ACUAD

Bba

Mod

Rccsn

Cdcc

KM

MpwEto

sidsTprst

F

A

R

0d

natomic Abnormalities of the Anterior Cingulateortex Before Psychosis Onset: An MRI Study ofltra-High-Risk Individuals

lex Fornito, Alison R. Yung, Stephen J. Wood, Lisa J. Phillips, Barnaby Nelson, Sue Cotton,ennis Velakoulis, Patrick D. McGorry, Christos Pantelis, and Murat Yücel

ackground: Abnormalities of the anterior cingulate cortex (ACC) are frequently implicated in the pathophysiology of psychotic disorders,ut whether such changes are apparent before psychosis onset remains unclear. In this study, we characterized prepsychotic ACCbnormalities in a sample of individuals at ultra-high-risk (UHR) for psychosis.

ethods: Participants underwent baseline magnetic resonance imaging and were followed-up over 12–24 months to ascertain diagnosticutcomes. Baseline ACC morphometry was then compared between UHR individuals who developed psychosis (UHR-P; n � 35), those whoid not (UHR-NP; n � 35), and healthy control subjects (n � 33).

esults: Relative to control subjects, UHR-P individuals displayed bilateral thinning of a rostral paralimbic ACC region that was negativelyorrelated with negative symptoms, whereas UHR-NP individuals displayed a relative thickening of dorsal and rostral limbic areas that wasorrelated with anxiety ratings. Baseline ACC differences between the two UHR groups predicted time to psychosis onset, independently ofymptomatology. Subdiagnostic comparisons revealed that changes in the UHR-P group were driven by individuals subsequently diag-osed with a schizophrenia spectrum psychosis.

onclusions: These findings indicate that anatomic abnormalities of the ACC precede psychosis onset and that baseline ACC differencesistinguish between UHR individuals who do and do not subsequently develop frank psychosis. They also indicate that prepsychotichanges are relatively specific to individuals who develop a schizophrenia spectrum disorder, suggesting they may represent a diagnosti-

ally specific risk marker.

ey Words: Bipolar disorder, depression, limbic system, mania,RI, prefrontal

ounting neuropathologic and magnetic resonance im-aging (MRI) evidence supports a primary role for ab-normalities of the anterior cingulate cortex (ACC) in the

athogenesis of psychotic disorders (1–8), but it remains unclearhether these abnormalities precede or follow psychosis onset.stablishing the timing of such neuroanatomic changes is criticalo determine whether they represent a risk marker for illnessnset or a secondary manifestation of the disease process.

To determine whether neuroanatomic differences repre-ent risk markers, it is necessary to identify and follow-upndividuals at elevated risk for psychosis to characterizeiagnostic outcomes. To our knowledge, only three such MRItudies incorporating ACC measures have been published.wo have investigated individuals at ultra-high-risk (UHR) forsychosis identified using state and trait criteria (9,10). Botheported reduced ACC gray matter in UHR individuals whoubsequently developed psychosis (UHR-P) compared withhose who did not (UHR-NP). Job and colleagues (11) found

rom the Department of Psychiatry (AF, SJW, DV, CP, MY), Melbourne Neu-ropsychiatry Centre, University of Melbourne, Carlton South; PACE Clinic(ARY, BN), ORYGEN Research Centre, Department of Psychiatry; Depart-ment of Psychology (LJP), School of Behavioural Science, Faculty ofMedicine, Dentistry, and Health Sciences; ORYGEN Research Centre (SC,PDM, MY), Department of Psychiatry, University of Melbourne; andHoward Florey Institute (CP), The University of Melbourne, Victoria,Australia.

ddress reprint requests to Alex Fornito, Ph.D., Melbourne NeuropsychiatryCentre, Levels 2 and 3, National Neuroscience Facility, 161 Barry Street,Carlton South, Vic 3053, Australia; E-mail: fornitoa@unimelb.edu.au.

eceived February 21, 2008; revised April 30, 2008; accepted May 29, 2008.

006-3223/08/$34.00oi:10.1016/j.biopsych.2008.05.032

reduced ACC gray matter in individuals at genetic high risk forschizophrenia compared with healthy control subjects (11),but these reductions were no more severe in those high-riskindividuals who eventually developed schizophrenia or sub-threshold psychotic symptoms (12).

One limitation of these reports is that baseline ACC measureswere not compared with healthy control subjects, making itunclear whether differences between high-risk people who doand do not develop psychosis reflect a putative abnormality.Second, the rates of transition to psychosis have been small (max-imum n � 23) (9), precluding the opportunity to investigatesubdiagnostic specificity. A third limitation is the use of voxel-basedmorphometry (VBM), which is restricted, by the image resolution(typically � 1 mm3), in its capacity to detect relatively subtlechanges. Further, VBM fails to account for the considerable inter-individual anatomic variability of the ACC, particularly with respectto the paracingulate sulcus (PCS), a highly variable structure (seeFigure 1) that influences regional volumes (13–15). Because thePCS is less frequent in patients with psychosis (16–18) and UHRindividuals (19), spurious group differences may arise if compar-ison groups are not matched for this morphologic variation.

In this study, we sought to address many of these limitationsby investigating ACC morphometry in a relatively large sample ofUHR individuals and healthy control subjects. We applied acortical surface-based protocol for parcellating the ACC (13,15),enabling measurement of regional gray matter volume and itsconstituent parameters—surface area and cortical thickness—with submillimeter precision (20). Importantly, control subjectsand UHR-NP individuals were matched to UHR-P participants forPCS morphology to ensure that any group differences were notattributable to variations in cortical folding patterns. This ap-proach enabled us to map prepsychosis ACC changes in greater

detail than previously attempted.

BIOL PSYCHIATRY 2008;64:758–765© 2008 Society of Biological Psychiatry

M

P

sMRvfCmboba

1pncwumboiUriad

F(sTttRtirthttm ness a

A. Fornito et al. BIOL PSYCHIATRY 2008;64:758–765 759

ethods and Materials

articipantsUHR individuals (n � 146) were recruited from the Per-

onal Assessment and Crisis Evaluation (PACE) Clinic inelbourne, Australia. The Comprehensive Assessment of At-isk Mental States (CAARMS) (21), a structured clinical inter-iew designed to assess prodromal symptomatology and riskor psychosis, was administered to all cases. Based onAARMS criteria, UHR individuals are characterized by one orore of the following: 1) attenuated psychotic symptoms; 2)rief, limited intermittent psychotic symptoms with spontane-us resolution; or 3) family history of psychosis accompaniedy a decline in general functioning (full criteria in ref. 22; seelso Supplement 1).

