v.2/HIVDR 2003-2010- 23 July - WHO1.Anti-HIV agents – therapeutic use. 2.Drug resistance, Viral...

HIV/AIDS Programme

For more information, contact:

World Health OrganizationDepartment of HIV/AIDS20, avenue Appia1211 Geneva 27Switzerland

E-mail: [email protected]

http://www.who.int/hiv/en/

ISBN 978 92 4 150393 8

WHO HIV DRUG RESISTANCE REPORT 2012


HIV/AIDS Programme

WHO Library Cataloguing-in-Publication Data

The HIV drug resistance report - 2012.

1.Anti-HIV agents – therapeutic use. 2.Drug resistance, Viral – drug effects. 3.HIV infections – epidemiology. 4.HIV infections – drug therapy 5.Population surveillance. 6.Sentinel surveillance. 7.Developing countries. 8.World health - statistics. I.World Health Organization.

ISBN 978 92 4 150393 8 (NLM classification: QV 268.5)

© World Health Organization 2012

All rights reserved. Publications of the World Health Organization are available on the WHO web site (www.who.int) or can be purchased from WHO Press, World Health Organization, 20 Avenue Appia, 1211 Geneva 27, Switzerland (tel.: +41 22 791 3264; fax: +41 22 791 4857; e-mail: [email protected]).

Requests for permission to reproduce or translate WHO publications – whether for sale or for noncommercial distribution – should be addressed to WHO Press through the WHO web site (http://www.who.int/about/licensing/copyright_form/en/index.html).

The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of the World Health Organization concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted lines on maps represent approximate border lines for which there may not yet be full agreement.

The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by the World Health Organization in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters.

All reasonable precautions have been taken by the World Health Organization to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall the World Health Organization be liable for damages arising from its use.

Design & layout: L’IV Com, Villars-sous-Yens, Switzerland

Printed in the USA

WHO HIV Drug Resistance Report 2012

1

Acronyms. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 2

Acknowledgements. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 3

Executive Summary. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 5

1. Introduction. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91.1 Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91.2 StructureoftheReport.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91.3 DeterminantsofHIVdrugresistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 9

1.3.1 Regimen-anddrug-specificfactors.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 91.3.2 Virusfactors.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101.3.3 Patientfactors.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 101.3.4 Programmaticfactors.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 10

1.4 WHO’sglobalHIVdrugresistancesurveillanceandmonitoringstrategy.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . 111.5 Noteondatasourcesandmethods.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 12

2. HIV drug resistance in high-income countries. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 172.1 DrugresistanceinARV-naïverecentlyorchronically-infectedpopulations. . . . . . . . . . . . . . . . . . . . . . . . . . . . 172.2 Acquireddrugresistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 18

3. Transmitted HIV drug resistance in low- and middle-income countries.. . . . . . . . . . . . . . . . . . . 203.1 Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 203.2 LiteraturereviewondrugresistanceamongARV-naïverecently-orchronically-infectedpopulations.. . . . . . . . . 203.3 WHOsurveystoassesstransmitteddrugresistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 21

3.3.1 Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 213.3.2ClassificationofWHOsurveysontransmittedHIVdrugresistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . 233.3.3Pooledanalysis. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 253.3.4PrevalenceofHIVdrugresistancemutations. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 27

4. Acquired drug resistance in low- and middle-income countries.. . . . . . . . . . . . . . . . . . . . . . . . . 294.1 Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 304.2Literaturereviewonacquireddrugresistanceinlow-andmiddle-incomecountries. . . . . . . . . . . . . . . . . . . . . 304.3 WHOsurveystoassessacquiredHIVdrugresistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 30

4.3.1 Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 314.3.2Drugresistancebeforeinitiationoffirst-lineantiretroviraltherapy(surveybaseline).. . . . . . . . . . . . . . . . 324.3.3Acquireddrugresistanceamongpeoplefailingfirst-lineantiretroviraltherapyat12months.. . . . . . . . . . . 35

4.3.3.1 Drugresistanceinpatientsfailingtherapyat12months. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374.3.3.2 Drugresistanceprevention. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374.3.3.3 Possibledrugresistance. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 374.3.3.4 PrevalenceandpatternsofHIVdrugresistanceamongpeopleexperiencingtreatment

failure12monthsafterinitiation. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 38

5. Early warning indicators. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415.1 Overview. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 415.2Revisedearlywarningindicatorsandtargets. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 43

6. Conclusions. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 45

Annexes. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48Annex1 Methodologicalnotes.. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 48Annex2 Supplementaltablesandfigures. . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . . 57

TABLE OF CONTENTS

2

3TC lamivudine

ABC abacavir

ART antiretroviraltherapy

AZT zidovudine

bPI boostedproteaseinhibitors

CRF circulatingrecombinantform

DNA deoxyribonucleicacid

d4T stavudine

ddI didanosine

EFV efavirenz

ETR etravirine

FTC emtricitabine

HIV humanimmunodeficiencyvirus

NNRTI non-nucleosidereverse-transcriptaseinhibitor

NRTI nucleosidereverse-transcriptaseinhibitor

NVP nevirapine

PCR polymerasechainreaction

PI proteaseinhibitors

PMTCT preventionofmother-to-childtransmissionofHIV

PR protease

RNA ribonucleicacid

RPV rilpivirine

RTI reversetranscriptaseinhibitor

RT-PCR reverse-transcriptasepolymerasechainreaction

SDRM surveillancedrugresistancemutation

TAM thymidineanaloguemutation

TDF tenofovir

ACRONYMS


3

T his reportwasproducedbySilviaBertagnolio(HIVDepartment,WorldHealthOrganization),MichaelR. Jordan

(HIVDepartment,WorldHealthOrganization), JhoneyBarcarolo (consultant)andNeilParkin (consultant,Data

FirstConsulting,Inc).SilviaBertagnoliodevelopedtheanalysisplanandprovidedoverallcoordination.Allmembers

ofthecoreteamcontributedtothereview,analysisandqualityassuranceofthedata,writingandeditingofthereportand

thepreparationoffiguresandtables.MartinaPenazzatoassistedwithresearchandreview,andAdilBahalimprovideddata

managementsupport.WethankJosPerriens,coordinatorofTechnologiesandCommoditiesTeam,forreviewingdraftsof

thismanuscriptatvariousstages.

