Pharmacogenetics of hypersensitivity drug reactions

(1)

Therapie(2017)72,231—243

Availableonlineat

ScienceDirect

www.sciencedirect.com

PHARMACOGENETICS

Pharmacogenetics

of

hypersensitivity

drug

reactions

Simone

Negrini

a,∗

,

Laurent

Becquemont

b

a_Department_of_internal_medicine,_clinical_immunology_unit,_centre_of_excellence_for

biomedicalresearch,universityofGenoa,16132Genoa,Italy

b_Pharmacology_department,_faculty_of_medicine_Paris-Sud,_UMR_1184,_CEA,_DSV/iMETI,

divisionofimmuno-virology,IDMIT,Insermcenterforimmunologyofviralinfectionsand

autoimmunediseases,hôpitalBicêtre,universityParisSud,Assistancepublique—Hôpitauxde

Paris,94275LeKremlin-Bicêtre,France

Received5July2016;accepted2September2016 Availableonline3January2017

KEYWORDS Hypersensitivitydrug reactions; HLA; Abacavir; Carbamazepine; Allopurinol

Summary Adversedrugreactionsareasignificantcauseofmorbidityandmortalityand rep-resentamajorburdenonthehealthcaresystem.Someofthosereactionsareimmunologically mediated (hypersensitivity reactions) and can beclinically subdivided into two categories: immediatereactions(IgE-related)anddelayedreactions(T-cell-mediated).Delayed hypersen-sitivityreactions includeboth systemicsyndromes and organ-specific toxicities andcan be triggered by a wide range of chemically diverse drugs. Recent studies have demonstrated astrong genetic associationbetween humanleukocyte antigen alleles andsusceptibilityto delayeddrughypersensitivity.Mostnotableexamplesincludehumanleukocyteantigen (HLA)-B*57:01alleleandabacavirhypersensitivitysyndromeorHLA-B*15:02andHLA-B*58:01alleles relatedtoseverecutaneousreactionsinducedbycarbamazepineandallopurinol,respectively. Thisreview aimstoexploreourcurrent understandinginthefield ofpharmacogenomicsof HLA-associateddrughypersensitivitiesanditstranslationintoclinicalpracticeforpredicting adversedrugreactions.

Abbreviations

ADRs adversedrugreactions APC antigen-presentingcell

DIHS drug-inducedhypersensitivitysyndrome

∗_{Corresponding}_author._Department_of_internal_medicine,

univer-sityofGenoa,vialeBenedettoXV,6,16132Genoa,Italy.

E-mailaddress:negrini@unige.it(S.Negrini).

EMA EuropeanMedicinesAgency FDA FoodandDrugAdministration HDRs hypersensitivitydrugreactions HIV humanimmunodeﬁciencyvirus HLA humanleukocyteantigen MPE maculopapulareruption NPV negativepredictivevalue PPV positivepredictivevalue SJS Stevens-Johnsonsyndrome TCR T-cellreceptor

http://dx.doi.org/10.1016/j.therap.2016.12.009

(2)

TEN toxicepidermalnecrolysis WHO WorldHealthOrganization

Hypersensitivity

drug

reactions

Adverse drug reactions (ADRs) are deﬁned by the World Health Organization (WHO) as an ‘‘unintended, noxious responsetoadrugthatoccursatadoseusuallyprescribed for humanpatients’’ [1].ADRs arerecognizedasa major healthproblem, intheUnited Statestheyareresponsible of6—7% of hospitalizationcasesand representthe fourth tosixthleadingcauseofdeathaccountingfor100,000fatal casesannually[2].InEurope,itisestimatedthat5%ofall hospitaladmissionand197,000deathsperyeararecaused by ADRs [3]. Along withthe mortalityand morbidity bur-den,itisestimatedthatthetotalcosttosocietyofADRsis $177billionandD 79billionintheUSAandEurope, respec-tively[3,4].Moreover,ADRsremainahugecostburdenfor pharmaceuticalindustryrepresentingthemaincauseofdrug withdrawalfromthemarket[5].

The historical pharmacologic classification of ADRs by Rawlins and Thompson classifies these into two main categories [6]. Type A reactions are predictable, dose-dependent,andrelatedtopharmacologicpropertiesofthe drug,and TypeBreactions, accounting for approximately 10% of all ADRs, are unpredictable, not dose-dependent andcorrespond tohypersensitivity drug reactions (HDRs). Occurring only in susceptible individuals Type BADRs are sometimestermed‘‘idiosyncratic’’.Since,terminologyused todefineHDRsisoftenconfusingduetoalackofcommonly accepteddefinitions,theEuropeanAcademyofAllergology and Clinical Immunology published a statement paper in ordertostandardizethenomenclatureofallergicdiseases, alsoincludingdrugallergy[7].Therevisednomenclaturehas beenfurtheradoptedbytheWorldAllergyOrganization[8]. Theproposedclassificationdefineas‘‘drugallergy’’those HDRs in which immunologic mechanisms may be demon-strated, while HDRs with symptoms and signs similar to realallergiesbutnottriggeredbyaspecificimmunological mechanismarereferredtoas‘‘non-allergicdrug hypersen-sitivity’’(e.g. non-specific histamine release, arachidonic acid pathway activation, bradykinin pathway alteration, etc.).Withregardtotheonsetofsymptoms,drugallergies may be further classified as immediate or delayed, sug-gestingalsothe immunologicalmechanism underlying the reaction[7,8].

Immunological

basis

of

hypersensitivity

drug

reactions

Immediate reactions occur immediately after drug expo-sure,areIgE-dependent,andderivefrommastcell degranu-lationandreleaseofpro-inflammatorymediators. Clinical manifestations include erythema, urticaria, angioedema, bronchospasm,andanaphylactic shock.Delayed hypersen-sitivity reactions generally occur days, or even weeks, afterdrug exposure andare mediatedby antigen-specific T lymphocytes. Examples of delayed-type HDRs include bothsystemicsyndromes (e.g.drugrash witheosinophilia andsystemicsymptoms)andorgan-specifictoxicities(e.g. hepatitis,pneumonitis,etc.).Delayed-typehypersensitivity

reactions correspondtotypeIVhypersensitivity according totheclassiﬁcation systemof Gell andCoombs[9].Since T-cellscanorchestratedifferentimmuneresponses result-ingin differentclinical entities,delayed reactions canbe further classiﬁed intoIVa (Th1 cells), IVb (Th2 cells), IVc (cytotoxicT-cell),andIVd(neutrophils)[10].

