: A Review. Current Pharmaceutical Design. 2012;18:5846–5852.

35. Maeda F, Keenan JP, Tormos JM, Topka H, Pascual-Leone A. Interindividual variability of the modulatory effects of repetitive transcranial magnetic stimulation on cortical excitability. Experimental Brain Research. 2000;133:425–430.

36. Brunoni AR, Ferrucci R, Fregni F, Boggio PS, Priori A. Transcranial direct current stimulation for the treatment of major depressive disorder: a summary of preclinical, clinical and translational findings. Progress in neuro-psychopharmacology, & biological psychiatry. 2012;39:9–16.

37. Mantovani A, Simpson HB, Fallon BA, Rossi S, Lisanby SH. Randomized sham-controlled trial of repetitive transcranial magnetic stimulation in treatment-resistant obsessive-compulsive disorder. The international journal of neuropsychopharmacology. 2010;13:217–227.

38. Watts BV, Landon B, Groft A, Young-Xu Y. A sham controlled study of repetitive transcranial magnetic stimulation for posttraumatic stress disorder. Brain Stimulation. 2012;5:38–43.

39. Berlim MT, Broadbent HJ, Van den Eynde F. Blinding integrity in randomized sham-controlled trials of repetitive transcranial magnetic stimulation for major depression: a systematic review and meta-analysis. The international journal of neuropsychopharmacology 2013;16:1173–1181.

40. Brunoni AR, Lopes M, Kaptchuk TJ, Fregni F. Placebo response of non-pharmacological and pharmacological trials in major depression: a systematic review and meta-analysis. PLoS One. 2009;4:e4824.

41. Chistyakov AV, Rubicsek O, Kaplan B, Zaaroor M, Klein E. Safety tolerability and preliminary evidence for antidepressant efficacy of theta-burst transcranial magnetic stimulation in patients with major depression. The international journal of neuropsychopharmacology 2010;13:387–393.

42. Iyer MB, Schleper N, Wassermann EM. Priming stimulation enhances the depressant effect of low-frequency repetitive transcranial magnetic stimulation. The Journal of neuroscience. 2003;23:10867–10872.

43. Vedeniapin A, Cheng L, George MS. Feasibility of simultaneous cognitive behavioral therapy and left prefrontal RTMS for treatment resistant depression. Brain Stimulation. 2010;3:207–210.

44. Rumi DO, et al. Transcranial magnetic stimulation accelerates the antidepressant effect of amitriptyline in severe depression: A double-blind placebo-controlled study. Biological Psychiatry. 2005;57:162–166.

402

45. Platz T, Rothwell JC. Brain stimulation and brain repair--rTMS: from animal experiment to clinical trials--what do we know. Restorative neurology and neuroscience. 2010;28:387–398.

403

AppendixII.This review was published in the Annals of the New York Academy of Science:

Dunlop, K., Hanlon, C.A., Downar, J. (2017). Noninvasive brain stimulation treatments for addiction and major depression. Annals of the New York Academy of Science, 1394(1): 31-54.

Abstract

Major depressive disorder (MDD) and substance use disorders (SUDs) are prevalent, disabling, and challenging illnesses for which new treatment options are needed, particularly in comorbid cases. Neuroimaging studies of the functional architecture of the brain suggest common neural substrates underlying MDD and SUDs. Intrinsic brain activity is organized into a set of functional networks, of which two are particularly relevant to psychiatry. The salience network (SN) is crucial for cognitive control and response inhibition, and deficits in SN function are implicated across a wide variety of psychiatric disorders, including MDD and SUDs. The ventromedial network (VMN) corresponds to the classic reward circuit, and pathological VMN activity for drug cues/negative stimuli is seen in SUDs/MDD. Noninvasive brain stimulation (NIBS) techniques, including rTMS and tDCS, have been used to enhance cortico–striatal–thalamic activity through the core SN nodes in the dorsal anterior cingulate cortex, dorsolateral prefrontal cortex, and anterior insula. Improvements in both MDD and SUD symptoms ensue, including in comorbid cases, via enhanced cognitive control. Inhibition of the VMN also appears promising in preclinical studies for quenching the pathological incentive salience underlying SUDs and MDD. Evolving techniques may further enhance the efficacy of NIBS for MDD and SUD cases that are unresponsive to conventional treatments.

Introduction

Major depressive disorder (MDD) and substance use disorders (SUDs) are challenging illnesses that produce significant burdens on patients and the healthcare system. Mental illness and SUDs are the leading worldwide cause of years lived with a disability (YLD),1 and MDD is the second leading psychiatric cause of YLD.2 The societal burden of MDD and SUDs have also dramatically increased over the last 20 years,2 emphasizing the importance of access to care and effective treatment options.

For MDD, the mainstays of conventional treatment are pharmacotherapy and psychotherapy. Studies of real-world effectiveness suggest that about one-third of patients will remit on an initial trial of antidepressant medication, while another one-third will remit after 1–3 additional medication trials.3,4 The remaining one-third of patients are labeled as having treatment-resistant depression (TRD), with a low likelihood of remission (10– 15%) on further trials.4 TRD affects approximately 2% of the general population.5 To address the challenge of treating TRD, combination therapies (antidepressant + antipsychotic, or antidepressant + anticonvulsant)6–9 and electroconvulsive therapy (ECT)10 have yielded promising clinical results. Even for TRD, however, these intensive interventions achieve varied remission rates between approximately 30%11 and 50%,6–8 and the relapse rate following ECT is 50% by 2 years.12

404

As with MDD, in SUD patients the mainstays of conventional treatment are pharmacotherapy and psychotherapy. Lifetime rates of drug use are 92% for alcohol, 74% for tobacco, 42% for cannabis, and 16% for cocaine, in the United States.13 The lifetime prevalence of alcohol dependence is 13%, and up to 3% for illicit substances.14–16 The most common psychotherapeutic approach to substance dependence is cognitive behavioral therapy.17,18 A meta-analysis of 34 randomized controlled trials on cognitive behavioral therapy for SUDs demonstrated that the average effect size was moderate (d = 0.45), with the highest effects for cannabis, cocaine, and opioid treatment.19 A form of behavioral therapy known as contingency management appears to be a particularly potent tool for multiple classes of SUD patients.20,21 Contingency management, however, requires that an individual is able to regulate/control their drug intake in order to get an alternative non-drug reinforcer. This may be difficult for many patients, as disruptions in response inhibition and in the neural circuitry required for response inhibition are hallmarks of addiction.22

From a pharmacotherapy perspective, the therapeutic approach varies with the substance being abused. Pharmacotherapy may complement behavioral approaches by replacing the abused substance with a less harmful substitute23,24 and thereby reduce the social and personal harm associated with the drug. However, dependences on several classes of drugs, including cocaine, have no approved pharmacotherapy, and relapse rates in the first 6 months after an outpatient treatment program are often higher than 75%.25 Some evidence also suggests that pharmacotherapy that interacts with neurotransmitter systems related to reward learning may enhance impulse control itself.26

High rates of treatment resistance and relapse represent a common challenge for MDD or SUDs. However, effective MDD and SUD treatment response is frequently hampered even further by the comorbidity of the two conditions. Among individuals with MDD, approximately 25– 40% have a comorbid SUD. Conversely, MDD is among the common psychiatric disorders that have high comorbidity with all types of SUDs.27,28 These patients report poorer response rates compared to their singly diagnosed counterparts from a 12-step program,29 from single-medication trials,30–32 and from cognitive behavioral therapy.33 Likewise, MDD patients with nicotine dependence also have increased difficulty with smoking cessation, with antidepressants having little influence on abstinence, and these patients are more likely to develop an episode of depression post–smoking cessation.34–37 Hence, treatment strategies should ideally accommodate the frequent comorbidity of these illnesses.

Treatment strategies for MDD and SUDs are traditionally approached separately and sequentially. Such an approach implicitly assumes that if the primary diagnosis is addressed, secondary diagnoses may resolve on their own.38 However, limited success rates for conventional approaches to MDD and SUDs raise several fundamental questions. Is it helpful to identify one of the disorders—MDD or SUD—as the primary disorder? Does the underlying pathophysiology of MDD and SUDs support separate treatment strategies in comorbid cases? Finally, can the neurobiology of SUDs and MDD suggest new treatment strategies that might address the problem of comorbidity, while improving remission and relapse rates for both disorders?

To answer these questions, we can take advantage of recent advances described in the neuroimaging literature on human brain function. These include an increasingly robust model of the functional neuroanatomy and network architecture of the brain, a vast body of neuroimaging

405

literature on the pathophysiology of SUDs and MDD in humans, and, finally, a set of automated, quantitative metaanalytic software that renders this enormous literature more tractable. Such tools allow us to detect consistent and statistically robust patterns by combining scans in thousands of patients and healthy individuals.

Of course, advances in the functional neuroanatomy of SUDs and MDD are of limited immediate clinical interest unless they can translate into anatomically specific therapeutic interventions. While it is difficult to aim conventional treatments at specific brain circuits, noninvasive brain stimulation (NIBS) is an emerging treatment modality that exerts neuroanatomically specific effects. There are two particular NIBS technologies that are quickly translating from research into clinical practice: repetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS). Repetitive TMS has held regulatory approval for MDD in many jurisdictions for several years and is being explored for SUDs in clinical trials. Transcranial DCS is a few years behind rTMS in translational progress but is likewise being explored with encouraging results in both MDD and SUDs. Unlike medications or therapy, the success of NIBS depends critically on the choice of target circuit and the type of stimulation (inhibitory or excitatory) applied to that circuit.39–41

This review is intended to serve four purposes. First, it will summarize key advances in our understanding of the functional architecture of the brain. Second, it will review how this emerging model of brain function relates to the pathophysiology behind SUDs and MDD. Third, it will review NIBS for the treatment of SUDs and MDD, particularly in cases where conventional treatments fail. Finally, it will review a number of promising areas for future research on NIBS in SUDs and MDD.

Functionalarchitectureofthehumanbrain

Functional networks of intrinsic brain activity

One of the key advances in the neuroimaging literature over the last 20 years is the demonstration that brain regions organize their activity into coherent functional networks.42 On functional magnetic resonance imaging (fMRI), these networks appear as correlations of the low-frequency fluctuations in blood oxygenation level–dependent (BOLD) signal between brain regions.43 Many networks were originally identified through data-driven methods for analyzing brain activity at rest and are termed resting-state networks. However, these networks reliably appear in ongoing brain activity during tasks as well,44 and meta-analyses of task-based activation also reveal consistent functional networks similar to those identified at rest.45

The precise number of functional networks and the functional role of each network are still being studied. An emerging consensus has been identified between 7 and 20 distinct functional networks.43 One recent resting-state fMRI analysis in 1000 healthy individuals found a stable seven network parcellation; these networks were further subdivided into a stable 17-network parcellation (Fig. 1).46 Many of these cortical networks also have correlated counterparts in the striatum47 and cerebellum,48 thus hinting that they may represent integrated, whole-brain functional circuits.

406

The default-mode network (DMN)49 is the best known and most studied of these functional networks. The DMN contains subcomponents that serve various introspective functions ranging from mind wandering,50 recollection and prospection,51 rumination,52 and self-reflection.51 Other networks act in opposition to the DMN and activate during behaviorally regulated task performance and externally focused cognition. These networks include frontoparietal networks (FPNs) and related areas sometimes known as the dorsal (DAN) and ventral attention networks (VAN).46 Other networks include lower-level sensory and motor cortices and ventromedial limbic networks that encompass the temporal pole and the orbitofrontal cortex (OFC)46 (Fig. 1).

Relevance of functional networks to MDD and SUDs Several functional networks have been studied extensively in MDD and SUDs,

particularly the DMN. However, there are two functional networks that are of particular interest in both SUDs and MDD, and it is worthwhile to briefly characterize these networks.

The first network of interest is the anterior cinguloinsular network (aCIN), or salience network (SN).46 Figure 2 visualizes the SN on the basis of a quantitative meta-analysis of SN neuroimaging studies analyzed using Neurosynth.45 The SN activates for salient sensory events,53,54 transitions from introspection to task performance,55 and during task initiation and switching.56 A remarkable recent study highlighted the central importance of the SN as a common neural substrate across psychiatric illness categories.57 The authors performed a meta-analysis of structural abnormalities across six psychiatric disorder categories, including MDD and SUDs, and found that all of them showed gray matter volume reductions in the dorsal anterior cingulate cortex (dACC) and anterior insula. Further analyses demonstrated that these areas belonged to a common functional network in healthy controls, both at rest and during task performance.57 This network corresponded almost exactly to the SN.45,46 Thus, of the 7–17 networks previously discussed, the SN merits particular attentionfor its role inMDD, SUDs, and other psychiatric pathophysiology.

The second network of interest is the ventromedial network (VMN), encompassing the nucleus accumbens (NAcc), medial OFC (mOFC), and ventromedial prefrontal cortex (VMPFC) (Fig. 3, created using Neurosynth33). This circuit is best known as the reward pathway and also includes components of the mediodorsal thalamus and midbrain dopaminergic structures. The VMN activates not only for rewards, but also for stimuli of other incentive value, including losses.58 The role of the VMN in SUDs is well documented,59,60 particularly for mediating abnormal incentive salience of drug cues, which is proposed to drive craving and relapse.61 VMN dysfunction is linked to anhedonia in MDD62 and activates paradoxically for negative stimuli, suggesting that negative cues have abnormal incentive salience in MDD, analogous to drug cues in SUDs.63

The opposed functions of the SN and VMN

The SN is critical for “switching” brain activity between the introspective DMN and the externally focused FPNs.64 It is also active during performance-related errors and task initiation.56,65 The functions of the SN map rather well on to the Research Domain Criteria (RDoC) construct of cognitive control: specifically, the subdomain of response inhibition/response selection. A quantitative fMRI meta-analysis using Neurosynth45 revealed a network closely matching the SN (Fig. 2). This function stands in opposition to the functions of

407

the VMN in mediating reward, incentive salience, and value assignment in similar meta-analyses using Neurosynth45 (Fig. 3). These two circuits have been proposed to play opposing roles in behavioral regulation: the VMN as a “drive network” mediating craving and urge, and the SN as a “gatekeeper network” mediating response selection and inhibition.66

If the roles of these networks are indeed opposed, we might predict one to be active when the other is not, and vice versa.67 In fact, this is precisely the case. For many of the functional networks previously discussed, there also exists a map of corresponding negatively correlated regions: the anti-network.68 The anti-network for the VMN, shown by qualitative meta-analysis, consists of the SN and corresponding striatothalamic partners. Conversely, the anti-network for the SN is the VMN, as evident in functional connectivity data45 (Fig. 4). Thus, the SN and VMN may play reciprocal roles as each other’s anti-networks, with opposing patterns of activity even in the resting brain. This functional architecture might explain the tendency for highly salient stimuli to reduce inhibitory control or, conversely, for cognitively demanding activities to attenuate urges, cravings, or emotions.

Coordination among the functional networks

Modeling the brain as a network can allow us to map information flow among its regions. Approaches to the study of the brain using graph theory, a set of mathematical techniques allowing formal analysis of network structures, consider individual brain regions as “nodes” and functional connections between correlated regions/nodes as “edges.” Such analyses identify clique-like “communities” of nodes and “hubs” that bridge together different communities within a larger network.69 This means that graph theoretical approaches can characterize the structural and functional interactions between networks and nodes.70

Reassuringly, graph theoretical analyses of the brain extract 10–12 communities that correspond to the same data-driven functional networks70 and also show how these communities connect to one another (Fig. 5A). Generally, sensory andmotor networks lie on the functional outskirts of the brain, as isolated functional units with few connections outside of their network. This makes sense given their roles as pathways for processing elementary sensory input and motor output.

What about other prominent networks, such as the DMN itself? The intuitive expectation is that the DMN should be situated centrally, in a crossroads position linking the other networks together. Instead, the DMN appears as a peripheral, closeknit module much like the visual cortex, but taking feedback from the VMN rather than the retina. The DMN’s position is not that of an integrative controller, but rather another specialized “think tank,” linking the incentive functions of the VMN to the high-level executive and cognitive functions of the FPNs, but not in close communication with most other networks (Fig. 5B).

Does another functional network hold the crossroads position? One recent study using graph theoretical methods identified nodes standing in critical hub positions between many networks. These nodes lay not in the DMN but in the SN itself.71 Close inspection of the whole-brain graph showed that the SN is in a gatekeeper position between the deliberative functions of the DMN and FPN and the behavioral output of the somatomotor cortex (Fig. 5B).

408

This overall functional architecture provides a framework for understanding the roles of the VMN, DMN, FPN, SN, and motor cortex in behavioral control (Fig. 5C). The trajectory of information flow originates within the basic drives of midbrain dopaminergic regions, which are elaborated into specific incentives or urges via the VMN and elaborated further into goals and mapped onto relevant scenarios via the DMN. The FPNs then develop these goals and scenarios into consummatory strategies, which are broken down into cognitions and behaviors that can be either inhibited or selected via the gatekeeper-like SN. If selected, responses finally map to the action-control systems of the somatomotor cortex to initiate overt behavior.

This urges-to-actions trajectory through the brain offers early and late points for pathology to arise in MDD or SUDs. Early pathology at the VMN–DMN interface would drive priorities away from core survival needs, assigning inappropriately high incentive salience to drug cues (in the setting of SUDs) or negative emotional cues (in the setting of MDD). Late pathology near the SN would interfere with response selection and inhibitory control, producing a pattern of chronic emotional liability, intrusive thinking, and impulsive behavior. These more elementary dysfunctions may cut across the diagnostic entities of MDD and SUDs, and may also shed light on MDD and SUD comorbidity as arriving from more fundamental disruptions in network architecture. Importantly, the neural overlap between MDD and SUDs may be an opportunity, rather than an obstacle, for interventions that can address both illnesses concurrently by targeting the underlying circuitry of the SN and VMN using NIBS.

Functionalarchitectureinpsychiatricdisorders

Neural circuit disruptions in SUDs

It is challenging to concisely summarize the neural circuits that are involved in all SUDs, since each substance has a unique pharmacologic profile and addiction involves multiple temporal phases, including regular controlled use, regular uncontrolled/habitual use, abstinence, and relapse. However, there appear to be common neurobiological pathways operating across substances of abuse, involving dysfunction of both the VMN and its anti-network, the SN.

Although patients with SUDs can become dependent on a variety of drugs that have diverse initial pharmacologic actions, nearly all drugs of abuse have a common final pathway in which they modulate reward circuitry in the ventral striatum–ventral tegmental area (VTA) pathway.72 For example, cocaine modulates this pathway directly by increasing dopamine transmission, while nicotine also has direct actions on the VTA via nicotinic receptors. Opiates and alcohol, however, modulate the ventral striatum and VTA indirectly via GABAergic disinhibition. While the majority of basic science research in addiction has focused on this subcortical reward pathway, recent work has demonstrated that many cortical and subcortical regions interact with this mesolimbic dopamine pathway, especially the VMN and its anti-network, the SN, as described below.

First, the SN is hypoactive in SUDs. Specifically, in SUD patients, the dACC and insula display hypoactivity to a variety of executive tasks, including the Stroop task,73,74 response inhibition75,76 (although this finding is mixed77), and emotion regulation.78 The dACC also

409

displays abnormal connectivity on resting-state and graph theoretical measures in SUDs.79 The abnormal network activation in the dACC may relate to the aberrant salience of substance-specific cues.80 SUD patients display lowered structural and functional integrity in both the dACC and insula.81,82 Although the role of the insula is still under debate, insular lesions have been reported to be associated with improved abstinence, possibly because disease-relevant cues are less salient.83,84 There is also an increase in dACC activation and an improvement in self-control behavior following abstinence, suggesting a causal role in SUD psychopathology.85

The anti-network of the SN, the VMN, is involved in limbic arousal in SUDs. Functional MRI studies have demonstrated that cue reactivity and craving are associated with elevated activity in the VMN, particularly in the ventral striatum, vmPFC, and OFC.86–88 Hyperactivation from the OFC is also seen during reward evaluation,89–91 risky decision making, personal relevance,92–95 and goal-driven processes associated with the ventral striatum. This abnormal VMN activity normalizes in smokers and cocaine users following treatment with varenicline/buproprion96–98 and methylphenidate,99 respectively. As in MDD, SUD patients also show VMN hypoactivation for natural rewards, compared to healthy controls.100

Connectivity between the VMN and SN is also impaired in SUDs, with SUD patients showing reduced corticostriatal VMN and SN resting-state functional connectivity.101,102 Furthermore, altered ACC corticocortical and corticostriatal connectivity between these circuits are related to SUD severity,101 high-risk behavior, and genetic polymorphisms associated with SUDs.103

Neural circuit disruptions in MDD

As with SUDs, MDD has been linked to a pattern of SN hypoactivity and VMN hyperactivity104 in a wide variety of network-based descriptions of the illness over the last decade.105,106 Ventral prefrontal activation in MDD predicted dorsal inactivation, in keeping with the framework of opposed SN and VMN activity.107 Inhibition of SN control mechanisms by pathological salience signals from the VMN may create a self-perpetuating imbalance, resulting in the persistent low mood of MDD rather than the transient low mood of normal sadness. MDD patients display a similar disruption of incentive salience for primary rewards108,109 and instead become attuned to disease-specific stimuli. For example, the rostral ACC and ventral striatum are hypoactive during reward feedback110–112 and hyperactive during self-referential negative processing.113,114 MDD patients also show an absence of ventral striatal and VMPFC/OFC inactivation for pleasant sights and tastes but heightened activation in the caudate in response to viewing aversive images.115 Altered responsesfrom the subgenual cingulate and OFC are also found in MDD during rumination and negative self-focus.116

As in SUDs, MDD patients show reduced activation in SN regions at rest. Reflecting this deficiency, MDD patients exhibit task-related hyperactivation from the dACC on many cognitive control tasks, including the Stroop117,118 and n-back tasks.119 The DMPFC is also inappropriately activated during positive affect processing in MDD, which normalizes after successful treatment.120 Additionally, the anterior insula and DMPFC are inappropriately active during negative affect.121 Increased activation after treatment also results in improved resting-state activity in the cingulate in MDD.122 On structural imaging, TRD patients tend to show

410

volumetric alterations in regions of the SN and cognitive deficits relative to their treated counterparts.123

Neural similarities in MDD/SUDs

As noted earlier, voxel-based morphometry studies of MDD and SUDs reveal a common neural substrate of decreased gray matter affecting the dACC and anterior insula—both key nodes of the SN.57 One interpretation of this neural overlap is that both SUDs and MDD feature deficits in self-regulation of impulses, cognition, emotion, and behavior—or in RDoC terms, response inhibition/selection. SUDs and MDD also share a common feature of inappropriate VMN activation as shown by fMRI. In SUDs, the VMN’s reward circuit activates in response to drug cues, despite their lack of primary survival value. In MDD, the VMN activates in response to negative rather than positive affective stimuli. One interpretation is that, for MDD, negative emotional stimuli acquire aberrant incentive salience, as with drug cues in SUDs. If so, both SUD and MDD patients also share an underlying pathology of distorted incentive salience.63

The shared neural substrates of SUDs and MDD may help to account for the high prevalence of MDD/SUD comorbidity. From a therapeutic standpoint, the shared neural circuitry between SUDs and MDD may also present an opportunity to address both disorders concurrently, using NIBS interventions. On this view, NIBS treatments should not be regarded as antidepressant or antisubstance use. Rather, they should be considered to offer two more nuanced approaches to treatment, cutting across diagnostic categories: enhancing cognitive control by targeting the nodes of the SN or relieving the distorted incentive salience of substance/negative cues by inhibiting the VMN.

One prediction of thisframework is that any NIBS interventions that target SN nodes (i.e., DLPFC, DMPFC, or anterior insula) with excitatory stimulation should exert a therapeutic effect across both MDD and SUDs, by enhancing cortico–striatal– thalamic connectivity through the SN on fMRI and by enhancing capacity for response selection/inhibition. Another prediction is that any NIBS interventions that target the VMN (by stimulating the frontal pole or VMPFC) should also exert a therapeutic effect across both MDD and SUDs by reduced cortico–striatal–thalamic connectivity through the VMN on fMRI, and by reducing incentive salience of drug cues/negative affective stimuli.

NIBSasatreatmentforMDD/SUDs

Repetitive TMS overview

Repetitive TMS applies powerful, focused, magnetic field pulses to target regions of the brain via a handheld induction coil placed against the scalp. By applying trains of pulses over several minutes, rTMS increases or decreases target brain region activity. While experimental applications of rTMS usually involve 1–3 sessions on a single day, therapeutic applications of rTMS typically require 20–30 daily sessions.124,125

411

Repetitive TMS appears to act through the mechanisms of synaptic plasticity: long-term potentiation and depression, although there is conflict among the effects of genetic polymorphisms, such as brain-derived neurotrophic factor, on these functions.126,127 Generally, the direction of the effect depends on the intensity, duration, and stimulation pattern.128,129 High-frequency stimulation (5– 20 Hz) is classically considered excitatory, while low frequency stimulation (1 Hz) is inhibitory.130,131 Some recent rTMS studies use stimulation patterns that mimic theta electroencephalography (EEG) rhythms, thought to be especially efficient for inducing plasticity.132,133 For example, two advantages of theta-burst stimulation (TBS)—classified as intermittent TBS (iTBS), which is generally excitatory, or continuous TBS (cTBS), which is generally inhibitory—are that the time required for a session of stimulation is rather brief and that it is as potent as conventional stimulation, both in model systems such as the motor cortex133 and in clinical applications.134,135

One current drawback to all forms of rTMS concerns the variability of effect: a certain percentage of individuals show neutral or inhibitory effects from excitatory stimulation and, conversely, some individuals show excitatory effects from inhibitory stimulation.128,136 This variability is observed both in model systems such as the motor cortex during single sessions and in therapeutic applications of rTMS over multiple sessions.137,138

Therapeutic effects of rTMS in neuropsychiatric disease may ensue through changes in cortico– striatal–thalamic circuits stemming from SN and VMN nodes. Positron emission tomography (PET) studies of rTMS reveal changes in striatal dopamine receptor occupancy following rTMS, with the changes localized to the specific region of the striatum that serves the cortical target (DMPFC and DLPFC) of stimulation.139,140 In keeping with this observation, dopamine agonists and antagonists can potentiate or block the effects of rTMS.141 Baseline cortico–striatal–thalamic functional connectivity on resting-state fMRI predicts treatment outcome across multiple conditions, including MDD,142 eating disorders,138 OCD,143 and movement disorders.144 Pre–post treatment changes in this measure also track outcomes across all of these disorders. Thus, rTMS is posited to exert therapeutic effects by enhancing or suppressing cortico–striatal–thalamic circuit integrity from the SN and/or VMN.