ThisworkwouldnothavebeenpossiblewithoutthecollaborationofnationalHIVprogrammemanagersandstaffwho

suppliedthesurveillancedatathatunderpinsthisreport.Developmentandpublicationofthisreportwerealsosupportedby

thegenerousfinancialcontributionoftheBillandMelindaGatesFoundation.WearesincerelygratefultoChristineRousseau

andChrisDuncombefortheircontinuedsupport.WeareindebtedtoDianeBennettforhersignificantcontributiontothe

developmentoftheWHOHIVdrugresistancesurveillanceandmonitoringstrategy,andtoKarenKelley,particularlyforher

importantworkonimprovingguidanceandtoolsforearlywarningindicators.

WewouldalsoliketothankAnnemarieWensing,ElliotRaizes,ScottHammer,RaphHamers,MarkWainberg(currentchair

oftheHIVResNet),forreviewingthereportandprovidingvaluablecommentsoninitialdrafts;JohnGregson,DanielDavis,

AndrewPhillips,andValentinaCambianoforthestatisticalanalysis;DanielDavisandRaviGuptafortheircontributionon

thesystematicreviewofacquiredandtransmitteddrugresistance;AndreaDeLuca,AnnaHearps,MarkWainberg,Scott

Hammer,andWataruSugiuraforcontributingtothesectionofresistanceinhigh-incomecountries;membersofHIVResNet

fortheirvaluableinputstodevelopthedrugresistancesurveymethods,inparticulartoMarkMyattandDonSutherland;

AnnemarieWensingandAnne-MiekeVandamme,JonathanSchapiro,RamiKantorandRobertShaferfortheircontributionto

theinterpretationofdatafromWHOsurveys;andTommyLiu,RobertShafer,Anne-MiekeVandamme,andTuliodeOliveira

forassistancewithsequenceanalysisanddrugresistanceinterpretations.

WesincerelythankTobiasF.RinkedeWit,RaphHamersandKimSigaloffformakingavailabledatafromthePharmAccess

AfricanStudiestoEvaluateResistance(PASER),andtheUnitedStatesCentersforDiseaseControlandPrevention(CDC)

foritscontributiontothedevelopmentandimplementationoftheHIVdrugresistancestrategy.

ThelaboratorymembersoftheWHOHIVDRNetworkhavebeenessentialtotheproductionofhigh-qualitydrugresistance

surveillancedata.TheauthorswouldliketothankChunfuYang(InternationalLaboratoryBranch,CDC/GAP,Atlanta),Martine

Peeters(LaboratoireRétrovirus IRD,Montpellier),andPaulSandstrom(PHAC,Ottawa) for theirsupport ingenotyping

specimensfortheWHOHIVdrugresistancesurveys.Inaddition,theauthorswouldliketothankalltheothermembersof

thenetworkforproducinghigh-qualitysurveillancedata:

• RamatouToureandChristianeAdje(RETROCI,Abidjan,Coted’Ivoire)

• AlmazAbebeandLibseworkMuluneh(HIVandOtherViralDiseaseResearch,EHNRI,AddisAbaba,Ethiopia)

• AvelinAghokeng(IMPM-IRD/CREMER,Yaounde,Cameroon)

• ClementZeh(KEMRI/CDCHIVResearchLaboratory,Kisumu,Kenya)

• CoumbaToureKaneandSouleymaneMboup(Bacteriology-VirologyUTHALeDantec,Dakar,Senegal)

• MarizaMorgadoand JoseCarlosCoutoFernandez (LaboratoryofAIDS andMolecular Immunology,Oswaldo Cruz

Foundation-FIOCRUZ,RiodeJaneiro,Brazil)

• LukeHanna(DepartmentofClinicalResearchTuberculosisResearchCentre(ICMR),Chennai,India)

• SrikanthTripathyandRameshParanjape(NationalAIDSResearchInstitute,IndianCouncilofMedicalResearch,Pune,India)

• SiriphanSaeng-aroon(NationalInstituteofHealth,DepartmentofMedicalSciences,Nonthaburi,Thailand)

• RuengpungSutthent(DeptMicrobiology,SirirajHospital,Bangkok,Thailand)

• ShaoYiming(DivisionofResearchonVirologyandImmunology(DRVI),NCAIDS,ChineseCenterforDiseaseControl

andPrevention,Beijing,China)

• PingZhong(ShanghaiMunicipalCenterforDiseaseControlandPrevention,Shanghai,China)

ACKNOWLEDGEMENTS

4

• HongShang(KeyLaboratoryofImmunologyofAIDS,MinistryofHealth,Shenyang,China)

• GillianHunt,JohannaLedwaba,andLynnMorris(AIDSVIRUSRESEARCHUNIT,NationalInstituteforCommunicable

Diseases,Johannesburg,SouthAfrica)

• ChrisParryandPontianoKaleebu(MRC/UVRIBasicSciencesLaboratory,Entebbe,Uganda)

• GeorgesDosSantos(ServicedeVirologieImmunologieCentreHospitalieretUniversitairedeFort-de-France,Fort-de-

France,Martinique)

• AnnaHearps,VickiEdouard,AdeleLee-WriedeandSuzanneCrowe(ClinicalResearchLaboratory,BurnetInstitutefor

MedicalResearchandPublicHealth,Melbourne,Australia)

• WendyStevens(CLSGenotypingLaboratory,JohannesburgGeneralHospital,Johannesburg,SouthAfrica)

• YasuhiroYamamura(AIDSResearchProgram-ImmunologyReferenceLaboratory,Ponce,PuertoRico)

• LeonMcNallyandPhilipCunningham(NSWStateReferenceLaboratoryforHIVandMolecularDiagnosticMedicine,

Sydney,Australia)

• JamesBrooks(NationalLaboratoryforHIVGenetics,PHAC,Ottawa,Canada)

• PatriciaPinsonandHerveFleury(LaboratoiredeVirologie,CHU,Bordeaux,France)

• RobSchuurman(DepartmentOfVirology,UniversityMedicalCenterUtrecht,Utrecht,Netherlands)

• CarmenMendozaandVincenteSoriano(InfectiousDiseasesDepartment,HospitalCarlosIII,Madrid,Spain)

• PatCaneandDeenanPillay(HealthProtectionAgency,London,UnitedKingdom)

WewouldalsoliketothankstafffromWHOoffices,includingEmilAsamoah-Odei,RichardBanda,FatimCham,Christopher

Dye,Guy-MichelGershy-Damet,NoreenJack,PatrickKombate-Noudjo,PaulAssimawePana,BrianPazvakavambwa,Giovanni

Ravasi,NirinaRazakasoa,PadminiSrikantiahandDongbaoYu.