Since drugs are generally too small to stimulate an immune response(molecular weights<1000Daltons), sev-eral,non-mutually exclusive, modelshave been proposed toexplainhowdrugscanbecomeimmunogenicandtrigger T-cells:the‘‘hapten/pro-haptenmodel’’,the‘‘Pi (pharma-cologicinteractionwithimmunereceptors)concept’’,the ‘‘alteredpeptiderepertoirehypothesis’’ andthe‘‘danger signal’’theory.The ‘‘hapten/pro-haptenmodel’’proposes that a chemically reactive drug, acting as hapten, can bind covalently to self-proteins (carriers) creating fully antigeniccomplexes. Theseneo-antigensareprocessedby an antigen-presenting cell (APC), loaded onto the HLA molecules and then presentedto appropriateT-cells.The pro-haptenisachemicallynon-reactivedrugthatbecomes reactive upon metabolism[11]. According to ‘‘pi-concept hypothesis’’,achemicallyinertdrug,unabletocovalently bind to proteins, is ableto structurally ‘‘ﬁt’’ intothe T-cell receptor. This interaction needs neither metabolism nor antigen processing. The initial stimulation of the T-cell receptor (TCR) is further supplemented by TCR-HLA interaction andprobably must take place in a context of hyper-reactiveT-cellswithalowthresholdlevelof activa-tion[12—14].The‘‘alteredpeptiderepertoirehypothesis’’ proposesthatthedrugbindstheantigen-bindinggrooveof HLAthusmodifying theantigen-bindingcleftthus altering the repertoire of self-peptides that are bound and pre-sented.SinceT-cellsareeducatedtobetoleranttoaspeciﬁc pool of peptidesduring thymic maturation, the presenta-tionoftheseneo-self-peptidesmayinduceT-cellactivation. Recentobservationsprovidestrongevidencethatthismodel isimplicatedinHDRsrelatedtoabacavirandcarbamazepine

[15].Lasthypothesisproposesthatthedrugitselfor conco-mitantsituations(e.g.viralinfections)canprovide‘‘danger signals’’ capable of upregulating costimulatory molecules andcytokinesininnateimmunecells, thusfacilitatingthe immuneactivation[11].

The human leukocyte antigen (HLA) molecules plays a crucial role in T-cells activation by presenting processed antigens to the T-cell receptor expressed on T lympho-cytes. Broadlyspeaking, therearetwomain types ofHLA molecules: the HLAclass I molecules, expressed on most nucleatedcells,andtheHLAclassII molecules,expressed byAPCs, suchasmonocytes ordendriticcells. HLAclassI moleculesareencodedbythreelociknownasHLA-A, HLA-B,andHLA-C;HLAclassIImoleculesareencodedbythree lociknownasHLA-DR,HLA-DQ,andHLA-DP.HLAclassIand classIImoleculesinitiatetheadaptiveimmuneresponseby presenting antigenstoCD8+(cytotoxic) andCD4+(helper) T-cells.BecausetheHLAmoleculesneedtopresentanhuge varietyof‘‘self’’and‘‘non-self’’peptides,theHLAgenes arebothnumerousandextremelypolymorphic.Takinginto accountthecrucialroleofHLAinimmuneresponseitisnot surprisingthatparticularHLAalleleshavebeenassociated withsusceptibilitytodiseasesinwhichtheimmunesystemis consideredtheprincipalmediator,likeinfectiousor autoim-munedisorders[16,17].Inthesameway,certainHLAalleles

(3)

Pharmacogeneticsofhypersensitivitydrugreactions 233 havebeenassociatedwithanincreasedriskofdelayedHDRs.

TheevidencethatcertainHLAallelesmayincreasetherisk of HDRs has promptedthe developmentof screening test strategiesandlabelingchangestodruginformationsheets. To date, the best-characterized HLA-HDRs associations includeabacavir,carbamazepine,andallopurinol.

Speciﬁc

examples

of

HLA-related

hypersensitivity

drug

reactions

Abacavir

Abacavir is a nucleoside analog reverse-transcriptase inhibitorusedaspartofcombinationtherapyforthe treat-ment of human immunodeﬁciency virus (HIV) infection. Abacaviris generallywell tolerated,but 5—8%ofpatients maydevelop an hypersensitivity reaction generallywithin the ﬁrst 6 weeks of treatment [18]. Abacavir hypersen-sitivity reaction is clinically characterized by fever, rash, constitutional, gastrointestinal and respiratory symptoms. Symptoms worsen with the continuation of abacavir and canbelife-threateningifthedrugisre-administeredafter discontinuation[19].EvenifAbacavirhypersensitivitywas reported in the pre-marketing phase of its development, only in 2002, twoindependent groups described a strong associationbetweentheHLA-B*57:01alleleand susceptibil-itytodevelopabacavirhypersensitivitysyndrome[20,21].

This association was successivelyreportedby different groups [22—25], but it was definitively confirmed in 2008 withtheresultsofthelargestrandomizedclinicaltrial per-formed todate inpharmacogenetics: the PREDICT-1trial. In this study, 1956 patients from 19 countries were ran-domized to either HLA-B*5701 geneticscreening (carriers ofHLA-B*5701didnotreceiveabacavir)ortothestandard of care (noscreening andall patientsreceived abacavir). Geneticscreening virtuallyeliminatedallimmunologically confirmed HDRs (0% in the screened group versus 2.7% in the control group), thus supporting a great translational potential of HLA-B*57:01 as screening test for abacavir-relatedHDR[26].Nevertheless,thegeneralisability ofthe PREDICT-1resultstoallraceswaslimitedbyitsprevalent Caucasian population (84% of the enrolled patients were Caucasian).TheSHAPEtrial,acase-controlstudyincluding similarpercentagesofblackandwhitepatients,revealeda 100%negativepredictivevalue (NPV)ofHLA-B*57:01 test-ingfor abacavirhypersensitivity both for blackand white patients,thusconfirmingthevalueofthescreeningacross different ethnicities [27]. A recent meta-analysis further confirmed that HLA-B*57:01 carriage is significantly asso-ciated withabacavir-induced hypersensitivity reactions in Whites,Blacks,andHispanics[28].TheNPVofHLA-B*57:01 screeningis100%(whichmeansthatHLA-B*57:01negative individuals will not develop hypersensitivity) making this testhighlyvaluableinpredictingtheriskofabacavir-related HDR. Nevertheless, the positive predictive value (PPV)of HLA-B*57:01genetictestingisestimatedtobearound50% implyingthat aroundhalfof allpatientsfound positiveat screeningactuallywilldeveloptheHDRifexposedto aba-cavir.HighNPV andlow PPVseems tobea characteristic featureofHLAscreeningtests(Table1)[29—33]suggesting

thatthesemarkersarenecessarybutnotsufﬁcientforADR onset,andothergenesand/orenvironmentalfactorsshould beinvolved.

Interestingly, HLA-B*57:01 has also been associated to an increased risk of drug-induced liver injury in patients treatedwith theantibiotic ﬂucloxacillin [34].The associ-ation of HLA-B*57:01 with both abacavir hypersensitivity andﬂucloxacillin-inducedhepatotoxicityindicatesthatthe same HLA allele lead to completely different reactions triggeredbychemicallyunrelated drugs.Inaddition, HLA-B*57:01hasbeenassociatedwithslowornon-progressionof HIVinfection[35—37].Takentogether,thesedataseemto suggestthatHLA-B*57:01isrelatedwithan‘‘hyperactive’’ immuneresponsethat,ononehandmayincreasetheriskto develophypersensitivityreactions,but,ontheotherhand, conferresistancetoinfections.

In light of the large number of evidences correlat-ing HLA-B*57:01 and abacavir-induced HDR, the genetic screeningprior toabacavirtreatment is recommendedby Food and Drug Administration (FDA), European Medicines Agency(EMA),andvirtuallyallinternationalHIVtreatment guidelines[32,33].Currently,HLA-B*57:01isoneofmost,if notthemost, commonly usedpharmacogeneticmarkerin clinicalpractice.

