Linking Pharmacovigilance with Pharmacogenetics

Drug Safety 2004; 27 (15): 1171-1184LEADING ARTICLE 0114-5916/04/0015-1171/$31.00/0

2004 Adis Data Information BV. All rights reserved.

Linking Pharmacovigilancewith PharmacogeneticsDavid W.J. Clark,1,2 Emma Donnelly,2 David M. Coulter,1 Rebecca L. Roberts3 andMartin A. Kennedy3

1 The New Zealand Intensive Medicines Monitoring Programme (IMMP), New ZealandPharmacovigilance Centre, Department of Preventive and Social Medicine, Dunedin School ofMedicine, Dunedin, New Zealand

2 Department of Pharmacology and Toxicology, Otago School of Medical Sciences, Universityof Otago, Dunedin, New Zealand

3 University of Otago, Gene Structure and Function Laboratory, Christchurch School ofMedicine and Health Sciences, Christchurch, New Zealand

The ability to identify individuals who are susceptible to adverse drug reac-Abstracttions (ADRs) has the potential to reduce the personal and population costs ofdrug-related morbidity. Some individuals may show an increased susceptibility tocertain ADRs through genetic polymorphisms that alter their responses to variousdrugs.

We wished to establish a methodology that would be acceptable to members ofthe general population and that would enable estimation of the risks that specificgenetic factors confer on susceptibility to specific ADRs. Buccal swabs wereselected as a minimally invasive method to obtain cells for DNA extraction. Wewished to determine whether DNA of sufficient quantity and quality could beobtained to enable genotyping for two different polymorphic genes that code forenzymes that are widely involved in drug disposition.

This article describes a small pilot study of methodology developed in the NewZealand Intensive Medicines Monitoring Programme (IMMP) to link prescriptionevent monitoring (PEM) studies with pharmacogenetics. The methodologyinvolves a nested case-control study design to investigate whether patients withgenetic variants in P-glycoprotein (P-gp) and cytochrome P450 (CYP) 2C9 aremore susceptible to psychiatric or visual disturbances following cyclo-oxygenase-2 inhibitor use (ADR signals identified in the IMMP database) thanmatched control patients taking the medication without experiencing any ADRs.

It was concluded that the use of buccal swabs is acceptable to patients andprovides DNA of sufficient quantity and quality for genotyping. Although nodifferences in the distribution of genotypes in the case and control populationswere found in this small study, case-control studies investigating genetic risks for

1172 Clark et al.

ADRs using drug cohorts from PEM studies are possible, and there are severalareas where population-based studies of genetic risk factors for ADRs are needed.

Examples are discussed where research in large populations is required urgent-ly. These are: (i) genetic variations affecting P-gp function; (ii) variations affect-ing drugs metabolised by CYP2C9 and other polymorphic CYP enzymes; (iii)genetic variation in β-adrenergic receptors and adverse outcomes from β-adre-noceptor agonist therapy; and (iv) genetic variation in cardiac cell membranepotassium channels and their association with long QT syndromes and seriouscardiac dysrhythmias.

Such studies will help to identify factors that increase the risk of unwantedoutcomes from drug therapy. They will also help to establish in what circum-stances genotyping should be performed prior to commencing drug treatment andin tailoring drug treatment for individual patients.

The ability to identify individuals who are sus- 1. The New Zealand Intensive Medicinesceptible to adverse drug reactions (ADRs) has the Monitoring Programmepotential to reduce the personal and population costsof drug-related morbidity. Genetic polymorphisms The New Zealand IMMP began in 1977[1] andmay alter the pharmacokinetics or pharmacodynam- was designed to play a significant part in monitoringics of a given drug in ways that affect the response to for previously unrecognised ADRs.[2] Generally,the drug concerned and create an increased suscepti- medicines are selected for inclusion in the IMMP ifbility to certain ADRs. they are of a new class and are likely to be used

This article is based on a presentation made as widely. A near complete record is obtained of allpart of a symposium on ‘Risk Factors in Pharmaco- patients who have been prescribed these medicinesgenetics’ during the Annual Scientific Meeting of by recording all prescriptions dispensed throughoutthe International Society of Pharmacovigilance the whole monitoring period. Monitoring is usually(ISoP) held in Marrakesh, Morocco in October continued for 4–5 years, which is sufficient time to2003. The paper describes the methodology devel- accumulate a cohort of adequate size for epidemio-oped in the New Zealand Intensive Medicines Moni- logical investigations. The methodology utilises thetoring Programme (IMMP) aimed at linking technique now known as PEM. A similar methodol-pharmacoepidemiology with pharmacogenetics in ogy is used by the DSRU in England.[3] In theorder to estimate the degree to which specific genet- IMMP, prescription records for the monitoredic factors alter susceptibility to some ADRs, and the medicines are received from dispensing pharmacistsresults from a small pilot study are given. Future in community and hospital pharmacies throughoutdirections are discussed and examples considered of the country. The details recorded in each drug cohortimportant areas where genetic studies in prescrip- includes prescribers’ names and addresses; thetion event monitoring (PEM) cohorts are opportune names, addresses, dates of birth and sex of patientsand are underway using the IMMP cohorts and the who have been prescribed the medicines; details ofcohorts from the Drug Safety Research Unit each prescription (medicine, dose, date and quantity(DSRU) in England. dispensed) and indication for treatment.

2004 Adis Data Information BV. All rights reserved. Drug Safety 2004; 27 (15)

Linking Pharmacovigilance with Pharmacogenetics 1173

Adverse events are defined as any untoward ex- er, to date, most investigations have focused onperience whether or not the event is thought to be variability in single genes predicted to contribute todrug related.[4] These events include changes in a inter-individual differences in drug responses (i.e.pre-existing condition or abnormally changed labo- pharmacogenetics).ratory values and all deaths. The method is based on

Several pharmacogenetic studies have providedsending questionnaires to the prescribers at regular

evidence that in specific populations some subjectsintervals following receipt and data entry of new

have particular single nucleotide polymorphismsprescription details. Doctors are requested to record(SNPs) or a set of closely linked SNPs on a chromo-all events noted in the patient records from the timesome (haplotypes) that influence their response toof commencement of the drug or the latest prescrip-specific drugs.[6-12] It is thus important to identifytion, whichever is appropriate. These forms arewhether specific SNPs or haplotypes may lead tocomputer generated and include all prescribing andclinically important differences in drug responsepatient details and the first and most recent dispens-

ing dates for reference. Doctors may also record and ADRs. We have previously demonstrated thatevents on IMMP duplicate prescriptions or send pharmacogenetic studies are feasible in patients re-spontaneous reports on the standard ‘yellow’ card. ported with adverse reactions.[13] However, because