The UHR cohort was regularly monitored over a minimum2-month period (mean � 13; maximum � 44 months). Partici-ants were then divided into individuals who did (UHR-P) or didot (UHR-NP) make the transition to psychosis using operationalriteria for determining psychosis onset (22). Final diagnosesere made by trained psychiatrists and clinical psychologistssing the Structured Clinical Interview for DSM-IV (23) andedical chart review. Of the 146 UHR individuals recruited, 41ecame psychotic over the follow-up period (UHR-P group). Sixf the scans acquired in this group were excluded because ofmage artifacts, leaving 35 UHR-P scans available for analysis.HR-NP individuals were then randomly selected from the

emaining 105 UHR participants and matched to each UHR-Pndividual according to PCS morphology (discussed later). Sexnd age were also matched as closely as possible (diagnostic

igure 1. (A) Example of the reconstructed white matter (yellow line) and pial (rROI) boundaries and their relationship with local variations in cortical folding. Tuperior rostral sulcus (SRS); bottom row presents a case with a “present” PCS1-weighted images with major sulci marked in black. Middle row presents ROIso indentations, whereas gyri correspond to protrusions. The posterior yellow linhe rostral and dorsal regions dorsally and rostral and subcallosal regions ventright column presents ROIs overlaid on the reconstructed pial surfaces. As can behe gray matter between the fundus of the cingulate sulcus (CS) and that of thef the SRS and CS were separate, the ACCP extended from the fundus of the Costroventral bank of the CS. The limbic ACC (ACCL) always comprised the grayhe white matter surface to facilitate tracing inside sulcal walls and then projecteow the ACCP is not visible from the pial surface in absent or continuous cases.

hickness was calculated by taking the distance from each vertex on the whitehese values (see ref. 24). Surface area was calculated by taking the sum of the are

atter surfaces. Gray matter volume was calculated as the product of the thick

etails in Supplement 1).

Healthy control subjects came from a similar sociodemo-graphic background to patients and were recruited by approach-ing ancillary hospital staff and family members and through localadvertisements. The final control sample was also selected froma larger cohort to match individually the UHR-P group for PCSmorphology, sex, and age. No participants had a history of signifi-cant head injury, intellectual disability, steroid or substance abuse,or neurologic disease, and control participants had no family historyof psychotic illness. Baseline symptomatology in UHR individualswas quantified using the Scale for the Assessment of NegativeSymptoms (SANS) (24), Brief Psychiatric Ratings Scale (BPRS)(25,26), and Hamilton Rating Scales for Depression and Anxiety(HRSD and HRSA, respectively; Table 1) (27,28).

Image AcquisitionScans were acquired using identical sequences on identical GE

Signa 1.5-Tesla scanners at Royal Melbourne Hospital (RMH) orRoyal Children’s Hospital (RCH), Victoria, Australia. A three-dimen-sional volumetric spoiled gradient recoil sequence generated 124contiguous coronal slices. Imaging parameters were time-to-echo �3.3 msec; time-to-repetition � 14.3 msec; flip angle � 30°; matrixsize � 256 � 256; field of view, 24 � 24 cm; voxel dimensions,.938 � .938 � 1.5 mm. The two scanners were calibrated fortnightlyusing the same proprietary phantom to ensure stability and accuracyof measurements. At the RMH site, 14 UHR-P, 9 UHR-NP, and 33control individuals were scanned; the remainder were scanned atthe RCH. In studies of other brain regions in this cohort, we havepreviously shown that scan site has no influence on brain volumes(29,30), and it had no significant effect on the current measures [graymatter volume: F(1,100) � .359, p � .550; surface area: F(1,100) �

e) surfaces overlaid on a T1-weighted image. (B) Examples of region-of-interestpresents a case with an “absent” paracingulate sulcus (PCS) and “continuous”

separate” SRS. Left column presents representative sagittal slices through thelaid on the reconstructed white matter surface. Sulci on this surface correspondesents the caudal border of the dorsal region. The anterior yellow line separatese middle yellow line represents the posterior border of the subcallosal region.

, if a PCS was present, the paralimbic anterior cingulate cortex(ACCP) compriseda PCS was absent, the ACCP was located on the dorsal bank of the CS. Similarly,

hat of the SRS. If the two sulci were continuous, the ACCP was located on ther between the callosal and cingulate sulci. The ROIs were always delineated on

o the pial surface to check for accuracy and consistency before continuing. Noteample of the triangular mesh used to reconstruct each cortical surface. Corticalr surface to the closest point on the pial surface and vice versa, and averaginge triangle faces included in each ROI, and averaging these for the pial and whitend area of each region.

ed linop rowand “over

e reprally. Th

seenPCS. IfS to tmatted ont(C) Exmattea of th

1.196, p � .277; cortical thickness: F(1,100) � .808, p � .371].

www.sobp.org/journal

C

atFowc

T

U

U

C

ts

T

SHANBSHHD

T

LAu

Usaa

pic

N

o

760 BIOL PSYCHIATRY 2008;64:758–765 A. Fornito et al.

w

lassification of Sulcal VariabilityBefore classifying PCS morphology, each participant’s im-

ge was stripped of extracerebral tissue and aligned to the N27emplate (31) through a 6° rigid-body transformation usingSL software (http://www.fmrib.ox.ac.uk/fsl). The incidencef the PCS and confluence of the superior rostral sulcus (SRS)ith the cingulate sulcus (CS) in each hemisphere were then

lassified according to established criteria (13,32) to guide

able 2. Mean (SD) Volume, Surface Area, and Cortical Thickness for Each G

d-ACCL

L R

HR-P (n � 35)Volume, mean (SD) mm3 1937.71 (599.88) 2321.44 (670.43) 1Area, mean (SD), mm2 677.07 (194.68) 797.46 (201.27)Thickness, mean (SD), mm 2.83 (.35) 2.82 (.25)

HR-NP (n � 35)Volume, mean (SD) mm3 1949.84 (626.22) 2290.71 (637.89) 1Area, mean (SD), mm2 655.74 (185.73) 771.51 (210.09)Thickness, mean (SD), mm 2.92 (.31) 2.91 (.30)

ontrol Subjects (n � 33)Volume, mean (SD) mm3 1848.58 (578.74) 2130.70 (541.27) 1Area, mean (SD), mm2 660.80 (177.12) 739.12 (173.02)Thickness, mean (SD), mm 2.73 (.31) 2.83 (.19)