Carbamazepine

Stevens-Johnson syndrome (SJS) and toxic epidermal necrolysis (TEN, also known as Lyell’s syndrome) are severemucocutaneousreactionscharacterizedbyextensive detachmentoftheskin.Theseskinreactionsareconsidered tobevariants ofthe samediseaseandtheyareclassiﬁed accordingtothedegreeofskininvolvementwhichis<10% inSJS,and>30%inTEN.Askindetachmentof10to30%of bodysurfaceareaisnamedSJS/TENoverlap[38].SJS/TEN arerarediseaseswithanincidenceof2—6cases/millionper yearbutbothareassociatedwithhighmorbidityand mor-tality(5%and50%mortalityforSJS forTEN,respectively)

[38,39].Fromapathophysiologicalpointsofview,SJS/TEN are attributable to autoreactive CD8+ cytotoxic lympho-cytes that cause keratinocytesdeath throughthe release ofcytotoxicmolecules,suchassoluble-FasL,granzyme B, perforin,granulysinandpro-apoptoticcytokines[40].Inthe vastmajorityof casesSJS/TENaretriggered bydrugsand severalmedicationsareconsideredathighriskofinducing SJS/TEN[41].

Carbamazepine is an aromatic amine anticonvulsant widely employed to treat epilepsy, bipolar disorder, and trigeminalneuralgia.Carbamazepineisgenerallysafe,but has been associated with maculopapular eruption (MPE), drugrashwitheosinophiliaandsystemicsymptoms(DRESS) syndrome (also known as DIHS [drug-induced hypersensi-tivitysyndrome]),and morerarely SJS/TEN[42]. In2004, Chung et al. reported a very strong association between carbamazepine-inducedSJS/TENandtheHLA-B*15:02allele inHan Chinesepatients from Taiwan. Theyobserved that all patients with carbamazepine-induced SJS/TEN carried the HLA-B*15:02 allele as compared to only 3% of drug-tolerantsubjectsand8.6% ofthegeneralpopulation[43]. Further studies have replicated this association in differ-entAsianareasincludingChina,Thailand,Malaysia,India, Vietnam,and Cambodia [44—53],and threerecent

(4)

meta-S.

Negrini,

L.

Becquemont

Table1 Mostrelevantgenotype-basedpharmacogenomicsrelationshipsbetweenHLAanddrughypersensitivity[29—33].

Drug Levelof

evidence

[29]

Allele Clinicalmanifestation Population RiskofADR Clinicalimplementation[29,32,33]

FrenchNationalPharmacogenetics network(RNPGx)recommendations Abacavir 1A HLA-B*57:01 DRESS Mixedpopulation Increased FDA:boxedwarning,test

recommended,EMA:testrequired, PMDA:labelinformation,HCSC: testrecommended,CPIC: alternativedrugforpositive patients,testrecommended DPWG:alternativedrugforpositive patients,virtuallyallHIV

TreatmentGuidelines:test required/recommended PPV=55%,NPV=100%forpatch testconﬁrmed,NNTtoprevent ‘‘1’’=32

RNPGxreco:essential Acetazolamide 4 HLA-B*59:01 SJS/TEN Asian(Korean) Increased RNPGxrecommendation:no

indicationinroutinecare Acetaminophen 3 HLA-DQB1*02:02 SJS/TEN Caucasian(Italian) Increased RNPGxreco:noindicationin

routinecare

Allopurinol 1A HLA-B*58:01 SJS/TEN MixedPopulation Increased Europeannationalcompetent authorities:warning(nogenotyping recommendations),ACR:

alternativedrugforHLA-B*58:01 positivepatients,CPIC:alternative drugforHLA-B*58:01positive patients

HLA-B*58:01test

PPV=3%,NPV=100%inHan Chinese,NNTtoprevent‘‘1’’=250 RNPGxreco:HLA-B*58:01

potentiallyusefulnamelyinAsian ancestrypopulations(HanChinese, ThaïandTaiwanese)

2B HLA-A*33:03 DrugHypersensitivity, SJS/TEN

Mixedpopulation Increased

3 HLA-A*02:01 SJS/TEN Asian Decreased

3 HLA-B*48:01 DrugHypersensitivity Asian Increased(not inallstudies)

3 HLA-C*03:02 Rash Asian(Korean) Increased

3 HLA-C*08:01 DrugHypersensitivity Mixedpopulation Increased(not inallstudies) 3 HLA-DQB1*05:02 SJS/TEN Caucasian(Italian) Increased

3 HLA-DR9;HLA-DR14 MPE Asian Increased

3 HLA-DRB1*03:01(inLD withHLA-B*58:01in somepopulations)

SJS/TEN Asian(HanChinese) Increased

3 HLA-DRB1*15:02, HLA-DRB1*13:02

SJS/TEN Caucasian(Italian) Increased

Aspirin 2B HLA-DPB1*03:01 Asthma Mixedpopulation Increased N/A

RNPGxreco:HLA-B*58:01:no indicationinroutinecare 3 HLA-DPB1*04:01 Asthma Mixedpopulation Decreased

(5)

P harmacogenetics of hypersensitivity drug reactions 235 Table1 (Continued) Drug Levelof evidence [29]

FrenchNationalPharmacogenetics network(RNPGx)recommendations Azathioprine(or

mercaptopurine)

3 HLA-DQA1*02:01, HLA-DRB1*07:01

Pancreatitis European Increased N/A

RNPGxreco:noindicationin routinecare

Carbamazepine 1A HLA-B*15:02 SJS/TEN MainlyAsiana _Increased _HLA-B*15:02

FDA:BoxedWarning,testrequired (patientswithAsianancestry), Europeannationalcompetent authorities:warning,test

recommended(patientswithAsian ancestry),PMDA:labelinformation, HCSC:testrecommended,CPNDS: drugcontraindicatedforpositive patients

Asian:PPV=1.8%,NPV=100%,NNT toprevent‘‘1’’=461

HLA-A*31:01

FDA:warning(nogenotyping recommendations),European nationalcompetentauthorities: warning(nogenotyping

recommendations),HCSC:test recommended,CPNDS:drug contraindicatedpositivepatients Japanese:PPV=12%,NPV=99%, NNTtoprevent‘‘1’’=67 Caucasian:PPV=43%,NPV=92%, NNTtoprevent‘‘1’’=47

RNPGxreco:HLA-B*15:02advisable inAsianancestrypopulations(Han ChineseandTaiwanese)

2A HLA-B*15:11 SJS/TEN Asian(Japanese, Korean,HanChinese)

Increased 2B HLA-A*31:01 DRESS,MPE,SJS/TEN Mixedpopulationa _Increased

2B HLA-B*40:01 SJS/TEN Asian(HanChinese, Taiwanese)

Decreased 2B HLA-C*03:02(inLD

withHLA-B*58:01in somepopulations)

SJS/TEN Mixedpopulation Increased

3 HLA-B*15:18 SJS/TEN Asian(Japanese) Increased 3 HLA-B*58:01 MEP Asian(HanChinese) Decreased 3 HLA-B*58:01 SJS/TEN Asian(HanChinese) Increased

(decreasedin onestudy) 3 HLA-B*59:01 SJS/TEN Asian(Japanese) Increased 3 HLA-B*13:01 DrugHypersensitivity Caucasian,Asian Increased(not

inallstudies)

3 HLA-B*46:01 SJS/TEN Asian Decreased

3 HLA-DRB1*03:01 MEP Asian(HanChinese) Increased

Clozapine 3 HLA-DRB3*02:02 Agranulocytosis Caucasian Increased N/ARNPGxreco:noindicationin routinecare