Because the IMMP databases include records of these earlier studies required physicians to collectmonitored medicines over a known time period, blood samples for DNA extraction, the number ofincluding those from patients who have not exper- patients willing to participate, and so subject them-ienced an adverse event, the incidence of events can selves to an invasive sampling procedure, was small.be calculated. The details of age, sex, dose, treat- In the pilot investigation described in this article, wement duration and indication for treatment allow the wished to assess whether buccal cell samples couldidentification of risk factors using cases and controls

yield enough DNA of sufficient quality for genotyp-from the individual cohorts. The study of risk factors

ing in larger population studies.can include genetic variation that may increase sus-

As part of the normal monitoring of rofecoxibceptibility to the adverse event.and celecoxib (cyclo-oxygenase [COX]-2 inhibi-

tors), the New Zealand IMMP has received a num-2. Investigations into Genetic Factors

ber of reports of visual[14] and psychiatric[15] distur-Affecting Drug Response

bances that appear after COX-2 inhibitor use and

resolve following withdrawal.[14] The strong evi-The field of research that uses large-scale, sys-

dence of causality from these reports and details oftematic genomic approaches to identify multiple

the types of visual and psychiatric disturbances as-drug response marker genes is known as

sociated with COX-2 inhibitor use have been pub-pharmacogenomics. It is likely that this approach

lished.[14,15] We hypothesised that there may be iden-will be increasingly used to identify genetic factorstifiable genetic factors that increased the risk of suchthat may influence an individual’s response to aevents in these patients. The pilot study was there-particular drug and thus assist in determining thefore designed to investigate whether selected genet-optimum drug regimen for that individual.ic polymorphisms contribute to the risk of develop-Pharmacogenomic approaches are also increasinglying visual or psychiatric disturbances following thebeing used by the drug industry to predict potentialuse of rofecoxib or celecoxib.problem areas during drug development.[5] Howev-


1174 Clark et al.

3. The Pilot Study The pilot case-control study was designed, in-volving patients from the IMMP who were currently

This study investigated the potential role of ge- taking rofecoxib or celecoxib, to test the hypothesisnetic polymorphisms in P-glycoprotein (P-gp) and that variant forms of P-gp and/or CYP2C9 are asso-cytochrome P450 (CYP) 2C9 in the visual and psy- ciated with the rare visual or psychiatric distur-chiatric disturbances experienced by some patients bances reported. Patients were genotyped for select-prescribed COX-2 inhibitors. ed SNPs in the genes encoding P-gp (ABCB1) and

National ethical approval was given for the use of CYP2C9 (CYP2C9). Previous experience usingthe IMMP data in this study and for the sampling DNA extracted from blood samples yielded a veryand genotyping of patients. low number of patients willing to participate.[13]

Therefore, it was considered essential for a study of3.1 Rationale this nature that another method of DNA collection

was adopted, and a simple buccal swab samplingFor most drug-related visual distubances and for method was selected.

psychiatric disturbances to occur, the drug con- The primary aims of the pilot study were: (i) tocerned must cross the blood-brain barrier. Several determine whether pharmacogenetic studies using avariants of the transporter protein P-gp show re- minimally invasive method of obtaining DNA areduced ability to extrude drugs and other substrates feasible, using a PEM database as a method offrom cells.[12,16] It is not yet known whether COX-2 recruiting cohorts of patients exposed to the sameinhibitors are substrates for P-gp, but if this is the drug, both with and without specific reactions; andcase, the presence of variant P-gp forms may lead to (ii) to determine whether a buccal swab samplingan increase in drug absorption or movement across method would enable the collection of sufficientthe blood-brain barrier with resulting high plasma or cells from inside the cheek for the extraction ofbrain concentrations; should the distribution of poor DNA and subsequent ABCB1 and CYP2C9 genotyp-P-gp transport efflux genotypes be greater in these ing.cases, this would highlight the need for in vitrostudies to confirm whether the COX-2 inhibitors are 3.2 Methodologysubstrates for P-gp. As prostacyclin plays an impor-tant role in the control of retinal blood flow[17] and 3.2.1 Recruitment of Patients

COX-2 inhibitors reduce prostacyclin synthesis in Subjects experiencing psychiatric or visual dis-the vascular endothelium,[18] an altered transport of turbances with celecoxib or rofecoxib that were ofcelecoxib or rofecoxib may contribute to the visual similar type to those previously reported by theor psychiatric disturbances. IMMP[14,15] were selected from the IMMP event

Patients experiencing the events may have inher- databases. Patients who were not experiencing anyited a gene that codes for variant CYP2C9 enzymes adverse events were selected from the IMMP co-that have been demonstrated to have a reduced abili- horts as controls. Prescribers were asked to indicatety to metabolise certain drugs, including celecox- whether patients (cases and controls) were takingib[19] and some other NSAIDs.[20,21] Therefore, it is specific listed medications (amiodarone, flucona-possible that genetic variability in proteins involved zole, paroxetine, probenecid, sertraline, trimetho-in the uptake or metabolism of COX-2 inhibitors prim and cotrimoxazole [trimethoprim/sul-may contribute to an increased risk of developing famethoxazole]) that might possibly interact with P-visual and/or psychiatric disturbances. gp or CYP2C9 substrates. None of the patients were



taking these medications during the times that results were sent to Dunedin, New Zealand, forcelecoxib or rofecoxib was prescribed. interpretation and analysis in collaboration with the

laboratory in Christchurch, New Zealand.The control patients were matched by date ofbirth (within 5 years), sex and geographical location.

3.2.3 Genotyping of SamplesWhere possible, two control patients who were pre-Genotyping for selected CYP2C9 (*1,*2 and *3)scribed the same COX-2 inhibitor by the same doc-

and ABCB1 (3435C>T) variants was performed us-tor were matched for each case. For each case, weing methods already established within the Christ-provided the patient’s doctor with the identificationchurch laboratory.[22,23] The method of buccal cellof three possible controls from the IMMP cohortssampling allows the recovery of sufficient cells fromand requested that he/she select the two most appro-which to isolate DNA. The DNA was extractedpriate. If the doctor could not identify any suitableusing the method described by Richards et al.,[24]

controls, we sought controls from other doctors lo-coded by number and date and stored at –20ºC untilcated in the same practice or general area. As someanalysed.patients had died and others were considered too

frail or ill to participate, a total of 37 cases and 913.2.4 Analysis of Results

controls were selected and invited to enrol.Differences in the frequency of variants between

cases and controls and between cases taking3.2.2 Obtaining the Buccal Cell Samplescelecoxib and rofecoxib were tested using relativeGeneral practitioners (GPs) were identified fromrisks with confidence intervals. Chi-squared teststhe IMMP database and contacted by mail with anwere also applied.explanation of the purpose of the investigation and

were asked for approval to approach their patient(s).Once the GP’s approval had been obtained, patients(37 cases and 91 controls) were contacted by mailwith an explanation of the study, invited to par-ticipate and asked to provide consent for their partic-ipation. The patients were informed that participa-tion would be at no cost to them and would involvemaking an appointment with their doctor so that aswab could be taken from inside their cheek toobtain a cell sample for DNA extraction.