Volume and area values are corrected for intracranial volume as detaileACCL, limbic anterior cingulate cortex; ACCP, paralimbic anterior cingula

ransition to psychosis onset during follow-up; UHR-P, ultra-high-risk indivi

able 1. Demographic Details

UHR-P(n � 35)a

ex, M/F (n) 21/14andedness, L/R/Mi (n) 3/31/0ge, mean (SD), years 19.30 (3.49) 1ART�IQ,c mean (SD) 93.86 (12.42) 97PRS� total,d mean (SD) 4.65 (3.06) 4ANS total, mean (SD) 30.80 (15.56) 21RSD total,e mean (SD) 20.79 (11.12) 1RSA total,e mean (SD) 15.73 (9.41) 1uration of symptoms before referral,median (min.–max.), yearsf

.97 (.01–8.08) .80

ime between scan and psychosisonset, median (min.–max.), years

.32 (.02–2.28)

BPRS, Brief Psychiatric Rating Scale, administered at intake; F, female; HRS, left-handed; M, male; Mi, mixed-handedness; NART-IQ, National Adult Ressessment of Negative Symptoms; UHR-NP, ultra-high-risk individuals wltra-high-risk individuals who made the transition to psychosis during follo

aEleven UHR-P individuals and 17 UHR-NP individuals were participatinHR-NP individuals were taking low-dose risperidone (mean dose � 1.3 mg

ix UHR-P and nine UHR-NP individuals were assigned to a control group recntipsychotic medication at the time of scanning. The breakdown of nontidepressants, 10 none, 15 data unavailable; UHR-NP: 11 antidepressants

bTwo UHR-P individuals could not be matched to a healthy controlaracingulate sulcus in both hemispheres (see Methods and Materials). P

ndividuals who could not be matched to control subjects did not alter the fiontrol subjects.

cOne UHR-P individual had a NART-IQ score � 70, but removing this parART-IQ data were unavailable for six UHR-P, four UHR-NP individuals, and

dRepresents sum of scores on BPRS scales assessing positive symptomsne UHR-P and one UHR-NP individual.

eData were unavailable for two UHR-P and two UHR-NP individuals.fGroup differences assessed using the Mann-Whitney U test.

ubcallosal divisions are denoted by the prefixes d-, r-, and s-, respectively.

ww.sobp.org/journal

region of interest (ROI) parcellation (Figure 1). UHR-P indi-viduals were subsequently assigned to one of four categoriesbased on their PCS classification in each hemisphere. Thesecategories were as follows: present in the left and absent in theright (n � 14), present in both hemispheres (n � 13), absentin both hemispheres (n � 3), or absent in the left and presentin the right (n � 5). UHR-NP individuals and control subjectswere then randomly selected from a larger database and

in Each Region-of-Interest

d-ACCP r-ACCL

L R L R

9 (705.75) 1265.11 (576.37) 1495.16 (1109.74) 1993.87 (889.37)3 (237.97) 451.80 (195.71) 471.07 (350.15) 632.24 (286.23)1 (.23) 2.74 (.24) 3.11 (.33) 3.12 (.33)

6 (624.87) 1225.50 (528.32) 1318.87 (700.90) 1849.46 (881.21)7 (201.23) 434.26 (183.75) 382.69 (202.58) 562.99 (278.48)2 (.21) 2.78 (.22) 3.36 (.37) 3.22 (.27)

0 (872.75) 1290.50 (590.93) 1441.14 (989.97) 1732.50 (958.81)4 (278.13) 448.00 (196.78) 445.48 (299.21) 538.02 (294.45)5 (.24) 2.83 (.21) 3.12 (.43) 3.13 (.23)

ethods and Materials.tex; L, left; R, right; UHR-NP, ultra-high-risk individuals who did not make thewho made the transition to psychosis during follow-up. Dorsal, rostral, and

NP5)a

Control Subjects(n � 33)b df F/t p

5 21/12/2 4/28/13.40) 20.97 (6.18) 2,100 1.19 .312.74) 100.14 (9.56) 2,89 2.23 .11.45) 1,66 .93 .347.21) 1,68 6.21 .02

9.26) 1,64 3.30 .078.06) 1,64 .29 .60–4.18) 481 .23

milton Rating Scale for Anxiety; HRSD, Hamilton Rating Scale for Depression;Test-Estimated Intelligence Quotient; R, right-handed; SANS, Scale for the

id not make the transition to psychosis onset during follow-up; UHR-P,.n intervention trial at the time of scanning. As such, five UHR-P and eight

) and engaged in cognitive-behavioral therapy at the time of scanning, andsupportive therapy alone. The remaining UHR participants were not taking

psychotic medication in the two UHR groups was as follows: UHR-P: 10ticonvulsant, 1 anticholinergic, 1 anxiolytic, 14 none, 7 data unavailable.ct because of insufficient control women classified as having a presentinary analyses indicated that removing the two UHR-P and two UHR-NPgs, and we report results for the analysis with 35 UHR-P, 35 UHR-NP, and 33

nt did not change the findings, and they were retained in the final analysis.ontrol subjects.cales 8, 9, 10, 11, and 15), following Ruggeri et al. Data were unavailable for

roup

649.9571.7

2.8

427.7490.7

2.8

572.8533.4

2.8

d in Mte corduals

UHR-(n � 3

20/12/29

9.91 (.32 (1.00 (2.03 (1

6.21 (4.58 (

(.03

A, Haadingho dw-up

g in a/day

eivingnanti, 1 ansubjerelimndin

ticipatwo c(i.e., s

ictm

C

cscwfmrpu

R

st(A(riacegwmtrcIu

S

mc

T

A. Fornito et al. BIOL PSYCHIATRY 2008;64:758–765 761

ndividually matched to each patient on the basis of this PCSategorization, sex, and age. Because PCS and SRS classifica-ions tend to be related, this procedure also resulted in goodatching for SRS morphology (15).

ortical Surface ReconstructionThe white (i.e., gray/white matter boundary) and pial (gray/

erebrospinal fluid boundary) cortical surfaces were recon-tructed using methods described in detail by Dale, Fischl, andolleagues (20,33), and as implemented in the Freesurfer soft-are package (http://surfer.nmr.mgh.harvard.edu). These sur-

ace representations enabled calculation of surface area, grayatter volume, and mean cortical thickness for each of six ACC

egions per hemisphere (discussed later) with submillimeterrecision (20) (see Figure 1). All surfaces were reconstructedsing the raw, unaligned images in native space.