Dapsone 2A HLA-B*13:01 Drughypersensitivity Asian Increased PPV=7.8%,NPV=99.8% NNTtoprevent‘‘1’’=84 RNPGxreco:noindicationin routinecare

3 HLA-A*02:01 Drughypersensitivity Asian(Japanese, Korean,Chinese)

Increased (decreasedin onestudy)

N/ARNPGxreco:noindicationin routinecare

(6)

S. Negrini, L. Becquemont Table1 (Continued) Drug Levelof evidence [29]

FrenchNationalPharmacogenetics network(RNPGx)recommendations Flucloxacillin 3 HLA-B*57:01 Drug-inducedliver

injury

Caucasian Increased PPV=0.12%,NPV=99.99% NNTtoprevent‘‘1’’=13,819 RNPGxreco:noindicationin routinecare

Lamotrigine 3 HLA-B*58:01 SJS/TEN Mixedpopulation Increased N/ARNPGxreco:noindicationin routinecare

3 HLA-B*15:02 SJS/TEN Asian Increased(only

foundin meta-analyses) Lapatinib 2B HLA-DQA1*02:01 Drug-inducedliver

injury

Mixedpopulation Increased HLA-DQA1*02:01or HLA-DRB1*07:01alleles carrier=FDA:warning(no genotypingrecommendations), EMA:warning(nogenotyping recommendations),HCSC:warning (nogenotypingrecommendations) RNPGxreco:potentiallyusefull 3 HLA-DRB1*07:01 Drug-inducedliver

injury

Mixedpopulation Increased Methazolamide 2A HLA-B*59:01 SJS/TEN Asian(Korean,

Japanese,Chinese)

Increased N/A

2B HLA-C*01:02 SJS/TEN Asian(Korean,

Chinese)

Increased Nevirapine 2A HLA-B*35:01 DrugHypersensitivity

(rash)

Mixedpopulation Increased (reportedtobe dependonCD4 T-cellscount) N/A RNPGxreco:noindicationin routinecare

2B HLA-DRB1*01:01 Drughypersensitivity (hepatitis)

Mixedpopulation Increased 3 HLA-B*35:01 Drughypersensitivity

(rash)

Asian(Thai) Increased (reportedtobe dependonCD4 T-cellscount) 3 HLA-DQB1*05:01 Drughypersensitivity

(various manifestations)

BlackorAfrican American

(7)

P harmacogenetics of hypersensitivity drug reactions 237 Table1 (Continued) Drug Levelof evidence [29]

FrenchNationalPharmacogenetics network(RNPGx)recommendations Nonsteroidal anti-inﬂammatory drugs 3 HLA-DR11 Anaphylactoid reactions

Spain Increased N/A

Oxcarbazepine 3 HLA-B*13:02 MEP Asian(Chinese) Increased N/A

3 HLA-B*15:02 MEP,SJS/TEN Asian Increased

3 HLA-B*15:19; HLA-B*15:27; HLA-B*27:09; HLA-B*38:02; HLA-B*48:04

MEP Asian(HanChinese) Increased

4

HLA-B*15:18:01/*40:01:01 genotype

SJS/TEN Mixedpopulation Increased N/A

RNPGxreco:noindicationin routinecare Pegylatedinterferon andribavirin 3 HLA-B*38:01 Non-responderto therapy

Egyptian N/A DPWG:informationaboutlower responseinHLA-B*44negative patients(nogenotyping recommendations)

3 HLA-B*44:02 Sustainedresponse Spain N/A

Phenobarbital 3 HLA-B*51:01 SJS/TEN Asian Increased N/ARNPGxreco:noindicationin routinecare

Phenytoin 1A HLA-B*15:02 SJS/TEN Asian Increased FDA:warning(nogenotyping

recommendations),CPIC:drug contraindicatedforB*15:02 positivepatients

HLA-B*15:02test:PPV=33%, NPV=100%

RNPGxreco:HLA-B*15:02advisable inAsianancestrypopulations(Han ChineseandTaiwanese)

3 HLA-B*13:01 SJS/TEN Asian Increased(not

increasedin onestudy) 3 HLA-B*56:02 DRESS IndigenousAustralian Increased Sulfasalazine 3 HLA-B*39:01;

HLA-B*13:01

DRESS Asian Increased N/ARNPGxreco:noindicationin routinecare

(8)

S. Negrini, L. Becquemont Table1 (Continued) Drug Levelof evidence [29]

FrenchNationalPharmacogenetics network(RNPGx)recommendations Thioamides

(carbimazole, methimazole, propylthiouracil)

2A HLA-B*38:02:01 Agranulocytosis Mixedpopulation Increased N/ARNPGxreco:noindicationin routinecare

Ticlopidine 3 HLA-B*44:03 Hepatotoxicityand cholestatic hepatotoxicity

Asian Increased N/ARNPGxreco:noindicationin routinecare

Trichloroethylene 3 HLA-B*13:01 Hypersensitivity dermatitis

Asian(Chinese) Increased N/ARNPGxreco:noindicationin routinecare

Ustekinumab 3 HLA-C*06:02 Betterresponse (psoriasis)

Caucasian N/A N/ARNPGxreco:noindicationin routinecare

Ximelagatran N/A DRB1*07:01, DQA1*02:01

Drug-inducedliver injury

European Increased Withdrawnin2006following reportsofhepatotoxicity Zonisamide 3 HLA-A*02:07:01 SJS/TEN Asian(Japanese) Increased N/ARNPGxreco:noindicationin

routinecare

ACR:American CollegeofRheumatology;CPIC:ClinicalPharmacogeneticsImplementationConsortium;CPNDS:Canadian PharmacogenomicsNetworkfor DrugSafety;DPWG:Dutch PharmacogeneticsWorkingGroup;DRESS:drugreactionwitheosinophiliaandsystemicsymptoms;EMA:EuropeanMedicinesAgency;FDA:USFoodandDrugAdministration;HCSC:Health Canada(SantéCanada);HDRs:hypersensitivitydrugreactions;MPE:maculopapulareruption;NPV:negativepredictivevalue;OR:oddsratio;PMDA:PharmaceuticalsandMedicalDevices Agency—Japan;PPV:positivepredictivevalue;SJS:Stevens-Johnsonsyndrome;TEN:toxicepidermalnecrolysis.RNPGxrecommendationsarecategorizedinto4categories:essential, advisable,potentiallyusefullandnoindicationinroutinecare.