5

A ttheendof2011,morethan8millionpeoplewerereceivingantiretroviraltherapyinlow-andmiddle-incomecountries,

adramatic26-foldincreasefromDecember2003.Althoughitcanbeminimized,somedegreeofHIVdrugresistance

isanticipatedtoemergeamongpeopleontreatmentevenwhenappropriateantiretroviraltherapyisprovidedand

highlevelsofadherenceareachieved.Therefore,WHOinitiatedglobalsurveillanceofHIVdrugresistancein2004inorderto

adequatelymonitortheemergenceofHIVdrugresistanceascountriesscaledupaccesstoantiretroviraltherapy.

ThisreportreviewsdataonHIVdrugresistanceinlow-andmiddle-incomecountriesbetween2003and2010andthreemain

conclusionsstandout.First,withtheexpansionoftreatmentachievedoverthelasteightyears,therearesignalsofincreasing

prevalenceoftransmittedHIVdrugresistance,particularlytonon-nucleosidereversetranscriptaseinhibitors(NNRTI),among

recentlyinfectedpopulationsintheareassurveyed.However,thoughincreasing,transmittedHIVdrugresistancehasnot

occurredatthehighlevelssomehadpredictedasaconsequenceoftherapidscale-upofantiretroviraltherapy.

Second,withrespecttoacquireddrugresistance,WHOsurveysindicatethat,ifpeopleareswitchedtosecond-lineregimens

soonaftervirologicalfailure,standardsecond-linetreatmentcombinationsarelikelytobeeffectiveforthemajorityofpatients

failingfirst-linetherapy.

Third,drugresistancesurveillanceprovidesimportantinformationontheeffectivenessofARTprogrammesandservices.

MonitoringofARTprogrammefunctioningthroughWHOHIVdrugresistanceearlywarning indicators in50countries

highlight the existence of important gaps in service delivery and programme performance, particularly with respect to

procurementandsupplysystems,adherenceandclinicretention.

AlthoughHIVdrugresistancedatafromlow-andmiddle-incomecountriesareincreasinglyavailable,lackofsurveillance

dataovertimesubstantiallylimitstheabilitytoassesstrendsinthesecountries.AsARTcoveragecontinuestogrow,national

programmesshouldperformroutinesurveillanceoftransmittedandacquiredHIVdrugresistancetooptimizeprogramme

planningandmanagementandtoinformantiretroviraltherapypolicy.

Drug resistance explained

HIVdrugresistancecanbecategorizedas:

• transmittedresistance,whichoccurswhenpreviouslyuninfectedindividualsareinfectedwithadrug-resistantvirus;and

• acquiredresistance,whichoccurswhenresistancemutationsemergebecauseofdrug-selectivepressureinindividuals

receivingantiretroviraltherapy.

It isessentialto implementroutinesurveillanceoftransmittedandacquiredHIVdrug

resistance. WHO transmitted drug resistance surveys alert programme managers to

the existence of drug-resistant HIV among recently infected populations in specific

geographicalareas.WHOsurveysofacquiredHIVdrugresistanceestimateprevalence

and patterns of resistance at treatment initiation, the proportion of people achieving

successful virological suppression at 12 months at sentinel sites and describe drug

resistanceinpopulationsexperiencingtreatmentfailure.1

MonitoringHIVdrugresistanceiscriticalforoptimalprogrammemanagementduetoits

importantpolicyimplications.DataonHIVdrugresistanceprovidethebasisforselecting

futurefirst-linetreatmentregimens,identifyingthemosteffectivesecond-linetherapiesfor

patientsfailingfirst-linecombinations,andforselectingoptimalapproachesforpreventing

mother-to-childtransmissionofHIVaswellasforpre-andpost-exposureprophylaxis.

1 WHOsurveystoassesstransmittedandacquireddrugresistancearenotintendedtobenationallyrepresentative.Additionally,areassurveyedmayvaryconsiderablyamongcountriesandacrosstime,sogeneralizationsmaynotbeappropriateorapplicable.

EXECUTIVE SUMMARY

Drug resistance in high-income countriesData suggest that 10–17% of ARV-naïve individuals treated in Australia, Japan, the United States of America and Europe are infected with virus resistant to at least one antiretroviral drug. These levels of drug resistance occurred early after antiretroviral therapy was introduced in many high-income countries in the late 1990s but have since plateaued. The proportion of people achieving treatment success (viral load suppression) has increased over time, thus reducing the emergence of acquired drug resistance and its subsequent transmission.

6

Transmitted drug resistance in low- and middle-income countries

Asystematicliteraturereviewsuggeststhattheprevalenceofdrugresistanceinselectlow-andmiddle-incomecountries

increasedbetween2003and2010,reachingapeakof6.6%in2009(95%confidenceinterval5.1%-8.3%).

PooledanalysisofdatafromWHOsurveys,whichtargetpeoplewhohavebeenrecentlyinfected,indicatesthatthereappears

tobeincreasinglevelsofresistancetoNNRTI,particularlyintheareassurveyedinAfrica,wheretheprevalenceofNNRTI

resistancereached3.4%(95%CI1.8%–5.2%)in2009.ThereisnoclearevidenceofincreasingHIVdrugresistancelevels

forotherdrugclasses.

Of72WHOsurveysoftransmitteddrugresistanceconductedbetween2004and2010,20(28%)wereclassifiedashaving

moderate(between5%and15%)prevalenceofresistance(Figure1).Theproportionofsurveyedareasreportingmoderate

levelsoftransmitteddrugresistanceincreasedfrom18%in2004-2006to32%in2007-2010(Table1).Thesefindings

deserveparticularattention.Ifconfirmedanddocumentedinmultipleareasofthesamecountry,immediateinvestigation

isrecommendedtounderstandtheirdeterminantsandpolicyimplications.

Figure 1 Geographical distribution of WHO surveys with moderate (between 5% and 15%) levels of drug resistance to any drug classa

Country reporting survey with moderate level of drug resistance to any class, 2007–2010

No data available or not participating in the survey

Not applicable

Country reporting survey with moderate level of drug resistance to any class, 2004-2010

No data available or not participating in the survey

Not applicable

2004–2010

a Areassurveyedvariedconsiderablyamongcountriesandovertime.

Number (%) of surveys with moderate (5–15%) prevalence

Year Totalsurveys Anydrugclass NNRTI NRTI PI

2004–2006 22 4(18%) 1(5%) 3(14%) 0(0%)

2007–2010 50 16(32%) 11(22%) 7(14%) 2(4%)

a Mid-pointperiod.