OI DelineationOur surface-based ACC parcellation protocol has been de-

cribed in detail elsewhere (13,15). Briefly, the protocol divideshe limbic (corresponding to Brodmann area 24) and paralimbicprimarily corresponding to Brodmann area 32) ACC (ACCL andCCP, respectively) into dorsal (d-ACCL and d-ACCP), rostralr-ACCL and r-ACCP), and subcallosal (s-ACCL and s-ACCP)egions, yielding six ROIs per hemisphere. Boundaries separat-ng the dorsal, rostral, and subcallosal regions were designed topproximate previously identified subdivisions specialized forognitive (dorsal), affective (rostral), and autonomic/negativemotion (subcallosal) processing (34,35). Boundaries distin-uishing the limbic from paralimbic ACC varied in accordanceith PCS and SRS variability, following postmortem work docu-enting how ACC cytoarchitecture changes with sulcal varia-

ions (36) (Figure 1; see ref. 13 for more details). We haveecently shown good reliabilities for this method (intraclassorrelation coefficients for all ROIs � .8, with most � .9) (15).ntracranial volume (ICV) was also estimated for each individualsing a previously described method (37).

tatistical AnalysesWe performed four types of analysis. First, regional gray

atter volumes, surface area, and cortical thickness wereompared using separate mixed within- and between-subjects

able 2. (continued)

r-ACCP

L R L

2576.41 (1159.16) 2468.07 (1246.66) 224.90 (141.67)902.46 (407.07) 864.37 (416.06) 90.89 (48.38)

2.77 (.35) 2.78 (.18) 2.34 (.43)

2570.57 (877.24) 2441.44 (1029.51) 194.47 (105.45)867.77 (287.92) 844.60 (348.39) 72.90 (33.77)

2.89 (.30) 2.83 (.21) 2.58 (.57)

2585.21 (1023.83) 2383.71 (933.05) 265.52 (189.37)857.83 (323.78) 810.33 (305.86) 97.74 (65.08)

2.88 (.26) 2.93 (.24) 2.59 (.45)

analysis of variance (ANOVA), with hemisphere (left or right),region (dorsal, rostral, and subcallosal), and cortex (ACCL orACCP) as within-subjects factors and group (UHR-P, UHR-NP,or control) as the between-subjects factor (adding sex did notchange the findings, so we report models without it). Maineffects and interactions were evaluated with � � .05, usingGreenhouse-Geisser corrected degrees of freedom. Post hoccontrasts were evaluated against a Bonferroni-adjusted alphato correct for multiple comparisons (i.e., the three possiblepairwise comparisons between the control, UHR-P, andUHR-NP groups). Effect sizes (Cohen’s d) are also reported forthese differences (negative values reflect reductions in pa-tients). In the second analysis, bivariate correlations betweengray matter changes and six clinical variables (baseline SANS,BPRS�, HRSD, HRSA, time between scan and psychosis onset,and duration of symptomatology before referral; see Table 1)were evaluated against a Bonferroni-adjusted alpha (adjustedfor six tests per group) using Pearson’s r. Spearman’s rho wasused for non-Gaussian variables. The third analysis used Coxregression to assess whether the ACC differences identified inthe ANOVA predicted time to psychosis onset in the UHR group,independently of shared variance with baseline symptom ratings.To prevent losing predictive information because of the conser-vative Bonferroni corrections applied in the ANOVA, and toavoid reliance on statistical significance as the sole inclusioncriterion, ROIs were selected for entry into the model if theyshowed an effect size of at least moderate magnitude (i.e., d �.4). Only UHR individuals were used in this analysis becausecontrol subjects did not complete symptom assessments. Thefourth analysis represented a preliminary investigation of subdi-agnostic differences. The UHR-P individuals were divided intothose who developed a schizophrenia spectrum (UHR-SZ: 19schizophrenia, 2 schizoaffective) or nonschizophreniform (UHR-OTHER: 7 psychotic depression, 6 bipolar I, 2 psychosis nototherwise specified) psychosis. We then ran separate ANOVAscomparing these two groups to their respective matched controland UHR-NP samples.

Gray matter volume and surface area estimates were cor-rected for ICV using previously described equations (38). Corticalthickness was not corrected for ICV because the anatomicsignificance of the relationship between the two is unclear (39).(Separate analyses that did correct for ICV yielded similar

CL s-ACCP

R L R

245.57 (150.50) 295.61 (164.61) 226.69 (171.69)83.49 (48.88) 99.23 (58.62) 74.51 (58.77)

2.87 (.51) 3.11 (.50) 3.12 (.58)

232.03 (125.03) 308.69 (163.80) 209.51 (123.04)77.04 (39.77) 95.54 (54.18) 66.16 (43.90)

2.97 (.48) 3.31 (.59) 3.37 (.59)

249.18 (144.49) 245.52 (156.47) 245.76 (176.79)80.50 (41.37) 80.42 (51.64) 78.03 (60.91)

3.00 (.63) 3.04 (.46) 3.28 (.46)

s-AC

www.sobp.org/journal

fQt

Fdrhopagcd

Fp(NA

762 BIOL PSYCHIATRY 2008;64:758–765 A. Fornito et al.

w

indings.) National Adult Reading Test—Estimated Intelligenceuotient (NART-IQ) (40) was not a significant covariate in any of

he analyses and so was not retained in the final models.

igure 2. (A) Effect sizes (Cohen’s d) for ultra-high-risk individuals whoeveloped psychosis (UHR-P) and who did not develop psychosis (UHR-NP)

elative to healthy control subjects in each subregion, collapsed acrossemispheres. Effect sizes for subgroups of UHR-P individuals who devel-ped either a schizophrenia spectrum psychosis (UHR-SZ) or nonschizo-hreniform (UHR-OTHER) psychosis relative to healthy control subjects arelso presented. (B) Effect sizes for the UHR-P, UHR-SZ, and UHR-OTHERroups relative to UHR-NP individuals. ACCL � Limbic anterior cingulateortex; ACCP � Paralimbic anterior cingulate cortex; d-, r-, and s- denoteorsal, rostral, and subcallosal divisions, respectively. * p � .05, corrected.

igure 3. Illustration of the location of anterior cingulate cortex (ACC) subsychosis (UHR-P) compared with (A) healthy control subjects and (B

C) Scatterplots illustrating the relationship between thickness in the rostraegative Symptoms (SANS) in the UHR-P (left) and UHR-NP (right) groups. (D

CC (r-ACCL) and baseline scores on the Hamilton Rating Scale for Anxiety (HRSA

ww.sobp.org/journal

Results

Preonset Differences Among the UHR-P, UHR-NP,and Control Groups

Group means for each measure are presented in Table 2.For both gray matter volume and surface area, there was nomain effect of group [volume: F (2,100) � .265, p � .768; area:F (2,100) � 1.366, p � .260], or any interaction between groupand region [volume: F (3.11,155.52) � .162, p � .927; area:F (2.98,148.93) � .367, p � .775], cortex [volume: F (2,100) �.364, p � .695; area: F (2,100) � .187, p � .830], or hemisphere[volume: F (2,100) � .414, p � .662; area: F (2,100) � .772, p �.465], nor was group involved in any three-way or higher-order interactions. In a secondary analysis, we calculated areaand volume z scores, estimated relative to the control group’smean and standard deviation, to control for interregionaldifferences in size (i.e., the volume and area of dorsal androstral regions were much larger than those of subcallosalareas). This analysis also found no significant group differ-ences.