(9)

Pharmacogeneticsofhypersensitivitydrugreactions 239 analyses that combined data obtain from different Asian

population all found odds ratios of approximately 80 for carbamazepine-induced SJS/TEN in patients carrying the HLA-B*15:02allele[54—56].A2011largeprospectivestudy, intended to evaluate the value of HLA-B*15:02 genotyp-ing before carbamazepine treatment, demonstrated that SJS/TEN did not develop in any of the treated patients if the subjects were HLA-B*1502-negative. Considered an expectedincidenceof0.23% basedonhistorical data,the geneticscreening successfully prevented 10 cases of HDR

[57].These resultspavedthewayfor theimplementation of HLA-B*15:02 screening before carbamazepine therapy inAsianpatients.Nevertheless, HLA-B*15:02hasnotbeen foundtobeariskfactorinCaucasianorJapanese popula-tions,probablybecauseoftheverylowprevalence ofthis allele in these ethnic group (1% or less). In European or Japaneserecentdatasuggeststhatcarbamazepine-related HDRs, includingSJS/TEN,MPE, and DRESS, areassociated with the presence of the HLA-A*31:01 allele [54,58—61]. Interestingly,inHan Chinese,HLA-A*31:01allelehasbeen associatedwithanincreasedriskofcarbamazepine-induced MPEorDRESS,whileHLA-B*15:02isrelatedonlytoSJS/TEN

[27,46,54].Theseobservationssuggestthatgenetic suscep-tibilitytocarbamazepine-relatedHDRs,at leastincertain populations,seemstobephenotype-specific.Beyond HLA-B*15:02 and HLA-A*31:01, other HLA alleles have been associated to an increased risk of carbamazepine-related HDRs in various populations (Table 1), nevertheless these studies arelimitedandneed furtherconfirmation. At this time, FDA recommends HLA-B*15:02 genetic testing prior toacarbamazepinebasedtreatmentinindividualsofAsian ethnicity[32,33].Awarninghasbeenaddedtothe prescrib-inginformationbyFDAindicatingthattherisksandbenefits ofusingcarbamazepineshouldbeweighedinHLA-A*31:01 allele carriers [33]. Different reports put HLA-B*15:02 in relationalsowithSJS/TENinducedbyaromatic antiepilep-tic drugs, other than carbamazepine, such as phenytoin, oxcarbazepine and lamotrigine [7,46,54,62]. These find-ings suggest that HLA-B*15:02 positive patients may show cross-reactivitytostructurallyrelatedanticonvulsants and consequentlyspecialattentionshouldbegivenwhen consid-eringantiepileptictreatmentforHLA-B*1502allelecarriers

[63]. Currently, FDA advise to avoid phenytoin as alter-native forcarbamazepine inHLA-B*1502positive patients, although it does not make speciﬁc recommendations for otherstructurally-relatedanticonvulsant[33].

Allopurinol

Allopurinolisaxanthineoxidaseinhibitorusedtotreat dis-ordersassociatedwithhyperuricemia,suchaschronicgout andtumorlysissyndrome.ThemostseriousADRsassociated toallopurinolareHDRs,whichincludeDRESSandSJS/TEN and occur in approximately 0.1—0.4% of treated patients [64]. Allopurinol is the most frequent cause of SJS/TEN inEurope,exceedingcarbamazepine andphenytoin,while in southeast Asia, is the second most common SJS/TEN-causativedrugaftercarbamazepine[45,65].

In 2005, a case-controlled study reported that a spe-ciﬁc HLA allele, HLA-B*58:01, was present in 100% of patients with allopurinol-induced HDRs compared to 15% of allopurinol-tolerant patients and 20% of healthy

con-trols [66]. Similar results have been replicated in other ethnic groups: strong association was reported in Han Chinese,Thais,JapaneseandKoreanswhileweaker associ-ationwasdescribedinEuropeans[67—72].Thisdifferences probably reﬂect the different prevalence of HLA-B*58:01 which is higher in Asian populations as compared to Europeans and/or additional contributing factors not yet elucidated.However,arecentmeta-analysisfoundastrong and signiﬁcant association the HLA-B*58:01 allele and allopurinol-induced SJS/TEN (OR 96.60 and OR 79.28 in studieswithmatched-controlorpopulation-control, respec-tively) in both Asian and non-Asian populations [73]. In the light of the strong association of HLA-B*58:01 with hypersensitivity reactions to allopurinol across different populations,ageneticscreeningofpatientspriorto initiat-ingallopurinolwillbemostlikelyadoptedinfuture.Todate, FDA and EMA have not implemented allopurinol labeling informationforHLA-B*58:01testing.TheClinical Pharmaco-geneticsImplementationConsortiumguidelinesrecommend thatallopurinolshouldnotbeprescribedtopatientswhoare positivefortheHLA-B*5801allelebutunderlinethat nega-tivetestingdoes notexcludethe possibilityof developing SCARs,especially in European populations [64]. The 2012 revisedguidelinesoftheAmericanCollegeofRheumatology alsorecommendHLA-B*58:01screeningpriortoallopurinol prescription,especiallyinthosepopulationswithhigh fre-quenciesof the allele, such as the Han Chinese or Thais

[74].

Conclusion

Abacavir,carbamazepine,andallopurinolarekeyexamples of pharmacogenetics implementation in routine medical practice(Table2)[75], but manyother HLA-related HDRs have been reported so far (Table 1). However, several considerations should be taken into account before such associationscouldbetranslatedincost-effectivescreening procedures,thesefactorshavebeen recentlyreviewedby PhillipsandMallal[76].Premisesforasuccessful implemen-tationofagivenHLApharmacogenetictestingintoroutine clinicalcarerequiresthat:drugtoxicityissevereand per-sistent,theNPV(andideallyalsothePPV)ofthetestreach 100%, culprit HLA allele is prevalent inside the screened population,andthe numberof patientsneeded totest in ordertopreventacaseislow.Inaddition,responsibledrug shouldhavegoodefﬁcacy,cost-effectivenessand, tolerabil-ity,intheabsenceofalternativedrug(s)withsimilarpositive characteristics. Flucloxacillin-related livertoxicity canbe citedasaparadigmaticexample,althoughstrongly associ-atedwithHLA-B*57:01allele(OR80),almost14,000patients shouldbetested toprevent asingle case of hepatotoxic-ity,makingthisscreeningneithercost-effectivenorfeasible

[34,77].Inviewof this,onlyfewHLA-disease-drug associ-ationsreportedtodate willbereasonablytranslatedinto routineclinicalpractice.

Nevertheless, clinical value of pharmacogenetic mark-ersis notonly limited to predict susceptibility toHDR in pre-treatment setting but may also be exploited astests ofexclusion,ashasbeenproposedfor there-introduction oflumiracoxib,ordifferentialdiagnosistools,ashasbeen suggested for ﬂucloxacillin-induced hepatitis [78]. Due

(10)

S.

Negrini,

L.

Becquemont

Table2 Maingeneticvariantsdiscussedinthisarticle.

Genea _Namea _Polymorphisma _Reference

rs

Commonname oftheallele

FrequencyofHLAallelesacross diversepopulations[75] Invitro function Invivo function Histocompatibility complex(Chromosome 6p21.3) Major histocompatibility complex,classI,A

*57:01 N/A HLA-B*57:01 WestEurope:5—10%,

Mediterranean:1—4%,UK:5—10%, MiddleEast:1—4%,USCaucasian: 5—10%,USAsia:1—3%,US African-American:2—3%,US Hispanic:1—5%,SouthAmerican Caucasian:5—10%,Sub-Saharan Africa:<1%,Australia:5—10%, Thailand:5—10%,China:<1%, Japan:<1%,India:5—15%

N/A N/A

*58:01 N/A HLA-B*58:01 WestEurope:1—4%,

Mediterranean:1—4%,UK:<1%, MiddleEast:5—10%,USCaucasian: 1—4%,USAsia:5—10%,US

African-American:5—10%,US Hispanic:1—4%,SouthAmerican Caucasian:1—4%,Sub-Saharan Africa:5—15%,Australia:1—4%, Thailand:5—15%,China:5—20%, Japan:<1%,India:5—15%

N/A N/A

*15:02 N/A HLA-B*15:02 WestEurope:<1%,Mediterranean: <1%,UK:<1%,MiddleEast:<1%, USCaucasian:<1%,USAsia: 5—10%,USAfrican-American:<1%, USHispanic:<1%,SouthAmerican Caucasian:<1%,Sub-Saharan Africa:<1%,Australia:<1%, Thailand:10—20%,China:5—20%, Japan:<1%,India:5—15%

N/A N/A

Froma practical pointofview,HLAgenotypingtestfor drughypersensitivity shouldbeinterpretedas follow.Since HLAexpression isco-dominant,resultsare reportedas either ‘‘positive’’or‘‘negative’’,withnointermediatephenotype.Theabsenceofthetestedalleles(reportedas‘‘negative’’onagenotypingtest)indicatesthatthepatienthasaverylow riskofHDR.Ontheopposite,‘‘positive’’result(heterozygoteandhomozygousvariants)meansthatthedrugiscontraindicatedduetotheincreasedriskofdrug-inducedhypersensitivity [39].N/A:notapplicable.