Table 1 Frequency of WHO surveys reporting moderate prevalence of transmitted HIV drug resistance, by period (before or after 2007)a


7

Availabledata–summarizedinFigure2–suggestthatthereisanassociationbetweenhigherlevelsofcoverageofantiretroviral

therapyandanincreasedprevalenceoftransmitteddrugresistancetoNNRTI,suchasnevirapineorefavirenz.However,

comparedwiththedramaticincreaseoftreatmentcoverage,theobservedriseinHIVdrugresistancewasmodest.This

impliesthattheexpansionofantiretroviraltherapydidnottriggerunexpectedincreasesintransmitteddrugresistancein

theareassurveyed.

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

0% 5% 10% 15% 20% 25% 30% 35% 40%

Prev

alen

ce o

f NN

RTI r

esis

tanc

e m

utat

ions

:%

of g

enot

ypes

Antiretroviral therapy coverage: % of people living with HIV receiving ART

Eastern AfricaSouthern AfricaWestern/Central AfricaSouth-East AsiaWestern PacificOther

Figure 2 Relationship between transmitted resistance to NNRTI drugs and antiretroviral therapy coverage

P-valueadjustedforregion=0.039;Odds-ratioper10%increaseinARTcoverage=1.49(95%C.I:1.07–2.08)

Acquired drug resistance in low- and middle-income countries

Accordingtodatafrom36WHOsurveysofacquiredHIVdrugresistanceassessingmorethan5000peoplein12low-and

middle-incomecountriesbetween2007and2010,theprevalenceofHIVdrugresistancetoanydrugamongpeoplestarting

antiretroviraltherapyrangedfrom4.8%(95%CI3.8%–6.0%)in2007to6.8%(95%CI4.8%–9.0%)in2010.

About90%ofpatientsaliveandontherapyat12monthsachievedtreatmentsuccess(viralloadsuppression).Amongpeople

withvirologicalfailure,72%hadresistance,mostlytonucleosidereversetranscriptaseinhibitors(NRTI)andNNRTIdrugs.The

remaining28%hadnoresistancemutationsandthereforeexperiencedtreatmentfailureforotherreasons,suchasverypoor

adherenceorextendedtreatmentinterruptions,andmayhavebeenswitchedtocostliersecond-lineregimensunnecessarily.

Theresistancepatternsobservedamongpeoplefailingfirst-linetreatmentafter12monthssuggestthat,ifthesepeoplewere

switchedtosecond-lineregimenssoonaftervirologicalfailure,standardsecond-linetherapies(consistingoftwonucleoside

classdrugsandaboostedproteaseinhibitor)wouldbeeffectiveinachievingviralloadsuppressioninthemajorityofcases.

Intotal,about18%ofthepeoplebeingtreatedwerelosttofollow-uporexperiencedtreatmentfailurewithoutresistance

mutationsbeingdetected.Giventhelikelihoodoftreatmentinterruptions,thesepeoplehaveahighprobabilityofharbouring

adrug-resistantvirus.Therefore,thetrueprevalenceofHIVdrugresistancecouldbeconsiderablyhigherthanthelevels

detectedinthesurveys.Thisunderscorestheneedforactivedefaultertracing,improvedpatientmonitoringandadherence

counselling.

8

Monitoring for prevention

KeytopreventingHIVdrugresistanceistheroutinemonitoringofprogrammaticfactorsknowntofavouritsemergence.

WHO’sglobalHIVdrugresistancesurveillanceandmonitoringstrategyrecommendsusingaminimumsetofHIVdrug

resistanceearlywarningindicatorsinalltreatmentsitestoidentifyfactorsknowntobeassociatedwithHIVdrugresistance

thatrequireimprovement,sothatcorrectiveactioncanbetakenattheclinicand/orprogrammelevel.Theindicatorsassess:

• howwellpopulationsareadherenttotherapy(on-timepillpick-up);

• whetherpharmaciesdispenseregimensthatarelikelytopromotetheemergenceofHIVdrugresistance,suchasmono-

ordualtherapy(dispensingpractices);

• whetherstock-outsofroutinelydispensedantiretroviralmedicinesoccur(drugsupplycontinuity);and

• theextenttowhichpeopleareretainedincareattheantiretroviralclinic-level.

Monitoringofanadditionalindicator,viralloadsuppressionat12months,isrecommendedatsiteswhereviralloadtesting

isroutinelyperformed.

Since 2004, 50 countries have piloted the monitoring of these indicators at select clinics. Although global trends or

conclusionscannotbeextrapolatedfromthesedata,aconsiderableproportionofclinicswerefoundtohaveimportantgaps

inservicedeliveryandprogrammeperformance,particularlywithrespecttoprocurementandsupplysystems,adherence

andclinicretention.

Conclusions

Asthecoverageofantiretroviraltherapycontinuestogrow,therearesignsofincreasedtransmitteddrugresistanceinthe

areassurveyed.Amongindividualsreceivingantiretroviraltherapy,acquireddrugresistancecontinuestohampertreatment

effectiveness.Nevertheless,availabledatasuggestthat,despitetherapidexpansionoftreatmentcoverage, increasesin

HIVdrugresistancehaveoccurredwithinexpectedlevelsintheareassurveyed,andnochangesinantiretroviraltreatment

guidelinesarewarrantedatthemoment.

Concertedactionisneededtopreservethefutureeffectivenessofantiretroviraltherapy.Treatmentprogrammesshould

monitorthequalityoftheservicestheydeliverbyusingtheearlywarningindicatorsforHIVdrugresistanceandundertaking

immediatecorrectiveactionwhenperformanceproblemsaredetected.Inaddition,programmesshouldperformroutine

surveillanceofHIVdrugresistanceamongpeopleexperiencingtreatmentfailureandamongrecently-infectedpopulations.

Despite intensive efforts, routine HIV drug resistance surveillance has not kept pace with the scale-up of treatment in

manycountries,limitingtheabilitytoreliablyassesstrendsovertime.WHO,throughitspartnernetworkofcollaborating

institutions,iscommittedtomonitoringHIVdrugresistancegloballyandtoadvocatingforexpandedroutinesurveillance

usingstandardizedmethodsandincreasedmobilizationofnationalandinternationalfundstosupportHIVdrugresistance

surveillance.


9

1.1 Overview

Combination antiretroviral therapy for HIV infection has

savedmillionsoflivessinceitwasintroduced.Ascoverage

of antiretroviral therapy continues to grow, some degree

of emergence and transmission of HIV drug resistance

is inevitable. Significant population-level HIV drug

resistance could potentially restrict future therapeutic

options and increase treatment costs by requiring new

andmoreexpensiveantiretroviral regimens.However,as

theexperiencesofmanycountriesdemonstrate,HIVdrug

resistance can be monitored and steps can be taken to

minimizeitsemergence.