For cortical thickness, there was a significant main effect ofgroup [F (2,100) � 5.276, p � .005] and an interaction betweengroup, region, and cortex [F (3.05,152.66) � 4.128, p � .007],but no interaction with hemisphere [F (3.19,159.37) � 1.288,p � .280]. Post hoc testing revealed that, relative to healthycontrol subjects, UHR-P individuals showed a significantthickness reduction in the bilateral r-ACCP (d � –.596 , p �.047, corrected). In contrast, UHR-NP individuals showedincreased thickness in the d-ACCL (d � .592, p � .049,corrected), with a trend in the r-ACCL (d � .584, p � .054,corrected), compared with control subjects. Comparisons be-tween UHR-P and UHR-NP individuals revealed a bilateralthickness reduction in the r-ACCL of the former group (d �–.630, p � .029, corrected), with a trend in the s-ACCP (d �-.578, p � .052, corrected). These differences are illustrated inFigures 2 and 3.

ns showing relative thinning in ultra-high risk-individuals who developedultra-high-risk individuals who did not develop psychosis (UHR-NP).

limbic ACC (r-ACCP) and baseline scores on the Scale for the Assessment oftterplots illustrating the relationship between thickness in the rostral limbic

regio) andl para) Sca

) in UHR-P (left) and UHR-NP (right) individuals.

C

br..tcbeiUd(s

P

Usabsyp9f2itttarr9.m9(

S

ctpsrpgasicsr

D

spdb

A. Fornito et al. BIOL PSYCHIATRY 2008;64:758–765 763

orrelations with Clinical VariablesIn the UHR-P group, there was a significant correlation between

ilateral r-ACCP thickness (the region showing significant thinningelative to healthy control subjects) and SANS ratings (r � –.465, p �03, corrected) that was not apparent in the UHR-NP group (r �169, p � .332, corrected; see Figure 3). The difference betweenhese two correlations (using Fisher’s r to z transform) was signifi-ant (p � .01). Conversely, a positive correlation was identifiedetween r-ACCL thickness (the region showing a significant differ-nce between UHR-P and UHR-NP individuals) and HRSA ratingsn UHR-NP individuals (r � .54, p � .01, corrected), but notHR-P participants (r � –.075, p � .679, corrected). Theifference between these two correlations was also significantp � .01). No other significant correlation between thickness andymptom or clinical measures was identified.

rediction of Psychosis Onset During Follow-UpFour regions showed an effect size � .4 in contrasts between

HR-P and UHR-NP individuals: the r-ACCL, r-ACCP, s-ACCL, and-ACCP. Thickness values were averaged across these regionsnd entered into a Cox regression as predictors, along withaseline SANS and HRSA scores (i.e., the only two symptomcales showing any correlations with ACC measures). This anal-sis revealed that thickness in these regions was a significantredictor of time to psychosis onset (Wald � 5.149, � � .196,5% confidence interval [CI] � .05–.80, p � .023), suggesting that,or every 1-mm decrease in thickness, there was an approximately0% increase in risk for psychosis (or, conversely, a 20% decreasen risk for psychosis for every 1-mm thickness increase, given thathe relationship is driven both by thinning in UHR-P individuals andhickening in UHR-NP individuals; see Discussion). The meanhickness difference between the two groups in these regions waspproximately .2 mm, suggesting the observed thickness changeselated to an average risk increase of approximately 4%. The SANSatings also predicted time to transition (Wald � 5.485, � � 1.027,5% CI � 1.00–.1.05, p � .019), but HRSA scores did not (Wald �304, � � 1.011, 95% CI � .97–1.05, p � .581). The thicknesseasures remained significant as predictors of outcome (� � .218,5% CI � .05–.99, p � .048) after the other symptom measuresHRSD and BPRS�) were added to the model.

ubdiagnostic DifferencesThe ANVOVA comparing UHR-SZ individuals to their matched

ontrol subjects and UHR-NP groups revealed a three-way interac-ion among diagnosis, region, and cortex [F(3.14,94.05) � 3.639,� .014]. Post hoc contrasts indicated that, compared with control

ubjects, the UHR-SZ group showed reduced thickness in the-ACCP (d � –.790, p � .046, corrected) and s-ACCL (d � –.782,� .049, corrected). Relative to UHR-NP individuals, the UHR-SZroup showed reduced r-ACCL (d � –.824, p � .035, corrected)nd s-ACCP (d � –.906, p � .017, corrected) thickness. Noignificant differences were identified between UHR-OTHERndividuals and their matched control or UHR-NP groups. Directomparison between UHR-SZ and UHR-OTHER individuals in aubset that could be matched for PCS morphology (n � 13)evealed no differences.

iscussion

Identifying the neurobiological changes that precede psycho-is onset represents an important step in understanding theathogenesis of psychotic disorders. In this study, we haveemonstrated that anatomic changes in the ACC 1) are apparent

efore the onset of frank psychosis, 2) distinguish between UHR

individuals who subsequently do and do not develop a psychoticepisode, and 3) are relatively specific to individuals who developa schizophrenia spectrum disorder. Consistent with previousfindings (7,15), our data also indicate that cortical thicknessrepresents a particularly sensitive metric for detecting anatomicchanges, possibly because it shows less variability across indi-viduals than area and volume (15). Importantly, our procedurefor matching groups for local sulcal and gyral variability ensuredthese changes are unlikely to reflect differences in corticalfolding patterns. Together these findings indicate that ACCabnormalities represent a risk marker of transition to psychosis.