Once informed consent had been received fromthe patients (consent was received from 17 cases and31 controls), the GPs were asked to return additionaldetails of the reported event, concurrent medicationsand conditions. They were then sent kits for thecollection of buccal cell samples (figure 1). Thesewere coded prior to being sent to the GPs so that thesubsequent genotyping was blind. The GPs wererequested to post the samples, using the reply paidcourier packs provided in the kit, to the geneticslaboratory for DNA extraction and genotyping. The

Fig. 1. Population genotyping kit sent to general practitioners (GPs)for obtaining buccal cell samples for extracting DNA. The kit con-tains a cytology brush, a 2mL screw top tube containing sodiumhydroxide solution, an instruction sheet for the GP and a prepaidcourier pack for sending the sample to the Gene Structure andFunction Laboratory, Christchurch School of Medicine and HealthSciences.


1176 Clark et al.

might be prescribed in the future, they should con-tact their GP. In addition to the letters containingtheir individual patients’ results, the GPs were alsosent a document with an explanation and list of othermedicines that might be affected by the geneticvariants of CYP2C9 and P-gp that were identifiedfor their patients.

3.3 Results

Of the 37 case and 91 control patients who wereapproached to participate in the study, genotypingwas carried out on 15 case and 24 control patients.Consent forms from the remaining nine patientswere received too late for the genotypes to be in-cluded in this study. The DNA obtained from buccalcells was of sufficient quality in all samples tested toenable ABCB1 and CYP2C9 genotyping (figure 2).In the small sample size genotyped, the distributionsof variant genotypes were similar between cases andcontrols and across drugs (table I).

3.4 Discussion

Genetic polymorphisms may contribute to inter-individual differences in drug response by alteringpharmacokinetics (e.g. where drug blood concentra-tions may be affected by variation in drug-metabolising enzymes or in transport proteins in-volved in drug uptake) or by altering pharmacody-namics (e.g. where variation in receptors, ion chan-nels or intracellular enzymes or in other cell/tissuecomponents may lead to altered drug response).

There are several studies that indicate that

1 2 3 4 M 5 6 7 8

bp

201138

63

bp

165135

30

M 1 2 3 4 5 6 7

a

b

Fig. 2. Genotyping assays for the detection of the ABCB1 polymor-phism 3435C>T and CYP2C9*3. (a) The region of the ABCB1 genecontaining the single nucleotide polymorphism 3435C>T was am-plified from patient buccal DNA using polymerase chain reaction(PCR). The resulting 201 base pair (bp) PCR products were di-gested with the restriction endonuclease DpnII, which only cuts theC allele to yield 138bp and 63bp fragments. Lanes 1, 6 and 8 arehomozygous for the T allele; lanes 3 and 4 are heterozygotes, andlanes 2, 5 and 7 are homozygous for the C allele. (b) Similarly, DNAsamples were screened for the CYP2C9*2 (representative gel notshown) and CYP2C9*3 alleles using PCR followed by digestion ofthe PCR products with different restriction endonucleases. Lanes 1,3 and 6 are heterozygous for CYP2C9*3; and lanes 2, 4, 5 and 7are homozygotes for the wildtype (CYP2C9*1) allele. NoCYP2C9*3 homozygotes were found in the pilot study. The PCRfragments for both assays were separated on 3% agarose andsized using a 25bp DNA ladder (M). polymorphisms in genes coding for proteins in-

volved in pharmacokinetic and pharmacodynamic3.2.5 Feedback to Patients and processes influence the effects of administeredGeneral Practitioners drugs, thereby altering susceptibility to ADRs.Letters were sent to the GPs and patients inform- However, most of these studies have compared

ing them of the individual results. Patient letters small, selected samples of the population, such asincluded a recommendation that if the patients those with or without particular variant genes (e.g.wished to discuss the wider implication of their studies comparing ‘extensive metaboliser’ withresults with respect to other medicines that they ‘poor metaboliser’ populations) and using short-



Table I. Results from genotyping cases taking celecoxib or rofecoxib and experiencing psychiatric or visual adverse effects and controlstaking celecoxib or rofecoxib without experiencing psychiatric or visual adverse effectsa

Event class Adverse effect Gene Case genotype Control genotype 1(genotype 2)

Celecoxib

Psychiatric Depression ABCB1 C/T C/T

CYP2C9 *1*1 *1*1

Depression ABCB1 C/T C/T

CYP2C *1*2 *1*1

Depression ABCB1 T/T C/T

CYP2C *1*1 *1*1

Hallucinations ABCB1 C/C T/T

CYP2C9 *1*2 *1*3

Suicidal ideation ABCB1 T/T C/C

CYP2C9 *1*1 *1*1

Visual Abnormal vision ABCB1 T/T T/T (C/T)

CYP2C9 *1*2 *1*1 (*1*2)

Visual field defect ABCB1 C/T T/T

CYP2C9 *1*1 *1*1

Rofecoxib

Psychiatric Confusion ABCB1 C/T C/T

CYP2C9 1/1 *1*2

Depression ABCB1 T/T C/T (C/C)

CYP2C9 *1*3 *1*3 (*1*1)

Restlessness ABCB1 C/T C/T (T/T)

CYP2C9 *1*1 *1*1 (*1*1)

Low concentration ABCB1 C/T C/T

CYP2C9 *1*1 *1*2

Delirium ABCB1 C/C T/T

CYP2C9 *1*3 *1*3

Anxiety ABCB1 T/T C/T

CYP2C9 *1*1 *1*3

Visual Blurred vision ABCB1 C/T C/T

CYP2C9 *1*1 *1*1

Teichopsia ABCB1 C/C C/C (C/T)