Insimpleterms,HIVdrugresistancerefers totheability

of HIV to replicate in the presence of drugs that usually

suppress its replication. HIV drug resistance is caused

by changes (mutations) in the virus’s genetic structure.

Mutations are very common in HIV because the virus

replicatesveryrapidlyanddoesnotcontain theproteins

neededtocorrectthemistakesitmakesduringthisprocess.

Assuch,somedegreeofHIVdrugresistanceisanticipated

to occur among people receiving treatment even when

appropriateregimensareprovidedandoptimaladherence

is achieved (1). Transmitted HIV drug resistance occurs

when previously uninfected individuals are infected with

drug-resistant virus, and acquired HIV drug resistance

developswhenmutationsemergeduetoviralreplication

inindividualsreceivingantiretroviraltherapy.

Thisreportaimstogenerallyassessthelevelsoftransmitted

and acquired drug resistance in select geographical

areas of low- and middle-income countries. It is based

on two distinct data sources: surveys performed to

assess transmitted and acquired drug resistance using

standardizedWHOmethods(WHOsurveys)andabroad

systematicreviewofthepublishedliteratureontransmitted

andacquireddrugresistance.Findingsfromthemonitoring

ofearlywarningindicatorsofHIVdrugresistancearealso

presentedanddiscussed.

1.2 Structure of the report

Thisreportisorganizedasfollows.

Chapter 1outlinestheobjectivesofthereport,discusses

the determinants of HIV drug resistance, and describes

the WHO global HIV drug resistance surveillance and

monitoringstrategy.

1. INTRODUCTION

Chapter 2providesanoverviewofHIVdrugresistancein

high-incomecountries.

Chapter 3discussesasystematicreviewandmeta-analysis

ofthepublishedliteratureonlevelsandtrendsoftransmitted

drugresistanceinselectareasoflow-andmiddle-income

countriesandpresentsdata fromsurveysof transmitted

drugresistanceconductedaccordingtostandardizedWHO

methods.

Chapter 4discussesasystematicreviewoflevelsofacquired

drug resistance in patients failing first-line antiretroviral

therapy inselect low-andmiddle-incomecountries,and

presents data from surveys of acquired drug resistance

conductedaccordingtostandardizedWHOmethods.

Chapter 5presentsfindingsfromthemonitoringofearly

warningindicatorsofHIVdrugresistance.

Chapter 6discussesoverallconclusions.

Annex 1presentsdetailednotesonthemethodsusedto

generateandinterpretthedatacontainedinthisreport.

Annex 2providessupplementaldataandtablesfromWHO

HIVdrugresistancesurveys.

1.3 Determinants of HIV drug resistance1

FactorscontributingtotheselectionofHIVdrugresistance

canbebroadlygroupedintofourcategories:1regimen-

anddrug-specific,2virus-related,3patient-specificand

4programmatic.

1.3.1 Regimen- and drug-specific factorsThe genetic barrier of an antiretroviral therapy regimen,

defined as the number of key mutations required to

overcomedrug-selectivepressure, isan important factor

intheemergenceofHIVdrugresistance.First-lineregimens

recommendedbyWHOforadultsandadolescentstypically

includeonenon-nucleosidereverse-transcriptaseinhibitor

(NNRTI), either nevirapine (NVP) or efavirenz (EFV),

combinedwithtwonucleosidereverse-transcriptase(NRTI)

backbonedrugs,typicallyzidovudineortenofovir,combined

witheitherlamivudineoremtricitabine(3).

1 Thissectionreliesextensivelyon:Bertagnolioetal.(2).

10

Although the efficacy of these regimens has been well

established(4–7) inbothhigh-incomeandlow-andmiddle-

incomecountries(8–10),arecognizedlimitationofNNRTI-

based regimens is their relatively lower genetic barrier

to resistance compared with regimens using boosted

protease inhibitors (bPI) in place of non-nucleosides.

AlthoughNNRTI-basedregimensdifferwithrespecttotheir

geneticbarrier,whichis influencedbytheaccompanying

NRTI component (5,11–13), they select significantly more

resistance than bPI-based regimens among people

experiencing treatment failure despite similar rates of

virologicalsuppression(14,15).

Suboptimal regimens, such as single-dose nevirapine for

preventingmother-to-childtransmission,andinappropriate

prescribingpracticesresultingintheuseofsingleandtwo-

drugantiretroviraltherapyregimens,canfurtherincrease

theriskofdevelopingHIVdrugemergence (16).

Interactions between drugs can favour the selection of

HIV drug resistance by reducing the concentration of

antiretroviral drugs to suboptimal levels. Rifampicin, for

example,hasbeenshowntoreducethelevelsofnevirapine

between 20% and 58% and efavirenz by 26% (17,18). In

addition,populationsexposedtoantiretroviraldrugsbefore

initiationfirst-lineantiretroviraltherapyarealsomorelikely

to carry pre-treatment resistance (19), leading to more

rapidvirologicalfailureandfurtheracquisitionofHIVdrug

resistance(20,21).

Theuseofcomplexregimenswithahighpillburdenalso

reduces adherence, thus favouring the selection of HIV

drug resistance (22,23). In contrast, the use of fixed-dose

combinations can improve adherence, facilitate rational

prescribingandstreamlinedrugprocurement(23).

1.3.2 Virus factorsEvidenceofpre-treatmentHIVdrugresistanceisstrongly

associatedwithvirological failureand furtheracquisition

of resistance after the first year of NNRTI-based first-

line antiretroviral therapy (20,24,25). Research has shown

thatindividualswithtransmittedHIVdrugresistanceare

expectedtoaccumulatemoreNRTIresistanceatthetime

of virological failure, leading to a growing number being

treatedwithaboostedPIandtwoNRTIwithpartialorno

activityatthetimeofswitchtosecond-linetherapy(26).

Moreover, the frequency and characteristics of mutation

patternsmayalsodifferacrossvirussubtypes.Forexample,

afterexposure tosingle-dosenevirapine,moreHIVdrug

resistanceisobservedinHIV-1subtypeDthaninsubtypeA

(27). Recent data suggest that increased rates of K65R

acquisitioninHIV-1subtypeCmaybecausedbythenature

ofthesubtypeCRNAtemplate (28).

1.3.3 Patient factorsAdherence toantiretroviral therapy iswell recognizedas

an essential component of individual and programmatic

treatmentsuccess.Pooradherencetoantiretroviraltherapy

is a predictor of virological failure (29–33), emergence of

HIVdrugresistance,diseaseprogression(34–36) anddeath

(37–39).Hence,sustainedscaleupofantiretroviraltherapy

dependsontheabilityofprogrammestodelivercareina

waythatminimizestreatmentinterruptionsthroughdrug

supplycontinuityandmaximizesadherence.