Transcranial DCS overview

Transcranial DCS applies lower-energy stimulation to the brain, using a montage of over two scalp electrodes, each typically 3–7 cm across. Constant-current stimulation is applied through the electrodes at intensities of just 1–2 mA. A tDCS session typically lasts 5–30 min, and a therapeutic course typically involves daily stimulation over 10–30 days.145

Transcranial DCS seeks to modulate the synaptic activity of target brain regions, rather than directly eliciting action potentials, as does rTMS.146 Anodal stimulation is classically considered to increase cortical excitability in the underlying brain region, while cathodal stimulation is considered to be inhibitory. However, as with rTMS, some individuals show the opposite pattern of effect, and variability remains a problematic issue.147 Newer patterns of stimulation include transcranial random noise stimulation (tRNS), as well as transcranial alternating current stimulation (tACS). The latter approach is considered promising, since stimulation can be tuned to match specific EEG frequency bands for more potent or more selective effects.148 In one notable recent example, investigators were able to induce lucid

412

dreaming in healthy subjects during rapid eye movement (REM) sleep by applying tACS at 20–40 Hz, but not at 2–12 Hz.149

The mechanisms of tDCS/tACS are still under investigation.150 However, the technique does appear to modulate the activity of resting-state networks on fMRI151 and may also modulate cortico–striatal functional connectivity.152 In these respects, its mechanisms may resemble those of rTMS to some degree, and, for therapeutic purposes, both techniques have been used to target similar brain regions in similar illnesses. However, at the physiological level, there are likely to be important mechanistic differences between the two approaches, the details of which require further study.146

Transcranial DCS is likely capable of stimulating many of the same regions as rTMS, although, once again, focal stimulation becomes more difficult for deeper structures. Electrical field simulations suggest that the direct effects of tDCS on neural activity are most prominent near superficial brain regions directly under the electrode, although deeper effects may be possible.153 Preclinical studies indicate that tDCS can modulate the excitability of the motor cortex and DLPFC,154 as well as deeper structures of the medial wall, such as the motor cortex of the lower limb,155 supplementary motor area,156 and DMPFC.157 There are also suggestions that tDCS may be capable of modulating the reward value of stimuli during task performance, suggesting engagement of deeper VMN nodes, such as the VMPFC and even the midbrain.158 Thus, many of the areas of interest for NIBS in MDD and SUDs are likely to be accessible to both tDCS and rTMS.

NIBS as a treatment for MDD

MDD is the original and best-studied therapeutic indication for rTMS. To date, dozens of large-scale randomized controlled trials have confirmed efficacy for rTMS in MDD, as summarized in recent meta-analyses.159,160 Response and remission rates in the most recent large studies are approximately 50–55% and 30–35%, respectively;161,162 real-world effectiveness studies report similar outcomes.159,160 Course lengths of 26–28 sessions are required for maximum effect;124 the poorer outcomes of early trials may be attributable in part to inadequate course lengths of 5–10 sessions.160

The most widely used rTMS protocol in MDD is high-frequency left DLPFC stimulation. Some studies alternatively used low-frequency right DLPFC rTMS,161 or both.162 Meta-analyses have not found marked outcome differences between these approaches,160 and no one stimulation pattern appears markedly superior.162 Few NIBS studies have sought targeted non-DLPFC nodes of the SN or VMN in MDD.163 One exception is the DMPFC,164 which was recently targeted with rTMS in a shamcontrolled study and in several open-label case series,142,165,166 with promising results. However, no rTMS trials have targeted the anterior insula or the VMPFC in MDD; these remain theoretically promising targets for intervention.

Mechanistic studies show that DLPFC or DMPFC stimulation may enhance cognitive control in healthy subjects and MDD patients. DLPFC or DMPFC excitatory rTMS in healthy subjects can enhance impulse control via delay-discounting tasks;167 likewise, rTMS of the presupplementary motor area, just posterior to the DMPFC, improves stop-signal task performance.168 Neurally, rTMS may enhance impulse control via strengthened DLPFC and DMPFCfrontal–striatal–cortical circuit integrity on fMRI,138 and, on PET, rTMS changes

413

dopamine receptor occupancy in these same striatal regions.169 In MDD, cortico–striatal–thalamic connectivity on fMRI predicts and correlates with treatment outcomes.142 These findings suggest that rTMS could be relieving MDD by enhancing cortico–striatal–thalamic circuit integrity in the nodes of the SN, thereby enhancing cognitive control over negative cognition and affect.

The literature on tDCS in MDD is more nascent. To date, 10 randomized controlled trials have been completed, with a recent meta-analysis supporting tDCS efficacy in MDD.145 Outcomes appeared comparable to antidepressant medication in one comparative trial.170 Again, 20–30 sessions may be required for maximal efficacy;171 the shorter, 10- session courses and relatively small sample sizes used in many tDCS trials may imply an underestimate of the true efficacy of the technique.145

To date, all tDCS trials in MDD have targeted the left DLPFC, using anodal, excitatory stimulation over the F3 EEG site; the inhibitory cathode has been variously applied to either the right DLPFC via the F4 EEG site or to the neighboring right supraorbital area near Fp2/F8.Mechanistically, anodal tDCS over the left DLPFC enhances cognitive control in healthy participants on a working memory task with negative emotional distracters.172 In a more direct demonstration, anodal DLPFC tDCS abolished the effect of negative emotional distractors on a working memory task in MDD patients,173 suggesting that the mechanisms of effect may likewise involve enhanced cognitive control. The functional anatomy of the SN suggests that similar effects might also be achieved via excitatory tDCS of the right DLPFC, the DMPFC, and the anterior insula. Frontopolar tDCS is also of interest for the anhedonic symptoms of MDD, since anodal stimulation of this site engages the VMPFC and VTA on fMRI, and enhances perceived attractiveness of faces.158 These regions are therefore important candidate targets for future tDCS trials in MDD.

NIBS as a treatment for SUDs and comorbid SUD/MDD

NIBS is attracting increasing interest as a novel therapeutic intervention in SUDs. To date, approximately 25 original research reports have been published on the efficacy of rTMS as a tool to decrease craving, along with at least six reviews on rTMS in addiction.174 The types of SUDs, the targets of stimulation, and the patterns of stimulation have varied across these studies. However, as in MDD, most so far have sought to enhance cognitive control mechanisms by targeting the nodes of the SN.174 The alternative approach, of attenuating craving and incentive salience via the VMN, is also beginning to be explored.175

For alcohol cravings, most studies have targeted the left or right DLPFC. In sham-controlled studies of rTMS, low-frequency right DLPFC stimulation reduced alcohol craving in some reports176 but not others,177,178 and high-frequency left DLPFC stimulation reduced attention to alcohol cues but did not reduce cravings.179 In open-label case reports, bilateral high-frequency rTMS reduced alcohol cravings in three patients,180 and low-frequency dACC rTMS likewise reduced refractory cravings in one patient; however, this patient eventually relapsed following rTMS.181 For DLPFC tDCS, left anodal/right cathodal stimulation reduced alcohol cravings in initial studies;182,183 however, relapse rates improved with this montage.184 Reversing the polarity to left cathodal/right anodal DLPFC tDCS yielded superior abstinence rates at 6 months, despite no differences in craving.185 Such findings are consistent

414

with a model in which SN stimulation exerts therapeutic effects by enhancing cognitive control rather than reducing cravings.

For stimulants (cocaine and methamphetamine), one open-label study reported reduced spontaneous craving with rTMS of the left DLPFC at high frequencies,186 while another found effects for right low-frequency, but not left high-frequency, DLPFC rTMS.187 Of note, rTMS of the left DLPFC at low frequencies has increased cue-induced craving,188 and anodal tDCS of the left DLPFC has increased risky decisionmaking in cocaine users,189 once again suggesting that the stimulation parameters may be important in determining whether cognitive control is enhanced or diminished. A recent study of tDCS in crack cocaine dependence accordingly reversed the polarity of stimulation to left cathodal/right anodal DLPFC tDCS, reporting a reduction in craving with five sessions of active but not sham stimulation.190

For nicotine, a slightly wider variety of targets have been studied across the SN, including the DLPFC and anterior insula. Left DLPFC rTMS, using low- or high-frequency stimulation, has reduced cravings in experimental studies.191,192 One clinical trial of high-frequency DLPFC rTMS reported reductions in the number of cigarettes smoked, even in the absence of changes in craving.193 A more recent trial targeted the anterior insula, as well as the DLPFC, bilaterally with a helmet-shaped deep rTMS coil, in a large sample of 115 patients.194 Thirteen sessions of high- but not low-frequency or sham rTMS reduced cigarette consumption and increased abstinence rates, with stronger effects when the patients were exposed to smoking cues during stimulation. However, there were no significant effects on craving despite the reduction in use, once again suggesting a mechanism of cognitive control enhancement rather than craving reduction per se.

With respect to tDCS in nicotine dependence, a preliminary study found that one session of active but not sham anodal bilateral DLPFC tDCS reduced cigarette cravings.195 Another preliminary study196 found a significant reduction in cigarettes smoked for active but not sham DLPFC stimulation using the left anodal/right cathodal polarity that had proven useful in alcohol and crack cocaine use, as above. A more recent study using the same type of tDCS197 found a reductionin cigarettes smoked after 5 days of active but not sham stimulation. Furthermore, the active group showed signs of enhanced cognitive control, in the form of a greater propensity to reject offers of cigarettes (but not money) in a decision-making task known as the Ultimatum game. On an important side note, at least one report has noted that nicotine patches abolish the effects of both anodal and cathodal tDCS in healthy volunteers,198 potentially posing a challenge to the therapeutic use of tDCS in tobacco cessation.

One final point concerns the potential of rTMS for treating comorbid MDD/SUDs in tandem, by addressing their common deficits in cognitive control. A recent study199 used high-frequency bilateral DLPFC rTMS via a deep helmet coil, and compared outcomes in patients with MDD versus MDD and alcohol dependence. One major difference between deep TMS and conventional rTMS is the coil geometry; deep TMS coils employ complex helmet-shaped windings that are able to reach deeper cortical structures.200 Depression scores improved 55% and 62% in the two groups, while scores on the Clinical Global Impression scale improved 67% and 78%. Improvement was actually significantly greater in the comorbid MDD/alcohol dependence patients than in the patients with MDD alone. It is also worth noting that a recent meta-analysis found that deep TMS appears to achieve antidepressant effects over multiple sessions.201 The suggestion is that the presence of SUDs may not interfere with the outcomes of

415

rTMS for MDD and indeed that the presence of comorbid SUDs might be an indicator of broader underlying deficits in cognitive control that render DLPFC rTMS more likely to be successful. This possibility warrants future investigation.

FuturedirectionsforNIBSinMDDandSUDs

Enhancing cognitive control

Cognitive control, and specifically response selection/inhibition, is presented in this review as one of the key transdiagnostic deficits across MDD and SUDs. Its associated functional network, the SN, maps closely onto a set of areas affected across multiple categories of psychiatric illness. Of its core cortical nodes, investigations have relied heavily on the DLPFC for NIBS in MDD and SUDs. The dACC and anterior insula, however, actually appear more prominently and consistently in the network than the DLPFC.57 Hence, it may be productive to pursue studies of dACC or anterior insula excitatory NIBS, under the hypothesis that these targets will surpass the DLPFC for enhancing cognitive control and thus achieve superior clinical outcomes inMDD and SUDs.

For the dACC, early evidence supports this hypothesis. In healthy controls, excitatory dorsomedial rTMS can reduce impulsivity on a delaydiscounting task,167 and there is a growing literature supporting clinical efficacyfor DMPFC rTMS across multiple disorders.142,165,166 DMPFC rTMS also shows initial promise for treating acute and chronic craving.202 Dorsal ACC activity is higher in response to addiction cues,93,203 especially when personally relevant,92,93,95 and during other tasks involving cognitive control and response inhibition.73,75,204

Reducing incentive salience

Reward circuitry pathology, and specifically distorted incentive salience, is also presented in this review as a common feature of MDD and SUDs. The associated network, VMN, is a candidate target for NIBS treatments. At present, rTMS and tDCS are unlikely to be able to stimulate the ventral striatum directly because of the depth of this structure. However, the cortical nodes of the VMN are not markedly deeper than the dACC or anterior insula. At least one deep rTMS coil has been designed to target the VMPFC,205 although it has not yet been used in MDD or SUDs. Likewise, at least one tDCS study has successfully modulated the VMPFC, along with the VMN circuit into the midbrain, enhancing perceived attractiveness of faces.158 In SUD patients, one study recently targeted the frontopolar cortex206 and found that high-frequency rTMS increased cueinduced cigarette craving, as would be expected for excitatory stimulation of this incentive pathway. Thus, stimulation of the VMN appears feasible and may exert effects on disease-relevant reward function.

Could inhibitory TMS to the VMPFC be beneficial for SUD patients?

While the majority of rTMS studies to date have focused on the DLPFC and its downstream targets in the dorsal striatum, a recent study demonstrated that it is also possible to

416

activate the VMPFC and its targets in the ventral striatum with TMS. Using integrated TMS/MRI scanning, they demonstrated that it is possible to differentially activate the frontal–striatal systems that govern executive control from those that govern limbic arousal by applying single-pulse TMS to the DLPFC and the VMPFC/frontal pole, respectively.66 TMS pulses applied to the frontal pole (EEG coordinate: FP1) led to elevated BOLD signal in the VMPFC, ventral striatum, and OFC—core regions of the VMN involved in limbic arousal and craving. TMS pulses applied to the DLPFC in healthy individuals at rest (EEG coordinate: F3) led to elevated BOLD signal in the DLPFC and dorsal striatum—core regions of the SN involved in executive control (Fig. 6A). Additionally, elevated DLPFC activity was accompanied by a reciprocal decrease in VMPFC activity, highlighting the reciprocal activity of these two networks.

Having demonstrated that it was possible to modulate activity in the VMPFC and ventral striatum, the authors performed a study in cocaine users175 in which they applied an inhibitory form of TMS (continuous theta burst stimulation) to the medial PFC (FP1) while the participants were engaged in a craving-induction task. Specifically, patients underwent a functional MRI scan immediately before and after a single session of real or sham cTBS. Immediately before the cTBS session, participants were asked to describe the last time that they used cocaine, using standardized techniques from exposure therapy. They were then primed and asked to think about this event while the cTBS was administered. It was demonstrated that, relative to sham, active stimulation significantly decreased stimulus-evoked BOLD signal in the VMPFC and the ventral striatum— critical brain regions for craving (Fig. 6B). Additionally, these data revealed a significant correlation between the TMS pulse intensity and the effect on these neural circuits (Fig. 6C). Thus, inhibitory rTMS to the frontal pole appears to be a successful strategy for engaging and attenuating the VMN target, especially when an inhibitory dose of rTMS is applied to a neural circuit that is in a primed state (e.g., thinking about drug cues). There are hints that this strategy may be more useful than SN excitatory stimulation in at least some patients with SUDs. Although the SN and VMN show reciprocal activity in healthy subjects, this may not be true in all individuals. In cocaine users, a recent MRI–TMS study showed that DLPFC rTMS did not elicit the usual pattern of reciprocal deacti vation of the VMN207 (Fig. 7). Thus, although some individuals may be able to achieve the desired suppression of the VMN indirectly during excitatory stimulation of the SN, other patients may require the VMN to be targeted directly. Investigating this hypothesis in MDD and SUD patients with neuroimaging methods will be an important area for future study.

Controlling the cognitive state during treatment

Few NIBS trials to date have controlled patients’ cognitive state during treatment. Brain activity during stimulation is known to have an important influence on the effects of rTMS,208–210 as evident even in the difference between resting and active motor threshold during motor cortex stimulation. Repetitive TMS of the SN for nicotine addiction is more effective in the presence of smoking cues.194 Conversely, in MDD, negative stimuli exposure during rTMS may disrupt the beneficial effects of treatment.209 With frontopolar stimulation, the effects of rTMS on cigarette craving were different, depending on whether the patient was presented with smoking or neutral cues during stimulation.206 Thus, future efforts to improve outcomes for NIBS in MDD/SUDs will likely benefit from more rigorous manipulation of the cognitive state during treatment.

417

Optimizing the treatment protocol

Available evidence concurs with clinical experience in suggesting that maximal effects of rTMS in MDD require 20–30 sessions of treatment, regardless of site or protocol.161,162,166 The limited evidence for tDCS likewise suggests that benefits continue to accumulate over 30 sessions in MDD.171 For this reason, early literature may underestimate the therapeutic potential of NIBS, and longer courses are required for future studies to establish the true effect size.

Long courses of treatment may negatively affect treatment adherence, as well as the scalability of NIBS as a treatment in the larger population. SUD patients respond better to briefer treatment interventions than to extended ones.211 However, more sessions need not require more days of treatment. Recent rTMS studies have begun to explore the use of multiple daily sessions separated by short intervals. These protocols are safe and tolerable, and have reduced the treatment course to 10 dayswith 2x daily stimulation,212 5 days with 4x daily stimulation,213 and, in one small case series, 2 days with 15 sessions.214 The optimal session number and interval have not been systematically investigated under randomized conditions, and this is an important area for future study. Accelerated regimens may be facilitated by cTBS and iTBS, which achieve similar effects as longer conventional protocols despite requiring only 1–3 min for delivery.134,166 Other products that have the potential to be clinically impactful include deep TMS201 and external trigeminal nerve stimulation.215

Reducing the variability of effect for NIBS

Although different NIBS protocols are classically considered to be excitatory or inhibitory, many individuals display effects opposite to the usual direction, or no effect at all. For 1-Hz rTMS, approximately 50% of individuals show excitation rather than inhibition;128,136 likewise, for 10-Hz rTMS, a similarly large proportion show inhibition rather than excitation.128 Variability is also problematic for cTBS and iTBS.216 Likewise, for tDCS, one study found that only 36% showed the classic pattern of excitatory effects for anodal and inhibitory for cathodal stimulation, and the opposite was true in 21%.147 The variability is not confined to the motor cortex: fMRI studies reveal similarly inconsistent effects of 1-Hz parietal rTMS on resting-state functional connectivity.136

This variability of effect likely impedes successful treatment outcome. Currently, there is no known reliable biological or neuroimaging marker to predict a priori which individuals will respond best to a particular stimulation target or stimulation protocol. However, group-level differences between responders and non-responders to treatment may help to elucidate these markers. For example, recent studies have found that 30 sessions of 10-Hz DMPFC rTMS strengthens cortico–striatal–thalamic resting-state connectivity in patients (responders) with low baseline connectivity, but weakens it in those with high baseline connectivity (non-responders).138 Thus, variability of NIBS effects must be addressed in future studies to improve treatment outcomes.

Better methods to control patient brain activity during stimulation may ameliorate some of the interindividual variability. For example, 20-Hz rTMS and newer forms of cTBS more consistently show effects both in the motor cortex128 and on resting-state functional connectivity.136 Quadripulse stimulation (QPS) uses 4-pulse bursts of stimulation that can be excitatory or inhibitory depending on the inter-pulse interval,217 and effects are longer lasting

418

and appear consistent across greater than 85–90% of individuals. Unfortunately, most rTMS devices cannot perform QPS without significant hardware modification, and QPS protocols may require 30 min of stimulation for optimal effects.218 Nonetheless, studies of novel rTMS protocols should be pursued to achieve better consistency in the effects.

Selecting and phenotyping patients

Important considerations in SUD trials are the type of substance dependence and the selection of SUD patients that are seeking treatment. Studies often do not disclose patient treatment-seeking status or whether patients are active substance users during NIBS treatments. Generally, patients are not abstinent for tobacco- or alcohol-related NIBS trials, while for illicit drug trials patients have completed detoxification. The active use of certain substances may influence the effect of NIBS. For example, tobacco use in healthy controls can reduce NIBS-induced inhibition of the motor cortex.198 In cigarette smokers, NIBS-induced facilitation was found to be abolished during nicotine withdrawal.219 Future studies should take into account these sources of heterogeneity, as they may alter clinical outcomes.

Diagnostic criteria are another important source of heterogeneity. MDD and SUDs are increasingly recognized to encompass a wide variety of subtypes, or endophenotypes. In MDD, for example, distinct endophenotypes have recently been proposed for patients with prominent memory impairment, neuroticism, cognitive control, and anhedonia.220 The latter three have recently been linked to VMN disconnectivity and, consequently, poor response to DMPFC rTMS.165 The presence of neutrally distinctive endophenotypes suggests that treatment targeted to individual pathology may improve success rates. For example, patients without cognitive control deficits, but with poor reward sensitivity, may be poor candidates for standard SN rTMS and might instead be better suited to stimulation of the VMN. Conversely, patients with mood liability, cognitivecontrol deficits, and multiple comorbidities, such as SUDs, might be identified as particularly good responders to SN NIBS.

A common practice in clinical trials of NIBS is to attempt to limit patient heterogeneity by excluding MDD patients with comorbid SUDs or active substance dependence. However, given that SN pathology may give rise to multiple comorbidities, this practice may be counterproductive in that it excludes good treatment candidates. A more productive approach may be to adopt generous inclusion criteria and to carefully characterize each patient before treatment.

Conclusions

Neuroimaging has revealed twofunctional networks playing key roles in MDD and SUD pathophysiology. The SN mediates cognitive control, and its hypofunction is a common feature across multiple psychiatric illnesses. The SN’s anti-network, the VMN, plays a key role in reward and incentive salience. Distorted incentive salience is a common feature of both MDD and SUDs. MDD/SUD comorbidity, and the tendency for symptoms of one disorder to exacerbate the other, can therefore be understood as reflecting a shared set of reciprocal neural substrates: inappropriate VMN activation, resulting in pathological incentives, and/or insufficient SN activation, resulting in an inability to exert cognitive control over those incentives. NIBS targeting the SN may enhance cognitive control, thereby relieving a deficit common to MDD and

419

SUDs. NIBS targeting the VMN is less well explored, but may attenuate pathological incentive salience, thereby reducing craving in SUDs and negative affect in MDD.

In conclusion, NIBS offers a promising new approach to treat MDD and SUDs. By strengthening cognitive control and quelling pathological incentive salience, NIBS may address underlying deficits common to both disorders, and may be particularly well suited to comorbid cases. Given the high prevalence and social impact of MDD and SUDs and high nonresponse rates, new treatment options will be a welcome development for clinicians and patients alike.

420

Figure 1. Functional networks in the intrinsic activity of the brain. (A) The intrinsic ongoing activity of the brain at rest or during task performance can be decomposed into sets or networks of brain regions that show correlated patterns of activation and deactivation over time. A set of at least seven distinct functional networks has been identified as consistently appearing in large datasets of up to 1000 individuals. However, these seven networks contain smaller sub-networks. A 17-network parcellation has been identified as stable across individuals. (B) The 17 resting-state networks identified by Yeo et al.46 include low-level visual and somatosensory cortical areas, higher-level networks involving premotor and sensory association areas, and larger frontoparietal networks involved in attention, cognition, and executive control. However, two networks (highlighted in dashed black lines) are of particular interest in MDD and SUDs: the more anterior of the two subnetworks of the ventral attention network and the ventromedial subnetwork of the limbic network. Adapted from Yeo et al.46

421

Figure 2. The salience network (SN). The reader is encouraged to replicate and explore the depicted networks in the Neurosynth tool.45 (A) The cingulo–opercular network from the parcellation of Figure 1 includes prominent nodes in the dorsal anterior cingulate cortex (dACC), anterior insula, dorsolateral prefrontal cortex (DLPFC), and inferior parietal lobule, as well as the dorsal anterior caudate nucleus. (B) A Neurosynth meta-analysis45 using the term “salience network” reveals the close correspondence of this network to the cingulo–opercular network identified above. Note that the mediodorsal thalamus can also be seen in the network in this analysis. (C) The areas identified as common sites of gray matter loss across MDD, SUDs, and several other categories of psychiatric disorders in a meta-analysis of 193 voxel-based morphometry studies57 correspond closely to SN nodes in the dACC and anterior insula. (D)

422

Resting-state functional connectivity maps seeding from the nodes in C reveal a network that corresponds closely to the rest of the SN, as seen in A and B.45 (E, F) Neurosynth meta-analyses using the terms “response inhibition” and “response selection” yield maps of activation that correspond closely to the SN, thus highlighting the role of the SN as a neural substrate for cognitive control.57

423

Figure 3. The ventromedial network (VMN). The reader is encouraged to replicate and explore the depicted networks in the Neurosynth tool.45 (A) The VMN consists of the cortical and striatal nodes of the ventral striatal–ventromedial prefrontal network from the parcellation of Figure 1. (B) A resting-state functional connectivity map seeded from the nucleus accumbens illustrates the strong connection to the ventromedial prefrontal cortex (VMPFC) and frontal pole.45 (C) A Neurosynth meta-analysis using the term “reward” reveals the classic reward circuit, including mesolimbic dopaminergic structures in the ventral tegmental area (VTA) and substantia nigra (SN), the ventral striatum, and a specific subregion of the VMPFC slightly posterior to the medial frontal pole.45 (D) A Neurosynth meta-analysis using the term “value” reveals the striatal and cortical components of the VMN, suggesting a broader role beyond reward to include valuation of incentives.45 (E) A meta-analysis of regions activated by drug cues in patients with addiction reveals a circuit corresponding closely to the VMN, illustrating the pathological distortion of reward value for drug cues in addiction.63 (F) A meta-analysis of regions activated by negative emotional stimuli in MDD reveals a similar signature of

424

pathological activation of reward-related areas for negative rather than positive cues, illustrating a common pathophysiology of distorted incentive salience in the VMN across SUDs and MDD.63

425

Figure 4. Reciprocal relationship of SN and VMN activity. The reader is encouraged to replicate and explore the depicted networks in the Neurosynth tool. Resting-state functional connectivity maps are generated using the Neurosynth tool45 from the seed coordinates indicated at left. Seeds in the anterior insula and dACC reveal a network of positively correlated regions throughout the other nodes of the SN, as expected (upper left). Notably, the anti-network of these SN seeds (i.e., regions showing negative rather than positive correlations) includes the key VMN nodes in the ventral striatum, VMPFC, and temporal poles (upper right), as may be seen by comparison with Figure 3. Conversely, seeds in the nucleus accumbens and VMPFC reveal a network of positively correlated regions corresponding to the VMN (lower right). The anti-network of these seeds corresponds well to the SN (lower left). In order to highlight the correspondence of the networks and anti-networks, blue colors are used for the SN networks and the VMN anti-networks, while orange colors are used for the VMN networks and SN anti-networks.