CYP2C9 *1*1 *1*1 (*1*2)

a Control genotypes 1 and 2 indicate matched controls. A further five controls were genotyped that had been matched to cases thatdid not consent to genotyping. The genotypes of these control patients were: C/C *1*1; C/T *1*1; CT/*1*2; TT/*1*1; and TT/*1*3. Inonly four cases were two matched controls per case able to be obtained for genotyping. Allelic variants that may lead to alteredplasma or CNS levels of substrates are shown in bold.

term dosage regimens.[25] Therefore, it should be In order to test the feasibility of such a populationhelpful to use epidemiological methods to estimate study, where the minimally invasive method of us-the risk of increased susceptibility to ADRs due to ing buccal swabs was used for obtaining DNA forpharmacogenetic factors, where comparisons are genotyping, we conducted a small pilot study, usingmade between matched populations of patients with case-control methodology. The pilot study investi-the selected adverse effect (cases) and those without gated whether selected SNPs within ABCB1 andany adverse effect (controls). CYP2C9, which might influence the pharmacokinet-


1178 Clark et al.

ics of the COX-2 inhibitors by (a) altering the pas- cases or controls. However, the number of casesage of drugs through biological membranes and/or patients (15 of the 37 approached [41%]) and con-(b) altering the metabolism of the drugs, were asso- trol patients (24 of the 91 approached [26%]) whociated with an increase in the susceptibility of pa- consented to participate and were genotyped wastients to psychiatric and visual disturbance. In the greatly in excess of those who agreed to participatecase and control samples genotyped (table I), sever- in our earlier genotyping study where DNA wasal variants were found that could potentially lead to extracted from peripheral blood.[13] A lower percent-altered substrate transport by P-gp or metabolism by age of controls than cases agreed to participate. ThisCYP2C9. may have been because of the lack of direct interest

due to not having experienced an adverse reaction,The results from this study suggest that geneticin spite of being supplied with the information thatvariability in P-gp or CYP2C9 may not contribute toknowledge of their genotypes might assist in thethe rare psychiatric or visual disturbances associatedchoice of other types of medications that could bewith celecoxib and rofecoxib. However, as the sam-needed in the future.ple size in this pilot study was small, there is a

possibility of type II error that would be minimised Because of time constraints, follow-up of pa-in a large population study. Furthermore, we only tients who had not returned consent forms was notexamined one of the many SNPs that have been carried out. However, as several patient consentdescribed in ABCB1 and we cannot rule out the forms were received following the closing date forpossibility that genetic variability in P-gp does im- the study, it is likely that telephone follow-up wouldpact on these COX-2 inhibitor-related ADRs. Two have improved the return rate. It is also possible thatrecent studies[26,27] investigating sequence diversity methods other than telephone follow-up could in-in ABCB1 have identified over 30 different SNP crease the return rate. In future investigations wehaplotypes of this gene, of which 14 were found to intend to approach patients by asking the prescribersinclude 3435C>T.[27] Results from these haplotyp- to co-sign and to forward the letters, describing theing studies strongly suggest that although screening intention of the study and inviting participation, tofor 3435C>T is a good starting point, to fully assess their patients. This was an Ethics Committee re-the potential impact of P-gp genetic variability on quirement for a future study and it may also improvedrug response it is necessary to screen patients for patient acceptance of the study objectives. Althoughhaplotypes rather than individual SNPs. we informed patients that doctors’ fees would be

paid by the IMMP, it may be possible that the offerDespite an apparent lack of association betweenof a financial reward to patients for participationselected P-gp and CYP2C9 variants and COX-2would increase enrolment. However, in some coun-inhibitor-induced psychiatric or visual disturbancestries, including New Zealand, the ethics of offering ain this study, we have demonstrated that a lessfinancial incentive to participate would be subject toinvasive method for obtaining DNA than using pe-intense scrutiny.ripheral blood samples is acceptable and that genetic

studies can be linked with pharmacovigilance data Linking pharmacogenetics and pharmacovigi-to provide information on genetic variation that may lance in epidemiological studies that use non-inva-alter susceptibility to ADRs. sive sampling methods has the potential to clarify

Because of health and other reasons, including the population risk of particular variant genes insome recent deaths, we were requested by the GPs increasing susceptibility to particular ADRs. PEMnot to approach several of the patients selected as databases, as in the DSRU in England and the



IMMP in New Zealand, have cohorts drawn from absorption, distribution and elimination of manythe general population who have been prescribed the drugs,[31,32] and this may contribute to serious ADRsmonitored drug(s), with information on outcomes of or interactions.[33,34] In addition, P-gp expression intherapy and other patient details. Patients with and endothelial cells of the CNS prevents the penetrationwithout particular adverse events can be identified of several drugs across the blood-brain barri-and their genotypes compared. We concur with the er.[12,35,36] The ABCB1 gene is highly polymorphicopinion expressed by Tucker[28] that “the challenge and many variant forms of P-gp show significantlynow is to assemble large, prospective, multidiscipli- reduced transporter function. One of the most com-nary, multicentre projects”. Case-control studies de- mon of these variant forms has a mutationsigned to assess the risk of selected variant SNPs (3435C>T) in exon 26 of the ABCB1 gene. Subjectsand haplotypes in predisposing towards adverse out- with the 3435T allele produce P-gp with alteredcomes from drug therapy in the ‘real world’ of substrate transport function compared with subjectsclinical practice will contribute to the evidence ob- with the 3435C allele. In some studies the 3435Ttained from future large-scale prospective studies. allele has been associated with increased P-gp func-

tion, whereas in other studies this variant appears to4. Future Directions encode a protein with impaired activity.[37-39] It is

possible that these discrepancies in relation toExamples of genes that are known to influence

3435T reflect differences in haplotypes. Our pilotdrug responses are provided in several reports.[28]

study sought to determine whether this knownWe have selected examples below where research

ABCB1 mutation is associated with increased sus-on the population risk of genetic variability in influ-

ceptibility to developing visual or psychiatric effectsencing therapeutic effect and the susceptibility to

in patients receiving celecoxib or rofecoxib.ADRs is underway or is needed urgently. Further

P-gp is known to transport many anticancerdetail is provided on genetic variability in P-gp anddrugs[12,16] and several other clinically importantCYP2C9: examples that may impact on pharma-drugs including some statins (atorvastatin, lovasta-cokinetics. Two further examples where genetic va-tin),[40] digoxin,[41,42] some calcium channel antago-riation may affect pharmacodynamics are also dis-nists (diltiazem, verapamil),[43] the angiotensin IIcussed.receptor antagonist losartan,[44,45] erythromycin,[40]