At the individual level, studies suggest that untreated

depression, active substance abuse, poor insight into

diseaseandtreatment,beinganadolescentoryoungadult,

ahigherpillburden,morefrequentdosingandforgetfulness

are associated with poor adherence (40). Adherence can

beespeciallychallengingamongchildren foravarietyof

reasons,includingdrugformulationsandpalatability(41–43).In addition, children are at greater risk of acquiring drug

resistance,sincetheyoftendependoncaregiversfortheir

treatment (44). If caregivers are themselves unwell, they

maynotbeabletoattendclinicvisitswiththechild,collect

medicationasneededorprovidethemwiththeirmedication

onschedule.OrphanslivingwithHIVfrequentlyfacethe

greatest challenges in terms of adherence. Although

orphans in institutional care typically have high levels of

adherence (since trained caregivers often provide care),

thosewhoareraised in thehouseholdsof relativeshave

pooreroutcomesandaremorelikelytodefaultorbenon-

adherenttocare(45).

Adherence may also be negatively affected by HIV-

associated stigma and discrimination (46,47). Notably,

peoplelivingwithHIVmayfearthattakingmedicationin

thepresenceofothersmayinadvertentlydisclosetheirHIV

status,thusdeterringthemfromadequatelyfollowingthe

regimensprescribed(48).

1.3.4 Programmatic factorsProgramme-levelfactors,suchaslimitedhumanresources,

inadequate infrastructure and weak supply management

systems, can also negatively affect treatment adherence

and retention in care and facilitate the emergence of

population-levelHIVdrugresistance.

The provision of chronic HIV care is still a challenge for

mosthealthsystemsinlow-andmiddle-incomecountries

as it requires robust and integrated systems to support

adherenceandtraceindividualswithunknowntreatment


11

outcomes.Overcrowdingandunderstaffingofantiretroviral

therapy clinics may further aggravate these constraints

by reducing the time dedicated for counselling and the

reinforcementofadherencemessages.Researchsuggests

thatreducingthequalityandintensityofpatientmonitoring

by antiretroviral therapy clinics may decrease retention,

leading to higher proportions of treatment interruptions

andmorepeoplewithunknowntreatmentoutcomes(49).

Moreover, fragile drug procurement and supply

management systems can result in drug stock-outs

and missed antiretroviral drug doses (50–52). Cost and

structural barriers, such as food insecurity and out-of-

pocketexpenditurefortransportormonitoringtests,can

equally lead to treatment interruptions and suboptimal

adherence(53,54).

The absence of routine viral load monitoring, which

is a more sensitive indicator of treatment failure than

clinical and immunological parameters, may lead some

people to experience prolonged periods of virological

failure before changing regimen (3,55). Although current

modelling of antiretroviral therapy effectiveness has not

reachedaconsensuswithrespecttotheimplementationof

systematicviralloadmonitoringinlow-andmiddle-income

countries (56–59),maintainingpeopleonafailingNNRTI-

basedregimensleadstotheaccumulationofmultipleNRTI

mutations (60).

1.4 WHO’s global HIV drug resistance surveillance and monitoring strategy

Understanding the emergence and transmission of

population-level HIV drug resistance and the interaction

between its various determinants require routine

surveillance, monitoring and evaluation, and operational

research.

Toadequatelymonitortheemergenceofdrugresistance,

WHO spearheaded the establishment of the global HIV

DrugResistanceNetwork(HIVResNet),comprisedofmore

than 50 international institutions, experts and national

HIV programme representatives. In collaboration with

HIVResNet and with support from the Bill & Melinda

Gates Foundation, WHO developed a global HIV drug

resistance surveillance and monitoring strategy (61). The

strategywasdesigned to informdecision-makingon the

optimal choice of antiretroviral regimens and to identify

any programmatic adjustments needed to minimize the

emergenceofHIVdrugresistance.Thestrategyhasthree

mainassessmentelements:(1)surveillanceoftransmitted

HIVdrugresistance in recently infectedpopulations, (2)

surveillanceofacquiredHIVdrugresistanceinpopulations

receivingantiretroviraltherapyand(3)monitoringofearly

warningindicatorsofHIVdrugresistance(61).

1. Surveillance of transmitted HIV drug resistance in

recently infected populations (62) (Chapter 3). The

WHO survey method for assessing transmitted HIV

drugresistanceclassifiesresistanceprevalenceaslow

(below5%),moderate(between5%and15%)orhigh

(over15%)inrecentlyinfectedpopulationsinaspecific

geographical area (62,63). Whenever possible, these

surveys use remnant specimens from the populations

of interest (e.g., young pregnant women) and data

from regularly performed serosurveys that estimate

HIV prevalence, which are already in place in most

countries.TransmittedHIVdrugresistancesurveysalert

programme planners to the existence of transmission

of drug-resistant HIV, and the results may inform the

selection of current regimens for preventing mother-

to-childtransmissionandfuturefirst-lineantiretroviral

therapyregimens.

2. Surveillance of acquired HIV drug resistance in

populations receiving antiretroviral therapy (64)(Chapter4).WHOprospectivesurveysofacquiredHIV

drugresistanceareperformedatsentinelantiretroviral

therapy clinics and estimate the prevalence and

patternsofHIVdrugresistanceinadultandpaediatric

populations experiencing antiretroviral therapy failure

(64). At each sentinel survey clinic, a cohort of people

initiatingfirst-linetherapyisformed.HIVdrugresistance

genotypingisperformedonpeopleinitiatingantiretroviral

therapy, and HIV-RNA is quantified at the time that

treatmentisswitchedtosecond-lineor12monthsafter

antiretroviral therapy initiation for people remaining

on first-line treatment. For people with detectable

virus (more than 1000 copies/ml), genotyping is

performed to characterize drug resistance mutations.

A threshold of 1000 copies/ml has been chosen to

characterizetreatmentfailurebecauseofthesensitivity

andreproducibilityofstandardcommercialgenotyping

assays.Surveyresultsprovidesite-specificassessments

ofviralloadsuppression,whichareparticularlyrelevant

to clinics and programmes in which viral load is not

routinelyperformed.1

1 Thedatapresentedhereinwereobtainedfromtheimplementationofthisprotocol.However,across-sectionalapproachtoassessingacquireddrugresistancehasbeendevelopedandiscurrentlybeingpilotedinNamibia(seeSection7inAnnex1).