426

Figure 5. Network architecture of the brain from incentive formation to behavioral execution. (A) The network architecture of the brain can be derived by placing nodes to cover the entire cerebrum and then extracting the intrinsic activity of these nodes over time. By connecting the well-correlated nodes with “edges,” a graph of the network architecture of the brain can be constructed. Within this architecture, the functional networks illustrated in Figure 1 appear as clusters of cliques, and the larger relationship between the networks can be seen.70 (B) A schematic derived from A illustrates a trajectory of information flow for behavioral control. This trajectory begins in the reward circuitry of the VMN, passing through the default-mode and frontoparietal networks and then the SN, before exiting the cerebrum via the nodes of the sensorimotor cortex to direct bodily movements. (C) This pathway of connections passes from one brain region to the next and allows the mapping of basic drives into specific incentives or cravings via the VMPFC and then the elaboration of these incentives into specific goals and scenarios via the default-mode network, followed by the refinement of these scenarios into specific strategies or plans for consummation of the goal. However, before these strategies can be executed as motor actions in the sensorimotor cortex, they must pass through the nodes of the SN, which thus sits in a gatekeeper position for response selection and behavioral inhibition. This functional architecture suggests that two points of intervention may be possible in SUDs and MDD: suppressing the pathological incentives early in this pathway at the VMPFC and/or strengthening the gatekeeper functions of response selection late in the pathway, via excitatory stimulation of the SN.

427

Figure 6. Preclinical evidence for targeting the VMN in SUDs. (A) TMS–fMRI studies reveal that stimulation of the DLPFC elicits activation in the corresponding corticostriatal circuit through the dorsal caudate nucleus, as well as other nodes of the SN. Stimulation over the frontal pole, in contrast, elicits activation in VMN nodes, including the VMPFC and ventral striatum.66 (B) Applying an inhibitory pattern of rTMS to the frontal pole (two trains of 1800 pulses of cTBS, 60 s apart) causes a reduction in TMS-evoked activation in the VMN and other limbic-network regions, including the ventral striatum and orbitofrontal cortex (OFC). The degree of inhibition is proportional to the intensity of stimulation. This evidence suggests that inhibitory stimulation of the frontal pole may successfully reduce activation in the cortical and striatal nodes of the VMN, which could have therapeutic value for reducing cravings in SUDs and the incentive salience of negative emotional cues in MDD.

428

Figure 7. Aberrant functional connectivity of theSN andVMN in cocaine users versus MDD patients.207 TMS of the DLPFC (F3 EEG site) during fMRI elicits local activation of the DLPFC itself, and reciprocal deactivation of the striatum and VMPFC, highlighting the reciprocal relationship of the SN and VMN in healthy controls. In cocaine users, however, TMS of the DLPFC elicited only local activation, suggesting a possible absence of the usual reciprocal relationship between SN and VMN may contribute to the pathophysiology of SUDs. If so, SN excitation alone may fail to inhibit the VMN and therefore may not exert the same degree of therapeutic effect in such cases. Instead, direct intervention to inhibit the VMN may be required.

429

References

1. Whiteford H.A., Degenhardt L., Rehm J., et al 2013. Global burden of disease attributable to mental and substance use disorders: findings from the global burden of disease study 2010. Lancet 382: 1575–1586. 2. Ferrari A.J., Charlson F.J., Norman R.E., et al 2013. Burden of depressive disorders by country, sex, age, and year: findings from the global burden of disease study 2010. PLoS Med. 10: e1001547. 3. Rush A.J. 2007. Limitations in efficacy of antidepressant monotherapy. J. Clin. Psychiatry 68(Suppl.): 8–10. 4. Rush A.J., Trivedi M.H., Wisniewski S.R., et al 2006. Acute and longer-term outcomes in depressed outpatients requiring one or several treatment steps: a STAR*D report. Am. J. Psychiatry 163: 1905–1917. 5. Nemeroff C.B. 2007. Prevalence and management of treatment-resistant depression. J. Clin. Psychiatry 68: 17–25. 6. Rocha F.L., Fuzikawa C., Riera R. & Hara C. 2012. Combination of antidepressants in the treatment of major depressive disorder: a systematic review and meta-analysis. J. Clin. Psychopharmacol. 32: 278–281. 7. Papakostas G.I., Shelton R.C., Smith J. & Fava M. 2007. Augmentation of antidepressants with atypical antipsychotic medications for treatment-resistant major depressive disorder: a meta-analysis. J. Clin. Psychiatry 68: 826–831. 8. Blier P., Ward H.E., Tremblay P., et al 2010. Combination of antidepressant medications from treatment initiation for major depressive disorder: a double-blind randomized study. Am. J. Psychiatry 167: 281–288. 9. Turner P. & Kantaria R.A. 2014. Systematic review and meta-analysis of the evidence base for add-on treatment for patients with major depressive disorder who have not responded to antidepressant treatment: a European perspective. J. Psychopharmacol. 28: 85–98. 10. Dierckx B., Heijnen W.T., van den Broek W.W. & Birkenhäger T.K. 2012. Efficacy of electroconvulsive therapy in bipolar versus unipolar major depression: a meta-analysis. Bipolar Disord. 14: 146–150. 11. Swan J.S., Macvicar R., Christmas D., et al 2014. Cognitive Behavioural Analysis System of Psychotherapy (CBASP) for chronic depression: clinical characteristics and six month clinical outcomes in an open case series. J. Affect. Disord. 152–154: 268–276. 12. Jelovac A., Kolshus E. & McLoughlin D.M.. 2013. Relapse following successful electroconvulsive therapy for major depression: a meta-analysis. Neuropsychopharmacology 38: 2467–2474. 13. Degenhardt L., Chiu W.T., Sampson N., et al 2008. Toward a global view of alcohol, tobacco, cannabis, and cocaine use: findings from the WHO world mental health surveys. PLoS Med. 5: 1053–1067. 14. U.S. Department of Health and Human Services (HHS). Substance Abuse and Mental Health Services Administration. 2010. Results from the 2010 National Survey on Drug Use and Health: Summary of National Findings. Vol. I 1–143. Rockville, MD: Substance Abuse and Mental Health Services Administration. 15. Grant B.F., Chou S.P., Goldstein R.B., et al 2008. Prevalence, correlates, disability, and comorbidity of DSM-IV borderline personality disorder: results from the Wave 2 National Epidemiologic Survey on Alcohol and Related Conditions. J. Clin. Psychiatry 69: 533–545. 16. Merikangas K.R. & McClair V.L. 2012. Epidemiology of substance use disorders. Hum. Genet. 131: 779–789. 17. Dutra L., Stathopoulou G., Basden S.L., et al 2008. A meta-analytic review of psychosocial interventions for substance use disorders. Am. J. Psychiatry 165: 179–187. 18. Magill M. & Ray La. 2009. Cognitive-behavioral treatment with adult alcohol and illicit drug users: a meta-analysis of randomized controlled trials. J. Stud. Alcohol Drugs 70: 516–527. 19. McHugh R.K., Hearon B.A. & Otto M.W. 2010. Cognitive behavioral therapy for substance use disorders. Psychiatr. Clin. North Am. 33: 511–525. 20. Prendergast M., Podus D., Finney J., et al 2006. Contingency management for treatment of substance use disorders: a meta-analysis. Addiction 101: 1546–1560. 21. Stitzer M. & Petry N. 2006. Contingency management for treatment of substance abuse. Annu. Rev. Clin. Psychol. 2: 411–434. 22. Morein-Zamir S. & Robbins T.W. 2014. Fronto-striatal circuits in response-inhibition: relevance to addiction. Brain Res 1628: 117–129. 23. Anton R.F., O'Malley S.S., Ciraulo D.A., et al 2006. Combined pharmacotherapies and behavioral interventions for alcohol dependence: the COMBINE study: a randomized controlled trial. JAMA 295: 2003–2017. 24. Johnson R.E., Jaffe J.H. & Fudala P.J. 1992. A controlled trial of buprenorphine treatment for opioid dependence. JAMA 267: 2750–2755. 25. Sinha R. 2011. New findings on biological factors predicting addiction relapse vulnerability. Curr. Psychiatry Rep. 13: 398–405.

430

26. Blum K., Braverman E.R., Holder J.M., et al 2000. Reward deficiency syndrome: a biogenetic model for the diagnosis and treatment of impulsive, addictive, and compulsive behaviors. J. Psychoactive Drugs 32(Suppl.): 1–112. 27. Hasin D.S., Goodwin R.D., Stinson F.S. & Grant B.F. 2005. Epidemiology of major depressive disorder: results from the National Epidemiologic Survey on Alcoholism and Related Conditions. Arch. Gen. Psychiatry 62: 1097–1106. 28. Kessler R.C., Chiu W.T., Demler O., et al 2005. Prevalence, severity, and comorbidity of 12-month DSM-IV disorders in the National Comorbidity Survey Replication. Arch. Gen. Psychiatry62: 617–627. 29. Kelly J., Mckellar J.D. & Moos R. 2003. Major depression in patients with substance use disorders: relationship to 12-step self-help involvement and substance use outcomes. Addiction 98: 499–508. 30. Nunes E.V. & Levin F.R. 2004. Treatment of depression in patients with alcohol or other drug dependence: a meta-analysis. JAMA 291: 1887–1896. 31. Nunes E.V., Sullivan M.A. & Levin F.R. 2004. Treatment of depression in patients with opiate dependence. Biol. Psychiatry 56: 793–802. 32. Raby Wilfrid N., Rubin Eric A., Garawi F., et al 2014. A randomized, double-blind, placebo-controlled trial of venlafaxine for the treatment of depressed cocaine-dependent patients. Am. J. Addict. 23: 68–75. 33. Riper H., Andersson G., Hunter S.B., et al 2014. Treatment of comorbid alcohol use disorders and depression with cognitive-behavioural therapy and motivational interviewing: a meta-analysis. Addiction 109: 394–406. 34. Glassman A.H., Covey L.S., Stetner F. & Rivelli S. 2001. Smoking cessation and the course of major depression: a follow-up study. Lancet 357: 1929–1932. 35. Covey L.S., Glassman A.H. & Stetner F. 1998. Cigarette smoking and major depression. J. Addict. Dis. 17: 35–46. 36. Covey L.S., Glassman A.H., Stetner F., et al 2002. A randomized trial of sertraline as a cessation aid for smokers with a history of major depression. Am. J. Psychiatry 159: 1731–1737. 37. Gierisch J.M., Bastian L.A., Calhoun P.S., et al 2012. Smoking cessation interventions for patients with depression: a systematic review and meta-analysis. J. Gen. Intern. Med. 27: 351–360. 38. Pettinati H.M., O'Brien C.P. & Dundon W.D. 2013. Current status of co-occurring mood and substance use disorders: a new therapeutic target. Am. J. Psychiatry 170: 23–30. 39. Fox M.D., Liu H. & Pascual-Leone A. 2013. Identification of reproducible individualized targets for treatment of depression with TMS based on intrinsic connectivity. Neuroimage 66: 151–160. 40. Kimbrell T.A., Little J.T. & Dunn R.T., et al 1999. Frequency dependence of antidepressant response to left prefrontal repetitive transcranial magnetic stimulation (rTMS) as a function of baseline cerebral glucose metabolism. Biol. Psychiatry 46: 1603–1613. 41. Speer A.M., Kimbrell T.A., Wassermann E.M., et al 2000. Opposite effects of high and low frequency rTMS on regional brain activity in depressed patients. Biol. Psychiatry 48: 1133–1141. 42. Sporns O. 2011. The human connectome: a complex network. Ann. N. Y. Acad. Sci. 1224: 109–125. 43. Damoiseaux J.S., Rombouts S.A.R.B., Barkhof F., et al 2006. Consistent resting-state networks across healthy subjects. Proc. Natl. Acad. Sci. U.S.A. 103: 13848–13853. 44. Smith S.M., Fox P.T., Miller K.L., et al 2009. Correspondence of the brain's functional architecture during activation and rest. Proc. Natl. Acad. Sci. U.S.A. 106: 13040–13045. 45. Yarkoni T., Poldrack R.A., Nichols T.E., et al 2011. Large-scale automated synthesis of human functional neuroimaging data. Nat. Methods 8: 665–670. 46. Thomas Yeo B.T., Krienen F.M., Sepulcre J., et al 2011. The organization of the human cerebral cortex estimated by intrinsic functional connectivity. J. Neurophysiol. 106: 1125–1165. 47. Choi E.Y., Yeo B.T.T. & Buckner R.L. 2012. The organization of the human striatum estimated by intrinsic functional connectivity. J. Neurophysiol. 108: 2242–2263. 48. Buckner R.L., Krienen F.M., Castellanos A., et al 2011. The organization of the human cerebellum estimated by intrinsic functional connectivity. J. Neurophysiol. 106: 2322–2345. 49. Greicius M.D., Krasnow B., Reiss A.L. & Menon V. 2003. Functional connectivity in the resting brain: a network analysis of the default mode hypothesis. Proc. Natl. Acad. Sci. U.S.A. 100: 253–258. 50. Kucyi A. & Davis K.D. 2014. Dynamic functional connectivity of the default mode network tracks daydreaming. Neuroimage 100: 471–480. 51. Spreng R.N. & Grady C.L. 2010. Patterns of brain activity supporting autobiographical memory, prospection, and theory of mind, and their relationship to the default mode network. J. Cogn. Neurosci. 22: 1112–1123. 52. Hamilton J.P., Furman D.J., Chang C., et al 2011. Default-mode and task-positive network activity in major depressive disorder: implications for adaptive and maladaptive rumination. Biol. Psychiatry 70: 327–333.

431

53. Downar J., Crawley A.P., Mikulis D.J. & Davis K.D. 2000. A multimodal cortical network for the detection of changes in the sensory environment. Nat. Neurosci. 3: 277–283. 54. Downar J., Crawley A.P., Mikulis D.J. & Davis K.D. 2001. The effect of task relevance on the cortical response to changes in visual and auditory stimuli: an event-related fMRI study. Neuroimage 14: 1256–1267. 55. Sridharan D., Levitin D.J. & Menon V. 2008. A critical role for the right fronto-insular cortex in switching between central-executive and default-mode networks. Proc. Natl. Acad. Sci. U.S.A. 105: 12569–12574. 56. Dosenbach N.U.F., Visscher K.M., Palmer E.D., et al 2006. A core system for the implementation of task sets. Neuron 50: 799–812. 57. Goodkind M., Eickhoff S.B., Oathes D.J., et al 2015. Identification of a common neurobiological substrate for mental illness. JAMA Psychiatry 5797: 305–315. 58. Cooper J.C. & Knutson B. 2008. Valence and salience contribute to nucleus accumbens activation. Neuroimage 39: 538–547. 59. Volkow N.D., Wang G.-J., Fowler J.S., et al 2011. Addiction: beyond dopamine reward circuitry. Proc. Natl. Acad. Sci. U.S.A. 108: 15037–15042. 60. Koob G.F. 2009. Dynamics of neuronal circuits in addiction: reward, antireward, and emotional memory. Pharmacopsychiatry 42 (Suppl.): S32–S41. 61. Berridge K.C. 2012. From prediction error to incentive salience: mesolimbic computation of reward motivation. Eur. J. Neurosci. 35: 1124–1143. 62. Tremblay L.K., Naranjo C.A., Graham S.J., et al 2005. Functional neuroanatomical substrates of altered reward processing in major depressive disorder revealed by a dopaminergic probe. Arch. Gen. Psychiatry 62: 1228–1236. 63. Diekhof E.K., Falkai P. & Gruber O. 2008. Functional neuroimaging of reward processing and decision-making: a review of aberrant motivational and affective processing in addiction and mood disorders. Brain Res. Rev. 59: 164–184. 64. Menon V. 2011. Large-scale brain networks and psychopathology: a unifying triple network model. Trends Cogn. Sci. 15: 483–506. 65. Dosenbach N.U.F., Fair D.A., Cohen A.L., et al 2008. A dual-networks architecture of top-down control. Trends Cogn. Sci. 12: 99–105. 66. Hanlon C.A., Canterberry M., Taylor J.J., et al 2013. Probing the frontostriatal loops involved in executive and limbic processing via interleaved TMS and functional MRI at two prefrontal locations: a pilot study. PLoS One 8: e67917. 67. Fox M.D., Snyder A.Z., Vincent J.L., et al 2005. The human brain is intrinsically organized into dynamic, anticorrelated functional networks. Proc. Natl. Acad. Sci. U.S.A. 102: 9673–9678. 68. Uddin L.Q., Kelly A.M., Biswal B.B., et al 2009. Functional connectivity of default mode network components: correlation, anticorrelation, and causality. Hum. Brain Mapp. 30: 625–637. 69. Bullmore E. & Sporns O.. 2009. Complex brain networks: graph theoretical analysis of structural and functional systems. Nat. Rev. Neurosci. 10: 186–198. 70. Power J.D., Cohen A.L., Nelson S.M., et al 2011. Functional network organization of the human brain. Neuron 72: 665–678. 71. Power J.D., Schlaggar B.L., Lessov-Schlaggar C.N. & Petersen S.E. 2013. Evidence for hubs in human functional brain networks. Neuron 79: 798–813. 72. Nestler E.J. 2005. Is there a common molecular pathway for addiction? Nat. Neurosci. 8: 1445–1449. 73. Moeller S.J., Hajcak G., Parvaz M.A., et al 2012. Psychophysiological prediction of choice: relevance to insight and drug addiction. Brain 135: 3481–3494. 74. Bolla K., Ernst M., Kiehl K., et al 2004. Prefrontal cortical dysfunction in abstinent cocaine abusers. J. Neuropsychiatry Clin. Neurosci. 16: 456–464. 75. Lee T.M.C., Zhou W.-H., Luo X.-J., et al 2005. Neural activity associated with cognitive regulation in heroin users: a fMRI study. Neurosci. Lett. 382: 211–216. 76. Hester R., Nestor L. & Garavan H.. 2009. Impaired error awareness and anterior cingulate cortex hypoactivity in chronic cannabis users. Neuropsychopharmacology 34: 2450–2458. 77. Connolly C.G., Foxe J.J., Nierenberg J., et al 2012. The neurobiology of cognitive control in successful cocaine abstinence. Drug Alcohol Depend. 121: 45–53. 78. Denier N., Gerber H., Vogel M., et al 2013. Reduction in cerebral perfusion after heroin administration: a resting state arterial spin labeling study. PLoS One 8: e71461. 79. Liu J., Liang J., Qin W., et al 2009. Dysfunctional connectivity patterns in chronic heroin users: an fMRI study. Neurosci. Lett. 460: 72–77. 80. Zhang X., Salmeron B.J., Ross T.J., et al 2011. Anatomical differences and network characteristics underlying smoking cue reactivity. Neuroimage 54: 131–141.

432

81. Radeloff D., Willmann K., Otto L., et al 2014. High-fat taste challenge reveals altered striatal response in women recovered from bulimia nervosa: a pilot study. World J. Biol. Psychiatry 15: 307–316. 82. Rando K., Hong K.I., Bhagwagar Z., et al 2011. Association of frontal and posterior cortical gray matter volume with time to alcohol relapse: a prospective study. Am. J. Psychiatry 168: 183–192. 83. Naqvi N.H. & Bechara A. 2010. The insula and drug addiction: an interoceptive view of pleasure, urges, and decision-making. Brain Struct. Funct. 214: 435–450. 84. Naqvi N.H., Rudrauf D., Damasio H. & Bechara A. 2007. Damage to the insula disrupts addiction to cigarette smoking. Science 315: 531–534. 85. Moeller S.J., Tomasi D., Woicik P.A., et al 2012. Enhanced midbrain response at 6-month follow-up in cocaine addiction, association with reduced drug-related choice. Addict. Biol. 17: 1013–1025. 86. Kilts C.D., Gross R.E., Ely T.D. & Drexler K.P.G. 2004. The neural correlates of cue-induced craving in cocaine-dependent women. Am. J. Psychiatry 161: 233–241. 87. Risinger R.C., Salmeron B.J., Ross T.J., et al 2005. Neural correlates of high and craving during cocaine self-administration using BOLD fMRI. Neuroimage 26: 1097–1108. 88. Li Q., Wang Y., Zhang Y., et al 2012. Craving correlates with mesolimbic responses to heroin-related cues in short-term abstinence from heroin: an event-related fMRI study. Brain Res. 1469: 63–72. 89. Rosenberg D.R., MacMaster F.P., Mirza Y., et al 2005. Reduced anterior cingulate glutamate in pediatric major depression: a magnetic resonance spectroscopy study. Biol. Psychiatry 58: 700–704. 90. Franklin T.R., Wang Z., Wang J., et al 2007. Limbic activation to cigarette smoking cues independent of nicotine withdrawal: a perfusion fMRI study. Neuropsychopharmacology 32: 2301–2309. 91. McClernon F.J., Hiott F.B., Huettel S.A. & Rose J.E. 2005. Abstinence-induced changes in self-report craving correlate with event-related FMRI responses to smoking cues. Neuropsychopharmacology 30: 1940–1947. 92. Engelmann J.M., Versace F., Robinson J.D., et al 2012. Neural substrates of smoking cue reactivity: a meta-analysis of fMRI studies. Neuroimage 60: 252–262. 93. Kühn S. & Gallinat J. 2011. Common biology of craving across legal and illegal drugs: a quantitative meta-analysis of cue-reactivity brain response. Eur. J. Neurosci. 33: 1318–1326. 94. Schacht J.P., Anton R.F. & Myrick H. 2013. Functional neuroimaging studies of alcohol cue reactivity: a quantitative meta-analysis and systematic review. Addict. Biol. 18: 121–133. 95. Moeller S.J. & Goldstein R.Z. 2014. Impaired self-awareness in human addiction: deficient attribution of personal relevance. Trends Cogn. Sci. 18: 635–641. 96. Franklin T., Wang Z., Suh J.J., et al 2011. Effects of varenicline on smoking cue–triggered neural and craving responses. Arch. Gen. Psychiatry 68: 516–526. 97. Menossi H.S., Goudriaan A.E., De Azevedo-Marques Périco C., et al 2013. Neural bases of pharmacological treatment of nicotine dependence – insights from functional brain imaging: a systematic review. CNS Drugs 27: 921–941. 98. Culbertson C.S., Bramen J., Cohen M.S., et al 2011. Effect of bupropion treatment on brain activation induced by cigarette-related cues in smokers. Arch. Gen. Psychiatry 68: 505–515. 99. Goldstein R.Z., Woicik P.A., Maloney T., et al 2010. Oral methylphenidate normalizes cingulate activity in cocaine addiction during a salient cognitive task. Proc. Natl. Acad. Sci. U.S.A. 107: 16667–16672. 100. Lubman D.I., Yücel M., Kettle J.W.L., et al 2009. Responsiveness to drug cues and natural rewards in opiate addiction: associations with later heroin use. Arch. Gen. Psychiatry 66: 205–212. 101. Zhang Y., Gong J., Xie C., et al 2015. Alterations in brain connectivity in three sub-regions of the anterior cingulate cortex in heroin-dependent individuals: evidence from resting state fMRI. Neuroscience 284: 998–1010. 102. Hu Y., Salmeron B.J., Gu H., et al 2015. Impaired functional connectivity within and between frontostriatal circuits and its association with compulsive drug use and trait impulsivity in cocaine addiction. JAMA Psychiatry 72: 584–592. 103. Hong L.E., Hodgkinson C.A., Yang Y., et al 2010. A genetically modulated, intrinsic cingulate circuit supports human nicotine addiction. Proc. Natl. Acad. Sci. U.S.A. 107: 13509–13514. 104. Koenigs M. & Grafman J. 2009. The functional neuroanatomy of depression: distinct roles for ventromedial and dorsolateral prefrontal cortex. Behav. Brain Res. 201: 239–243. 105. Mayberg H.S., Lozano A.M., Voon V., et al 2005. Deep brain stimulation for treatment-resistant depression. Neuron 45: 651–660. 106. Brody A.L., Saxena S., Mandelkern M.A., et al 2001. Brain metabolic changes associated with symptom factor improvement in major depressive disorder. Biol. Psychiatry 50: 171–178. 107. Hamilton J.P., Chen G., Thomason M.E., et al 2011. Investigating neural primacy in Major Depressive Disorder: multivariate Granger causality analysis of resting-state fMRI time-series data. Mol. Psychiatry 16: 763–772.