4.1 Pharmacokinetic Variation fexofenadine,[46] nortriptyline[47] and phenytoin.[33]

These examples have been selected because, insome cases, studies (referenced after drug names)4.1.1 Variations Affecting P-Glycoprotein Function

have associated P-gp genetic polymorphisms withP-gp is a transmembrane protein efflux pumpadverse effects or potentially important effects onencoded by the ABCB1 gene (formerly known aspharmacokinetics. Therefore, they are potentiallyMDR1). As discussed earlier, P-gp functions to ex-important candidate drugs for population studiestrude many drugs and other ‘foreign’ lipophilicaimed at assessing the association of variant P-gpcompounds from the cells back into the extracellularand ADRs. More complete lists of drugs that arespace, thereby preventing substrates from accumu-known substrates for P-gp may be obtained fromlating in critical organs such as the brain.[29,30] P-gpSchwab et al.[16] and Marzolini et al.[12] It is impor-has been isolated in endothelial cells of the gut,tant that the risks of genetic variation in the functionkidney, liver and other organs.[12] Because of this, it

has an important role in determining the extent of of P-gp on adverse outcomes of drugs, including


1180 Clark et al.

adverse interactions between drugs, are determined system ADRs associated with celecoxib, comparedin large populations. with patients identified as carriers for CYP2C9 al-

leles other than CYP2C9*3 (Layton D, DSRU, per-4.1.2 Variations Affecting Drugs Metabolised by sonal communication).Cytochrome P450 (CYP) 2C9 and OtherPolymorphic CYP Enzymes 4.2 Pharmacodynamic VariationThe CYP enzymes are a group of haemoproteins

responsible for catalysing the phase I metabolism of 4.2.1 Genetic Variation in β-Adrenergic Receptors:almost all drugs.[48] Over recent years emphasis has Association with Adverse Outcomes frombeen placed on the clinical implications of genetic β-Adrenoceptor Agonist Therapy

variability in CYP2D6 because this enzyme is in- β-Adrenoceptor agonists are the most commonlyvolved in the metabolism of a wide range of current- used medication for rapid bronchodilation duringly prescribed drugs.[49] acute asthma episodes. However, serious adverse

outcomes may occur with the use of β-agonists,CYP2C9 is one of the four known members of theincluding long-acting β-agonists.[58-60] These mayhuman 2C family and, although it has received lessinvolve the respiratory, cardiovascular or neuromus-attention than CYP2D6, it is one of the most abun-cular systems.dant CYPs found in the human liver.[50-55] CYP2C9

accounts for about 20% of the total liver CYP con- Of particular note are adverse respiratory out-tent and is responsible for the metabolic clearance of comes that occur following the use of long-acting β-about 10% of all drugs currently in use.[20,54] agonists, such as salmeterol and formoterol.[61]

When these drugs are used regularly, some patientsSeveral polymorphisms have been observed infind that they do not get their usual relief from short-the CYP2C9 enzyme.[54,56] Of these polymorphisms,acting β-agonist therapy. In this respect, there istwo alleles referred to as CYP2C9*2 and CYP2C9*3conclusive evidence that subsensitivity oflead to the reduced function of the enzyme. Muta-β2-adrenergic receptors develops when long-actingtions that alter amino acids in the substrate recogni-β-agonists are used regularly.[62] A number of SNPstion sites of CYP2C9 proteins (such as the *3 allele)in the β2-adrenoceptor (ADRB2) gene have beencan markedly reduce the metabolic ability of thedescribed.[63] To date, the most important of theseenzyme[54,56] and have clinically significant effectsappear to confer variation in both the acute re-on the pharmacokinetics of some drugs.[57] Althoughsponses to β2-receptor stimulation and in their pro-our small pilot study did not confirm that visual andpensity for desensitisation and downregulation.[63-65]psychiatric adverse reactions are associated with

reduced CYP2C9 metabolism, it remains possible Lima et al.[66] reported that bronchodilator re-that this is an important risk for susceptibility to sponses, resulting from acute administration ofadverse gastrointestinal events, including gastroin- salbutamol, are enhanced in patients with an argi-testinal bleeding and ulceration that may follow nine substitution in position 16 of the ADRB2 genecelecoxib and other NSAID use. The risk that genet- (Gly>Arg polymorphism). This raised the possibili-ic variants of CYP enzymes pose for NSAID and ty that the continuous use of short-acting β2-agonistsother drug use needs to be assessed using case- or of long-acting β-agonists such as salmeterol orcontrol studies in large populations. In the UK, the formoterol may result in downregulation of theDSRU is commencing a study to investigate wheth- β2-receptors. Israel et al.[8] and Taylor et al.[58] haveer patients identified as being hetero- or homozy- shown a consistent relationship between the pres-gous for CYP2C9*3 have a higher risk of alimentary ence of the Arg-16 allele and deterioration in lung



function during regular treatment with the short- Alteration in the rate of cardiac cell membrane repo-acting β-agonist salbutamol. In the study by Taylor larisation is reflected by a prolongation of the QTet al.,[58] the frequency of exacerbations of asthma interval, a common cause of serious dysrhythmiaswas increased in patients who were homozygous for including torsade de pointes.[69] The two most com-the Arg-16 allele. In addition, the benefits of regular mon types of inherited long QT syndromes (LQT1,salmeterol treatment were much less in homozygous LQT2) are due to variant genes that affect the func-Arg-16 patients than in the other genotypes. There- tion of cardiac cell membrane potassium channels:fore, there is strong evidence to support the hypothe- KCNQ1 (formerly known as KVLQT1) and KCNH2sis that adverse outcomes may be determined by (formerly known as HERG). A third, less common,ADRB2 genotype. Research is planned in the IMMP type of long QT syndrome (LQT3) is due to variantsto determine the risk for the development of adverse of a gene (SCN5A) that codes for sodium channelsoutcomes from specific genetic variations following (INa).[70]

treatment with long-acting β-agonists.K+ currents that pass through specialised K+