ACC Abnormalities Preceding Psychosis OnsetComparison of UHR-P individuals with healthy control sub-

jects revealed that thinning of the r-ACCP precedes the onset offrank psychosis. The localization of these changes is interestingin light of our recent work demonstrating bilateral thinning of theparalimbic region in first episode schizophrenia (8), and longi-tudinal research in childhood-onset schizophrenia suggesting theearliest postonset changes appear in paralimbic regions andspread to engulf the limbic ACC over a 5-year period (41). Wehave also found evidence for longitudinal gray matter reductionsin dorsal paralimbic regions during the transition to psychosis(9). Together these findings suggest a timetable for the progres-sion of ACC abnormalities in schizophrenia, whereby the earliestreductions emerge in the rostral paralimbic region during theprodrome, extend across the dorsal and subcallosal paralimbicregions following psychosis onset, and spread to encompasslimbic regions with continued illness duration.

Evidence that the r-ACCP plays an important role in socialcognitive processing, particularly mentalizing others’ intentions(42), suggests abnormalities in the area lead to difficultiesinteracting with others. This may underlie the increasing socialwithdrawal that is characteristic of the psychosis prodrome(43,44). Recent data suggest genetic high-risk individuals display-ing subthreshold psychotic symptoms show poorer performanceon theory-of-mind tasks than those without such symptoms (45).In this context, our finding of an inverse correlation betweenr-ACCP thickness and negative symptoms in the UHR-P groupsuggests the functional consequences of these anatomic changesinclude a diminished engagement with the external world.However, determining whether the r-ACCp changes cause eleva-tions in negative symptoms or whether both are manifestationsof some other trait marker requires further longitudinal investi-gation. Given that we found no correlation between r-ACCP

thickness and length of symptoms before referral, the thicknesschanges appear related to the magnitude rather than duration ofprodromal symptomatology.

ACC Changes Predictive of the Transition to PsychosisComparison between UHR-P and UHR-NP individuals indi-

cated that changes in the r-ACCL and, to a lesser extent, therostral paralimbic and limbic and paralimbic subcallosal regions,differentiated between the two groups. Cox regression indicatedthese thickness differences predicted time to psychosis onsetindependently of any correlations with baseline symptom rat-ings. Examination of these differences relative to healthy controlsubjects indicated the association was driven by abnormal thick-ening in the UHR-NP group, in addition to thinning in the UHR-Pgroup, suggesting that baseline differences between high-riskindividuals who do and do not subsequently become psychoticmay result from abnormalities in either group, not just prepsy-

chotic individuals. Our previous study of UHR individuals (9) did

www.sobp.org/journal

natwpsb(pdfAtaop

utomaegMbtdp

S

dsUUsU

rbiirbdfhftohwu

L

tsStmd

764 BIOL PSYCHIATRY 2008;64:758–765 A. Fornito et al.

w

ot examine control subjects, and although Borgwardt et al. (10)nd Job et al. (12) both employed control groups in their studies,he former only compared them to the high-risk cohort as ahole (irrespective of diagnostic outcome), with the latter com-aring longitudinal, but not baseline, changes relative to controlubjects (although an earlier study examined baseline differencesetween control subjects and the high-risk group as a whole)11). Our current findings therefore caution against attributingreonset differences between high-risk individuals who do ando not subsequently develop psychosis to abnormalities in theormer group unless healthy control subjects are also examined.n intriguing possibility arising from these data is that abnormal

hickening in the UHR-NP group confers increased resiliencegainst transition to psychosis, although the positive correlationbserved between thickness and anxiety levels suggests that itredisposes to other psychopathology.

The differences between the UHR-P and UHR-NP groups arenlikely to be due to variations in their baseline clinical charac-eristics, given that ACC measures predicted time to psychosisnset independently of any shared variance with symptomeasures. The two groups were well-matched demographically,

nd while a small proportion of UHR individuals were alsonrolled in an intervention trial, the relative proportions in eachroup receiving each intervention type were similar (Table 1).oreover, the dissociation observed with respect to correlationsetween regional ACC abnormalities and symptom measures inhe two groups suggests that the anatomic changes are related toistinct pathophysiologic processes related to their divergentsychiatric outcomes.

ubdiagnostic DifferencesThe changes observed in the UHR-P group were primarily

riven by individuals who went on to develop a schizophreniapectrum psychosis. Effect sizes relative to both the control andHR-NP groups were larger when analyses were restricted to theHR-SZ group, and the differences were generally in the sameubregions. In contrast, effect sizes for contrasts involving theHR-OTHER group were relatively small.The only other study to investigate preonset brain changes

elated specifically to risk for schizophrenia found no evidence foraseline differences between those who eventually developed thellness and those who did not (12). One reason for this discrepancys that thickness, but not volume, may represent a specific marker ofisk for psychosis, consistent with reports that other, nonvolumetricrain changes predict transition (46). Further reasons for ouriscrepant findings include our attempt to account for the con-ounding influence of PCS variability and the possibility that geneticigh-risk individuals may show different neurobiological changesrom those at clinical high risk. Although our multifaceted approacho defining risk is likely to be more representative of the spectrumf psychosis patients (i.e., not all such individuals possess a familyistory), it is possible that different risk categories are associatedith distinct neuroanatomic profiles. Our sample size did not permits to stratify analyses by risk category.

imitationsImages were acquired on two scanners, although it is unlikely

hat this influenced our findings for several reasons. First, thecanners were of an identical make and calibrated fortnightly.econd, we have previously shown that differences betweenhese two scanners have no effect on a variety of neuroanatomiceasures (29,30), including those used in this study. Third, if the

ifferences were due to scan site effects, it is unclear why UHR-P

ww.sobp.org/journal

and UHR-NP should display a distinct and regionally specificprofile of anatomic changes relative to healthy control subjects,why these should be correlated with specific symptom dimen-sions, and why they should occur for thickness but not volume orsurface area. Finally, the cortical surface reconstruction tech-niques and resulting measures we used are robust to variationsacross scanners and acquisition protocols (20,47).