433

108. Nestler E.J. & Carlezon W.A. 2006. The mesolimbic dopamine reward circuit in depression. Biol. Psychiatry 59: 1151–1159. 109. Tremblay L.K., Naranjo C.A., Graham S.J., et al 2005. Functional neuroanatomical substrates of altered reward processing in major depressive disorder revealed by a dopaminergic probe. Arch. Gen. Psychiatry 62: 1228–1236. 110. Robinson O.J., Cools R., Carlisi C.O., et al 2012. Ventral striatum response during reward and punishment reversal learning in unmedicated major depressive disorder. Am. J. Psychiatry 169: 152–159. 111. Pizzagalli D.A., Holmes A.J., Dillon D.G., et al 2009. Reduced caudate and nucleus accumbens response to rewards in unmedicated individuals with major depressive disorder. Am. J. Psychiatry166: 702–710. 112. Knutson B., Bhanji J.P., Cooney R.E., et al 2008. Neural responses to monetary incentives in major depression. Biol. Psychiatry 63: 686–692. 113. Wagner G., Koch K., Schachtzabel C., et al 2013. Self-referential processing influences functional activation during cognitive control: an fMRI study. Soc. Cogn. Affect. Neurosci. 8: 828–837. 114. Wagner G., Schachtzabel C., Peikert G. & Bär K.-J. 2015. The neural basis of the abnormal self-referential processing and its impact on cognitive control in depressed patients. Hum. Brain Mapp. 36: 2781–2794. 115. McCabe C., Cowen P.J. & Harmer C.J. 2009. Neural representation of reward in recovered depressed patients. Psychopharmacology (Berl.) 205: 667–677. 116. Cooney R.E., Joormann J., Eugène F., et al 2010. Neural correlates of rumination in depression. Cogn. Affect. Behav. Neurosci. 10: 470–478. 117. Wagner G., Sinsel E., Sobanski T., et al 2006. Cortical inefficiency in patients with unipolar depression: an event-related fMRI study with the Stroop task. Biol. Psychiatry 59: 958–965. 118. Holmes A.J. & Pizzagalli D.A. 2008. Response conflict and frontocingulate dysfunction in unmedicated participants with major depression. Neuropsychologia 46: 2904–2913. 119. Harvey P.O., Fossati P., Pochon J.B., et al 2005. Cognitive control and brain resources in major depression: an fMRI study using the n-back task. Neuroimage 26: 860–869. 120. Bermpohl F., Walter M., Sajonz B., et al 2009. Attentional modulation of emotional stimulus processing in patients with major depression—alterations in prefrontal cortical regions. Neurosci. Lett. 463: 108–113. 121. Mayberg H.S., Liotti M., Brannan S.K., et al 1999. Reciprocal limbic-cortical function and negative mood: converging PET findings in depression and normal sadness. Am. J. Psychiatry 156: 675–682. 122. Pizzagalli D., Pascual-Marqui R.D., Nitschke J.B., et al 2001. Anterior cingulate activity as a predictor of degree of treatment response in major depression: evidence from brain electrical tomography analysis. Am. J. Psychiatry 158: 405–415. 123. Li C.-T., Lin C.-P., Chou K.-H., et al 2010. Structural and cognitive deficits in remitting and non-remitting recurrent depression: a voxel-based morphometric study. Neuroimage 50: 347–56. 124. Carpenter L.L., Janicak P.G., Aaronson S.T., et al 2012. Transcranial magnetic stimulation (TMS) for major depression: a multisite, naturalistic, observational study of acute treatment outcomes in clinical practice. Depress. Anxiety 29: 587–596. 125. Solvason H.B., Husain M., Fitzgerald P.B., et al 2014. Improvement in quality of life with left prefrontal transcranial magnetic stimulation in patients with pharmacoresistant major depression: acute and six month outcomes. Brain Stimul. 7: 219–225. 126. Wang H., Wang X. & Scheich H. 1996. LTD and LTP induced by transcranial magnetic stimulation in auditory cortex. Neuroreport 7: 521–525. 127. Antal A., Chaieb L., Moliadze V., et al 2010. Brain-derived neurotrophic factor (BDNF) gene polymorphisms shape cortical plasticity in humans. Brain Stimul. 3: 230–237. 128. Maeda F., Keenan J.P., Tormos J.M., et al 2000. Interindividual variability of the modulatory effects of repetitive transcranial magnetic stimulation on cortical excitability. Exp. Brain Res. 133: 425–430. 129. Pascual-Leone A., Tormos J.M., Keenan J., et al 1998. Study and modulation of human cortical excitability with transcranial magnetic stimulation. J. Clin. Neurophysiol. 15: 333–343. 130. Chen R., Classen J., Gerloff C., et al 1998. Depression of motor cortex excitability by low-frequency transcranial magnetic stimulation. Neurology 48: 1398–1403. 131. Pascual-Leone A., Valls-Solé J., Wassermann E.M. & Hallett M. 1994. Responses to rapid-rate transcranial magnetic stimulation of the human motor cortex. Brain 117(Pt 4): 847–858. 132. Huang Y.-Z., Edwards M.J., Rounis E., et al 2005. Theta burst stimulation of the human motor cortex. Neuron 45: 201–206. 133. Huang Y.-Z., Rothwell J.C., Lu C.-S., et al 2009. The effect of continuous theta burst stimulation over premotor cortex on circuits in primary motor cortex and spinal cord. Clin. Neurophysiol. 120: 796–801. 134. Li C.-T., Chen M.-H., Juan C.-H., et al 2014. Efficacy of prefrontal theta-burst stimulation in refractory depression: a randomized sham-controlled study. Brain 137(Pt 7): 2088–2098.

434

135. Bakker N., Shahab S., Giacobbe P., et al 2015. rTMS of the dorsomedial prefrontal cortex for major depression: safety, tolerability, effectiveness, and outcome predictors for 10 Hz versus intermittent theta-burst stimulation. Brain Stimul. 8: 208–215. 136. Eldaief M.C., Halko M.A., Buckner R.L. & Pascual-Leone A. 2011. Transcranial magnetic stimulation modulates the brain's intrinsic activity in a frequency-dependent manner. Proc. Natl. Acad. Sci. U.S.A. 108: 21229–21234. 137. Nettekoven C., Volz L.J., Leimbach M., et al 2015. Inter-individual variability in cortical excitability and motor network connectivity following multiple blocks of rTMS. Neuroimage 118: 209–218. 138. Dunlop J., Woodside B., Lam E., et al 2015. Increases in frontostriatal connectivity are associated with response to dorsomedial repetitive transcranial magnetic stimulation in refractory binge/purge behaviors. Neuroimage Clin. 8: 611–618. 139. Pogarell O., Koch W., Pöpperl G., et al 2007. Acute prefrontal rTMS increases striatal dopamine to a similar degree as d-amphetamine. Psychiatry Res. 156: 251–255. 140. Pogarell O., Koch W., Pöpperl G., et al 2006. Striatal dopamine release after prefrontal repetitive transcranial magnetic stimulation in major depression: preliminary results of a dynamic [123I] IBZM SPECT study. J. Psychiatr. Res. 40: 307–314. 141. Monte-Silva K., Ruge D., Teo J.T., et al 2011. D2 receptor block abolishes θ burst stimulation-induced neuroplasticity in the human motor cortex. Neuropsychopharmacology 36: 2097–2102. 142. Salomons T.V., Dunlop K., Kennedy S.H., et al 2014. Resting-state cortico-thalamic-striatal connectivity predicts response to dorsomedial prefrontal rTMS in major depressive disorder. Neuropsychopharmacology 39: 488–498. 143. Dunlop K., Woodside B., Olmsted M., et al 2015. Reductions in cortico-striatal hyperconnectivity accompany successful treatment of obsessive-compulsive disorder with dorsomedial prefrontal rTMS. Neuropsychopharmacology. 144. Strafella A.P., Ko J.H., Grant J., et al 2005. Corticostriatal functional interactions in Parkinson's disease: a rTMS/[11C]raclopride PET study. Eur. J. Neurosci. 22: 2946–2952. 145. Meron D., Hedger N., Garner M. & Baldwin D.S. 2015. Transcranial direct current stimulation (tDCS) in the treatment of depression: systematic review and meta-analysis of efficacy and tolerability. Neurosci. Biobehav. Rev. 57: 46–62. 146. Brunoni A.R., Nitsche M.A., Bolognini N., et al 2012. Clinical research with transcranial direct current stimulation (tDCS): challenges and future directions. Brain Stimul. 5: 175–195. 147. Wiethoff S., Hamada M. & Rothwell J.C. 2014. Variability in response to transcranial direct current stimulation of the motor cortex. Brain Stimul. 7: 468–475. 148. Reato D., Rahman A., Bikson M. & Parra L.C. 2013. Effects of weak transcranial alternating current stimulation on brain activity-a review of known mechanisms from animal studies. Front. Hum. Neurosci. 7: 687. 149. Voss U., Holzmann R., Hobson A., et al 2014. Induction of self awareness in dreams through frontal low current stimulation of gamma activity. Nat. Neurosci. 17: 810–812. 150. Stagg C.J. & Nitsche M.A. 2011. Physiological basis of transcranial direct current stimulation. Neuroscientist 17: 37–53. 151. Keeser D., Meindl T., Bor J., et al 2011. Prefrontal transcranial direct current stimulation changes connectivity of resting-state networks during fMRI. J. Neurosci. 31: 15284–15293. 152. Polanía R., Paulus W. & Nitsche M.A. 2012. Modulating cortico-striatal and thalamo-cortical functional connectivity with transcranial direct current stimulation. Hum. Brain Mapp. 33: 2499–2508. 153. Edwards D., Cortes M., Datta A., et al 2013. Physiological and modeling evidence for focal transcranial electrical brain stimulation in humans: a basis for high-definition tDCS. Neuroimage74: 266–275. 154. Fregni F., Boggio P.S., Nitsche M., et al 2005. Anodal transcranial direct current stimulation of prefrontal cortex enhances working memory. Exp. Brain Res. 166: 23–30. 155. Bocci T., Marceglia S., Vergari M., et al 2015. Transcutaneous spinal direct current stimulation modulates human corticospinal system excitability. J. Neurophysiol. 114: 440–446. 156. Vollmann H., Conde V., Sewerin S., et al 2013. Anodal transcranial direct current stimulation (tDCS) over supplementary motor area (SMA) but not pre-SMA promotes short-term visuomotor learning. Brain Stim. 6: 101-7. 157. Hsu T.-Y., Tseng L.-Y., Yu J.-X., et al 2011. Modulating inhibitory control with direct current stimulation of the superior medial frontal cortex. Neuroimage 56: 2249–2257. 158. Chib V.S., Yun K., Takahashi H. & Shimojo S. 2013. Noninvasive remote activation of the ventral midbrain by transcranial direct current stimulation of prefrontal cortex. Transl. Psychiatry 3: e268.

435

159. Kedzior K., Azorina V. & Reitz S. 2014. More female patients and fewer stimuli per session are associated with the short-term antidepressant properties of repetitive transcranial magnetic stimulation (rTMS): a meta-analysis of 54 sham-controlled studies published between 1997–2013. Neuropsychiatr. Dis. Treat. 10: 727–756. 160. Berlim M.T., van den Eynde F., Tovar-Perdomo S. & Daskalakis Z.J. 2014. Response, remission and drop-out rates following high-frequency repetitive transcranial magnetic stimulation (rTMS) for treating major depression: a systematic review and meta-analysis of randomized, double-blind and sham-controlled trials. Psychol. Med. 44: 225–239. 161. Brunelin J., Jalenques I., Trojak B., et al 2014. The efficacy and safety of low frequency repetitive transcranial magnetic stimulation for treatment-resistant depression: the results from a large multicenter French RCT. Brain Stimul. 7: 855–863. 162. Fitzgerald P.B., Hoy K., Gunewardene R., et al 2011. A randomized trial of unilateral and bilateral prefrontal cortex transcranial magnetic stimulation in treatment-resistant major depression. Psychol. Med. 41: 1187–1196. 163. Downar J. & Daskalakis Z.J. 2013. New targets for rTMS in depression: a review of convergent evidence. Brain Stimul. 6: 231–240. 164. Kreuzer P.M., Schecklmann M., Lehner A., et al 2015. The ACDC pilot trial: targeting the anterior cingulate by double cone coil rTMS for the treatment of depression. Brain Stimul. 8: 240–246. 165. Downar J., Geraci J., Salomons T.V., et al 2014. Anhedonia and reward-circuit connectivity distinguish nonresponders from responders to dorsomedial prefrontal repetitive transcranial magnetic stimulation in major depression. Biol. Psychiatry 76: 176–185. 166. Bakker N., Shahab S., Giacobbe P., et al 2015. rTMS of the dorsomedial prefrontal cortex for major depression: safety, tolerability, effectiveness, and outcome predictors for 10 Hz versus intermittent theta-burst stimulation. Brain Stimul. 8: 205–215. 167. Cho S.S., Kosimori Y., Aminian K., et al 2015. Investing in the future: stimulation of the medial prefrontal cortex reduces discounting of delayed rewards. Neuropsychopharmacology 40: 546–553. 168. Watanabe T., Hanajima R., Shirota Y., et al 2015. Effects of rTMS of pre-supplementary motor area on fronto basal ganglia network activity during stop-signal task. J. Neurosci. 35: 4813–4823. 169. Strafella A.P., Paus T., Barrett J. & Dagher A. 2001. Repetitive transcranial magnetic stimulation of the human prefrontal cortex induces dopamine release in the caudate nucleus. J. Neurosci. 21: 1–4. 170. Valiengo L., Benseñor I.M., Goulart A.C., et al 2013. The sertraline versus electrical current therapy for treating depression clinical study (select-TDCS): results of the crossover and follow-up phases. Depress. Anxiety 30: 646–653. 171. Loo C.K., Alonzo A., Martin D., et al 2012. Transcranial direct current stimulation for depression: 3-week, randomised, sham-controlled trial. Br. J. Psychiatry 200: 52–59. 172. Plewnia C., Schroeder P.A., Kunze R., et al 2015. Keep calm and carry on: improved frustration tolerance and processing speed by transcranial direct current stimulation (tDCS). PLoS One 10: e0122578. 173. Wolkenstein L. & Plewnia C. 2013. Amelioration of cognitive control in depression by transcranial direct current stimulation. Biol. Psychiatry 73: 646–651. 174. Gorelick D.A., Zangen A. & George M.S. 2014. Transcranial magnetic stimulation in the treatment of substance addiction. Ann. N. Y. Acad. Sci. 1327: 79–93. 175. Hanlon C.A., Dowdle L.T., Austelle C.W., et al 2015. What goes up, can come down: novel brain stimulation paradigms may attenuate craving and craving-related neural circuitry in substance dependent individuals. Brain Res 1628: 199–209. 176. Mishra B.R., Nizamie S.H., Das B. & Praharaj S.K. 2010. Efficacy of repetitive transcranial magnetic stimulation in alcohol dependence: a sham-controlled study. Addiction 105: 49–55. 177. Herremans S.C., Baeken C., Vanderbruggen N., et al 2012. No influence of one right-sided prefrontal HF-rTMS session on alcohol craving in recently detoxified alcohol-dependent patients: results of a naturalistic study. Drug Alcohol Depend. 120: 209–213. 178. Herremans S.C., Vanderhasselt M.-A., De Raedt R. & Baeken C. 2013. Reduced intra-individual reaction time variability during a Go-NoGo task in detoxified alcohol-dependent patients after one right-sided dorsolateral prefrontal HF-rTMS session. Alcohol Alcohol. 48: 552–557. 179. Höppner J., Broese T., Wendler L., et al 2011. Repetitive transcranial magnetic stimulation (rTMS) for treatment of alcohol dependence. World J. Biol. Psychiatry 12(Suppl.): 57–62. 180. Rapinesi C., Kotzalidis G.D., Serata D., et al 2013. Efficacy of add-on deep transcranial magnetic stimulation in comorbid alcohol dependence and dysthymic disorder: three case reports. Prim. Care Companion CNS Disord. 15: PCC.12m01438. 181. De Ridder D., Vanneste S., Kovacs S., et al 2011. Transient alcohol craving suppression by rTMS of dorsal anterior cingulate: an fMRI and LORETA EEG study. Neurosci. Lett. 496: 5–10.

436

182. Boggio P.S., Sultani N., Fecteau S., et al 2008. Prefrontal cortex modulation using transcranial DC stimulation reduces alcohol craving: a double-blind, sham-controlled study. Drug Alcohol Depend. 92: 55–60. 183. Nakamura-Palacios E.M., de Almeida Benevides M.C., da Penha Zago-Gomes M., et al 2012.Auditory event-related potentials (P3) and cognitive changes induced by frontal direct current stimulation in alcoholics according to Lesch alcoholism typology. Int. J. Neuropsychopharmacol. 15: 601–616. 184. da Silva M.C., Conti C.L., Klauss J., et al 2013. Behavioral effects of transcranial direct current stimulation (tDCS) induced dorsolateral prefrontal cortex plasticity in alcohol dependence. J. Physiol. (Paris) 107: 493–502. 185. Klauss J., Penido Pinheiro L.C., Silva Merlo B.L., et al 2014. A randomized controlled trial of targeted prefrontal cortex modulation with tDCS in patients with alcohol dependence. Int. J. Neuropsychopharmacol. 17: 1793–1803. 186. Politi E., Fauci E., Santoro A. & Smeraldi E. 2008. Daily sessions of transcranial magnetic stimulation to the left prefrontal cortex gradually reduce cocaine craving. Am. J. Addict. 17: 345–346. 187. Camprodon J.A., Martínez-Raga J., Alonso-Alonso M., et al 2007. One session of high frequency repetitive transcranial magnetic stimulation (rTMS) to the right prefrontal cortex transiently reduces cocaine craving. Drug Alcohol Depend. 86: 91–94. 188. Li X., Malcolm R.J., Huebner K., et al 2013. Low frequency repetitive transcranial magnetic stimulation of the left dorsolateral prefrontal cortex transiently increases cue-induced craving for methamphetamine: a preliminary study. Drug Alcohol Depend. 133: 641–646. 189. Gorini A., Lucchiari C., Russell-Edu W. & Pravettoni G. 2014. Modulation of risky choices in recently abstinent dependent cocaine users: a transcranial direct-current stimulation study. Front. Hum. Neurosci. 8: 661. 190. Batista E.K., Klauss J., Fregni F., et al 2015. A randomized placebo-controlled trial of targeted prefrontal cortex modulation with bilateral tDCS in patients with crack-cocaine dependence. Int. J. Neuropsychopharmacol. 18. 191. Hayashi T., Hyun J., Strafella A.P. & Dagher A. 2013. Dorsolateral prefrontal and orbitofrontal cortex interactions during self-control of cigarette craving. Proc. Natl. Acad. Sci. U.S.A. 110: 4422–4427. 192. Li X., Hartwell K.J., Owens M., et al 2013. Repetitive transcranial magnetic stimulation of the dorsolateral prefrontal cortex reduces nicotine cue craving. Biol. Psychiatry 73: 714–720. 193. Eichhammer P. 2003. High-frequency repetitive transcranial magnetic stimulation decreases cigarette smoking. J. Clin. Psychiatry 64: 951–953. 194. Dinur-Klein L., Dannon P., Hadar A., et al 2014. Smoking cessation induced by deep repetitive transcranial magnetic stimulation of the prefrontal and insular cortices: a prospective, randomized controlled trial. Biol. Psychiatry 76: 742–749. 195. Fregni F., Liguori P., Fecteau S., et al 2008. Cortical stimulation of the prefrontal cortex with transcranial direct current stimulation reduces cue-provoked smoking craving: a randomized, sham-controlled study. J. Clin. Psychiatry 69: 32–40. 196. Boggio P.S., Liguori P., Sultani N., et al 2009. Cumulative priming effects of cortical stimulation on smoking cue-induced craving. Neurosci. Lett. 463: 82–86. 197. Fecteau S., Agosta S., Hone-Blanchet A., et al 2014. Modulation of smoking and decision-making behaviors with transcranial direct current stimulation in tobacco smokers: a preliminary study. Drug Alcohol Dep. 140: 78–84. 198. Thirugnanasambandam N., Grundey J., Adam K., et al 2011. Nicotinergic impact on focal and non-focal neuroplasticity induced by non-invasive brain stimulation in non-smoking humans. Neuropsychopharm. 36: 879-86. 199. Rapinesi C., Curto M., Kotzalidis G.D., et al 2015. Antidepressant effectiveness of deep Transcranial Magnetic Stimulation (dTMS) in patients with Major Depressive Disorder (MDD) with or without Alcohol Use Disorders (AUDs): a 6-month, open label, follow-up study. J. Affect. Disord. 174: 57–63. 200. Roth Y., Zangen A. & Hallett M. 2002. A coil design for transcranial magnetic stimulation of deep brain regions. J. Clin. Neurophysiol. 19: 361–370. 201. Kedzior K., Gellersen H., Brachetti A. & Berlim M. 2015. Deep transcranial magnetic stimulation (DTMS) in the treatment of major depression: an exploratory systematic review and meta-analysis. J. Affect. Dis. 187: 73–83. 202. Ceccanti M. & Inghilleri M. 2015. Deep TMS on alcoholics: effects on cortisolemia and dopamine pathway modulation. A pilot study. Can. J. Psychiatry 290: 283–290. 203. Goudriaan A.E., De Ruiter M.B., Van Den Brink W., et al 2010. Brain activation patterns associated with cue reactivity and craving in abstinent problem gamblers, heavy smokers and healthy controls: an fMRI study. Addict. Biol. 15: 491–503. 204. Fu L.P., Bi G.H., Zou Z.T., et al 2008. Impaired response inhibition function in abstinent heroin dependents: an fMRI study. Neurosci. Lett. 438: 322–326. 205. Lev-Ran S., Shamay-Tsoory S.G., Zangen A. & Levkovitz Y. 2012. Transcranial magnetic stimulation of the ventromedial prefrontal cortex impairs theory of mind learning. Eur. Psychiatry27: 285–289.

437

206. Rose J.E., McClernon F.J., Froeliger B., et al 2011. Repetitive transcranial magnetic stimulation of the superior frontal gyrus modulates craving for cigarettes. Biol. Psychiatry 70: 794–799. 207. Hanlon C.A., Canterberry M., Taylor J., et al 2013. Functional and structural integrity of frontal-striatal circuits in cocaine users: an interleaved TMS-BOLD & DTI study: Abstract Biol. Psych. 73: 94S. 208. Hallett M. 2007. Transcranial magnetic stimulation: a primer. Neuron 55: 187–199. 209. Isserles M., Rosenberg O., Dannon P., et al 2011. Cognitive-emotional reactivation during deep transcranial magnetic stimulation over the prefrontal cortex of depressive patients affects antidepressant outcome. J. Affect. Disord. 128: 235–242. 210. Amiaz R., Levy D., Vainiger D., et al 2009. Repeated high-frequency transcranial magnetic stimulation over the dorsolateral prefrontal cortex reduces cigarette craving and consumption. Addiction 104: 653–660. 211. Moyer A., Finney J.W., Swearingen C.E. & Vergun P. 2002. Brief interventions for alcohol problems: a meta-analytic review of controlled investigations in treatment-seeking and non-treatment-seeking populations. Addiction 97: 279–292. 212. McGirr A., Van den Eynde F., Tovar-Perdomo S., et al 2015. Effectiveness and acceptability of accelerated repetitive transcranial magnetic stimulation (rTMS) for treatment-resistant major depressive disorder: an open label trial. J. Affect. Disord. 173: 216–220. 213. Baeken C., Marinazzo D., Wu G.-R., et al 2014. Accelerated HF-rTMS in treatment-resistant unipolar depression: Insights from subgenual anterior cingulate functional connectivity. World J. Biol. Psychiatry 15: 286–297. 214. Holtzheimer P.E., McDonald W.M., Mufti M., et al 2010. Accelerated repetitive transcranial magnetic stimulation for treatment-resistant depression. Depress. Anxiety 27: 960–963. 215. Cook I.A., Schrader L.M., DeGiorgio C.M., et al 2013. Trigeminal nerve stimulation in major depressive disorder: acute outcomes in an open pilot study. Epilepsy Behav. 28: 221–226. 216. Hamada M., Murase N., Hasan A., et al 2013. The role of interneuron networks in driving human motor cortical plasticity. Cereb. Cortex 23: 1593–1605. 217. Hamada M. & Ugawa Y. 2010. Quadripulse stimulation–a new patterned rTMS. Restor. Neurol. Neurosci. 28: 419–424. 218. Hamada M., Terao Y., Hanajima R., et al 2008. Bidirectional long-term motor cortical plasticity and metaplasticity induced by quadripulse transcranial magnetic stimulation. J. Physiol. 586: 3927–3947. 219. Grundey J., Thirugnanasambandam N., Kaminsky K., et al 2012. Neuroplasticity in cigarette smokers is altered under withdrawal and partially restituted by nicotine exposition. J. Neurosci. 32: 4156–4162. 220. Goldstein B.L. & Klein D.N. 2014. A review of selected candidate endophenotypes for depression. Clin. Psychol. Rev. 34: 417–427.

438

AppendixIII.This study was published in the Frontiers in Neuroscience:

Dunlop, K., Woodside, B., Downar, J. (2015). Targeting Neural Endophenotypes of Eating Disorders with Non-invasive Brain Stimulation. Frontiers in Neuroscience, (10): 30.

Abstract

Abstract

The term “eating disorders” (ED) encompasses a wide variety of disordered eating and compensatory behaviors, and so the term is associated with considerable clinical and phenotypic heterogeneity. This heterogeneity makes optimizing treatment techniques difficult. One class of treatments is non-invasive brain stimulation (NIBS). NIBS, including repetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS), are accessible forms of neuromodulation that alter the cortical excitability of a target brain region. It is crucial for NIBS to be successful that the target is well selected for the patient population in question. Targets may best be selected by stepping back from conventional DSM-5 diagnostic criteria to identify neural substrates of more basic phenotypes, including behavior related to rewards and punishment, cognitive control, and social processes. These phenotypic dimensions have been recently laid out by the Research Domain Criteria (RDoC) initiative. Consequently, this review is intended to identify potential dimensions as outlined by the RDoC and the underlying behavioral and neurobiological targets associated with ED. This review will also identify candidate targets for NIBS based on these dimensions and review the available literature on rTMS and tDCS in ED. This review systematically reviews abnormal neural circuitry in ED within the RDoC framework, and also systematically reviews the available literature investigating NIBS as a treatment for ED.

Introduction

The term “eating disorders” (ED) encompasses a wide variety of disordered eating and compensatory behaviors that inappropriately alter the patient's body shape or weight, or the subjective experience of one's own body shape or weight. According to recent studies, the lifetime prevalence of EDs is 5.7% for females, and 1.2% in males (Golden et al., 2003; Hudson et al., 2007; Smink et al., 2014). The lifetime prevalence of the top three EDs according to the DSM-5 diagnostic criteria is 2.3, 1.7, and 0.8% for adolescent binge eating disorder (BED), anorexia nervosa (AN), and bulimia nervosa (BN), respectively (Golden et al., 2003; Hudson et al., 2007; Smink et al., 2014). BED is associated with recurrent episodes of binging, typically during negative affect (Leehr et al., 2015), and with the absence of inappropriate compensatory behaviors to avoid weight gain. Both AN and BN are associated with disturbances in the subjective experience of one's own body shape or weight; this phenotype is also known as body dysmorphia. BN is also defined by recurrent episodes of binge eating, with inappropriate compensatory behaviors to avoid weight gain; such behaviors include vomiting, excessive exercise, laxative misuse or diuretic misuse. In contrast, AN is defined by the persistent

439

restriction of food, an intense fear of gaining weight, and a significantly low body weight for one's developmental stage. AN has two subtypes, restricting (ANR) and binge-eating/purging (ANBP), with the latter distinguished from the former by the presence of binges and/or purges.

Despite a low lifetime prevalence rate relative to other psychiatric disorders, EDs carry a significant burden of illness, both socially and individually. Treatment capacity in specialized ED programs is presently inadequate to meet demand (Hart et al., 2011), and for patients who do manage to access specialized programs, economic difficulties and high costs often hamper treatment adherence (Gatt et al., 2014). EDs are also associated with a high mortality rate; for one, approximately 10% of AN sufferers will die within 10 years of disease onset (Sullivan, 1995). According to a recent meta-analysis, the overall standard mortality ratio (SNR) for AN is 5.86, higher than schizophrenia (2.8), bipolar disorder (2.1), and major depression (1.6) (Arcelus et al., 2011). Conventional ED treatments, including pharmacotherapy, and in- and out-patient behavioral therapies, are associated with suboptimal recovery rates (~50% for AN; ~45% for BN; ~50–70% for BED), high relapse rates (ranging from 9 to 65%), and high chronicity (~20% will develop a chronic disorder; Olmsted et al., 2005; Mitchell et al., 2007; Shapiro et al., 2007; Carter et al., 2012; Hay et al., 2012; Hilbert et al., 2012; Amianto et al., 2015). ANBP, in particular, has the poorest prognosis of the eating disorders (Steinhausen and Weber, 2009). EDs are also highly co-morbid with other psychiatric disorders, such as major depression and obsessive-compulsive disorder, whose presence negatively impacts treatment outcomes (Godart et al., 2003; Crane et al., 2007; Mischoulon et al., 2011). Thus, new treatment approaches are urgently needed, especially for the substantial proportion of ED patients who are unresponsive to conventional treatment strategies.