As in the case of ABCB1, although individual channels are the predominant outward currents inSNPs in ADRB2 may influence drug response, heart muscle involved in membrane repolarisation.haplotypes may be more clinically relevant. The most predominant K+ current responsible forHaplotype analysis of the ADRB2 gene and its pro- ventricular repolarisation is known as IK, of whichmoter region takes into account the possibility that there are two distinct components: IKr and IKs (rapidcombinations of SNPs may confer particular pheno- and slow). Several mutations exist in the genes thattypic effects.[67] In New Zealand, the most common code for the potassium channel proteins, KCNQ1functionally relevant haplotypes are 4/4, 2/2, 4/6, 2/ and KCNH2. Individuals carrying these mutant4 and 2/6[68] and the frequency of specific genes have reduced function in transporting K+ fromhaplotypes is sufficiently common that they can be the cells and are therefore susceptible to long QTidentified in modestly sized groups of subjects. interval and, when taking drugs that also potentiallyTherefore, genetic analyses in populations using inhibit IK currents, are at a high risk of QT prolonga-short- and long-acting β-agonists should provide tion and serious arrhythmias.[71] Lists of such drugsevidence relating adverse outcomes in patients tak- may be found in the literature[70,72] and Internet sitesing inhaled long-acting β-agonists to combination (e.g. http://www.torsades.org), which are updated ashaplotypes, as well as to the important SNPs. In new examples come to light. They include wellspite of advances in asthma management, asthma known examples, such as terfenadine, cisapride, an-morbidity and mortality remain high. In view of this tipsychotic medications and also drugs such as sibu-and because of concerns regarding serious adverse tramine and dexfenfluramine,[73] where investiga-outcomes following long-acting β-agonist use,[60]

tions in animals have provided an indication of thedetermination in populations of specific genetic va- potential to prolong QT interval.riation in β-receptors and the risks following β-

More research is needed to identify further drugsagonist use is needed urgently.associated with this serious problem and to measurethe risk of a long corrected QT interval (QTc) with4.2.2 Genetic Variation in Cardiac Cell Membranethese drugs in large populations. Currently, thePotassium Channels: Association with Long QT

Syndromes and Serious Cardiac Dysrhythmias DSRU, in collaboration with St. George’s HospitalSeveral drugs are associated with the occurrence Medical School (London, UK), is conducting such

of rare but potentially serious cardiac dysrhythmias. research (i.e. the DARE [Drug-Induced Arrhythmia


1182 Clark et al.

Risk Evaluation] study). This research will system- Acknowledgementsatically document and follow up incident cases of

A generous grant from the New Zealand Pharmacy Edu-drug-induced ventricular arrhythmia in England and cation and Research Foundation has enabled us to undertakecompare them with matched controls. In addition, the experimental work described in this article. We are also

grateful to Janelle Ashton for invaluable assistance with thisblood samples from case and control patients will bestudy and to Deborah Layton and Andrew Boshier for furtheranalysed for mutations and polymorphisms of theinformation on pharmacogenetic studies initiated in the Drugpotassium and sodium channel genes implicated inSafety Research Unit. Dr Kennedy is a Senior Research

the long QT and similar syndromes.[74] Results from Fellow of the Health Research Council of New Zealand. Drthis and similar research should greatly increase the Roberts is the recipient of a Health Sciences Career Develop-

ment Award of the University of Otago. The New Zealandknowledge of this rare, but potentially serious, car-Intensive Medicines Monitoring Programme (IMMP) hasdiac risk of some medicines.been supported financially by Medsafe of the NZ Ministry ofHealth and some pharmaceutical companies, in particular,

5. Conclusion Merck Research Laboratories, USA.

In spite of a number of association studies whereReferences

a link has been demonstrated between particular 1. McQueen EG. Intensified adverse drug reaction reportingpolymorphisms in genes and drug response, there scheme. N Z Med J 1977; 85 (589): 477

2. Coulter DM. Signal generation in the New Zealand Intensivehas been a lack of large population studies whereMedicines Monitoring Programme: a combined clinical and

other factors have been controlled for that may statistical approach. Drug Saf 2002; 25 (6): 433-93. Inman WH. Postmarketing surveillance of adverse drug reac-modify drug responses. The pilot study outlined in

tions in general practice. II: prescription-event monitoring atthis article demonstrates that case-control studies the University of Southampton. BMJ (Clin Res Ed) 1981; 282investigating the risk of genetic variation in the (6271): 1216-7

4. Coulter DM. The New Zealand Intensive Medicines Monitoringwider population, using databases held by centresProgramme in pro-active safety surveillance. Pharmacoepi-

undertaking PEM studies, are possible. PEM demiol Drug Saf 2000; 9: 273-805. Brazell C, Freeman A, Mosteller M. Maximizing the value ofdatabases, such as those held by the IMMP for New

medicines by including pharmacogenetic research in drug de-Zealand and the DSRU for England, are able to velopment and surveillance. Br J Clin Pharmacol 2002; 53 (3):identify cases and matched controls. Application of 224-31

6. Aithal GP, Day CP, Kesteven PJ, et al. Association ofbuccal cell sampling, as described in our pilot study,polymorphisms in the cytochrome P450 CYP2C9 with warfa-

should enhance the recruitment of patients to such rin dose requirement and risk of bleeding complications. Lan-cet 1999; 353 (9154): 717-9studies and enable the collection of sufficient DNA

7. Drysdale CM, McGraw DW, Stack CB, et al. Complex promot-for successful genotyping.er and coding region beta 2-adrenergic receptor haplotypesalter receptor expression and predict in vivo responsiveness.There are many areas of concern in pharmacoge-Proc Natl Acad Sci U S A 2000; 97 (19): 10483-8netics that are urgently in need of population studies,

8. Israel E, Drazen JM, Liggett SB, et al. The effect of polymorph-including those outlined in this article. Such studies isms of the beta (2)-adrenergic receptor on the response to

regular use of albuterol in asthma. Am J Respir Crit Care Medwill contribute to the evidence on whether or not2000; 162 (1): 75-80

selected genetic factors represent a risk in the popu- 9. Higashi MK, Veenstra DL, Kondo LM, et al. Association be-tween CYP2C9 genetic variants and anticoagulation-relatedlation of unwanted outcomes from drug therapy.outcomes during warfarin therapy. JAMA 2002; 287 (13):Furthermore, the studies will help to establish in1690-8

which circumstances genotyping should be per- 10. Evans WE, McLeod HL. Pharmacogenomics: drug disposition,drug targets, and side effects. N Engl J Med 2003; 348 (6):formed prior to commencing drug treatment and538-49

may also help in tailoring drug treatment for individ- 11. Weinshilboum R. Inheritance and drug response. N Engl J Medual patients. 2003; 348 (6): 529-37



12. Marzolini C, Paus E, Buclin T, et al. Polymorphisms in human 29. Brinkmann U, Roots I, Eichelbaum M. Pharmacogenetics of theMDR1 (P-glycoprotein): recent advances and clinical rele- human drug-transporter gene MDR1: impact of polymorph-vance. Clin Pharmacol Ther 2004; 75 (1): 13-33 isms on pharmacotherapy. Drug Discov Today 2001; 6 (16):