We individually matched groups for PCS morphology tocontrol for the confounding influence of ACC anatomic variabil-ity (13,15). This matching procedure introduced no samplingbiases in UHR-P group, which comprised all UHR-P individualsfor whom MRI data of sufficient quality were available. Althoughhealthy control subjects and UHR-NP individuals were selectedfrom a larger sample to match the UHR-P individuals for PCSmorphology, there were no significant differences between thosewho were and were not selected in terms of age or NART-IQ. Inany case, the matching procedure would have minimized groupdifferences, because they were chosen on the basis of possessingsimilar brain morphology. However, individuals do not presentin clinical settings after having been selected for some neuroana-tomic characteristic. Thus although our findings indicate that ACCchanges are able to discriminate between UHR individuals who doand do not subsequently develop psychosis independently ofdifferences in baseline symptom profiles or variations in corticalfolding patterns, their practical utility in predicting which high-riskindividuals will make the transition to psychosis remains a topic forfurther investigation in unselected samples. We also note thatalthough some UHR-NP individuals may have developed psychosisafter our follow-up period, the majority of transitions occur withinthe first year (48). We are currently conducting a longer-termfollow-up of this cohort to assess their outcomes.

In summary, we have demonstrated that anatomic abnormal-ities of the ACC are present before psychosis onset and arerelatively specific to risk for schizophrenia spectrum disorders.Further work that examines the longitudinal progression of thesechanges and integrates them with changes observed in otherbrain regions (10,12,49) will increase the predictive power ofneuroanatomic risk markers and elucidate underlying patho-genic mechanisms.

Neuroimaging analysis was facilitated by the Neuropsychia-try Imaging Laboratory managed by Ms. Bridget Soulsby at theMelbourne Neuropsychiatry Centre and supported by Neuro-sciences Victoria. The research was supported by the MelbourneNeuropsychiatry Centre (Sunshine Hospital), Department of Psy-chiatry, the University of Melbourne, the National Health andMedical Research Council (ID 236175; 350241), and the IanPotter Foundation. AF was supported by a JN Peters Fellowshipand a National Health and Medical Research Council (NHMRC)CJ Martin Fellowship (ID: 454797). SJW was supported by aNHMRC Clinical Career Development Award and a NationalAlliance for Research on Schizophrenia and Depression YoungInvestigator Award. MY was supported by a NHMRC ClinicalCareer Development Award (ID: 509345).

The authors reported no biomedical financial interests orpotential conflicts of interest.

Supplementary material cited in this article is availableonline.

1. Benes FM (1993): Relationship of cingulate cortex to schizophrenia andother psychiatric disorders. In: Vogt B A, Gabriel M, editors. Neurobiologyof Cingulate Cortex and Limbic Thalamus: A Comprehensive Handbook.

Boston: Birkhäuser.

1

1

1

1

1

1

1

1

1

1

2

2

2

2

2

2

A. Fornito et al. BIOL PSYCHIATRY 2008;64:758–765 765

2. Yücel M, Wood SJ, Fornito A, Riffkin J, Velakoulis D, Pantelis C (2003):Anterior cingulate dysfunction: Implications for psychiatric disorders?J Psychiatry Neurosci 28:350 –354.

3. Sanders GS, Gallup GG, Heinsen H, Hof PR, Schmitz C (2002): Cognitivedeficits, schizophrenia, and the anterior cingulate cortex. Trends CognSci 6:190 –192.

4. Goldstein JM, Goodman JM, Seidman LJ, Kennedy DN, Makris N, Lee H,et al. (1999): Cortical abnormalities in schizophrenia identified by struc-tural magnetic resonance imaging. Arch Gen Psychiatry 56:537–547.

5. Benes FM, Vincent SL, Todtenkopf M (2001): The density of pyramidaland nonpyramidal neurons in anterior cingulate cortex of schizo-phrenic and bipolar subjects. Biol Psychiatry 50:395– 406.

6. Harrison PJ (2002): The neuropathology of primary mood disorder. Brain125:1428 –1449.

7. Fornito A, Malhi GS, Lagopoulos J, Ivanovski B, Wood SJ, Saling MM, et al.(2008): Anatomical abnormalities of the anterior cingulate and paracingu-late cortex in patients with bipolar I disorder. Psychiatry Res 162:123–132.

8. Fornito A, Yucel M, Wood SJ, Adamson C, Velakoulis D, Saling MM,et al. (2008): Surface-based morphometry of the anterior cingulate cor-tex in first episode schizophrenia. Hum Brain Mapp 29:478 – 489.

9. Pantelis C, Velakoulis D, McGorry PD, Wood SJ, Suckling J, Phillips LJ, etal. (2003): Neuroanatomical abnormalities before and after onset ofpsychosis: A cross-sectional and longitudinal MRI comparison. Lancet361:281–288.

0. Borgwardt SJ, Riecher-Rossler A, Dazzan P, Chitnis X, Aston J, Drewe M,et al. (2007): Regional gray matter volume abnormalities in the at riskmental state. Biol Psychiatry 61:1148 –1156.

1. Job DE, Whalley HC, McConnell S, Glabus M, Johnstone EC, Lawrie SM(2003): Voxel-based morphometry of grey matter densities in subjectsat high risk of schizophrenia. Schizophr Res 64:1–13.

2. Job DE, Whalley HC, Johnstone EC, Lawrie SM (2005): Grey matterchanges over time in high risk subjects developing schizophrenia. Neu-roimage 25:1023–1030.

3. Fornito A, Whittle S, Wood SJ, Velakoulis D, Pantelis C, Yucel M (2006):The influence of sulcal variability on morphometry of the human ante-rior cingulate and paracingulate cortex. Neuroimage 33:843– 854.

4. Paus T, Otkay N, Caramanos Z, MacDonald D, Zijdenbos A, D’Avirro D,et al. (1996): In vivo morphometry of the intrasulcal gray matter in thehuman cingulate, paracingulate, and superior rostral sulci: Hemisphericasymmetries, gender differences and probability maps. J Comp Neurol376:664 – 673.

5. Fornito A, Wood SJ, Whittle S, Fuller J, Adamson C, Saling MM, et al.(2008): Variability of the paracingulate sulcus and morphometry of themedial frontal cortex: Associations with cortical thickness, surface area,volume, and sulcal depth. Hum Brain Mapp 29:222–236.

6. Yücel M, Stuart GW, Maruff P, Wood SJ, Savage GR, Smith DJ, et al. (2002):Paracingulate morphologic differences in males with establishedschizophrenia: A magnetic resonance imaging morphopmetric study.Biol Psychiatry 52:15–23.

7. Le Provost JB, Bartres-Faz D, Paillere-Martinot ML, Artiges E, Pappata S,Recasens C, et al. (2003): Paracingulate sulcus morphology in men withearly-onset schizophrenia. Br J Psychiatry 182:228 –232.

8. Fornito A, Malhi GS, Lagopoulos J, Ivanovski B, Wood SJ, Velakoulis D,et al. (2007): In vivo evidence for early neurodevelopmental anomaly ofthe anterior cingulate cortex in bipolar disorder. Acta Psychiatr Scand116:467– 472.