(11)

Pharmacogeneticsofhypersensitivitydrugreactions 241 to their pharmacological unpredictability and very low

incidence, HDRs are usually recognized in the post-marketingphaseof the druglife cycleleading toproduct withdrawal after important investment hasbeen madein researchanddevelopment.AsystematicDNAbankingduring drugdevelopmentphasecouldallow,ifaHDRoccurs,to ret-rospectivelyresearchasusceptibilitygenomicbiomarker.In thiscontext,successfulidentiﬁcationofapredictivemarker could avert withdrawal of the drug from the market by limiting the use of the drug to genetically tolerant sub-jects[79].Lastbutnotleast,progressesinbasicresearchon immunologicalmechanismsinvolvedinHDRscouldimprove drugdesign,andlaythegroundworkforthedevelopmentof invitrotestscapabletoidentifyharmfulcompoundsin pre-clinicalphasesofdrugdevelopment.Theseimplementations wouldincreasedrugsafetybypredictingHDRsbeforetheir useinman,and,atthesametime,wouldgreatlyreducethe costofdrugdevelopment[80].

Disclosure

of

interest

Theauthorsdeclarethattheyhavenocompetinginterest.

References

[1]WHOWorldHealthOrganization.International drug monitor-ing.Theroleofhospital.Geneva,Switzerland:Technicalreport series;1966.p.425.

[2]LazarouJ,PomeranzBH,CoreyPN.Incidenceofadversedrug reactionsinhospitalizedpatients:ameta-analysisof prospec-tivestudies.JAMA1998;279(15):1200—5.

[3]TheEuropeanCommissiondocument‘‘Strengthening pharma-covigilancetoreduceadverseeffectsofmedicines’’.Brussels; 2008[MEMO/08/782].

[4]Ernst FR,GrizzleAJ. Drug-related morbidityand mortality: updatingthecost-of-illnessmodel.JAmPharmAssoc(Wash) 2001;41(2):192—9.

[5]Wysowski DK, Swartz L. Adverse drug event surveillance anddrug withdrawalsintheUnitedStates, 1969—2002: the importanceofreportingsuspectedreactions.ArchInternMed 2005;165(12):1363—9.

[6]RawlinsMD,ThompsonJW.Pathogenesisofadversedrug reac-tions.In:OxfordUniversityPress,editor.Textbookofadverse drugreactions.OxfordUniversityPressed;1977.

[7]JohanssonSG,HourihaneJO,BousquetJ,Bruijnzeel-KoomenC, DreborgS,HaahtelaT,etal.Arevisednomenclatureforallergy. AnEAACIpositionstatementfromtheEAACInomenclaturetask force.Allergy2001;56(9):813—24.

[8]Johansson SG, Bieber T, Dahl R, Friedmann PS, Lanier BQ, LockeyRF,etal.Revisednomenclatureforallergyforglobal use: report of the nomenclature review Committee of the World Allergy Organization, October 2003. J Allergy Clin Immunol2004;113(5):832—6.

[9]CoombsPRGP. Classiﬁcationofallergicreactionsresponsible for clinicalhypersensitivity and disease. Clinical aspects of immunology.Oxford,England:OxfordUniversityPress;1968. p.575—96.

[10] PichlerWJ.Delayeddrughypersensitivityreactions.AnnIntern Med2003;139(8):683—93.

[11] PichlerWJ,AdamJ,DaubnerB,GentinettaT,KellerM,YerlyD. Drughypersensitivityreactions:pathomechanismandclinical symptoms.MedClinNorthAm2010;94(4):645—64[xv].

[12] Schnyder B, Mauri-Hellweg D, Zanni M, Bettens F, Pich-ler WJ. Direct, MHC-dependent presentation of the drug

sulfamethoxazole to human alphabeta T cell clones. J Clin Invest1997;100(1):136—41.

[13]Naisbitt DJ, Farrell J, Wong G, Depta JP, Dodd CC, Hop-kins JE, et al. Characterization of drug-speciﬁc T cells in lamotrigine hypersensitivity. J Allergy Clin Immunol 2003;111(6):1393—403.

[14]Zanni MP,vonGreyerz S,Schnyder B,BranderKA, FrutigK, Hari Y, et al. HLA-restricted, processing- and metabolism-independentpathwayofdrugrecognitionbyhumanalphabeta Tlymphocytes.JClinInvest1998;102(8):1591—8.

[15]PavlosR,MallalS,OstrovD,PompeuY,PhillipsE.Fever,rash, and systemicsymptoms:understandingtheroleofvirus and HLAinseverecutaneousdrugallergy.JAllergyClinImmunol Pract2014;2(1):21—33.

[16]BlackwellJM,JamiesonSE,BurgnerD.HLAandinfectious dis-eases.ClinMicrobiolRev2009;22(2):370—85.

[17]GoughSC,SimmondsMJ.TheHLARegionandautoimmune dis-ease:associationsandmechanismsofaction.CurrGenomics 2007;8(7):453—65.

[18]Hetherington S,McGuirk S,Powell G, CutrellA,Naderer O, SpreenB,etal.Hypersensitivityreactionsduringtherapywith thenucleosidereversetranscriptaseinhibitorabacavir.Clinical therapeutics2001;23(10):1603—14.

[19]EscautL,LiotierJY,AlbengresE,CheminotN,VittecoqD. Aba-cavirrechallengehastobeavoidedincaseofhypersensitivity reaction.AIDS1999;13(11):1419—20.

[20]Hetherington S, Hughes AR, Mosteller M, Shortino D, Baker KL, Spreen W, et al. Genetic variations in HLA-B region and hypersensitivity reactions to abacavir. Lancet 2002;359(9312):1121—2.

[21]MallalS,NolanD,WittC,MaselG,MartinAM,MooreC,etal. Associationbetween presence ofHLA-B*5701,HLA-DR7, and HLA-DQ3 and hypersensitivity to HIV-1 reverse-transcriptase inhibitorabacavir.Lancet2002;359(9308):727—32.

[22]Hughes DA, VilarFJ, Ward CC, AlﬁrevicA, ParkBK, Pirmo-hamedM.Cost-effectivenessanalysisofHLAB*5701genotyping in preventing abacavir hypersensitivity. Pharmacogenetics 2004;14(6):335—42.

[23]Rauch A, NolanD, MartinA, McKinnon E,Almeida C,Mallal S. Prospectivegeneticscreeningdecreasestheincidence of abacavirhypersensitivity reactionsintheWesternAustralian HIVcohortstudy.ClinInfecDis2006;43(1):99—102.