Neuromodulation technologies are beginning to emerge as a promising new treatment option for treatment resistant ED patients. The potential usefulness of these techniques was recently illustrated in a pilot study using subgenual cingulate deep brain stimulation (DBS) to achieve symptomatic improvements in severe and treatment-refractory AN (Lipsman et al., 2013). Although potentially powerful, DBS remains for the moment a fairly invasive treatment, and is available only to small volumes of patients in specialist neurosurgical centers. A more accessible alternative is non-invasive brain stimulation (NIBS), including techniques such as repetitive transcranial magnetic stimulation (rTMS) and transcranial direct current stimulation (tDCS). rTMS uses rapid pulses of an electromagnetic field to elicit action potentials in the target area of cortex. tDCS uses a weaker intensity electrical stimulus, delivered by scalp electrodes, to modulate cortical excitability in the underlying regions. Both NIBS strategies attempt to alter the cortical excitability of a target brain region to normalize particular disorder-specific phenotypes. Cortical targets are typically selected based on abnormal structural or functional attributes in the disorder relative to healthy controls. Appropriate cortical targeting using NIBS is critical for optimal treatment efficacy (Fox et al., 2013). Therefore, a proper understanding of the neural substrates, as well as the cognitive and behavioral phenotypes accompanying these substrates, is crucial for optimizing future treatments.

Two major issues associated with NIBS as a treatment for ED are the tremendous heterogeneity in the cognitive and behavioral phenotypes of patients within this illness category, and the dynamic course of the illness, in which patients can switch from one ED diagnosis to another over time (Garfinkel et al., 1995; Keel and Mitchell, 1997; Lilenfeld et al., 1998; Sullivan et al., 1998; Strober et al., 2000; Bulik et al., 2005; Milos et al., 2005). This

440

variability within a single diagnosis and this malleability of symptoms is a likely contributor to the limited clinical efficacy of NIBS (and in conventional treatments more generally) for ED.

Two possible solutions to address this heterogeneity are genomic methods, such as phenotypic linkage analyses, as well as neuroimaging methods. Such tools stratify patients on underlying behavioral, genetic, and neuropathological dimensions rather than self-reported symptoms alone. Therefore, these tools may be useful to identify the underlying behavioral and neuropathological endophenotypes related to more basic dimensions of behavior, independent of DSM-5 diagnoses. Such analyses and resulting endophenotypes can also be related to the behavioral and circuit-based dimensions of the recently described Research Domain Criteria (Insel et al., 2010) (RDoC). The RDoC is a recent strategy aimed at integrating basic neuroscientific knowledge with clinical diagnoses by first describing fundamental behaviors, described below, as dimensions. These dimensions are then used to describe the pathological behaviors of psychiatric disorders. By using the RDoC schema in combination with neuroimaging and phenotypic linkage methods, we may be able to identify sufficient stimulatory targets addressing specific phenotypes such as restrictive behavior or binging, regardless of DSM-5 diagnosis. For NIBS treatments, diagnostic systems must be capable of parsing this heterogeneity using endophenotypes so we may select the optimal stimulation target for a particular behavioral marker, or neural substrate.

Here, we will review NIBS as a treatment for the three most prevalent forms of ED: AN, BN, and BED. First, we will posit candidate dimensions as outlined by the RDoC and their underlying behavioral and neurobiological targets associated with ED as potential candidates for NIBS. Second, we will review the available literature on rTMS and tDCS as possible treatments for ED. Lastly, we will discuss the current limitations of the NIBS-ED field, and opportunities of future study and development.

GoingBeyondtheDSM-5Diagnosis:HowCanWeMaximizeEfficacy?

As discussed above, one of the major obstacles in ED diagnosis and treatment is the heterogeneity within each diagnostic category; conversely, comparisons of clinical and psychological features across patients suggest that there is significant overlap between ED diagnoses (Garfinkel et al., 1995; Lilenfeld et al., 1998; Sullivan et al., 1998; Strober et al., 2000). Compounding this problem is the evolution of the illness, such that patients may transition from one diagnostic category to another over time (Bulik et al., 2005). For example, it is estimated that approximately 50% of patients initially diagnosed with ANR will develop binge/purge behaviors, and approximately 30% of BN patients have a history of AN (Keel and Mitchell, 1997). In another study following DSM-IV-diagnosed AN and BN, only one third of subjects retained their original diagnosis after 30 months (Milos et al., 2005). To improve diagnostic consistency and treatment efficacy may require us to identify more stable, more granular, and more biologically based subgroups, or endophenotypes, within the ED population.

Some classification efforts have focused on a single DSM diagnosis. AN has been subdivided into 3 stable classes based on co-occurring symptoms: fat-phobic ANR, fat-phobic ANBP, and non-fat-phobic ANR (Wildes et al., 2013). BN has been subdivided based on

441

personality attributes (affective-perfectionistic, impulsive and low-comorbid psychopathology clusters Wonderlich et al., 2005) and based on presenting symptoms (binging, purging, and bingeing-purging, Striegel-Moore et al., 2005).

A number of latent class (LCA) and latent profile analyses (LPA) have been performed on symptomatic and personality factors to stratify endophenotypes spanning AN and BN. One symptom-based LCA found optimal fitting for a 4-group classification. ANR, ANBP/BN, ANR without OCD, and BN with only vomiting as purging were the four groups identified (Keel et al., 2004). Another symptom-based LPA identified 4 ED classes: binging with multiple types of compensatory behavior; binging with only vomiting as compensatory behavior; binging without purging; and low/normal weight with excessive exercise (Eddy et al., 2009).

As evidenced above, there now exist a variety of different proposals for how best to subcategorize ED patients, within and across DSM-5 diagnoses. How, therefore, can we converge upon a system that offers maximum clinical usefulness? One potentially fruitful method would be to better characterize the heterogeneity among ED patients in biological terms, using techniques such as positron emission tomography (PET), electroencephalography (EEG) and magnetic resonance imaging (MRI) to identify distinct neurobiological substrates underlying the different subgroups within ED. Clinical endophenotypes could then be tied to neurobiological substrates, which could in turn serve as targets for individually- or phenotypically-tailored treatment strategies.

Such an approach would also allow us to describe illnesses in dimensional rather than categorical terms. For example, the influential RDoC framework (Insel et al., 2010) includes dimensional constructs such as positive valence, negative valence, cognitive systems, social processes, and arousal and regulatory systems (for a review of how RDoC dimensions relate to ED neurobiology, see Wildes and Marcus, 2015, Table 1). Many endophenotypes, previously identified as symptom clusters in the ED population, can be framed parsimoniously as the result of pathology affecting these dimensions, either singly or in combination (Figure 1). An “RDoC formulation” of our ED endophenotypes carries the advantage of pointing toward specific cognitive processes, neural pathways, neurotransmitter systems, molecular targets, or genes that might be targeted for therapeutic effect. For the purposes of this review, we will confine our discussion to potential novel uses of NIBS to target specific neural pathways that are associated with RDoC constructs, as they relate to specific endophenotypes within the ED population.

RDoCDomainsasEDEndophenotypesandNIBSTargets

For the following section, a systematic review was completed using PubMed (NIH, http://www.ncbi.nlm.nih.gov/pubmed), with searches containing the following terms: first, clinical terms for the three ED diagnoses in this review (bulimia nervosa, anorexia nervosa, and binge eating disorder), and second, RDoC related terms as discussed in a recent review on RDoC cognitive systems (Morris and Cuthbert, 2012).

NegativeValenceSystems

442

Negative valence systems are activated in response to aversive stimuli, and include fear, anxiety and loss-related behaviors. In a recent meta-analysis investigating neural activations for negative and positive affect, negative valence was associated with greater activation in the amygdala and anterior insula (Lindquist et al., 2015). The lateral orbitofrontal cortex (OFC) is also associated with negative valence, particularly during the anticipation and receipt of punishment (Ursu et al., 2008).

A number of studies support the role of negative valence systems in ED, mainly in behaviors associated with negative affect, sensitivity to punishment, anxiety, harm avoidance, and response to the receipt of punishment (Figure 1). For example, behavioral measures of negative affect and negative urgency are the two most predictive features before a binge episode in both BED and BN (Bohon and Stice, 2012; Berg et al., 2015; Leehr et al., 2015; Racine et al., 2015). On functional neuroimaging, BN patient reported negative affect is related to neural responsivity during the anticipation of a food reward in both the striatum and insula (Bohon and Stice, 2012). This relation suggests that negative affect and food-reward are inappropriately coupled in this disorder. More generally, BN patients also have higher neural responses to negative body image descriptors (Miyake et al., 2010), in areas associated with the regulation and inhibition of fear and emotional processing circuits, including the dorsomedial prefrontal cortex (DMPFC) (Kühn et al., 2011; Åhs et al., 2015). These findings shed light on the role of negative attentional bias in the psychopathology of bulimic-type disorders.

Restrictive subtypes of ED also show hypersensitivity on measures related to negative valence systems. Behaviorally, exaggerated harm avoidance and sensitivity to punishment are typically associated with forms of AN (Harrison et al., 2010). Similarly, on fMRI, AN patients display increased neural activation in right anterior insula and DLPFC during pain anticipation, and exaggerated responses to punishment (pain and monetary losses) in the DLPFC, and the anterior, mid-, and motor cingulate (Bischoff-Grethe et al., 2013; Strigo et al., 2013; Bar et al., 2015). Cowdrey and colleagues also found an exaggerated response to an aversive taste and sight of food in the insula, striatum and ACC (Cowdrey et al., 2011). Trait-anxiety is also a common feature of AN, and is associated with the exaggerated activity of fear-related circuits to food and body-related cues. Regions of exaggerated response to symptom-provoking stimuli include the amygdala, hippocampus, insula, ACC, and medial PFC (Ellison et al., 1998; Frank et al., 2002, 2012b; Seeger et al., 2002; Uher et al., 2004; Friederich et al., 2010; Vocks et al., 2010). Finally, at the receptor level, PET imaging reveals increased striatal dopamine binding potential and altered cingulate serotonergic (increased 5-HT1A, but decreased 5-HT2A) binding potential is associated with harm avoidance in AN (Bailer et al., 2004, 2007; Frank et al., 2005).

SummaryofPotentialNegativeValenceTargets

Both bulimic and restrictive-type EDs display some form of negative valence abnormality on behavioral and neuroimaging modalities (Figure 2). In ED with a binging component, it appears that negative affect and food-reward responsivity are intimately coupled via the exaggerated response of the amygdala, insula and DMPFC. Restriction-related EDs display a similar pattern in the amygdala, right anterior insula, DLPFC and mPFC accompanying aspects of harm avoidance and receipt of punishment. Frontal regions, particularly the medial PFC and DMPFC, are thought to inhibit activity of the basolateral amygdala (BLA) (Cho et al., 2013; Felix-Ortiz et al., 2015).

443

Hyperactivation of the DMPFC, DLPFC, and anterior insula during negative valence paradigms has two possible interpretations. First, these areas may be inhibiting BLA activity, but insufficiently, in which case excitatory stimulation may be beneficial. Alternatively, these areas may actually be inappropriately driving BLA activity in a top-down fashion, in which case inhibitory stimulation would be preferable. A key study illustrated these opposite mechanisms in emotion regulation in healthy controls vs. MDD patients (Johnstone et al., 2007): during emotional reappraisal, limbic frontal regions suppressed amygdala activity in controls, but counterproductively increased amygdala activity in MDD.

For NIBS interventions, direct suppression of the amygdala is challenging due to its deep location; strategies aimed at damping negative valence systems will therefore likely target in prefrontal cortex and insula. Excitatory prefrontal NIBS has been recently shown to attenuate amygdala-dependent negative processing in healthy controls (Baeken et al., 2010; Guhn et al., 2014), and this strategy may be best in “bottom-up” pathology, where emotional reappraisal/self-regulation systems are underactive rather than pathologically hyperactive (i.e., in BN and BED). Conversely, where negative valence systems are driven by “top-down” pathology, and self-regulation is if anything excessive, inhibitory stimulation may be preferable. Inhibitory NIBS of the DMPFC and lateral OFC both show promise in obsessive-compulsive disorder (Mantovani et al., 2010; Nauczyciel and Drapier, 2012; Dunlop K. et al., 2015), and these strategies may be better suited to AN-R, particularly in cases with comorbid OCD.

PositiveValenceSystems

Positive valence systems encompass neural circuits related to motivation, reward seeking, and habit formation behaviors. According to a recent meta-analysis in healthy controls, positive stimuli are associated with activity in the VMPFC and ACC (Lindquist et al., 2015). All three major EDs, AN, BN, and BED, have been previously shown to be altered in this dimension (Figure 1).

From a behavioral perspective, AN patients show diminished sensitivity to conventional reward, as evident on psychometric measures (Harrison et al., 2010; Glashouwer et al., 2014) and delay discounting tasks (Steinglass et al., 2012). From a neurobiological perspective, ANR patients likewise display a blunted neural response to food reward in the insula and striatum (Wagner et al., 2008), decreased response to food images in the insula (Holsen et al., 2012; Oberndorfer et al., 2013b), and altered striatal activation during a reward-learning paradigm (Wagner et al., 2007). In a recent fMRI study of delay discounting in AN patients and healthy controls, AN patients had a marked preference for delayed rewards, associated with lower activation in the striatum and dorsal ACC during decision-making; these behavioral and neural abnormalities normalized to control levels after treatment (Decker et al., 2014). However, another study found that weight restoration did not affect choice behavior on a delay discounting task (Ritschel et al., 2015), suggesting that a preference for delayed over immediate rewards may be an endophenotypic feature in low-BMI individuals. In either case, the identified striatal and prefrontal regions are all involved in the motivational aspect of reward and food-reward processing.

There is also evidence that reward evaluation is altered in AN, in which secondary (contextual) rewards such as exercise and dietary restriction carry higher reward value relative to food or other primary rewards (Schebendach et al., 2007; Klein et al., 2010). The so-called

444

“reward contamination theory” of AN posits a pathological re-configuration of the patient's reward system through stress-induced activation of the mesolimbic dopamine system, via ventral tegmental area opioid receptors. In this framework, AN behaviors essentially represent a maladaptive, but well-entrenched type of habit-formation (Keating et al., 2012; Walsh, 2013).

The findings that support this theory suggest that there is altered motivational salience for disease-related stimuli. For example, AN patients tend to rate physical exercise as “pleasant,” more so than food (Giel et al., 2013). In fact, food-reward in AN activates a weight-gain fear response (i.e., negative valence systems) in the amygdala and extrastriate body rather than positive valence systems from the striatum, orbitofrontal cortex, and ACC (Vocks et al., 2011). The DLPFC is also hyperactive in response to images of food and the anticipation of reward, suggesting the presence of enhanced cognitive control over food cues and reward (Ehrlich et al., 2015; Sanders et al., 2015). ANR patients also have a high prevalence of comorbid OCD (Torresan et al., 2013). The level of compulsivity predicts the reactivity of the superior frontal gyrus, ACC and striatum and deactivation of the PFC to images of high-calorie foods (Rothemund et al., 2011), and lowered right DLPFC activity is seen in response to obsessive-compulsive symptom provocation in AN (Suda et al., 2014). Thus, hypofunctioning of primary reward systems (and potentially, hyperfunctioning of secondary/contextual reward systems) may be important target processes in ANR.

In contrast disorders in the BN/BED spectrum are often associated with elevated primary reward valuation and reward sensitivity. These are typically associated with a higher willingness to work for a food reward (Schebendach et al., 2013), as well as higher impulsivity (Manwaring et al., 2011; Chan et al., 2013; Mole et al., 2015) At the neural level, BN and BED patients show increased activity for reward receipt in areas including the medial OFC, ventral striatum and insula (Schienle et al., 2009; Frank et al., 2011, 2012a; Radeloff et al., 2012; Weygandt et al., 2012; Oberndorfer et al., 2013a). BED patients display hyperactivations in the ventral striatum and inferior frontal gyrus during reward anticipation, and reduced medial PFC activity during a monetary incentive delay task (Balodis et al., 2014, 2013a). On PET imaging, areas like the insula, PFC and ventral striatum, associated with reward-motivation and food-reward processing, have altered serotonergic and dopaminergic binding in BN (Broft et al., 2012; Galusca et al., 2014). An important associated feature may also be deficient behavioral self-regulation and impulsivity. BN patients also show reduced activation in anticipation of a food reward is seen in ACC and right anterior insula; lower ACC activity predicts how much the patient will overeat (Frank et al., 2006; Bohon and Stice, 2011). Parallels have been drawn between the neural substrates of BN/BED and addiction, due to the similar alterations to motivation and reward-related circuitry on fMRI and task-based paradigms between the two psychopathologies (Filbey et al., 2012).

SummaryofPotentialPositiveValenceTargets

In terms of positive valence systems, it appears that both restrictive and binging phenotypes of ED display alterations in incentive salience that is potentially modulated by the opioid system (Keating et al., 2012; Giuliano and Cottone, 2015; Figure 3). In the case of ANR, conventional primary rewards appear to be devalued in favor of pathological secondary or contextual rewards, such as starvation and excessive exercise. A broader preference for long-term/contextual over immediate primary rewards is also apparent in choice behavior during delay discounting. Neurally, the primary reward systems of the ventral striatum and ventromedial

445

prefrontal cortex appear hypoactive, while contextual or secondary reward systems operating through lateral orbitofrontal and lateral temporal regions appear hyperactive. Hyperactivity in lateral orbitofrontal pathways is also strongly associated with OCD, and with compulsivity in general (Ahmari et al., 2013; Beucke et al., 2013). This finding would be consistent with the broader phenotype of ANR. Neurally-based strategies in ANR might therefore include enhancing primary reward value via medial prefrontal-striatal pathways, or attenuating secondary reward value via lateral prefrontal-striatal pathways. For instances where conventional rewards are less valued than maladaptive ones (restrictive, fat-phobic ED), inhibitory NIBS over lateral networks for maladaptive secondary rewards, and excitatory NIBS over medial networks for conventional rewards, may be a possible therapeutic protocol to realign incentive-salience mechanisms to normal, adaptive functioning.

In the case of binge/purge-related EDs, repeated exposures to the transient reward value of food intake (or the transient anti-anxiety effect of purging) would cause these behaviors to acquire pathologically high incentive value (especially in the presence of negative urgency), via neural mechanisms that parallel those of addiction. Effective strategies would therefore parallel those for substance addiction: enhancing cognitive/impulse control over urges to binge and purge, or suppressing urge intensity.

NIBS strategies for enhancing cognitive control involve excitatory stimulation of the nodes of the salience network, including the DLPFC, dACC, and insula (Dunlop et al., accepted). Each of these targets have demonstrated efficacy in substance dependence (Mishra et al., 2010; De Ridder et al., 2011; Meng et al., 2014), with effects apparently mediated by enhanced control rather than reduced urge. Recently, excitatory rTMS over the dACC has been reported to reduce symptoms in treatment-resistant binge/purge ED, via enhanced integrity of frontostriatal circuits in the salience network (Dunlop K. et al., 2015).

NIBS may also be capable of suppressing urge, by targeting frontopolar and ventromedial sites. In one preclinical rTMS study, substance use disorder patients underwent inhibitory rTMS over the ventral frontal pole during a task evoked a cue-related craving response. A single session of inhibitory rTMS reduced the severity of craving in this group relative to sham, and stimulation proved capable of engaging core reward nodes in the ventral striatum, as well as the associated ventromedial prefrontal regions (Hanlon et al., 2013, 2015). Urge suppression via inhibitory ventromedial prefrontal stimulation has yet to be attempted in ED, but would be a reasonable strategy to complement excitatory salience-network stimulation in binge/purge-related ED populations.

CognitiveSystems

The cognitive systems dimension refers to processes responsible for cognitive processing, including attention, perception, memory, language, and cognitive control. In healthy control studies, these behaviors are associated with activity in the DMPFC, DLPFC, and anterior insula (Albares et al., 2014; Cho et al., 2014; Luo et al., 2014; Reineberg et al., 2015). These networks tend to be associated with the central executive and salience resting-state networks (Reineberg et al., 2015), responsible for response selection and inhibition.

Abnormal cognitive control mechanisms are evident in most ED populations (Figure 1). On the one hand, BN and BED-type diagnoses tend to display reduced capacity for impulse and

446

cognitive control. This is particularly evident for disease-relevant stimuli (Wu et al., 2013), but is also apparent for positive and negative emotional valence images (Tapajóz P de Sampaio et al., 2015), suggesting a broader endophenotype of deficient cognitive and behavioral control. In fact, binge episodes are partially defined by the individual's loss of control during eating, and impulse control disorders (ICD) are common comorbidities (Fernández-Aranda et al., 2008). Purging behaviors are also associated with higher levels of impulsivity, and different forms of purging may represent separate manifestations of compulsivity and impulsivity (Hoffman et al., 2012).

On the other hand, restrictive-type EDs tend to show a different profile of cognitive control abnormalities. Cognitive control capacity may appear above normal levels in certain domains, such as temporal discounting (Steinglass et al., 2012). However, cognitive control may be abnormal in certain specific domains related to the illness; for example, for negative valence images (Tapajóz P de Sampaio et al., 2015), food stimuli (Oberndorfer et al., 2013b; Sanders et al., 2015), or body-image-related stimuli (Lee et al., 2014). AN patients also have altered cognitive control depending on the reward valence of the object, as the impulse control networks are overly activated for physical exercise relative to food images in a go/no-go task (Kullmann et al., 2014). AN patients also show a reduced ability to switch to an optimal decision-making strategy, called cognitive flexibility (Zastrow et al., 2009).

From a neural perspective, impulsive-type deficits on response control tasks are related to lower frontostriatal activations. BED patients show reduced activity in the inferior frontal gyrus, ventromedial PFC and insula during the Stroop task, and this diminished activity is associated with poor dietary restraint (Balodis et al., 2013b). BED prefrontal hypoactivity has also been correlated with psychometric measures of attentional impulsiveness and a disease-relevant go/no-go task (Hege et al., 2014). BN patients show hypoactivity in frontostriatal circuitry during cognitive control tasks like the Simon Spatial Incompatibility task; affected areas include the inferior frontal gyrus, striatum, ACC, OFC, DLPFC, and middle frontal gyrus (Marsh et al., 2009, 2011; Celone et al., 2011). On the go/no-go task, adolescent BN and ANBP patients display hyperactivations in the ACC and right DLPFC, albeit without impaired task performance relative to controls (Lock et al., 2011).

AN patients also show hypoactivity in frontostriatal circuits from the medial PFC on a response inhibition task related to cognitive control deficits (Oberndorfer et al., 2011; Wierenga et al., 2014), but hyperconnectivity to a response inhibition task that used exercise-related stimuli as its cue (Kullmann et al., 2014). Additionally, AN patients also display poorer performance on cognitive flexibility tasks, and this performance is reflected by lower activity in frontostriatal circuits through the thalamus, ventral striatum, ACC, middle frontal gyrus, and ventrolateral PFC (Zastrow et al., 2009; Sato et al., 2013; Garrett et al., 2014; Wildes et al., 2014; Lao-Kaim et al., 2015). On resting-state fMRI, higher thalamo-cortical functional connectivity through the DLPFC and anterior PFC is associated with poorer performance on the Stroop task and working memory (Biezonski et al., 2015). Thus, domain-specific abnormalities of cognitive control are evident at both the behavioral and the neural level in AN.

SummaryofPotentialCognitiveControlTargets

Both restricting- and binge/purge-type EDs show deficits on tasks related to cognitive control, including behavioral inhibition, working memory, selective attention, and cognitive flexibility (Figures 1, 4). Generally, BED displays poorer response inhibition and lower activity

447

in the inferior frontal gyrus and ventromedial PFC, both of which are accessible via excitatory forms of NIBS. BN and ANBP display lower activity in the inferior frontal gyrus, ACC, OFC, and DLPFC; all but the OFC are easily accessible for excitatory NIBS. As noted earlier, excitatory NIBS of salience-network nodes in DLPFC, DMPFC, and anterior insula appears to improve cognitive control and impulsivity even in healthy controls (Cho et al., 2014, 2010; Meng et al., 2014). Enhanced cognitive control, via improved frontostriatal connectivity through these salience-network nodes, may mediate recently reported improvements in binge and purge behaviors with excitatory DMPFC-rTMS (Dunlop K. et al., 2015). Similar effects via similar mechanisms should be expected for excitatory rTMS targeting DLPFC and anterior insula.

For AN, neural correlates of cognitive control show considerable variability depending on the task and valence of stimuli. On the one hand, AN patients in some studies show broad deficits of cognitive control and flexibility, and hypoactivity of the frontostriatal circuitry, during many tasks related to cognitive control; hence, excitatory NIBS might be beneficial if combined with cognitive tasks during stimulation. On the other hand, patients sometimes show the reverse pattern of hyperconnectivity and excessive cognitive control/compulsivity in these same circuits, within illness-specific domains; excitatory stimulation may therefore be unhelpful, or could potentially exacerbate illness. In keeping with this concern, high-frequency DMPFC-rTMS was recently reported to exert a paradoxical inhibitory effect on frontostriatal connectivity in a subpopulation of ED patients with high baseline connectivity; these patients showed symptomatic worsening rather than improvement (Dunlop K. et al., 2015). Thus, targeting cognitive control in AN-R may require a more nuanced approach than is the case for binge-purge symptoms.

SocialProcessingSystems

Social processing systems refer to circuits involved in social communication, and the perception and understanding of oneself and others. Targets identified in healthy controls include the insula, responsible for interoception (Craig, 2002); the temporoparietral junction, for theory of mind-related processing (Saxe and Kanwisher, 2003); and higher-order visual processing regions, for processing one's own and others' faces (Hummel et al., 2013).