12

Monitoring of early warning indicators for HIV drug

resistance(61)(Chapter5).Earlywarningindicatorsmonitor

factors at individual clinics known to create situations

favourable to the emergence of HIV drug resistance.

Withoutrequiringdrugresistancetesting,themonitoring

of early warning indicators provides the context for

interpreting the results from surveys of transmitted and

acquired HIV drug resistance. The timely identification

of clinics with suboptimal performance helps tailor

appropriate interventions that can potentially optimize

careandtreatmentandreducetheriskofpopulation-level

HIVdrugresistanceemergence.

Inadditiontothesethreekeyassessmentelements,WHO

has developed a comprehensive HIV drug resistance

laboratorystrategy,whichincludeslaboratorymembership

and rigorous quality assurance of genotyping data to

supportpublichealthsurveillance(65). Asof2011,27testing

laboratories for HIV drug resistance had been granted

membership(Figure1.1).

1.5 Note on data sources and methods

Aggregatelevelsandtrendsdiscussedinthemeta-analyses

performed based on data from published studies on

transmittedandacquireddrugresistance(excludingWHO

surveys)shouldbeconsideredinlightoftheheterogeneity

of study methods and countries assessed. Many of the

studies included were performed using distinct methods

Figure 1.1 HIV drug resistance testing laboratories designated for public health surveillance by the WHO, 2011

^^

^

^

^^^^

^ ^

^̂̂

^^

^

^

^

^ ^

^

^

^̂

^

^

^

^

^

^

^^^

^ Laboratories undergoing assessment

^ Laboratories designated by WHO for HIV drug resistance surveillance

Not applicable

and may differ with respect to the population assessed

(such as recent or chronic infections), sampling frame

(suchasconsecutive,convenientorrandomselectionfrom

thegeneralpopulation)andthelaboratorymethodsused

(such as dried blood spot or plasma specimens and the

genotypingmethodsused).

Individual studies may also have been influenced by

factors such as antiretroviral therapy coverage, variation

in HIV subtypes, quality of care at antiretroviral therapy

programmes and sites, country income levels and the

structure or organization of health services. Therefore,

prevalenceestimatesmaynotbenationallyor regionally

representative. Moreover, studies included in the meta-

analyses reported resistance data according to any of

the internationally recognized drug resistance mutation

lists. Therefore variation in how mutations were defined

mayhave influenced individual study resultsand,hence,

aggregateanalyses.Thismaybethecaseparticularlyfor

estimatesofPIresistance.

The number of low- and middle-income countries with

availabledataonHIVdrugresistanceremainslimited.This

impliesthattheresultsandconclusionspresentedinthis

report may be biased towards programmes with above-

average performance, as the implementation of surveys

and/orstudiesondrugresistancecanitselfindicateabove-

averageconcernwithprogrammaticqualityandtreatment

success.


13

AlthoughearlywarningindicatorsofHIVdrugresistance

were designed to be nationally representative, reported

resultsreflectapilotandscale-upphaseandaretherefore

unlikelytobetypicalofacountry’santiretroviraltreatment

programme functioning. WHO surveys to assess

transmittedandacquireddrugresistancearenotintended

tobenationallyrepresentative.Additionally,areassurveyed

varied considerably, among countries and across time,

so generalizations may not be appropriate or applicable.

Nevertheless,theirresultsshouldbeinterpretedasanalert

toprogrammemanagersthatresistancetransmissionand

acquisitionareoccurringinspecificgeographicalareasof

a country and that, depending on observed levels, wider

policyactionmaybewarranted.

Regionsrefer,unlessotherwisenoted,toWHO’sstandard

regionalgrouping (66).Asthisreportfocusesonlow-and

middle-income countries, the term “Latin America and

theCaribbean”isusedinsteadofRegionoftheAmericas.

Forthepurposesofthisreport,subregionalgroupingsfor

theWHOAfricanRegion(central,eastern,southernand

western Africa) are used to highlight, when appropriate,

patterns applicable or specific to a subset of countries

(Section 1 in Annex 1 provides detailed sub-regional

countrygrouping).Asia refers tocountries in theSouth-

EastAsiaRegionandWesternPacificRegioncombined.All

confidenceintervalsquotedareatthe95%level.

14

REFERENCES

1. RichmanDDetal.TheprevalenceofantiretroviraldrugresistanceintheUnitedStates.AIDS,2004,18:1393–1401.

2. BertagnolioSetal.DeterminantsofHIVdrugresistanceandpublichealthimplicationsinlow-andmiddle-incomecountries.AntiviralTherapy,inpress.

3. AntiretroviraltherapyforHIVinfectioninadultsandadolescents:recommendationsforapublichealthapproach.2010revision.Geneva,WorldHealthOrganization,2010(http://www.who.int/hiv/pub/arv/adult2010/en/index.html,accessed28June2012).

4. Gallant JE et al. Efficacy and safety of tenofovir DF vs stavudine in combination therapy in antiretroviral-naive patients: a 3-yearrandomizedtrial.JAMA,2004,292:191–201.

5. SoriaAetal.Resistanceprofilesafterdifferentperiodsofexposuretoafirst-lineantiretroviralregimeninaCamerooniancohortofHIVtype-1–infectedpatients.AntiviralTherapy,2009,14:339–347.

6. GallantJEetal.TenofovirDF,emtricitabine,andefavirenzvs.zidovudine, lamivudine,andefavirenzforHIV.NewEnglandJournalofMedicine,2006,354:251–260.

7. SmithCJetal.TherateofviralreboundafterattainmentofanHIVload<50copies/mlaccordingtospecificantiretroviraldrugsinuse:resultsfromamulticentercohortstudy.JournalofInfectiousDiseases,2005,192:1387–1397.

8. BraitsteinPetal.MortalityofHIV-1-infectedpatientsinthefirstyearofantiretroviraltherapy:comparisonbetweenlow-incomeandhigh-incomecountries.Lancet,2006,367:817–824.

9. NachegaJBetal.Efavirenzversusnevirapine-basedinitialtreatmentofHIVinfection:clinicalandvirologicaloutcomesinsouthernAfricanadults.AIDS,2008,22:2117–2125.

10. PalombiLetal.Incidenceandpredictorsofdeath,retention,andswitchtosecond-lineregimensinantiretroviral-treatedpatientsinsub-SaharanAfricansiteswithcomprehensivemonitoringavailability.ClinicalInfectiousDiseases,2009,48:115–122.