[24]WatersLJ,MandaliaS,GazzardB,NelsonM.Prospective HLA-B*5701screeningandabacavirhypersensitivity:asinglecentre experience.AIDS2007;21(18):2533—4.

[25]ZucmanD,TruchisP,MajerholcC,StegmanS,Caillat-Zucman S.Prospectivescreeningforhumanleukocyteantigen-B*5701 avoids abacavir hypersensitivity reaction in the ethnically mixed French HIV population. JAcquir ImmuneDeﬁc Syndr 2007;45(1):1—3.

[26]MallalS,PhillipsE,CarosiG,MolinaJM,WorkmanC,Tomazic J,etal.HLA-B*5701screeningforhypersensitivitytoabacavir. NEnglJMed2008;358(6):568—79.

[27]Hung SI, Chung WH, Jee SH, Chen WC, Chang YT, Lee WR,et al. Genetic susceptibilityto carbamazepine-induced cutaneous adversedrugreactions.Pharmacogenet Genomics 2006;16(4):297—306.

[28]Sousa-PintoB,Pinto-RamosJ,CorreiaC,Goncalves-CostaG, Gomes L, Gil-Mata S, et al. Pharmacogenetics of abacavir hypersensitivity: a systematic review and meta-analysis of the association with HLA-B*57:01. J Allergy Clin Immunol 2015;136(4)[1092-4e3].

[29]Whirl-CarrilloM,McDonaghEM,HebertJM,GongL,SangkuhlK, ThornCF,etal.Pharmacogenomicsknowledgeforpersonalized medicine.ClinPharmacolTher2012;92(4):414—7.

[30]PavlosR,MallalS,OstrovD,BuusS,MetushiI,PetersB,etal. Tcell-mediatedhypersensitivityreactionstodrugs.AnnuRev Med2015;66:439—54.

(12)

[31]Yip VL, Marson AG, Jorgensen AL, Pirmohamed M, Alﬁre-vic A. HLA genotype and carbamazepine-induced cutaneous adverse drugreactions:a systematicreview.ClinPharmacol Ther2012;92(6):757—65.

[32]EMA. European Medicines Agency. http://www.ema.europa.

eu/.[Accessed18November2016].

[33]FDA.U.S.FoodandDrugAdministration.http://www.fda.gov/ [Accessed18November2016].

[34]DalyAK,DonaldsonPT,BhatnagarP,ShenY,Pe’erI,Floratos A, et al. HLA-B*5701 genotype is a major determinant of drug-induced liver injury due to ﬂucloxacillin. Nat Genet 2009;41(7):816—9.

[35]MiguelesSA,SabbaghianMS,ShupertWL,BettinottiMP, Mar-incola FM,Martino L,et al.HLAB*5701 ishighlyassociated with restriction of virus replication in a subgroup of HIV-infectedlongtermnonprogressors.ProcNatlAcadSciUSA 2000;97(6):2709—14.

[36]FellayJ,ShiannaKV,GeD,ColomboS,LedergerberB,WealeM, etal.Awhole-genomeassociationstudyofmajordeterminants forhostcontrolofHIV-1.Science2007;317(5840):944—7. [37]UKCollaborativeHIVCohortStudySteeringCommittee.HLA

B*5701status,diseaseprogression,andresponseto antiretro-viraltherapy.AIDS2013;27(16):2587—92.

[38]MockenhauptM.ThecurrentunderstandingofStevens-Johnson syndrome and toxic epidermal necrolysis. Expert Rev Clin Immunol2011;7(6):803—13[quiz14-5].

[39]Downey A, JacksonC, HarunN, Cooper A. Toxic epidermal necrolysis:reviewofpathogenesisandmanagement.JAmAcad Dermatol2012;66(6):995—1003.

[40]DaoRL, SuSC,ChungWH.Recentadvancesof pharmacoge-nomics insevere cutaneous adverse reactions:immune and nonimmunemechanisms.AsiaPacAllergy2015;5(2):59—67. [41]Harr T, French LE.Toxic epidermal necrolysis and

Stevens-Johnsonsyndrome.OrphanetJRareDis2010;5:39.

[42]Barbarino JM, Kroetz DL, Klein TE, Altman RB. Phar-mGKB summary: very important pharmacogene information for human leukocyte antigen B. Pharmacogenet genomics 2015;25(4):205—21.

[43]ChungWH,HungSI,HongHS,HsihMS,YangLC,HoHC,etal. Medical genetics: a marker for Stevens-Johnson syndrome. Nature2004;428(6982):486.

[44]Chung WH, Hung SI. Recent advances in the genetics and immunologyofStevens-Johnsonsyndromeandtoxicepidermal necrosis.JDermatolSci2012;66(3):190—6.

[45]ChengCY,SuSC,ChenCH,ChenWL,DengST,ChungWH.HLA associationsandclinicalimplicationsinT-cellmediateddrug hypersensitivityreactions:anupdatedreview.JImmunolRes 2014;2014:565320.

[46]Man CB, Kwan P, Baum L, Yu E, Lau KM, Cheng AS, etal.AssociationbetweenHLA-B*1502alleleand antiepilep-ticdrug-inducedcutaneousreactionsinHanChinese.Epilepsia 2007;48(5):1015—8.

[47]WuXT,HuFY,AnDM,YanB,JiangX,KwanP,etal. Associa-tionbetweencarbamazepine-inducedcutaneousadversedrug reactionsandtheHLA-B*1502alleleamongpatientsincentral China.EpilepsyBehav2010;19(3):405—8.

[48]WangQ,ZhouJQ,ZhouLM,ChenZY,FangZY,ChenSD,etal. Association between HLA-B*1502allele and carbamazepine-inducedseverecutaneousadversereactionsinHanpeopleof southernChinamainland.Seizure2011;20(6):446—8. [49]TassaneeyakulW,TiamkaoS,JantararoungtongT,ChenP,Lin

SY, Chen WH, et al. Association between HLA-B*1502 and carbamazepine-inducedseverecutaneousadversedrug reac-tionsinaThaipopulation.Epilepsia2010;51(5):926—30. [50]LocharernkulC,LoplumlertJ,LimotaiC,KorkijW,DesudchitT,

TongkobpetchS,etal.Carbamazepineandphenytoininduced Stevens-JohnsonsyndromeisassociatedwithHLA-B*1502allele inThaipopulation.Epilepsia2008;49(12):2087—91.

[51]Then SM, Rani ZZ, Raymond AA, Ratnaningrum S, Jamal R. Frequency of the HLA-B*1502 allele contributing to carbamazepine-inducedhypersensitivityreactionsinacohort ofMalaysianepilepsypatients. AsianPac JAllergy Immunol 2011;29(3):290—3.

[52]MehtaTY,Prajapati LM,Mittal B,Joshi CG,ShethJJ, Patel DB,etal.AssociationofHLA-B*1502alleleand carbamazepine-induced Stevens-Johnson syndrome among Indians. Indian J DermatolVenereolLeprol2009;75(6):579—82.

[53]KulkantrakornK,TassaneeyakulW, TiamkaoS, Jantararoung-tong T, Prabmechai N, Vannaprasaht S, et al. HLA-B*1502 stronglypredictscarbamazepine-inducedStevens-Johnson syn-drome and toxic epidermal necrolysisin Thai patientswith neuropathicpain.PainPract2012;12(3):202—8.