This dimension has received less attention in the ED literature relative to positive/negative valence systems and cognitive control (Figure 1). However, it may have relevance in AN patients, who show higher levels of alexithymia, deficits in visceral sensory perception or “interoception” (Craig, 2002; Strigo et al., 2013), and distorted perceptions of body shapes (Suchan et al., 2013). AN patients with higher levels of alexithymia show lower ACC, PCC, and right temporoparietal junction (TPJ) activation during social decision-making tasks (Miyake et al., 2009, 2012; McAdams and Krawczyk, 2011). More specifically, ANR patients display altered anterior and dorsal mid-insula activations based on the modality of interoception they are attending to (Kerr et al., 2015). On resting-state fMRI, AN patients also display increased functional connectivity from the anterior insula to the default mode network associated with self-reported problems with interoceptive awareness, suggesting a heightened level of cognitive control toward interoceptive processes (Boehm et al., 2014). AN patients also have altered neural responses to visually-presented body shapes, particularly in areas associated with visual processing and reward: the ventral striatum, extrastriate body area (EBA), DLPFC, parietal regions, medial PFC, and fusiform gyrus (Cowdrey et al., 2012; Spangler and Allen,

448

2012; Castellini et al., 2013; Fladung et al., 2013; Suchan et al., 2013; Suda et al., 2013; Fonville et al., 2014). Finally, two recent studies have also identified areas of abnormal activation in response to benevolent and malevolent social relationships. During benevolent social relations, AN patients tend to display reductions in DMPFC, possibility related to lowered reward valence for social reward and interaction (McAdams et al., 2015; Via et al., 2015).

In summary, AN patients may show deficits across multiple domains related to self-perception (alexithymia, interoception, and body shape perception) and social function (interpersonal interaction, theory of mind; Figures 1, 5). The latter function has been successfully enhanced with excitatory DMPFC-rTMS in autism-spectrum disorder (Enticott et al., 2011, 2014). During social interactions, AN patients likewise tend to display DMPFC hypoactivity during social interaction, and so excitatory stimulation over this region may worth exploring. For self-perception, relevant targets include anterior insula (alexithymia), posterior insula (interoception), TPJ and EBA (social cue perception, body shape perception). NIBS has successfully targeted each of these regions in other applications (Ciampi de Andrade et al., 2012; Dinur-Klein et al., 2014; Donaldson et al., 2015). Excitatory stimulation of the insula and TPJ may be worth exploring for alexithymia and deficits in interoception. Conversely, inhibitory stimulation of the TPJ and EBA may be worth exploring for aberrant self- and body perception.

NIBSTechniquesasTherapeuticInterventionsinED

For the following section, a systematic review was completed using PubMed (NIH, http://www.ncbi.nlm.nih.gov/pubmed), with searches containing the following terms: first, clinical terms for the three ED diagnoses in this review and related phenotypes (BN, AN, BED, binging, purging, excessive exercise), and second, NIBS related terms (rTMS, TMS, tDCS).

NIBSOverview:rTMSandtDCS

rTMS applies powerful, focused magnetic field pulses over the scalp to elicit action potentials in the underlying region of cortex. Typically, treatment sessions occur once daily, for a total of 20–30 daily sessions (Carpenter et al., 2012; Solvason et al., 2014). rTMS mechanisms are thought to involve synaptic plasticity via long-term potentiation or depression, with the direction of effect dependent on the stimulation intensity, duration, and pattern (Pascual-Leone et al., 1998; Maeda et al., 2000). Higher frequency stimulation (5–20 Hz) is usually considered to be excitatory, while low frequency (< 1 Hz) stimulation is considered inhibitory (Pascual-Leone et al., 1994; Chen et al., 1997). More recently, however, considerable heterogeneity on electrophysiological, neuroimaging, and clinical measures has been found for most if not all patterns of rTMS (Maeda et al., 2000; Eldaief et al., 2011; Dunlop K. et al., 2015; Dunlop K. et al., 2015; Nettekoven et al., 2015).

tDCS, on the other hand, uses a constant, low amplitude current to modulate cortical excitability, rather than eliciting action potentials directly. As with rTMS, sessions typically occur daily, for a total of 10–30 sessions (Meron et al., 2015). While the mechanisms of tDCS are still debated, it is likely that modulated cortical excitability also elicits subtle effects on synaptic plasticity via long-term potentiation and depression (Brunoni et al., 2012). Anodal

449

stimulation is considered excitatory, and cathodal stimulation inhibitory. However, as with rTMS, both types of tDCS display considerable inter-individual variability in their effects (Wiethoff et al., 2014). Newer variants such as transcranial alternating current stimulation (tACS), may exert more consistent, frequency-specific effects (Voss et al., 2014); however, their therapeutic potential is poorly understood at present.

NIBSasaTreatmentforBEDandFoodCraving

To date, the majority of published NIBS-ED studies have focused on female patients with abnormally high food craving or urge to eat, as opposed to a specific formal DSM-5 diagnosis (Tables 2, 3; McClelland et al., 2013; Grall-Bronnec and Sauvaget, 2014; Val-Laillet et al., 2015). These preclinical studies typically involve a single session of stimulation, with subjectively rated cue-induced craving as the primary outcome. With rTMS, two studies reported contradictory results for 10 Hz stimulation of the left DLPFC rTMS: one study (n = 28) found decreased craving after active vs. sham stimulation (Uher et al., 2005), while the other (n = 10) found that active stimulation was no better than sham in terms of cue-induced craving control (Barth et al., 2011). The studies differed in stimulation parameters, however, and enrolled only healthy participants who self-reported having strong food cravings, but did not carry a formal ED diagnosis. Hence, it may be difficult to extrapolate these findings to the effects of a full therapeutic course of 20–30 sessions in patients with pathological deficits of self-control and a formal ED diagnosis.

There is also a growing body of literature investigating DLPFC-tDCS as a method to reduce craving and food intake. In four published studies recruiting individuals with strong food cravings, a single session of anodal right DLPFC/cathodal left DLPFC tDCS was able to reduce cue-induced craving, reduce food intake, and improve the participants' ability to resist food relative to sham-tDCS (Fregni et al., 2008; Goldman et al., 2011; Kekic et al., 2014; Lapenta et al., 2014). Future work involving tDCS should employ multiple sessions as opposed to a single session in a randomized, sham-controlled setting, as a treatment for the inappropriate eating patterns associated with BED. Studies in populations carrying a formal ED diagnosis, with significant functional impairment and distress, are also needed.

NIBSasaTreatmentforBN

The earliest publication of rTMS as a potential treatment for BN is a case report of a patient diagnosed with comorbid depression and BN who achieved an unexpected remission of binge and purge symptoms and depressive improvements after 10 sessions of 20 Hz rTMS over the left DLPFC (Hausmann et al., 2004; McClelland et al., 2013; Table 2). Follow-up studies involving high frequency left DLPFC rTMS have been mixed: one group found that a single session reduced the urge to eat, the number of binges 24 h post-rTMS, and salivary cortisol levels (Van den Eynde et al., 2010; Claudino et al., 2011), while another study found no difference between active- and sham-rTMS after 15 sessions of 20 Hz rTMS over the left DLPFC (Walpoth et al., 2008). A more recent study applied a single session of excitatory left DLPFC-rTMS in 8 female patients with BN, and reported reduced subjective ratings of craving post-rTMS, along with lower cerebral oxygenation in the DLPFC on near-infrared spectroscopy (Sutoh et al., 2016). These findings hint at the potential promise of DLPFC-rTMS for BN, which would be in keeping with the much more extensive literature demonstrating that this intervention enhances cognitive control in healthy subjects (Cho et al., 2010), and patient populations (Van

450

den Eynde et al., 2010), with therapeutic effects in mechanistically related disorders such as addiction (Gorelick et al., 2014).

More recently, our group has shifted the rTMS stimulation target from the DLPFC to the DMPFC, as a potential treatment for major depression (Downar et al., 2014; Salomons et al., 2014; Bakker et al., 2015). As with first case report of DLPFC-rTMS for BN, we too found an unexpected remission of chronic treatment refractory binge and purge symptoms in an MDD patient with comorbid BN, following 20 sessions of 10 Hz DMPFC-rTMS. The onset of effect was rapid, occurring in the first week of treatment, and was maintained for 9 weeks post-treatment (Downar et al., 2012). In a follow-up, open-label series of 10 Hz DMPFC-rTMS in 28 ED patients with binge/purge behaviors, we noted ≥ 50% symptom reduction in 57%. On resting-state fMRI, we found increased resting-state functional connectivity in fronto-striatal salience network circuits (through DMPFC, anterior insula, and ventral striatum) specifically in the treatment responders but not non-responders (Dunlop K. et al., 2015), consistent with similar findings for DMPFC-rTMS in MDD and obsessive-compulsive disorder (Salomons et al., 2014; Dunlop K. et al., 2015). These findings suggest that DMPFC-rTMS may improve bulimic symptoms through an improvement of top-down cognitive control over urges, via frontostriatal circuits through salience-network nodes. Future work should include a sham-controlled arm, along with behavioral measures to better characterize the cognitive domains mediating the therapeutic effects of DMPFC-rTMS in BN.

NIBSasaTreatmentforAN

To date, there are few published sham-controlled trials on tDCS and rTMS as treatments for AN (Bainbridge and Brown, 2014; McClelland et al., 2013). One preclinical study in a small sample of AN patients (n = 10) applied a single session of 10 Hz left DLPFC-rTMS, with patients reporting less anxiety and less feeling full and feeling fat (Table 2; Van den Eynde et al., 2013). An open-label case series in 5 AN patients applied 20 sessions of excitatory DLPFC-rTMS, reporting improvements in anxiety, feeling fat/full and urge to restrict/exercise over the course of treatment, enduring to 6 months; however, these effects had waned by 12-months post-treatment (McClelland et al., 2016). Another open-label series in 7 AN patients applied 10 sessions of anodal left DLPFC tDCS (Table 3), reporting improvements on the Eating Disorders Inventory (EDI) and the Eating Attitude Test (EAT) (Khedr et al., 2014). Although, these early publications are promising, further preliminary work in larger groups, with a longer course and sham control, must be performed to determine whether rTMS and tDCS are efficacious treatments for AN.

ConsiderationsforFutureStudies

PatientSelection

In an attempt to limit heterogeneity, inclusion criteria for NIBS studies in ED patients are often based on DSM-5 diagnostic categories. However, as noted above, DSM-5 diagnoses still encompass substantial heterogeneity, and may conflate neurobiologically distinct endophenotypes. Future studies enrolling ED patients for NIBS trials should make efforts to at

451

least characterize the underlying phenotypes within the clinical populations they are treating, and ideally should target a specific endophenotype associated with a specific neural substrate. Such studies should also measure behavioral or biological markers of this endophenotype to assess whether the target process was successfully engaged, and whether the engaged process did indeed mediate any observed symptomatic improvements.

InterventionParameters

Several treatment parameters are important to consider when designing NIBS studies in ED. First, treatment parameters (protocol, total number of sessions, and number of sessions per day) needs to be selected, keeping in mind both patient convenience and therapeutic efficacy. In the older MDD-NIBS literature, 20–30 sessions of once daily rTMS is the standard protocol, with sessions lasting up to 45–60 min. However, such schedules are onerous for patients and limit overall clinic capacity. More recent studies have begun to explore much briefer protocols, such as 1–3 min theta-burst stimulation (Li et al., 2014), which have been reported to achieve equivalent or superior outcomes (Bakker et al., 2015). Other protocols, such as quadripulse stimulation (QPS), have been reported to achieve much more consistent effects across individuals (Huang et al., 2005; Tsutsumi et al., 2014). Still other recent MDD trials have delivered multiple sessions per day (up to five sessions a day), to complete the full course in 4–10 days rather than the usual 4–6 weeks (Holtzheimer et al., 2010; Baeken et al., 2014). Future ED rTMS trials should make use of these innovations to reduce patient burden, increase capacity or consistency, and accelerate the pace of improvement.

ConcurrentTasksorTherapies

Another consideration for NIBS trials for ED is whether stimulation should be applied concurrently with psychotherapy or a specific cognitive/behavioral task, as opposed to simply during rest. This is especially the case if NIBS protocols are designed based on RDoC dimensions, and targets cortical regions based on abnormal activation on certain tasks. As discussed above, many areas, including the ACC/mPFC, DLPFC, insula, inferior frontal gyrus, and ventrolateral PFC are hyperactive to some tasks, but hypoactive in others. With stimulation during rest, it is difficult to assess or constrain the activation state of the underlying cortical target. Having the patient perform illness-specific cognitive task has now been shown to enhance (or reduce) the therapeutic effects of rTMS across several different indications. For example, reading trauma-related scripts during rTMS enhanced efficacy for PTSD (Isserles et al., 2013); undergoing rTMS in the presence of substance cues enhances efficacy in addiction (Dinur-Klein et al., 2014). Analogous approaches may be helpful in ED.

TreatmentTarget

A final consideration for ED-NIBS concerns the feasibility of the proposed target. Although, targets such as DLPFC, DMPFC, OFC, and TPJ have now been targeted in a variety of studies, others such as ventromedial prefrontal cortex or anterior insula may be difficult to reach without specially designed coils, and without also stimulating overlying structures. More feasibility studies are needed to assess how well that these areas can be engaged with rTMS and tDCS (Chib et al., 2013).

452

Another consideration during target selection is determining the appropriate stimulation intensity in the case of rTMS. For example, treatment intensity is determined by measuring the resting motor threshold of region of cortex directly posterior to stimulation site; in these cases, resting motor threshold is determined by the activation of the thumb or big toe for DLPFC and DMPFC, respectively (Schutter and van Honk, 2006; Hallett, 2007). It is therefore unclear for novel stimulatory sites what would be the most appropriate and reliable sites to determine optimal stimulation intensity. Studies using finite element modeling may also be helpful for optimizing stimulator placement and intensity (Nitsche et al., 2012).

The effects of rTMS also dramatically decrease the farther the site is from the scalp surface (Kozel et al., 2000), and so it is likely that stimulation intensity will have to be quite large for deep targets such as anterior insula or VMPFC. If this is the case, it is likely that pain tolerability will be a factor. In addition, trigeminal nerve pain, scalp pain, and headaches are common adverse effects associated with rTMS (Machii et al., 2006; Rossi and Hallett, 2009). Tolerability will need to be maintained when stimulating these novel targets, particularly in scalp regions with trigeminal innervation, such as the frontopolar, orbitofrontal, or temporopolar regions. This may be challenging for more intense rTMS protocols, although helmet-shaped “deep TMS” coil geometries may be somewhat helpful in allowing deep stimulation of these regions while maintaining tolerability (Roth et al., 2007). Certain targets (e.g., OFC, frontopolar cortex) may be more amenable to tDCS, which is relatively painless compared to rTMS. Another non-invasive technique worthy of future investigation is cutaneous non-invasive vagus nerve stimulation, which is also delivered via external electrodes. Its more invasive counterpart, surgically-implanted vagus nerve stimulation has recently shown some efficacy for medication-resistant depression (Ben-Menachem et al., 2015; Grimonprez et al., 2015).

Finally, stimulating multiple targets in a single session might be the optimal way to address all the abnormal behavioral dimensions in a given ED patient. Different ED symptom dimensions map to different cortical targets, and so confining stimulation to a single target may be insufficient to address multi-dimensional ED pathology. For example, in BN, excitatory stimulation of the DMPFC/insula combined with inhibitory stimulation of the VMPFC may be a more optimal strategy for enhancing cognitive control while reducing urge intensity. “Deep TMS” coils have been designed to stimulate multiple targets simultaneously (Dinur-Klein et al., 2014), and multi-channel coils allow different protocols at different targets simultaneously (Roth et al., 2014). However, the therapeutic effects of sequential vs. simultaneous stimulation have not yet been compared directly. Further research should be done to describe the safety, tolerability, clinical efficacy, and neural mechanisms of stimulating multiple targets, either sequentially or simultaneously.

Conclusion

Neuroimaging, psychometric, and behavioral findings are converging upon a new approach to classifying psychiatric disorders, including EDs, in terms of endophenotypes or symptom dimensions. New proposed frameworks, such as the RDoC, seek to describe EDs in terms of dysfunction in specific underlying brain functions such as cognitive control, positive and negative valence, and social/self-related cognition. These functions in turn are gradually

453

being linked to specific neurobiological processes, described at multiple levels spanning clinical symptomatology, behavioral task performance, neuroimaging studies of macro-scale network function, and cellular, molecular, and genetic mechanisms. With the advent of anatomically focal NIBS interventions, a “neuroanatomical formulation” of ED pathology may become relevant not only for basic science, but for clinical care.

At present, neuroanatomical, endophenotypic, and RDoC formulations of ED pathology must be considered tentative and preliminary. However, from available literature, it does appear that some of the tremendous and dynamic heterogeneity of symptoms in the ED population can be understood parsimoniously in terms of dysfunction in a few key cognitive systems and their associated neural circuits. For example, in BN and BED, binge and purge behaviors may acquire pathologically strong incentive salience by mechanisms similar to addiction; impaired cognitive control in turn renders binge/purge urges hard to resist, particularly during negative affect. NIBS strategies designed for addiction (e.g., enhancing cognitive control via salience-network stimulation and damping urge intensity via ventromedial stimulation) may be helpful in this setting. In ANR, this strategy may be less helpful; instead, targeting pathologically overactive negative-valence systems may address the excessive valuation of secondary over primary rewards, and the underlying compulsivity. NIBS strategies developed for OCD (such as inhibitory stimulation of the OFC and DMPFC) may be more helpful in this setting. Ancillary NIBS strategies for AN may also target distortions of body image, alexithymia, and deficits of interoception via insular, TPJ, and EBA stimulation. However, it must be acknowledged that nearly all of these approaches are at present theoretically based, and lacking even in preclinical support. The field is urgently in need of future studies in clinical populations, with adequate sample sizes and sham controls, and using endophenotypic markers to validate or refute the proposed mechanisms of action for NIBS in EDs.

To conclude, patients with EDs stand to benefit tremendously from ongoing progress in three areas: symptom characterization, diagnostic formulation, and targeted intervention. Recent initiatives will allow us to make better sense of the heterogeneity of ED pathology, both across individuals and within individuals over time. As we improve our abilities to identify robust symptom clusters, link those clusters to neural substrates, and target those substrates with NIBS interventions, treatment outcomes will improve. These advances need not occur at the expense of existing and well-validated treatment strategies involving medications, psychotherapy, and behavior modification. Rather, they will likely work in a synergistic fashion to complement and facilitate our existing treatment strategies: enhancing the cognitive control that is a prerequisitive for successful cognitive-behavioral treatments in BN, or reducing the compulsivity and rigidity that hampers behavior modification in AN. Given the considerable patient burden and chronicity of EDs, these advances in treatment options will be a welcome change for patients, families and clinicians alike.

454

Table 1. Overview of the 5 Research Domain Criteria domains as adapted from Insel et al. (2010) and Morris and Cuthbert (2012).

455

Figure 1. Cognitive and behavioral phenotypes by RDoC dimension (Insel et al., 2010) for anorexia nervosa, bulimia nervosa, and binge eating disorder (Schebendach et al., 2007, 2013; Fernández-Aranda et al., 2008; Zastrow et al., 2009; Harrison et al., 2010; Klein et al., 2010; Miyake et al., 2010; Manwaring et al., 2011; Bohon and Stice, 2012; Hoffman et al., 2012; Steinglass et al., 2012; Chan et al., 2013; Giel et al., 2013; Strigo et al., 2013; Wu et al., 2013; Glashouwer et al., 2014; Kullmann et al., 2014; Mole et al., 2015; Berg et al., 2015; Racine et al., 2015; Tapajóz P de Sampaio et al., 2015). NR, Natural Rewards.

456

Figure 2. Candidate NIBS targets that address abnormal phenotypes related to the RDoC negative valence dimension. (A) Candidate negative valence NIBS targets for anorexia nervosa (AN) (Ellison et al., 1998; Frank et al., 2002, 2012b; Seeger et al., 2002; Uher et al., 2004; Friederich et al., 2010; Vocks et al., 2010; Cowdrey et al., 2011; Bischoff-Grethe et al., 2013; Strigo et al., 2013; Bär et al., 2015). The dorsolateral prefrontal cortex (DL) is abnormally hyperactive for pain anticipation and the receipt of punishment. The anterior cingulate cortex (ACC) is hyperactive for aversive food stimuli, the receipt of punishment, and anxiety. Finally, the anterior insula (IN) is abnormally hyperactive during anxiety and the anticipation of pain. (B) Candidate negative valence NIBS targets for bulimia nervosa (BN) and binge eating disorder (BED). The ACC is abnormally activated for negative words about the body (Miyake et al., 2010), while the insula is hyperactive during negative affect (Bohon and Stice, 2012).

457

Figure 3. Candidate NIBS targets that address abnormal phenotypes related to the RDoC positive valence dimension. (A) Candidate positive valence NIBS targets for anorexia nervosa (AN) (Wagner et al., 2007, 2008; Rothemund et al., 2011; Vocks et al., 2011; Holsen et al., 2012; Oberndorfer et al., 2013b; Torresan et al., 2013; Decker et al., 2014; Suda et al., 2014; Ehrlich et al., 2015; Sanders et al., 2015). The dorsolateral prefrontal cortex (DL) is both hyperactive when the participant views images of food, but hypoactive during symptom, particularly OCD-related, provocation. The anterior cingulate cortex (ACC) is also differentially activated; it is hyperactive when the participant views images of food, but hypoactive when the participant delays a reward. Also, the insula (IN) is hypoactive when the participant views images of food. (B) Candidate positive valence NIBS targets for bulimia nervosa (BN) (Frank et al., 2006, 2011; Bohon and Stice, 2011; Broft et al., 2012; Radeloff et al., 2012; Weygandt et al., 2012; Oberndorfer et al., 2013a; Galusca et al., 2014). The ACC is hypoactive during reward anticipation, and this hypoactivity predicts later overeating. The orbitofrontal cortex (OFC) is hyperactive during the receipt of a reward. The IN is both hyperactive during the receipt of a reward, but hypoactive during reward anticipation. (C)Candidate positive valence NIBS targets for binge eating disorder (BED) (Schienle et al., 2009; Frank et al., 2012a; Weygandt et al., 2012; Balodis et al., 2013a, 2014). Both the OFC and the IN are abnormally hyperactive during the receipt of a reward, while the inferior frontal gyrus (IFG) is hyperactive during reward anticipation.

458

Figure 4. Candidate NIBS targets that address abnormal phenotypes related to the RDoC cognitive control dimension. (A) Candidate cognitive control NIBS targets for anorexia nervosa (AN) (Oberndorfer et al., 2011; Sato et al., 2013; Garrett et al., 2014; Wierenga et al., 2014; Wildes et al., 2014; Biezonski et al., 2015; Lao-Kaim et al., 2015). The dorsolateral prefrontal cortex (DL) is both hyperactive during interference control tasks (such as the Stroop task), and for working memory, but hypoactive during cognitive flexibility and set-shifting tasks. The anterior cingulate cortex (ACC) is hypoactive during response inhibition tasks, while the ventrolateral prefrontal cortex (VL) is hypoactive during cognitive flexibility tasks. (B)Candidate cognitive control NIBS targets for bulimia nervosa (BN) (Marsh et al., 2009; Rossi and Hallett, 2009; Celone et al., 2011; Lock et al., 2011). Both the ACC and the DL are hyperactive during response inhibition tasks, but hypoactive during interference control tasks, while both the orbitofrontal cortex (OFC) and inferior frontal gyrus (IFG) are hypoactive during inference control tasks. (C) Candidate cognitive control NIBS targets for binge eating disorder (BED) (Balodis et al., 2013b; Hege et al., 2014). The ventromedial prefrontal cortex (VM), insula (IN), and IFG are abnormally hypoactive during interference control, poor dietary restraint, impulsivity, and response inhibition.

459

Figure 5. Candidate NIBS targets that address abnormal phenotypes related to the RDoC social processing dimension in anorexia nervosa (AN) (Cowdrey et al., 2012; Miyake et al., 2012; Spangler and Allen, 2012; Castellini et al., 2013; Fladung et al., 2013; Suda et al., 2013; Boehm et al., 2014; Fonville et al., 2014; Kerr et al., 2015; McAdams et al., 2015; Via et al., 2015). The anterior cingulate cortex (ACC), posterior cingulate cortex (PCC), and temporoparietal junction (TPJ) are abnormally hypoactive during deficits in social decision-making and alexithymia, while low insula (IN) activity is related to deficits in interoceptive awareness. The dorsolateral prefrontal cortex (DL) is both abnormally hyperactive when the participant views oversized images of themselves, but hypoactive when viewing images depicting body-checking behavior. The fusiform face area (FFA) is both abnormally hyperactive when the participant views highly emotional facial expressions, but hypoactive when viewing distorted body shapes and images depicting body-checking behavior. The extrastriate body area (EBA) is both abnormally hyperactive when the participant views images of their own body, but hypoactive when those images are distorted.

Table 2. Overview of the available ED-rTMS literature.

460

Table 3. Overview of the available ED-tDCS literature.

References

Ahmari, S. E., Spellman, T., Douglass, N. L., Kheirbek, M. A., Simpson, H. B., Deisseroth, K., et al. (2013). Repeated cortico-striatal stimulation generates persistent OCD-like behavior. Science340, 1234–1239. Åhs, F., Kragel, P. A., Zielinski, D. J., Brady, R., and LaBar, K. S. (2015). Medial prefrontal pathways for the contextual regulation of extinguished fear in humans. Neuroimage 122, 262–271.