9. Yücel M, Wood SJ, Phillips LJ, Stuart GW, Smith DJ, Yung A, et al. (2003):Morphology of the anterior cingulate cortex in young men at ultra-highrisk of developing a psychotic illness. Brit J Psychiatry 182:518 –524.

0. Fischl B, Dale AM (2000): Measuring the thickness of the human cerebralcortex from magnetic resonance images. Proc Natl Acad Sci U S A 97:11050 –11055.

1. Yung AR, Yuen HP, McGorry PD, Phillips LJ, Kelly D, Dell’Olio M, et al.(2005): Mapping the onset of psychosis: The Comprehensive Assess-ment of At-Risk Mental States. Aust N Z J Psychiatry 39:964 –971.

2. Yung A, Phillips L, McGorry PD (2004): Treating Schizophrenia in theProdromal Phase. London: Taylor & Francis.

3. First MB, Spitzer RL, Gibbon M, Williams JB (1998): Structured ClinicalInterview for DSM-IV. Washington, DC: American Psychiatric Press.

4. Andreasen NC (1982): Negative symptoms in schizophrenia. Definitionand reliability. Arch Gen Psychiatry 39:784 –788.

5. Overall JE, Gorham DR (1962): The Brief Psychiatric Rating Scale. PsycholRep 10:790 – 812.

26. McGorry PD, Goodwin RJ, Stuart GW (1988): The development, use, andreliability of the brief psychiatric rating scale (nursing modification)—anassessment procedure for the nursing team in clinical and research set-tings. Compr Psychiatry 29:575–587.

27. Hamilton M (1959): The assessment of anxiety states by rating. Br JPsychiatry 32:50 –55.

28. Hamilton M (1960): A rating scale for depression. J Neurol NeurosurgPsychiatry 23:56 – 62.

29. Velakoulis D, Wood SJ, Wong MT, McGorry PD, Yung A, Phillips L, et al.(2006): Hippocampal and amygdala volumes according to psychosisstage and diagnosis: A magnetic resonance imaging study of chronicschizophrenia, first-episode psychosis, and ultra-high-risk individuals.Arch Gen Psychiatry 63:139 –149.

30. Phillips LJ, Velakoulis D, Pantelis C, Wood S, Yuen HP, Yung AR, et al.(2002): Non-reduction in hippocampal volume is associated with higherrisk of psychosis. Schizophr Res 58:145–58.

31. Holmes CJ, Hoge R, Collins L, Woods R, Toga AW, Evans AC (1998):Enhancement of MR images using registration for signal averaging.J Comput Assist Tomogr 22:324 –333.

32. Yücel M, Stuart GW, Maruff P, Velakoulis D, Crowe SF, Savage G, et al.(2001): Hemispheric and gender-related differences in the gross mor-phology of the anterior cingulate/paracingulate cortex in normal volun-teers: An MRI morphometric study. Cereb Cortex 11:17–25.

33. Dale AM, Fischl B, Sereno MI (1999): Cortical surface-based analysis I.Segmentation and surface reconstruction. Neuroimage 9:179 –194.

34. Bush G, Luu P, Posner MI (2000): Cognitive and emotional influences inanterior cingulate cortex. Trends Cogn Sci 4:215–222.

35. Devinsky O, Morrell MJ, Vogt BA (1995): Contributions of anterior cingu-late cortex to behaviour. Brain 118:279 –306.

36. Vogt BA, Nimchinsky EA, Vogt LJ, Hof PR (1995): Human cingulate cor-tex: Surface features, flat maps, and cytoarchitecture. J Comp Neurol359:490 –506.

37. Eritaia J, Wood SJ, Stuart GW, Bridle N, Dudgeon P, Maruff P, et al. (2000):An optimized method for estimating intracranial volume from mag-netic resonance images. Magn Reson Med 44:973–977.

38. Free SL, Bergin PS, Fish DR, Cook MJ, Shorvon SD, Stevens JM (1995):Methods for normalization of hippocampal volumes measured withMR. AJNR Am J Neuroradiol 16:637– 643.

39. Rakic P (1988): Specification of cerebral cortical areas. Science 241:170 –176.

40. Nelson HE, O’Connell A (1978): Dementia: The estimation of premorbidintelligence levels using the new adult reading test. Cortex 14:234 –244.

41. Vidal CN, Rapoport JL, Hayashi KM, Geaga JA, Sui Y, McLemore LE, et al.(2006): Dynamically spreading frontal and cingulate deficits mapped inadolescents with schizophrenia. Arch Gen Psychiatry 63:25–34.

42. Amodio DM, Frith CD (2006): Meeting of minds: The medial frontalcortex and social cognition. Nat Rev Neurosci 7:268 –277.

43. Moller P, Husby R (2000): The initial prodrome in schizophrenia: Search-ing for naturalistic core dimensions of experience and behavior. Schizo-phr Bull 26:217–232.

44. Jackson HJ, McGorry PD, Dudgeon P (1995): Prodromal symptoms ofschizophrenia in first-episode psychosis: Prevalence and specificity.Compr Psychiatry 36:241–250.

45. Marjoram D, Miller P, McIntosh AM, Cunningham Owens DG, JohnstoneEC, Lawrie S (2006): A neuropsychological investigation into “theory ofmind” and enhanced risk of schizophrenia. Psychiatry Res 144:29 –37.

46. Harris JM, Moorhead TW, Miller P, McIntosh AM, Bonnici HM, Owens DG,et al. (2007): Increased prefrontal gyrification in a large high-risk cohortcharacterizes those who develop schizophrenia and reflects abnormalprefrontal development. Biol Psychiatry 62:722–729.

47. Han X, Jovicich J, Salat D, van der Kouwe A, Quinn B, Czanner S, et al.(2006): Reliability of MRI-derived measurements of human cerebral cor-tical thickness: The effects of field strength, scanner upgrade and man-ufacturer. Neuroimage 32:180 –194.

48. Yung AR, Phillips LJ, Yuen HP, McGorry PD (2004): Risk factors for psy-chosis in an ultra high-risk group: Psychopathology and clinical fea-tures. Schizophr Res 67:131–142.

49. Garner B, Pariante CM, Wood SJ, Velakoulis D, Phillips L, Soulsby B, et al.(2005): Pituitary volume predicts future transition to psychosis in indi-viduals at ultra-high risk of developing psychosis. Biol Psychiatry 58:417–

423.

www.sobp.org/journal