[54]GroverS,KukretiR.HLAallelesandhypersensitivityto carba-mazepine:anupdatedsystematicreviewwithmeta-analysis. PharmacogenetGenomics2014;24(2):94—112.

[55]TangamornsuksanW,ChaiyakunaprukN,SomkruaR,Lohitnavy M, Tassaneeyakul W. Relationship between the HLA-B*1502 alleleandcarbamazepine-inducedStevens-Johnsonsyndrome andtoxicepidermalnecrolysis:asystematicreviewand meta-analysis.JAMADermatol2013;149(9):1025—32.

[56]Bloch KM, Sills GJ, Pirmohamed M, Alﬁrevic A. Pharma-cogenetics of antiepileptic drug-induced hypersensitivity. Pharmacogenomics2014;15(6):857—68.

[57]Chen P, Lin JJ, Lu CS, Ong CT, Hsieh PF, Yang CC, et al.Carbamazepine-induced toxic effects and HLA-B*1502 screeninginTaiwan.NEnglJMed2011;364(12):1126—33. [58]OzekiT,MushirodaT,YowangA,TakahashiA,KuboM,Shirakata

Y,etal.Genome-wideassociationstudyidentiﬁesHLA-A*3101 alleleasageneticriskfactorforcarbamazepine-induced cuta-neousadversedrugreactionsinJapanesepopulation.HumMol Genet2011;20(5):1034—41.

[59]McCormackM,AlﬁrevicA,BourgeoisS,FarrellJJ, Kasperavi-ciuteD,CarringtonM,etal.HLA-A*3101and carbamazepine-inducedhypersensitivityreactionsinEuropeans.NEnglJMed 2011;364(12):1134—43.

[60]Alﬁrevic A, Jorgensen AL, Williamson PR, Chadwick DW, Park BK,Pirmohamed M.HLA-B locus in Caucasian patients with carbamazepine hypersensitivity. Pharmacogenomics 2006;7(6):813—8.

[61]Niihara H, Kakamu T, Fujita Y, Kaneko S, Morita E. HLA-A31stronglyassociateswithcarbamazepine-inducedadverse drug reactions but not with carbamazepine-induced lym-phocyteproliferation ina Japanese population.JDermatol 2012;39(7):594—601.

[62]HungSI,ChungWH,LiuZS,ChenCH,HsihMS,HuiRC,etal. Common risk allele in aromatic antiepileptic-drug induced Stevens-Johnsonsyndrome andtoxic epidermalnecrolysisin HanChinese.Pharmacogenomics2010;11(3):349—56. [63]LiaoWP,ShiYW,MinFL.HLA-B*1502screeningandtoxiceffects

ofcarbamazepine.NEngl JMed2011;365(7):672—3 [author reply3].

[64]Hershﬁeld MS, Callaghan JT, Tassaneeyakul W, Mushiroda T, Thorn CF, Klein TE, et al. Clinical pharmacogenetics implementation consortium guidelines for human leukocyte antigen-Bgenotypeandallopurinoldosing.ClinPharmacolTher 2013;93(2):153—8.

[65]Lee MT, Mahasirimongkol S, Zhang Y, Suwankesawong W, Chaikledkaew U, Pavlidis C, et al. Clinical application of pharmacogenomics:theexample ofHLA-baseddrug-induced toxicity.PublicHealthGenomics2014;17(5—6):248—55. [66]HungSI,ChungWH,LiouLB,ChuCC,LinM,HuangHP,etal.

HLA-B*5801alleleasa geneticmarker forseverecutaneous adversereactionscausedbyallopurinol.ProcNatlAcadSciU SA2005;102(11):4134—9.

[67]Tohkin M, Kaniwa N, Saito Y, Sugiyama E, Kurose K, NishikawaJ,etal.Awhole-genomeassociationstudyofmajor

(13)

Pharmacogeneticsofhypersensitivitydrugreactions 243 determinants for allopurinol-related Stevens-Johnson

syn-dromeand toxic epidermal necrolysisin Japanese patients. PharmacogenomicsJ2013;13(1):60—9.

[68] TassaneeyakulW,JantararoungtongT,ChenP,LinPY,Tiamkao S,Khunarkornsiri U,et al. Strongassociation between HLA-B*5801andallopurinol-inducedStevens-Johnsonsyndromeand toxicepidermalnecrolysisinaThaipopulation.Pharmacogenet Genomics2009;19(9):704—9.

[69] CaoZH,WeiZY,ZhuQY,ZhangJY,YangL,QinSY,etal. HLA-B*58:01alleleisassociatedwithaugmentedriskforbothmild andseverecutaneousadversereactionsinducedbyallopurinol inHanChinese.Pharmacogenomics2012;13(10):1193—201.

[70] KaniwaN,SaitoY, AiharaM,MatsunagaK,TohkinM,Kurose K,etal.HLA-BlocusinJapanesepatientswithanti-epileptics andallopurinol-related Stevens-Johnsonsyndrome and toxic epidermalnecrolysis.Pharmacogenomics2008;9(11):1617—22.

[71] KangHR,JeeYK,Kim YS,LeeCH, JungJW,Kim SH,etal.

Positive and negative associations of HLA class I alleles with allopurinol-induced SCARs in Koreans. Pharmacogenet Genomics2011;21(5):303—7.

[72] LonjouC,BorotN,SekulaP,LedgerN,ThomasL, HalevyS, etal.AEuropeanstudyofHLA-BinStevens-Johnsonsyndrome andtoxicepidermalnecrolysisrelatedtoﬁvehigh-riskdrugs. PharmacogenetGenomics2008;18(2):99—107.

[73] Somkrua R, Eickman EE, Saokaew S, Lohitnavy M, Chaiyakunapruk N. Association of HLA-B*5801 allele and

allopurinol-induced Stevens Johnson syndrome and toxic epidermalnecrolysis:asystematicreviewandmeta-analysis. BMCMedGenet2011;12:118.

[74]KhannaD,FitzgeraldJD, KhannaPP,BaeS,Singh MK,Neogi T, etal. 2012American CollegeofRheumatologyguidelines formanagementofgout.Part1:systematicnonpharmacologic andpharmacologictherapeuticapproachestohyperuricemia. ArthritisCareRes(Hoboken)2012;64(10):1431—46.

[75]Gonzalez-GalarzaFF,TakeshitaLY,SantosEJ,KempsonF,Maia MH,daSilvaAL,etal.Allelefrequencynet2015update:new features forHLAepitopes,KIRand diseaseandHLAadverse drugreactionassociations.NucleicAcidsRes2015;43:D784—8 [Databaseissue].

[76]PhillipsEJ,MallalSA.Pharmacogeneticsofdrug hypersensitiv-ity.Pharmacogenomics2010;11(7):973—87.

[77]KarlinE,PhillipsE.Genotypingforseveredrughypersensitivity. CurrAllergyAsthmaRep2014;14(3):418.

[78]AlﬁrevicA,PirmohamedM.Druginducedhypersensitivityand theHLAcomplex.Pharmaceuticals2011;4(1):69—90.

[79]BecquemontL.HLA:apharmacogenomicssuccessstory. Phar-macogenomics2010;11(3):277—81.

[80]WhiteKD,ChungWH,HungSI,MallalS,PhillipsEJ.Evolving modelsof theimmunopathogenesis ofTcell-mediateddrug allergy: the role of host, pathogens, and drug response. J AllergyClinImmunol2015;136(2):219—34[quiz35].