461

Albares, M., Lio, G., Criaud, M., Anton, J.-L., Desmurget, M., and Boulinguez, P. (2014). The dorsal medial frontal cortex mediates automatic motor inhibition in uncertain contexts: evidence from combined fMRI and EEG studies. Hum. Brain Mapp. 35, 5517–5531. Amianto, F., Ottone, L., Abbate-Daga, G., and Fassino, S. (2015). Binge-eating disorder diagnosis and treatment: a recap in front of DSM-5. BMC Psychiatry 15:6. Arcelus, J., Mitchell, A. J., Wales, J., and Nielsen, S. (2011). Mortality rates in patients with anorexia nervosa and other eating disorders. A meta-analysis of 36 studies. Arch. Gen. Psychiatry68, 724–731. Baeken, C., De Raedt, R., Van Schuerbeek, P., Vanderhasselt, M. A., De Mey, J., Bossuyt, A., et al. (2010). Right prefrontal HF-rTMS attenuates right amygdala processing of negatively valenced emotional stimuli in healthy females. Behav. Brain Res. 214, 450–455. Baeken, C., Marinazzo, D., Wu, G.-R., Van Schuerbeek, P., De Mey, J., Marchetti, I., et al. (2014). Accelerated HF-rTMS in treatment-resistant unipolar depression: Insights from subgenual anterior cingulate functional connectivity. World J. Biol. Psychiatry, 15, 286–297. Bailer, U. F., Frank, G. K., Henry, S. E., Price, J. C., Meltzer, C. C., Mathis, C. A., et al. (2007). Exaggerated 5-HT1A but Normal 5-HT2A Receptor Activity in Individuals Ill with Anorexia Nervosa. Biol. Psychiatry 61, 1090–1099. Bailer, U., Price, J., Meltzer, C. C., Mathis, C. A., Frank, G. K., Weissfeld, L., et al. (2004). Altered 5-HT(2A) receptor binding after recovery from bulimia-type anorexia nervosa: relationships to harm avoidance and drive for thinness. Neuropsychology 29, 1143–1155. Bainbridge, K., and Brown, A. (2014). rTMS as a treatment for anorexia nervosa. Brain Stimul. 7, 149–150. Bakker, N., Shahab, S., Giacobbe, P., Blumberger, D. M., Daskalakis, Z. J., Kennedy, S. H., et al. (2015). rTMS of the dorsomedial prefrontal cortex for major depression: safety, tolerability, effectiveness, and outcome predictors for 10 Hz versus intermittent theta-burst stimulation. Brain Stimul. 8, 205–215. Balodis, I. M., Grilo, C. M., Kober, H., Worhunsky, P. D., White, M. A., Stevens, M. C., et al. (2014). A pilot study linking reduced fronto-Striatal recruitment during reward processing to persistent bingeing following treatment for binge-eating disorder. Int. J. Eat. Disord. 47, 376–384. Balodis, I. M., Kober, H., Worhunsky, P. D., White, M. A., Stevens, M. C., Pearlson, G. D., et al. (2013a). Monetary reward processing in obese individuals with and without binge eating disorder. Biol. Psychiatry 73, 877–886. Balodis, I. M., Molina, N. D., Kober, H., Worhunsky, P. D., White, M. A., Sinha, R., et al. (2013b). Divergent neural substrates of inhibitory control in binge eating disorder relative to other manifestations of obesity. Obesity 21, 367–377. Bär, K.-J., de la Cruz, F., Berger, S., Schultz, C. C., and Wagner, G. (2015). Structural and functional differences in the cingulate cortex relate to disease severity in anorexia nervosa. J. Psychiatry Neurosci. 40, 269–279. Bar, K.-J., de la Cruz, F., Berger, S., Schultz, C., and Wagner, G. (2015). Structural and functional differences in the cingulate cortex relate to disease severity in anorexia nervosa. J. Psychiatry Neurosci. 40, 269–279. Barth, K. S., Rydin-Gray, S., Kose, S., Borckardt, J. J., O'Neil, P. M., Shaw, D., et al. (2011). Food cravings and the effects of left prefrontal repetitive transcranial magnetic stimulation using an improved sham condition. Front. Psychiatry 2:9. Ben-Menachem, E., Revesz, D., Simon, B. J., and Silberstein, S. (2015). Surgically implanted and non-invasive vagus nerve stimulation: a review of efficacy, safety and tolerability. Eur. J. Neurol.22, 1260–1268. Berg, K. C., Crosby, R. D., Cao, L., Crow, S. J., Engel, S. G., Wonderlich, S., et al. (2015). Negative affect prior to and following overeating-only, loss of control eating-only, and binge eating episodes in obese adults. Int. J. Eat. Disord. 48, 641–653. Beucke, J. C., Sepulcre, J., Talukdar, T., Linnman, C., Zschenderlein, K., Endrass, T., et al. (2013). Abnormally high degree connectivity of orbitofrontal cortex in obsessive-compulsive disorder. JAMA Psychiatry 70, 619–629. Biezonski, D., Cha, J., Steinglass, J., and Posner, J. (2015). Evidence for thalamocortical circuit abnormalities and associated cognitive dysfunctions in underweight individuals with anorexia nervosa. Neuropsychopharmacology. Bischoff-Grethe, A., McCurdy, D., Grenesko-Stevens, E., Irvine, L. E. Z., Wagner, A., Yau, W.-Y. W., et al. (2013). Altered brain response to reward and punishment in adolescents with Anorexia nervosa. Psychiatry Res. 214, 331–340. Boehm, I., Geisler, D., King, J. A., Ritschel, F., Seidel, M., Deza Araujo, Y., et al. (2014). Increased resting state functional connectivity in the fronto-parietal and default mode network in anorexia nervosa. Front. Behav. Neurosci. 8:346. Bohon, C., and Stice, E. (2011). Reward abnormalities among women with full and subthreshold bulimia nervosa: a functional magnetic resonance imaging study. Int. J. Eat. Disord. 44, 585–595. Bohon, C., and Stice, E. (2012). Negative affect and neural response to palatable food intake in bulimia nervosa. Appetite 58, 964–970.

462

Broft, A., Shingleton, R., Kaufman, J., Liu, F., Kumar, D., Slifstein, M., et al. (2012). Striatal dopamine in bulimia nervosa: a PET imaging study. Int. J. Eat. Disord. 45, 648–656. Brunoni, A. R., Nitsche, M. A., Bolognini, N., Bikson, M., Wagner, T., Merabet, L., et al. (2012). Clinical research with transcranial direct current stimulation (tDCS): challenges and future directions. Brain Stimul. 5, 175–195. Bulik, C. M., Bacanu, S.-A., Klump, K. L., Fichter, M. M., Halmi, K. A., Keel, P., et al. (2005). Selection of eating-disorder phenotypes for linkage analysis. Am. J. Med. Genet. B. Neuropsychiatr. Genet. 139B, 81–87. Carpenter, L. L., Janicak, P. G., Aaronson, S. T., Boyadjis, T., Brock, D. G., Cook, I. A., et al. (2012). Transcranial magnetic stimulation (TMS) for major depression: a multisite, naturalistic, observational study of acute treatment outcomes in clinical practice. Depress. Anxiety 29, 587–596. Carter, J. C., Mercer-Lynn, K. B., Norwood, S. J., Bewell-Weiss, C. V., Crosby, R. D., Woodside, D. B., et al. (2012). A prospective study of predictors of relapse in anorexia nervosa: implications for relapse prevention. Psychiatry Res. 200, 518–523. Castellini, G., Polito, C., Bolognesi, E., D'Argenio, A., Ginestroni, A., Mascalchi, M., et al. (2013). Looking at my body. similarities and differences between anorexia nervosa patients and controls in body image visual processing. Eur. Psychiatry 28, 427–435. Celone, K. A., Thompson-Brenner, H., Ross, R. S., Pratt, E. M., and Stern, C. E. (2011). An fMRI investigation of the fronto-striatal learning system in women who exhibit eating disorder behaviors. Neuroimage 56, 1749–1757. Chan, T. W. S., Ahn, W.-Y., Bates, J. E., Busemeyer, J. R., Guillaume, S., Redgrave, G. W., et al. (2013). Differential impairments underlying decision making in anorexia nervosa and bulimia nervosa: a cognitive modeling analysis. Int. J. Eat. Disord., 47, 157–167. Chen, R., Classen, J., Gerloff, C., Celnik, P., Wassermann, E. M., Hallett, M., et al. (1997). Depression of motor cortex excitability by low-frequency transcranial magnetic stimulation. Neurology 48, 1398–1403. Chib, V. S., Yun, K., Takahashi, H., and Shimojo, S. (2013). Noninvasive remote activation of the ventral midbrain by transcranial direct current stimulation of prefrontal cortex. Transl. Psychiatry 3, e268. Cho, J.-H., Deisseroth, K., and Bolshakov, V. Y. (2013). Synaptic encoding of fear extinction in mPFC-amygdala circuits. Neuron 80, 1491–1507. Cho, S. S., Ko, J. H., Pellecchia, G., Van Eimeren, T., Cilia, R., and Strafella, A. P. (2010). Continuous theta burst stimulation of right dorsolateral prefrontal cortex induces changes in impulsivity level. Brain Stimul. 3, 170–176. Cho, S. S., Kosimori, Y., Aminian, K., Obeso, I., Rusjan, P., Lang, A., et al. (2014). Investing in the future: stimulation of the medial prefrontal cortex reduces discounting of delayed rewards. Neuropsychopharmacology. 40, 546–553. Ciampi de Andrade, D., Galhardoni, R., Pinto, L. F., Lancelotti, R., Rosi, J., Marcolin, M. A., et al. (2012). Into the island: a new technique of non-invasive cortical stimulation of the insula. Neurophysiol. Clin. 42, 363–368. Claudino, A. M., Van den Eynde, F., Stahl, D., Dew, T., Andiappan, M., Kalthoff, J., et al. (2011). Repetitive transcranial magnetic stimulation reduces cortisol concentrations in bulimic disorders. Psychol. Med. 41, 1329–1336. Cowdrey, F. A., Harmer, C. J., Park, R. J., and McCabe, C. (2012). Neural responses to emotional faces in women recovered from anorexia nervosa. Psychiatry Res. 201, 190–195. Cowdrey, F. A., Park, R. J., Harmer, C. J., and McCabe, C. (2011). Increased neural processing of rewarding and aversive food stimuli in recovered anorexia nervosa. Biol. Psychiatry 70, 736–743. Craig, A. D. (2002). How do you feel? Interoception: the sense of the physiological condition of the body. Nat. Rev. Neurosci. 3, 655–666. Crane, A., Roberts, M. E., and Treasure, J. (2007). Are obsessive−compulsive personality traits associated with a poor outcome in anorexia nervosa? A systematic review of randomized controlled trials and naturalistic. Int. J. Eat. Disord. 40, 581–588. Decker, J. H., Figner, B., and Steinglass, J. E. (2014). On weight and waiting: delay discounting in anorexia nervosa pretreatment and posttreatment. Biol. Psychiatry, 78, 606–614. De Ridder, D., Vanneste, S., Kovacs, S., Sunaert, S., and Dom, G. (2011). Transient alcohol craving suppression by rTMS of dorsal anterior cingulate: an fMRI and LORETA EEG study. Neurosci. Lett. 496, 5–10. Dinur-Klein, L., Dannon, P., Hadar, A., Rosenberg, O., Roth, Y., Kotler, M., et al. (2014). Smoking cessation induced by deep repetitive transcranial magnetic stimulation of the prefrontal and insular cortices: a prospective, randomized controlled trial. Biol. Psychiatry 76, 742–749. Donaldson, P. H., Rinehart, N. J., and Enticott, P. G. (2015). Noninvasive stimulation of the temporoparietal junction: a systematic review. Neurosci. Biobehav. Rev. 55, 547–572. Downar, J., Geraci, J., Salomons, T. V., Dunlop, K., Wheeler, S., McAndrews, M. P., et al. (2014). Anhedonia and reward-circuit connectivity distinguish nonresponders from responders to dorsomedial prefrontal repetitive transcranial magnetic stimulation in major depression. Biol. Psychiatry 76, 176–185.

463

Downar, J., Sankar, A., Giacobbe, P., Woodside, B., and Colton, P. (2012). Unanticipated rapid remission of refractory bulimia nervosa, during high-dose repetitive transcranial magnetic stimulation of the dorsomedial prefrontal cortex: a case report. Front. Psychiatry 3:30. Dunlop, K., Woodside, B., Lam, E., Olmsted, M., Colton, P., Giacobbe, P., et al. (2015). Increases in frontostriatal connectivity are associated with response to dorsomedial repetitive transcranial magnetic stimulation in refractory binge/purge behaviors. Neuroimage Clin. 8, 611–618. Dunlop, K., Hanlon, C. A., and Downar, J. (accepted). Non-invasive brain stimulation treatments for addiction major depression. Ann. N.Y. Acad. Sci. Addict. Rev. Dunlop, K., Woodside, B., Olmsted, M., Colton, P., Giacobbe, P., and Downar, J. (2015). Reductions in cortico-striatal hyperconnectivity accompany successful treatment of obsessive-compulsive disorder with dorsomedial prefrontal rTMS. Neuropsychopharmacology. Eddy, K. T., Crosby, R. D., Keel, P. K., Wonderlich, S. A., le Grange, D., Hill, L., et al. (2009). Empirical identification and validation of eating disorder phenotypes in a multisite clinical sample. J. Nerv. Ment. Dis. 197, 41–49. Ehrlich, S., Geisler, D., Ritschel, F., King, J., Seidel, M., Boehm, I., et al. (2015). Elevated cognitive control over reward processing in recovered female patients with anorexia nervosa. J. Psychiatry Neurosci. 40, 307–315. Eldaief, M. C., Halko, M. A., Buckner, R. L., and Pascual-Leone, A. (2011). Transcranial magnetic stimulation modulates the brain's intrinsic activity in a frequency-dependent manner. Proc. Natl. Acad. Sci. U.S.A. 108, 21229–21234. Ellison, Z., Foong, J., Howard, R., Bullmore, E., Williams, S., and Treasure, J. (1998). Functional anatomy of calorie fear in anorexia nervosa. Lancet 352, 1192. Enticott, P. G., Fitzgibbon, B. M., Kennedy, H. A., Arnold, S. L., Elliot, D., Peachey, A., et al. (2014). A double-blind, randomized trial of deep repetitive transcranial magnetic stimulation (rTMS) for autism spectrum disorder. Brain Stimul. 7, 206–211. Enticott, P. G., Kennedy, H. A., Zangen, A., and Fitzgerald, P. B. (2011). Deep repetitive transcranial magnetic stimulation associated with improved social functioning in a young woman with an autism spectrum disorder. J. ECT 27, 41–43. Felix-Ortiz, A. C., Burgos-Robles, A., Bhagat, N. D., Leppla, C. A., and Tye, K. M. (2015). Bidirectional modulation of anxiety-related and social behaviors by amygdala projections to the medial prefrontal cortex. Neuroscience. Fernández-Aranda, F., Pinheiro, A. P., Thornton, L. M., Berrettini, W. H., Crow, S., Fichter, M. M., et al. (2008). Impulse control disorders in women with eating disorders. Psychiatry Res. 157, 147–157. Filbey, F. M., Myers, U. S., and DeWitt, S. (2012). Reward circuit function in high BMI individuals with compulsive overeating: similarities with addiction. Neuroimage 63, 1800–1806. Fladung, A.-K., Schulze, U. M. E., Schöll, F., Bauer, K., and Grön, G. (2013). Role of the ventral striatum in developing anorexia nervosa. Transl. Psychiatry 3, e315. Fonville, L., Giampietro, V., Surguladze, S., Williams, S., and Tchanturia, K. (2014). Increased BOLD signal in the fusiform gyrus during implicit emotion processing in anorexia nervosa. Neuroimage Clin. 4, 266–273. Fox, M. D., Liu, H., and Pascual-Leone, A. (2013). Identification of reproducible individualized targets for treatment of depression with TMS based on intrinsic connectivity. Neuroimage 66, 151–160. Frank, G. K., Bailer, U. F., Henry, S. E., Drevets, W., Meltzer, C. C., Price, J. C., et al. (2005). Increased dopamine D2/D3 receptor binding after recovery from anorexia nervosa measured by positron emission tomography and [11C]raclopride. Biol. Psychiatry 58, 908–912. Frank, G. K., Kaye, W. H., Meltzer, C. C., Price, J. C., Greer, P., McConaha, C., et al. (2002). Reduced 5-HT2A receptor binding after recovery from anorexia nervosa. Biol. Psychiatry 52, 896–906. Frank, G. K., Wagner, A., Achenbach, S., McConaha, C., Skovira, K., Aizenstein, H., et al. (2006). Altered brain activity in women recovered from bulimic-type eating disorders after a glucose challenge: a pilot study. Int. J. Eat. Disord. 39, 76–79. Frank, G. K. W., Reynolds, J. R., Shott, M. E., Jappe, L., Yang, T. T., Tregellas, J. R., et al. (2012a). Anorexia nervosa and obesity are associated with opposite brain reward response. Neuropsychopharmacology 37, 2031–2046. Frank, G. K. W., Reynolds, J. R., Shott, M. E., and O'Reilly, R. C. (2011). Altered temporal difference learning in bulimia nervosa. Biol. Psychiatry 70, 728–735. Frank, G. K. W., Roblek, T., Shott, M. E., Jappe, L. M., Rollin, M. D. H., Hagman, J. O., et al. (2012b). Heightened fear of uncertainty in anorexia and bulimia nervosa. Int. J. Eat. Disord. 45, 227–232. Fregni, F., Orsati, F., Pedrosa, W., Fecteau, S., Tome, F. A. M., Nitsche, M. A., et al. (2008). Transcranial direct current stimulation of the prefrontal cortex modulates the desire for specific foods. Appetite 51, 34–41.

464

Friederich, H.-C., Brooks, S., Uher, R., Campbell, I. C., Giampietro, V., Brammer, M., et al. (2010). Neural correlates of body dissatisfaction in anorexia nervosa. Neuropsychologia 48, 2878–2885. Galusca, B., Sigaud, T., Costes, N., Redoute, J., Massoubre, C., and Estour, B. (2014). Wide impairment of cerebral serotoninergic activity but inter-individual heterogeneity in bulimia nervosa patients: a pilot [(18)F]MPPF/PET study. World J. Biol. Psychiatry 15, 599–608. Garfinkel, P. E., Lin, E., Goering, P., Spegg, C., Goldbloom, D. S., Kennedy, S., et al. (1995). Bulimia nervosa in a Canadian community sample: prevalence and comparison of subgroups. Am. J. Psychiatry 152, 1052–1058. Garrett, A., Lock, J., Datta, N., Beenhaker, J., Kesler, S. R., and Reiss, A. L. (2014). Predicting clinical outcome using brain activation associated with set-shifting and central coherence skills in anorexia nervosa. J. Psychiatr. Res. 57, 26–33. Gatt, L., Jan, S., Mondraty, N., Horsfield, S., Russell, J., Laba, T., et al. (2014). The household economic burden of eating disorders and adherence to treatment in Australia. BMC Psychiatry29:338. Giel, K. E., Kullmann, S., Preißl, H., Bischoff, S. C., Thiel, A., Schmidt, U., et al. (2013). Understanding the reward system functioning in anorexia nervosa: crucial role of physical activity. Biol. Psychol. 94, 575–581. Giuliano, C., and Cottone, P. (2015). The role of the opioid system in binge eating disorder. CNS Spectr. 20, 537–545. Glashouwer, K. A., Bloot, L., Veenstra, E. M., Franken, I. H. A., and de Jong, P. J. (2014). Heightened sensitivity to punishment and reward in anorexia nervosa. Appetite 75, 97–102. Godart, N. T., Flament, M. F., Curt, F., Perdereau, F., Lang, F., Venisse, J., et al. (2003). Anxiety disorders in subjects seeking treatment for eating disorders: a DSM-IV controlled study. Psychiatry Res. 117, 245–258. Golden, N. H., Katzman, D. K., Kreipe, R. E., Stevens, S. L., Sawyer, S. M., Rees, J., et al. (2003). Eating disorders in adolescents: position paper of the society for adolescent medicine. J. Adolesc. Heal. 33, 496–503. Goldman, R. L., Borckardt, J. J., Frohman, H. A., O'Neil, P. M., Madan, A., Campbell, L. K., et al. (2011). Prefrontal cortex transcranial direct current stimulation (tDCS) temporarily reduces food cravings and increases the self-reported ability to resist food in adults with frequent food craving. Appetite 56, 741–746. Gorelick, D. A., Zangen, A., and George, M. S. (2014). Transcranial magnetic stimulation in the treatment of substance addiction. Ann. N.Y. Acad. Sci. 1327, 79–93. Grall-Bronnec, M., and Sauvaget, A. (2014). The use of repetitive transcranial magnetic stimulation for modulating craving and addictive behaviours: a critical literature review of efficacy, technical and methodological considerations. Neurosci. Biobehav. Rev. 47C, 592–613. Grimonprez, A., Raedt, R., Baeken, C., Boon, P., and Vonck, K. (2015). The antidepressant mechanism of action of vagus nerve stimulation: evidence from preclinical studies. Neurosci. Biobehav. Rev. 56, 26–34. Guhn, A., Dresler, T., Andreatta, M., Müller, L. D., Hahn, T., Tupak, S. V., et al. (2014). Medial prefrontal cortex stimulation modulates the processing of conditioned fear. Front. Behav. Neurosci. 8:44. Hallett, M. (2007). Transcranial magnetic stimulation: a primer. Neuron 55, 187–199. Hanlon, C. A., Canterberry, M., Taylor, J., Devries, W., and George, M. (2013). “Functional and structural integrity of frontal-striatal circuits in cocaine users: an interleaved TMS-BOLD & DTI study,” in Annual Meeting of the Society for Biological Psychiatry, Vol. 256 (San Francisco, CA). Hanlon, C. A., Dowdle, L. T., Austelle, C. W., Devries, W., Mithoefer, O., Badran, B. W., et al. (2015). What goes up, can come down: Novel brain stimulation paradigms may attenuate craving and craving-related neural circuitry in substance dependent individuals. Brain Res. 1628(Pt A), 199–209. Harrison, A., O'Brien, N., Lopez, C., and Treasure, J. (2010). Sensitivity to reward and punishment in eating disorders. Psychiatry Res. 177, 1–11. Hart, L. M., Granillo, M. T., Jorm, A. F., and Paxton, S. J. (2011). Unmet need for treatment in the eating disorders: a systematic review of eating disorder specific treatment seeking among community cases. Clin. Psychol. Rev. 31, 727–735. Hausmann, A., Mangweth, B., Walpoth, M., Hoertnagel, C., Kramer-Reinstadler, K., Rupp, C. I., et al. (2004). Repetitive transcranial magnetic stimulation (rTMS) in the double-blind treatment of a depressed patient suffering from bulimia nervosa: a case report. Int. J. Neuropsychopharmacol. 7, 371–373. Hay, P. J., Touyz, S., and Sud, R. (2012). Treatment for severe and enduring anorexia nervosa: a review. Aust. N. Z. J. Psychiatry 42, 1136–1144. Hege, M. A., Stingl, K. T., Kullmann, S., Schag, K., Giel, K. E., Zipfel, S., et al. (2014). Attentional impulsivity in binge eating disorder modulates response inhibition performance and frontal brain networks. Int. J. Obes. (Lond). 39, 353–360. Hilbert, A., Bishop, M. E., Stein, R. I., Tanofsky-Kraff, M., Swenson, A. K., Welch, R. R., et al. (2012). Long-term efficacy of psychological treatments for binge eating disorder. Br. J. Psychiatry 200, 232–237.

465

Hoffman, E. R., Gagne, D. A., Thornton, L. M., Klump, K. L., Brandt, H., Crawford, S., et al. (2012). Understanding the association of impulsivity, obsessions, and compulsions with binge eating and purging behaviours in anorexia nervosa. Eur. Eat. Disord. Rev. 20, e129–e136. Holsen, L., Lawson, E., Blum, J., Ko, E., Makris, N., Fazeli, P., et al. (2012). Food motivation circuitry hypoactivation related to hedonic and nonhedonic aspects of hunger and satiety in women with active anorexia nervosa and weight-restored women with anorexia nervosa. J. Psychiatry Neurosci. 37, 322–232. Holtzheimer, P. E., McDonald, W. M., Mufti, M., Kelley, M. E., Quinn, S., Corso, G., et al. (2010). Accelerated repetitive transcranial magnetic stimulation for treatment-resistant depression. Depress. Anxiety 27, 960–963. Huang, Y.-Z., Edwards, M. J., Rounis, E., Bhatia, K. P., and Rothwell, J. C. (2005). Theta burst stimulation of the human motor cortex. Neuron 45, 201–206. Hudson, J. I., Hiripi, E., Pope, H. G., and Kessler, R. C. (2007). The prevalence and correlates of eating disorders in the National Comorbidity Survey Replication. Biol. Psychiatry 61, 348–358. Hummel, D., Rudolf, A. K., Brandi, M.-L., Untch, K.-H., Grabhorn, R., Hampel, H., et al. (2013). Neural adaptation to thin and fat bodies in the fusiform body area and middle occipital gyrus: an fMRI adaptation study. Hum. Brain Mapp. 34, 3233–3246. Insel, T., Cuthbert, B., Garvey, M., Heinssen, R., Pine, D. S., Quinn, K., et al. (2010). Research domain criteria (RDoC): toward a new classification framework for research on mental disorders. Am. J. Psychiatry 167, 748–751. Isserles, M., Shalev, A. Y., Roth, Y., Peri, T., Kutz, I., Zlotnick, E., et al. (2013). Effectiveness of deep transcranial magnetic stimulation combined with a brief exposure procedure in post-traumatic stress disorder–a pilot study. Brain Stimul. 6, 377–383. Johnstone, T., van Reekum, C. M., Urry, H. L., Kalin, N. H., and Davidson, R. J. (2007). Failure to regulate: counterproductive recruitment of top-down prefrontal-subcortical circuitry in major depression. J. Neurosci. 27, 8877–8884. Keating, C., Tilbrook, A. J., Rossell, S. L., Enticott, P. G., and Fitzgerald, P. B. (2012). Reward processing in anorexia nervosa. Neuropsychologia 50, 567–575. Keel, P. K., Fichter, M., Quadflieg, N., Bulik, C. M., Baxter, M. G., Thornton, L., et al. (2004). Application of a latent class analysis to empirically define eating disorder phenotypes. Arch. Gen. Psychiatry 61, 192–200. Keel, P. K., and Mitchell, J. E. (1997). Outcome in bulimia nervosa. Am. J. Psychiatry 154, 313–321. Kekic, M., McClelland, J., Campbell, I., Nestler, S., Rubia, K., David, A. S., et al. (2014). The effects of prefrontal cortex transcranial direct current stimulation (tDCS) on food craving and temporal discounting in women with frequent food cravings. Appetite 78, 55–62. Kerr, K., Moseman, S., Avery, J., Bodurka, J., Zucker, N., and Kyle Simmons, W. (2015). Altered insula activity during visceral interoception in weight-restored patients with anorexia nervosa. Neuropsychopharmacology. 41, 521–528. Khedr, E., Elfetoh, N., Ali, A., and Noamany, M. (2014). Anodal transcranial direct current stimulation over the dorsolateral prefrontal cortex improves anorexia nervosa: a pilot study. Restor. Neurol. Neurosci. 32, 789–797. Klein, D. A., Schebendach, J. E., Gershkovich, M., Bodell, L., Foltin, R., and Walsh, B. T. (2010). Behavioral assessment of the reinforcing effect of exercise in women with anorexia nervosa: further paradigm development and data. Int. J. Eat. Disord. 43, 611–618. Kozel, F. A., Nahas, Z., deBrux, C., Molloy, M., Lorberbaum, J. P., Bohning, D., et al. (2000). How coil-cortex distance relates to age, motor threshold, and antidepressant response to repetitive transcranial magnetic stimulation. J. Neuropsychiatry Clin. Neurosci. 12, 376–384. Kühn, S., Gallinat, J., and Brass, M. (2011). “Keep Calm and Carry On”: structural correlates of expressive suppression of emotions. PLoS ONE 6:e16569. Kullmann, S., Giel, K. E., Hu, X., Bischoff, S. C., Teufel, M., Thiel, A., et al. (2014). Impaired inhibitory control in anorexia nervosa elicited by physical activity stimuli. Soc. Cogn. Affect. Neurosci. 9, 917–923. Lao-Kaim, N., Fonville, L., Giampietro, V., Williams, S., Simmons, A., and Tchanturia, K. (2015). Aberrant function of learning and cognitive control networks underlie inefficient cognitive flexibility in anorexia nervosa: a cross-sectional fMRI study. PLoS ONE 10:e124027. Lapenta, O. M., Sierve, K., de Macedo, E. C., Fregni, F., and Boggio, P. S. (2014). Transcranial direct current stimulation modulates ERP-indexed inhibitory control and reduces food consumption. Appetite 83, 42–48. Lee, S., Ran Kim, K., Ku, J., Lee, J. H., Namkoong, K., and Jung, Y. C. (2014). Resting-state synchrony between anterior cingulate cortex and precuneus relates to body shape concern in anorexia nervosa and bulimia nervosa. Psychiatry Res. 221, 43–48. Leehr, E. J., Krohmer, K., Schag, K., Dresler, T., Zipfel, S., and Giel, K. E. (2015). Emotion regulation model in binge eating disorder and obesity–a systematic review. Neurosci. Biobehav. Rev. 49, 125–134.