835-913. Clark D, Morgan A, Hananeia L, et al. Drug metabolism ge-30. Seelig A. A general pattern for substrate recognition by P-notypes and their association with adverse drug reactions in

glycoprotein. Eur J Biochem 1998; 251 (1-2): 252-61selected populations: a pilot study of methodology. Drug Saf2000; 9: 393-400 31. Benet LZ, Izumi T, Zhang Y, et al. Intestinal MDR transport

proteins and P-450 enzymes as barriers to oral drug delivery. J14. Coulter DM, Clark DW, Savage RL. Celecoxib, rofecoxib, andControl Release 1999; 62 (1-2): 25-31acute temporary visual impairment. BMJ 2003; 327 (7425):

32. Thiebaut F, Tsuruo T, Hamada H, et al. Cellular localization of1214-5the multidrug-resistance gene product P-glycoprotein in nor-15. Coulter DM. Acute psychiatric reactions with COX-2 inhibitors.mal human tissues. Proc Natl Acad Sci U S A 1987; 84 (21):Prescriber Update 2002; 23 (2): 217735-816. Schwab M, Eichelbaum M, Fromm MF. Genetic polymorph-

33. Kerb R, Aynacioglu AS, Brockmoller J, et al. The predictiveisms of the human MDR1 drug transporter. Annu Revvalue of MDR1, CYP2C9, and CYP2C19 polymorphisms forPharmacol Toxicol 2003; 43: 285-307phenytoin plasma levels. Pharmacogenomics J 2001; 1 (3):17. Haefliger IO, Meyer P, Flammer J, et al. The vascular endotheli-204-10um as a regulator of the ocular circulation: a new concept in

34. Greiner B, Eichelbaum M, Fritz P, et al. The role of intestinal P-ophthalmology? Surv Ophthalmol 1994; 39 (2): 123-32glycoprotein in the interaction of digoxin and rifampin. J Clin18. Clark DWJ, Layton D, Shakir SAW. Do some inhibitors ofInvest 1999; 104 (2): 147-53cyclooxygenase-2 (COX-2) increase the risk of

35. Cordon-Cardo C, O’Brien JP, Casals D, et al. Multidrug-resis-thromboembolic events? Linking pharmacology withtance gene (P-glycoprotein) is expressed by endothelial cells atpharmacoepidemiology. Drug Saf 2004; 27 (7): 427-56blood-brain barrier sites. Proc Natl Acad Sci U S A 1989; 86

19. Tang C, Shou M, Rushmore TH, et al. In-vitro metabolism of(2): 695-8

celecoxib, a cyclooxygenase-2 inhibitor, by allelic variant36. Schinkel AH, Wagenaar E, Mol CA, et al. P-glycoprotein in theforms of human liver microsomal cytochrome P450 2C9:

blood-brain barrier of mice influences the brain penetrationcorrelation with CYP2C9 genotype and in-vivoand pharmacological activity of many drugs. J Clin Investpharmacokinetics. Pharmacogenetics 2001; 11 (3): 223-351996; 97 (11): 2517-24

20. Miners JO, Birkett DJ. Cytochrome P4502C9: an enzyme of37. Sakaeda T, Nakamura T, Horinouchi M, et al. MDR1 genotype-

major importance in human drug metabolism. Br J Clinrelated pharmacokinetics of digoxin after single oral adminis-

Pharmacol 1998; 45 (6): 525-38tration in healthy Japanese subjects. Pharm Res 2001; 18 (10):

21. Brenner SS, Herrlinger C, Dilger K, et al. Influence of age and 1400-4cytochrome P450 2C9 genotype on the steady-state disposition

38. Kim RB, Leake BF, Choo EF, et al. Identification of functional-of diclofenac and celecoxib. Clin Pharmacokinet 2003; 42 (3):

ly variant MDR1 alleles among European Americans and283-92

African Americans. Clin Pharmacol Ther 2001; 70 (2): 189-9922. Martin JH, Begg EJ, Kennedy MA, et al. Is cytochrome P450 39. Fellay J, Marzolini C, Meaden ER, et al. Response to antire-

2C9 genotype associated with NSAID gastric ulceration? Br J troviral treatment in HIV-1-infected individuals with allelicClin Pharmacol 2001; 51 (6): 627-30 variants of the multidrug resistance transporter 1: a

23. Roberts RL, Joyce PR, Mulder RT, et al. A common P-glyco- pharmacogenetics study. Lancet 2002; 359 (9300): 30-6protein polymorphism is associated with nortriptyline-induced 40. Kim RB, Wandel C, Leake B, et al. Interrelationship betweenpostural hypotension in patients treated for major depression. substrates and inhibitors of human CYP3A and P-glycoprote-Pharmacogenomics J 2002; 2 (3): 191-6 in. Pharm Res 1999; 16 (3): 408-14

24. Richards B, Skoletsky J, Shuber AP, et al. Multiplex PCR 41. Hoffmeyer S, Burk O, von Richter O, et al. Functionalamplification from the CFTR gene using DNA prepared from polymorphisms of the human multidrug-resistance gene: mul-buccal brushes/swabs. Hum Mol Genet 1993; 2 (2): 159-63 tiple sequence variations and correlation of one allele with P-

25. Daly AK, Day CP. Candidate gene case-control association glycoprotein expression and activity in vivo. Proc Natl Acadstudies: advantages and potential pitfalls. Br J Clin Pharmacol Sci U S A 2000; 97 (7): 3473-82001; 52 (5): 489-99 42. Johne A, Kopke K, Gerloff T, et al. Modulation of steady-state

26. Tang K, Ngoi SM, Gwee PC, et al. Distinct haplotype profiles kinetics of digoxin by haplotypes of the P-glycoprotein MDR1and strong linkage disequilibrium at the MDR1 multidrug gene. Clin Pharmacol Ther 2002; 72 (5): 584-94transporter gene locus in three ethnic Asian populations. 43. Morita N, Yasumori T, Nakayama K. Human MDR1 polymor-Pharmacogenetics 2002; 12 (6): 437-50 phism: G2677T/A and C3435T have no effect on MDR1

27. Kroetz DL, Pauli-Magnus C, Hodges LM, et al. Sequence transport activities. Biochem Pharmacol 2003; 65 (11): 1843-diversity and haplotype structure in the human ABCB1 52(MDR1, multidrug resistance transporter) gene. 44. Sekino K, Kubota T, Okada Y, et al. Effect of the singlePharmacogenetics 2003; 13 (8): 481-94 CYP2C9*3 allele on pharmacokinetics and pharmacodynam-

28. Tucker G. Pharmacogenetics: expectations and reality. BMJ ics of losartan in healthy Japanese subjects. Eur J Clin2004; 329 (7456): 4-6 Pharmacol 2003; 59 (8-9): 589-92