466

Li, C.-T., Chen, M.-H., Juan, C.-H., Huang, H.-H., Chen, L.-F., Hsieh, J.-C., et al. (2014). Efficacy of prefrontal theta-burst stimulation in refractory depression: a randomized sham-controlled study. Brain 137, 2088–2098. Lilenfeld, L. R., Kaye, W. H., Greeno, C. G., Merikangas, K. R., Plotnicov, K., Pollice, C., et al. (1998). A controlled family study of anorexia nervosa and bulimia nervosa: psychiatric disorders in first-degree relatives and effects of proband comorbidity. Arch. Gen. Psychiatry 55, 603–610. Lindquist, K. A., Satpute, A. B., Wager, T. D., Weber, J., and Barrett, L. F. (2015). The brain basis of positive and negative affect: evidence from a meta-analysis of the human neuroimaging literature. Cereb. Cortex. Lipsman, N., Woodside, D. B., Giacobbe, P., Hamani, C., Carter, J. C., Norwood, S. J., et al. (2013). Subcallosal cingulate deep brain stimulation for treatment-refractory anorexia nervosa: a phase 1 pilot trial. Lancet 381, 1361–1370. Lock, J., Garrett, A., Beenhakker, J., and Reiss, A. L. (2011). Aberrant brain activation during a response inhibition task in adolescent eating disorder subtypes. Am. J. Psychiatry 168, 55–64. Luo, Y., Qin, S., Fernández, G., Zhang, Y., Klumpers, F., and Li, H. (2014). Emotion perception and executive control interact in the salience network during emotionally charged working memory processing. Hum. Brain Mapp. 35, 5606–5616. Machii, K., Cohen, D., Ramos-Estebanez, C., and Pascual-Leone, A. (2006). Safety of rTMS to non-motor cortical areas in healthy participants and patients. Clin. Neurophysiol. 117, 455–471. Maeda, F., Keenan, J. P., Tormos, J. M., Topka, H., and Pascual-Leone, A. (2000). Interindividual variability of the modulatory effects of repetitive transcranial magnetic stimulation on cortical excitability. Exp. Brain Res. 133, 425–430. Mantovani, A., Simpson, H. B., Fallon, B. A., Rossi, S., and Lisanby, S. H. (2010). Randomized sham-controlled trial of repetitive transcranial magnetic stimulation in treatment-resistant obsessive-compulsive disorder. Int. J. Neuropsychopharmacol. 13, 217–227. Manwaring, J. L., Green, L., Myerson, J., Strube, M. J., and Wilfley, D. E. (2011). Discounting of Various types of rewards by women with and without binge eating disorder: evidence for general rather than specific differences. Psychol. Rec. 61, 561–582. Marsh, R., Horga, G., Wang, Z., Wang, P., Klahr, K. W., Berner, L. A., et al. (2011). An fMRI study of self-regulatory control and conflict resolution in adolescents with bulimia nervosa. Am. J. Psychiatry 168, 1210–1220. Marsh, R., Steinglass, J. E., Gerber, A. J., Graziano O'Leary, K., Wang, Z., Murphy, D., et al. (2009). Deficient activity in the neural systems that mediate self-regulatory control in bulimia nervosa. Arch. Gen. Psychiatry 66, 51–63. McAdams, C. J., and Krawczyk, D. C. (2011). Impaired neural processing of social attribution in anorexia nervosa. Psychiatry Res. 194, 54–63. McAdams, C. J., Lohrenz, T., and Montague, P. R. (2015). Neural responses to kindness and malevolence differ in illness and recovery in women with anorexia nervosa. Hum. Brain Mapp.36, 5207–5219. McClelland, J., Bozhilova, N., Campbell, I., and Schmidt, U. (2013). A systematic review of the effects of neuromodulation on eating and body weight: evidence from human and animal studies. Eur. Eat. Disord. Rev. 21, 436–455. McClelland, J., Kekic, M., Campbell, I. C., and Schmidt, U. (2016). Repetitive transcranial magnetic stimulation (rTMS) treatment in enduring anorexia nervosa: a case series. Eur. Eat. Disord. Rev. 24, 157–163. Meng, Z., Liu, C., Yu, C., and Ma, Y. (2014). Transcranial direct current stimulation of the frontal-parietal-temporal area attenuates smoking behavior. J. Psychiatr. Res. 54, 19–25. Meron, D., Hedger, N., Garner, M., and Baldwin, D. S. (2015). Transcranial direct current stimulation (tDCS) in the treatment of depression: systematic review and meta-analysis of efficacy and tolerability. Neurosci. Biobehav. Rev. 57, 46–62. Milos, G., Spindler, A., Schnyder, U., and Fairburn, C. G. (2005). Instability of eating disorder diagnoses: prospective study. Br. J. Psychiatry 187, 573–578. Mischoulon, D., Eddy, K. T., Keshaviah, A., Dinescu, D., Ross, S. L., Kass, A. E., et al. (2011). Depression and eating disorders: treatment and course. J. Affect. Disord. 130, 470–477. Mishra, B. R., Nizamie, S. H., Das, B., and Praharaj, S. K. (2010). Efficacy of repetitive transcranial magnetic stimulation in alcohol dependence: a sham-controlled study. Addiction105, 49–55. Mitchell, J. E., Agras, S., and Wonderlich, S. (2007). Treatment of bulimia nervosa: where are we and where are we going? Int. J. Eat. Disord. 40, 95–101. Miyake, Y., Okamoto, Y., Onoda, K., Shirao, N., Mantani, T., and Yamawaki, S. (2009). Neural correlates of alexithymia in response to emotional stimuli: a study of anorexia nervosa patients. Hiroshima J. Med. Sci. 58, 1–8.

467

Miyake, Y., Okamoto, Y., Onoda, K., Shirao, N., Okamoto, Y., Otagaki, Y., et al. (2010). Neural processing of negative word stimuli concerning body image in patients with eating disorders: an fMRI study. Neuroimage 50, 1333–1339. Miyake, Y., Okamoto, Y., Onoda, K., Shirao, N., Okamoto, Y., and Yamawaki, S. (2012). Brain activation during the perception of stressful word stimuli concerning interpersonal relationships in anorexia nervosa patients with high degrees of alexithymia in an fMRI paradigm. Psychiatry Res. 201, 113–119. Mole, T. B., Irvine, M. A., Worbe, Y., Collins, P., Mitchell, S. P., Bolton, S., et al. (2015). Impulsivity in disorders of food and drug misuse. Psychol. Med. 45, 771–782. Morris, S. E., and Cuthbert, B. N. (2012). Research Domain Criteria: cognitive systems, neural circuits, and dimensions of behavior. Dialogues Clin. Neurosci. 14, 29–37. Nauczyciel, C., and Drapier, D. (2012). Repetitive transcranial magnetic stimulation in the treatment of obsessive-compulsive disorder. Rev. Neurol. (Paris). 168, 655–661. Nettekoven, C., Volz, L. J., Leimbach, M., Pool, E.-M., Rehme, A. K., Eickhoff, S. B., et al. (2015). Inter-individual variability in cortical excitability and motor network connectivity following multiple blocks of rTMS. Neuroimage 118, 209–218. Nitsche, M. A., Koschack, J., Pohlers, H., Hullemann, S., Paulus, W., and Happe, S. (2012). Effects of frontal transcranial direct current stimulation on emotional state and processing in healthy humans. Front. Psychiatry 3:58. Oberndorfer, T. A., Frank, G. K. W., Simmons, A. N., Wagner, A., McCurdy, D., Fudge, J. L., et al. (2013a). Altered insula response to sweet taste processing after recovery from anorexia and bulimia nervosa. Am. J. Psychiatry 170, 1143–1151. Oberndorfer, T. A., Kaye, W. H., Simmons, A. N., Strigo, I. A., and Matthews, S. C. (2011). Demand-specific alteration of medial prefrontal cortex response during an inhibition task in recovered anorexic women. Int. J. Eat. Disord. 44, 1–8. Oberndorfer, T., Simmons, A., McCurdy, D., Strigo, I., Matthews, S., Yang, T., et al. (2013b). Greater anterior insula activation during anticipation of food images in women recovered from anorexia nervosa versus controls. Psychiatry Res. 214, 132–141. Olmsted, M. P., Kaplan, A. S., and Rockert, W. (2005). Defining remission and relapse in bulimia nervosa. Int. J. Eat. Disord. 38, 1–6. Pascual-Leone, A., Tormos, J. M., Keenan, J., Tarazona, F., Cañete, C., and Catalá, M. D. (1998). Study and modulation of human cortical excitability with transcranial magnetic stimulation. J. Clin. Neurophysiol. 15, 333–343. Pascual-Leone, A., Valls-Solé, J., Wassermann, E. M., and Hallett, M. (1994). Responses to rapid-rate transcranial magnetic stimulation of the human motor cortex. Brain 117(Pt 4), 847–858. Racine, S. E., Burt, S. A., Keel, P. K., Sisk, C. L., Neale, M. C., Boker, S., et al. (2015). Examining associations between negative urgency and key components of objective binge episodes. Int. J. Eat. Disord. 48, 527–531. Radeloff, D., Willmann, K., Otto, L., Lindner, M., Putnam, K., Leeuwen, S., et al. (2012). High-fat taste challenge reveals altered striatal response in women recovered from bulimia nervosa: a pilot study. World J. Biol. Psychiatry, 15, 307–316. Reineberg, A. E., Andrews-Hanna, J. R., Depue, B. E., Friedman, N. P., and Banich, M. T. (2015). Resting-state networks predict individual differences in common and specific aspects of executive function. Neuroimage 104, 69–78. Ritschel, F., King, J. A., Geisler, D., Flohr, L., Neidel, F., Boehm, I., et al. (2015). Temporal delay discounting in acutely ill and weight-recovered patients with anorexia nervosa. Psychol. Med. 45, 1229–1239. Rossi, S., and Hallett, M. (2009). Safety, ethical considerations, and application guidelines for the use of transcranial magnetic stimulation in clinical practice and research. Clin. Neurophysiol.120, 2008–2039. Roth, Y., Amir, A., Levkovitz, Y., and Zangen, A. (2007). Three-dimensional distribution of the electric field induced in the brain by transcranial magnetic stimulation using figure-8 and deep H-coils. J. Clin. Neurophysiol. 24, 31–38. Roth, Y., Levkovitz, Y., Pell, G. S., Ankry, M., and Zangen, A. (2014). Safety and characterization of a novel multi-channel TMS stimulator. Brain Stimul. 7, 194–205. Rothemund, Y., Buchwald, C., Georgiewa, P., Bohner, G., Bauknecht, H. C., Ballmaier, M., et al. (2011). Compulsivity predicts fronto striatal activation in severely anorectic individuals. Neuroscience 197, 242–250. Salomons, T. V., Dunlop, K., Kennedy, S. H., Flint, A., Geraci, J., Giacobbe, P., et al. (2014). Resting-state cortico-thalamic-striatal connectivity predicts response to dorsomedial prefrontal rTMS in major depressive disorder. Neuropsychopharmacology 39, 488–498. Sanders, N., Smeets, P., van Elburg, A., Danner, U., van Meer, F., Hoek, H., et al. (2015). Altered food-cue processing in chronically ill and recovered women with anorexia nervosa. Front. Behav. Neurosci. 9:46.

468

Sato, Y., Saito, N., Utsumi, A., Aizawa, E., Shoji, T., Izumiyama, M., et al. (2013). Neural basis of impaired cognitive flexibility in patients with anorexia nervosa. PLoS ONE 8:e61108. Saxe, R., and Kanwisher, N. (2003). People thinking about thinking people. The role of the temporo-parietal junction in “theory of mind.” Neuroimage 19, 1835–1842. Schebendach, J., Broft, A., Foltin, R., and Walsh, B. T. (2013). Can the reinforcing value of food be measured in bulimia nervosa? Appetite 62, 70–75. Schebendach, J. E., Klein, D. A., Foltin, R. W., Devlin, M. J., and Walsh, B. T. (2007). Relative reinforcing value of exercise in inpatients with anorexia nervosa: model development and pilot data. Int. J. Eat. Disord. 40, 446–453. Schienle, A., Schäfer, A., Hermann, A., and Vaitl, D. (2009). Binge-Eating disorder: reward sensitivity and brain activation to images of food. Biol. Psychiatry 65, 654–661. Schutter, D. J. L. G., and van Honk, J. (2006). A standardized motor threshold estimation procedure for transcranial magnetic stimulation research. J. ECT 22, 176–178. Seeger, G., Braus, D. F., Ruf, M., Goldberger, U., and Schmidt, M. H. (2002). Body image distortion reveals amygdala activation in patients with anorexia nervosa – a functional magnetic resonance imaging study. Neurosci. Lett. 326, 25–28. Shapiro, J. R., Berkman, N. D., Brownley, K. A., Sedway, J. A., Lohr, K. N., and Bulik, C. M. (2007). Bulimia nervosa treatment: a systematic review of randomized controlled trials. Int. J. Eat. Disord. 40, 321–336. Smink, F. R. E., Van Hoeken, D., Oldehinkel, A. J., and Hoek, H. W. (2014). Prevalence and severity of DSM-5 eating disorders in a community cohort of adolescents. Int. J. Eat. Disord. 47, 610–619. Solvason, H. B., Husain, M., Fitzgerald, P. B., Rosenquist, P., McCall, W. V., Kimball, J., et al. (2014). Improvement in quality of life with left prefrontal transcranial magnetic stimulation in patients with pharmacoresistant major depression: acute and six month outcomes. Brain Stimul.7, 219–225. d Spangler, D. L., and Allen, M. D. (2012). An fMRI investigation of emotional processing of body shape in bulimia nervosa. Int. J. Eat. Disord. 45, 17–25. Steinglass, J. E., Figner, B., Berkowitz, S., Simpson, H. B., Weber, E. U., and Walsh, B. T. (2012). Increased capacity to delay reward in anorexia nervosa. J. Int. Neuropsychol. Soc. 18, 773–780. Steinhausen, H.-C., and Weber, S. (2009). The outcome of bulimia nervosa: findings from one-quarter century of research. Am. J. Psychiatry 166, 1331–1341. Striegel-Moore, R. H., Franko, D. L., Thompson, D., Barton, B., Schreiber, G. B., and Daniels, S. R. (2005). An empirical study of the typology of bulimia nervosa and its spectrum variants. Psychol. Med. 35, 1563–1572. Strigo, I. A., Matthews, S. C., Simmons, A. N., Oberndorfer, T., Klabunde, M., Reinhardt, L. E., et al. (2013). Altered insula activation during pain anticipation in individuals recovered from anorexia nervosa: evidence of interoceptive dysregulation. Int. J. Eat. Disord. 46, 23–33. Strober, M., Freeman, R., Lampert, C., Diamond, J., and Kaye, W. (2000). Controlled family study of anorexia nervosa and bulimia nervosa: evidence of shared liability and transmission of partial syndromes. Am. J. Psychiatry 157, 393–401. Suchan, B., Bauser, D. S., Busch, M., Schulte, D., Grönemeyer, D., Herpertz, S., et al. (2013). Reduced connectivity between the left fusiform body area and the extrastriate body area in anorexia nervosa is associated with body image distortion. Behav. Brain Res. 241, 80–85. Suda, M., Brooks, S. J., Giampietro, V., Friederich, H. C., Uher, R., Brammer, M. J., et al. (2013). Functional neuroanatomy of body checking in people with anorexia nervosa. Int. J. Eat. Disord.46, 653–662. Suda, M., Brooks, S. J., Giampietro, V., Uher, R., Mataix-Cols, D., Brammer, M. J., et al. (2014). Provocation of symmetry/ordering symptoms in Anorexia nervosa: a functional neuroimaging study. PLoS ONE 9:e97998. Sullivan, P. F. (1995). Mortality in anorexia nervosa. Am. J. Psychiatry 152, 1073–1074. Sullivan, P. F., Bulik, C. M., and Kendler, K. S. (1998). The epidemiology and classification of bulimia nervosa. Psychol. Med. 28, 599–610. Sutoh, C., Koga, Y., Kimura, H., Kanahara, N., Numata, N., Hirano, Y., et al. (2016). Repetitive transcranial magnetic stimulation changes cerebral oxygenation on the left dorsolateral prefrontal cortex in bulimia nervosa: a near-infrared spectroscopy pilot study. Eur. Eat. Disord. Rev. 24, 83–88. Tapajóz P de Sampaio, F., Soneira, S., Aulicino, A., Harris, P., and Allegri, R. F. (2015). Emotional reactivity to social stimuli in patients with eating disorders. Psychiatry Res. 229, 887–894. Torresan, R. C., Ramos-Cerqueira, A. T. A., Shavitt, R. G., do Rosário, M. C., de Mathis, M. A., Miguel, E. C., et al. (2013). Symptom dimensions, clinical course and comorbidity in men and women with obsessive-compulsive disorder. Psychiatry Res. 209, 186–195. Tsutsumi, R., Hanajima, R., Terao, Y., Shirota, Y., Ohminami, S., Shimizu, T., et al. (2014). Effects of the motor cortical quadripulse transcranial magnetic stimulation (QPS) on the contralateral motor cortex and interhemispheric interactions. J. Neurophysiol. 111, 26–35.

469

Uher, R., Murphy, T., Brammer, M. J., Dalgleish, T., Phillips, M. L., Ng, V. W., et al. (2004). Medial prefrontal cortex activity associated with symptom provocation in eating disorders. Am. J. Psychiatry 161, 1238–1246. Uher, R., Yoganathan, D., Mogg, A., Eranti, S. V., Treasure, J., Campbell, I. C., et al. (2005). Effect of left prefrontal repetitive transcranial magnetic stimulation on food craving. Biol. Psychiatry58, 840–842. Ursu, S., Clark, K. A., Stenger, V. A., and Carter, C. S. (2008). Distinguishing expected negative outcomes from preparatory control in the human orbitofrontal cortex. Brain Res. 1227, 110–119. Val-Laillet, D., Aarts, E., Weber, B., Ferrari, M., Quaresima, V., Stoeckel, L. E., et al. (2015). Neuroimaging and neuromodulation approaches to study eating behavior and prevent and treat eating disorders and obesity. NeuroImage. Clin. 8, 1–31. Van den Eynde, F., Claudino, A. M., Mogg, A., Horrell, L., Stahl, D., Ribeiro, W., et al. (2010). Repetitive transcranial magnetic stimulation reduces cue-induced food craving in bulimic disorders. Biol. Psychiatry 67, 793–795. Van den Eynde, F., Guillaume, S., Broadbent, H., Campbell, I. C., and Schmidt, U. (2013). Repetitive transcranial magnetic stimulation in anorexia nervosa: a pilot study. Eur. Psychiatry28, 98–101. Via, E., Soriano-Mas, C., Sánchez, I., Forcano, L., Harrison, B. J., Davey, C. G., et al. (2015). Abnormal social reward responses in anorexia nervosa: an fMRI study. PLoS ONE 10:e0133539. Vocks, S., Busch, M., Grönemeyer, D., Schulte, D., Herpertz, S., and Suchan, B. (2010). Neural correlates of viewing photographs of one's own body and another woman's body in anorexia and bulimia nervosa: an fMRI study. J. Psychiatry Neurosci. 35, 163–176. Vocks, S., Herpertz, S., Rosenberger, C., Senf, W., and Gizewski, E. R. (2011). Effects of gustatory stimulation on brain activity during hunger and satiety in females with restricting-type anorexia nervosa: an fMRI study. J. Psychiatr. Res. 45, 395–403. Voss, U., Holzmann, R., Hobson, A., Paulus, W., Koppehele-Gossel, J., Klimke, A., et al. (2014). Induction of self awareness in dreams through frontal low current stimulation of gamma activity. Nat. Neurosci. 17, 810–812. Wagner, A., Aizenstein, H., Mazurkewicz, L., Fudge, J., Frank, G. K., Putnam, K., et al. (2008). Altered insula response to taste stimuli in individuals recovered from restricting-type anorexia nervosa. Neuropsychopharmacology 33, 513–523. Wagner, A., Aizenstein, H., Venkatraman, V. K., Fudge, J., May, J. C., Mazurkewicz, L., et al. (2007). Altered reward processing in women recovered from anorexia nervosa. Am. J. Psychiatry164, 1842–1849. Walpoth, M., Hoertnagl, C., Mangweth-Matzek, B., Kemmler, G., Hinterhölzl, J., Conca, A., et al. (2008). Repetitive transcranial magnetic stimulation in bulimia nervosa: preliminary results of a single-centre, randomised, double-blind, sham-controlled trial in female outpatients. Psychother. Psychosom. 77, 57–60. Walsh, B. T. (2013). The enigmatic persistence of anorexia nervosa. Am. J. Psychiatry 170, 477–484. Weygandt, M., Schaefer, A., Schienle, A., and Haynes, J. D. (2012). Diagnosing different binge-eating disorders based on reward-related brain activation patterns. Hum. Brain Mapp. 33, 2135–2146. Wierenga, C. E., Ely, A., Bischoff-Grethe, A., Bailer, U. F., Simmons, A. N., and Kaye, W. H. (2014). Are extremes of consumption in eating disorders related to an altered balance between reward and inhibition? Front. Behav. Neurosci. 8:410. Wiethoff, S., Hamada, M., and Rothwell, J. C. (2014). Variability in response to transcranial direct current stimulation of the motor cortex. Brain Stimul. 7, 468–475. Wildes, J. E., Forbes, E. E., and Marcus, M. D. (2014). Advancing research on cognitive flexibility in eating disorders: the importance of distinguishing attentional set-shifting and reversal learning. Int. J. Eat. Disord. 47, 227–230. Wildes, J. E., Forbush, K. T., and Markon, K. E. (2013). Characteristics and stability of empirically derived anorexia nervosa subtypes: towards the identification of homogeneous low-weight eating disorder phenotypes. J. Abnorm. Psychol. 122, 1031–1041. Wildes, J., and Marcus, M. (2015). Application of the Research Domain Criteria (RDoC) framework to eating disorders: emerging concepts and research. Curr. Psychiatry Rep. 17, 1–10. Wonderlich, S. A., Crosby, R. D., Joiner, T., Peterson, C. B., Bardone-Cone, A., Klein, M., et al. (2005). Personality subtyping and bulimia nervosa: psychopathological and genetic correlates. Psychol. Med. 35, 649–657. Wu, M., Hartmann, M., Skunde, M., Herzog, W., and Friederich, H.-C. (2013). Inhibitory control in bulimic-type eating disorders: a systematic review and meta-analysis. PLoS ONE 8:e83412. Zastrow, A., Kaiser, S., Stippich, C., Walther, S., Herzog, W., Tchanturia, K., et al. (2009). Neural correlates of impaired cognitive-behavioral flexibility in anorexia nervosa. Am. J. Psychiatry 166, 608–616.

470

Copyright Acknowledgements This thesis acknowledges the following copyright permissions: Elsevier Limited for permission to reproduce the following published article:

Dunlop, K., Woodside. B., Lam, E., Olmsted, M., Colton, P., Giacobbe, P., Downar, J. (2015).

“Increases in frontostriatal connectivity are associated with response to dorsomedial repetitive

transcranial magnetic stimulation in refractory binge/purge behaviors.” NeuroImage: Clinical 8:

611-8.

The Nature Publishing Group for permission to reproduce the following published article:

Dunlop, K., Woodside, B., Olmsted, M., Colton, P., Giacobbe, P., Downar, J. (2016).

“Reductions in cortico-striatal hyperconnectivity accompany successful treatment of obsessive-

compulsive disorder with dorsomedial prefrontal rTMS.” Neuropsychopharmacology 41(5):

1395-403.

Frontiers Editorial for permission to reproduce the following published article: Dunlop, K.,

Woodside, B., Downar, J. (2015). Targeting Neural Endophenotypes of Eating Disorders with

Non-invasive Brain Stimulation. Frontiers in Neuroscience, (10): 30.

Wiley Periodicals Inc. on behalf of The New York Academy of Sciences for permission to

reproduce the following published article: Dunlop, K., Hanlon, C.A., Downar, J. (2017).

Noninvasive brain stimulation treatments for addiction and major depression. Annals of the New

York Academy of Science, 1394(1): 31-54.

Journal of Visualized Experiments for permission to reproduce the following published

article: Dunlop, K., Gaprielian, P., Blumberger, D., Daskalakis, Z.J., Kennedy, S.H., Giacobbe,

P., Downar, J. (2015). MRI-guided dmPFC-rTMS as a Treatment for Treatment-resistant Major

Depressive Disorder. Journal of Visualized Experiments, (102): 53129.

Identifying transdiagnostic predictors and mechanisms of ...

Documents

Transcript of Identifying transdiagnostic predictors and mechanisms of ...