1184 Clark et al.

45. Soldner A, Benet LZ, Mutschler E, et al. Active transport of the 61. Lipworth BJ, Aziz I. Bronchodilator response to albuterol afterangiotensin-II antagonist losartan and its main metabolite EXP regular formoterol and effects of acute corticosteroid adminis-3174 across MDCK-MDR1 and caco-2 cell monolayers. Br J tration. Chest 2000; 117 (1): 156-62Pharmacol 2000; 129 (6): 1235-43

62. Lipworth BJ. Airway subsensitivity with long-acting beta 2-46. Cvetkovic M, Leake B, Fromm MF, et al. OATP and P-glyco-

agonists: is there cause for concern? Drug Saf 1997; 16 (5):protein transporters mediate the cellular uptake and excretion

295-308of fexofenadine. Drug Metab Dispos 1999; 27 (8): 866-7163. Green SA, Turki J, Bejarano P, et al. Influence of beta 2-47. Uhr M, Steckler T, Yassouridis A, et al. Penetration of ami-

adrenergic receptor genotypes on signal transduction in humantriptyline, but not of fluoxetine, into brain is enhanced in micewith blood-brain barrier deficiency due to mdr1a P-glycopro- airway smooth muscle cells. Am J Respir Cell Mol Biol 1995;tein gene disruption. Neuropsychopharmacology 2000; 22 (4): 13 (1): 25-33380-7

64. Liggett SB. Polymorphisms of the beta2-adrenergic receptor48. Nebert DW, Gonzalez FJ. P450 genes: structure, evolution, and and asthma. Am J Respir Crit Care Med 1997; 156 (4 Pt 2):

regulation. Annu Rev Biochem 1987; 56: 945-93S156-62

49. Bertilsson L, Dahl ML, Dalen P, et al. Molecular genetics of65. Liggett SB. Genetics of beta 2-adrenergic receptor variants inCYP2D6: clinical relevance with focus on psychotropic drugs.

asthma. Clin Exp Allergy 1995; 25 Suppl. 2: 89-94Br J Clin Pharmacol 2002; 53 (2): 111-2266. Lima JJ, Thomason DB, Mohamed MH, et al. Impact of genetic50. Goldstein JA, de Morais SM. Biochemistry and molecular biol-

polymorphisms of the beta2-adrenergic receptor on albuterology of the human CYP2C subfamily. Pharmacogenetics 1994;4 (6): 285-99 bronchodilator pharmacodynamics. Clin Pharmacol Ther

1999; 65 (5): 519-2551. Lasker JM, Wester MR, Aramsombatdee E, et al. Characteriza-tion of CYP2C19 and CYP2C9 from human liver: respective 67. Littlejohn MD, Taylor DR, Miller AL, et al. Determination ofroles in microsomal tolbutamide, S-mephenytoin, and

beta2-adrenergic receptor (ADRB2) haplotypes by a mul-omeprazole hydroxylations. Arch Biochem Biophys 1998; 353

tiplexed polymerase chain reaction assay [abstract]. Hum(1): 16-28Mutat 2002; 20 (6): 479

52. Miners J. CYP2C9 polymorphism: impact on tolbutamide68. Taylor DR, Kennedy MA. Genetic variation of the beta (2)-pharmacokinetics and response. Pharmacogenetics 2002; 12

adrenoceptor: its functional and clinical importance in(2): 91-2bronchial asthma. Am J Pharmacogenomics 2001; 1 (3): 165-53. McCrea JB, Cribb A, Rushmore T, et al. Phenotypic and geno-74typic investigations of a healthy volunteer deficient in the

conversion of losartan to its active metabolite E-3174. Clin 69. Roden DM. Drug-induced prolongation of the QT interval. NPharmacol Ther 1999; 65 (3): 348-52

Engl J Med 2004; 350 (10): 1013-2254. Goldstein JA. Clinical relevance of genetic polymorphisms in

70. Escande D. Pharmacogenetics of cardiac K (+) channels. Eur Jthe human CYP2C subfamily. Br J Clin Pharmacol 2001; 52

Pharmacol 2000; 410 (2-3): 281-7(4): 349-5571. Guzey C, Spigset O. Genotyping of drug targets: a method to55. Gill HJ, Tjia JF, Kitteringham NR, et al. The effect of genetic

predict adverse drug reactions? Drug Saf 2002; 25 (8): 553-60polymorphisms in CYP2C9 on sulphamethoxazole N-hydrox-ylation. Pharmacogenetics 1999; 9 (1): 43-53 72. Haddad PM, Anderson IM. Antipsychotic-related QTc prolon-

56. Stubbins MJ, Harries LW, Smith G, et al. Genetic analysis of the gation, torsade de pointes and sudden death. Drugs 2002; 62human cytochrome P450 CYP2C9 locus. Pharmacogenetics (11): 1649-711996; 6 (5): 429-39

73. Hu S, Wang S, Gibson J, et al. Inhibition of delayed rectifier K+57. Furuya H, Fernandez-Salguero P, Gregory W, et al. Genetic

channels by dexfenfluramine (Redux). J Pharmacol Exp Therpolymorphism of CYP2C9 and its effect on warfarin mainte-

1998; 287 (2): 480-6nance dose requirement in patients undergoing anticoagulationtherapy. Pharmacogenetics 1995; 5 (6): 389-92 74. Davies M, Behr E, Carter N, et al. Methodology of the Drug-

Induced Arrythmia Risk Evaluation (DARE) study [abstract].58. Taylor DR, Drazen JM, Herbison GP, et al. Asthma exacerba-Pharmacoepidemiol Drug Saf 2003; 12: S115tions during long term beta agonist use: influence of beta (2)

adrenoceptor polymorphism. Thorax 2000; 55 (9): 762-7

59. Green SA, Rathz DA, Schuster AJ, et al. The Ile164 beta (2)-Correspondence and offprints: Dr David W.J. Clark, Depart-adrenoceptor polymorphism alters salmeterol exosite binding

and conventional agonist coupling to G (s). Eur J Pharmacol ment of Preventive and Social Medicine, Otago Medical2001; 421 (3): 141-7

School, NZ Pharmacovigilance Group, P.O. Box 913, Dune-60. Mann M, Chowdhury B, Sullivan E, et al. Serious asthma

din, New Zealand.exacerbations in asthmatics treated with high-dose formoterol.E-mail: [email protected] 2003; 124 (1): 70-4


Linking Pharmacovigilance with Pharmacogenetics

Documents

Transcript of Linking Pharmacovigilance with Pharmacogenetics