11. MargotNAetal.DevelopmentofHIV-1drugresistancethrough144weeksinantiretroviral-naivesubjectsonemtricitabine,tenofovirdisoproxilfumarateandefavirenzcomparedwithlamivudine/zidovudineandefavirenzinstudyGS-01-934.JournalofAcquiredImmuneDeficiencySyndromes,2009,52:209–221.

12. MaseratiRetal.EmergingmutationsatvirologicalfailureofHAARTcombinationscontainingtenofovirandlamivudineoremtricitabine.AIDS,2010,24:1013–1018.

13. SvicherVetal.Differentevolutionofgenotypicresistanceprofilestoemtricitabineversuslamivudineintenofovir-containingregimens.JournalofAcquiredImmuneDeficiencySyndromes,2010,55:336–344.

14. Cozzi-LepriAetal.Therateofaccumulationofnonnucleosidereversetranscriptaseinhibitor(NNRTI)resistanceinpatientskeptonavirologicallyfailingregimencontaininganNNRTI.HIVMedicine,2012,13:62–72.

15. GuptaRetal.EmergenceofdrugresistanceinHIVtype1-infectedpatientsafterreceiptoffirst-linehighlyactiveantiretroviraltherapy:asystematicreviewofclinicaltrials.ClinicalInfectiousDiseases,2008,47:712–722.

16. ChomatAMetal.Knowledge,beliefs,andhealthcarepracticesrelatingtotreatmentofHIVinVellore,India.AIDSPatientCareandSTDs,2009,23:477–484.

17. MaartensG,DecloedtE,CohenK.Effectivenessandsafetyofantiretroviralswithrifampicin:crucialissuesforhigh-burdencountries.AntiviralTherapy,2009,14:1039–1043.

18. Boulle A et al. Outcomes of nevirapine- and efavirenz-based antiretroviral therapy when coadministered with rifampicin-basedantituberculartherapy.JAMA,2008,300:530–539.

19. AndreottiMetal.ResistancemutationpatternsinplasmaandbreastmilkofHIV-infectedwomenreceivinghighly-activeantiretroviraltherapyformother-to-childtransmissionprevention.AIDS,2007,21:2360–2362.

20. WittkopLetal.EffectoftransmitteddrugresistanceonvirologicalandimmunologicalresponsetoinitialcombinationantiretroviraltherapyforHIV(EuroCoord-CHAINjointproject):aEuropeanmulticohortstudy.LancetInfectiousDiseases,2011,377:1580–1587.

21. JordanMR.AssessmentsofHIVdrugresistancemutationsinresource-limitedsettings.ClinicalInfectiousDiseases,2011,52:1058–1060.

22. MaggioloFetal.Once-a-daytherapyforHIVinfection:acontrolled,randomizedstudyinantiretroviral-naiveHIV-1-infectedpatients.AntiviralTherapy,2003,8:339–346.

23. JudayTetal.FactorsassociatedwithcompleteadherencetoHIVcombinationantiretroviraltherapy.HIVClinicalTrials,2011,12:71–78.

24. SigaloffKCetal.HighprevalenceoftransmittedantiretroviraldrugresistanceamongnewlyHIVtype1diagnosedadultsinMombasa,Kenya.AIDSResearchandHumanRetroviruses,2012[Epubaheadofprint].

25. HamersRLetal.EffectofpretreatmentHIV-1drugresistanceonimmunological,virological,anddrug-resistanceoutcomesoffirst-lineantiretroviraltreatmentinsub-SaharanAfrica:amulticentrecohortstudy.LancetInfectiousDiseases,2011,12:307–317.

26. BartlettJAetal.Lopinavir/ritonavirmonotherapyaftervirologicfailureoffirst-lineantiretroviraltherapyinresource-limitedsettings.AIDS,2012,Epub2012/03/24.

27. HauserAetal.Emergenceandpersistenceofminordrug-resistantHIV-1variants inUgandanwomenafternevirapinesingle-doseprophylaxis.PLoSOne,2011,6:e20357.

28. McCollDJetal.Prevalence,genotypicassociationsandphenotypiccharacterizationofK65R,L74VandotherHIV-1RTresistancemutationsinacommercialdatabase.AntiviralTherapy,2008,13:189–197.

29. LucasGM,ChaissonRE,MooreRD.Highlyactiveantiretroviraltherapyinalargeurbanclinic:riskfactorsforvirologicfailureandadversedrugreactions.AnnalsofInternalMedicine,1999,131:81–87.

30. NachegaJBetal.Adherencetononnucleosidereversetranscriptaseinhibitor-basedHIVtherapyandvirologicoutcomes.AnnalsofInternalMedicine,2007,146:564–573.


15

31. ArnstenJHetal.Antiretroviral therapyadherenceandviralsuppression inHIV-infecteddrugusers:comparisonofself-reportandelectronicmonitoring.ClinicalInfectiousDiseases,2001,33:1417–1423.

32. ShuterJetal.HIV-infectedpatientsreceivinglopinavir/ritonavir-basedantiretroviraltherapyachievehighratesofvirologicsuppressiondespiteadherencerateslessthan95%.JournalofAcquiredImmuneDeficiencySyndromes,2007,45:4–8.

33. MartinMetal.Relationshipbetweenadherence level, typeof theantiretroviral regimen,andplasmaHIVtype1RNAviral load:aprospectivecohortstudy.AIDSResearchandHumanRetroviruses,2008,24:1263–1268.

34. BangsbergDRetal.Adherence-resistancerelationshipsforproteaseandnon-nucleosidereversetranscriptaseinhibitorsexplainedbyvirologicalfitness.AIDS,2006,20:223–231.

35. HarriganPRetal.PredictorsofHIVdrug-resistancemutationsinalargeantiretroviral-naivecohortinitiatingtripleantiretroviraltherapy.JournalofInfectiousDiseases,2005,191:339–347.

36. BangsbergDRetal.Non-adherencetohighlyactiveantiretroviraltherapypredictsprogressiontoAIDS.AIDS,2001,15:1181–1183.

37. HoggRSetal.Intermittentuseoftriple-combinationtherapyispredictiveofmortalityatbaselineandafter1yearoffollow-up.AIDS,2002,16:1051–1058.

38. NachegaJBetal.AdherencetohighlyactiveantiretroviraltherapyassessedbypharmacyclaimspredictssurvivalinHIV-infectedSouthAfricanadults.JournalofAcquiredImmuneDeficiencySyndromes,2006,43:78–84.

39. WoodEetal.ImpactofbaselineviralloadandadherenceonsurvivalofHIV-infectedadultswithbaselineCD4cellcounts≤200cells/ml.AIDS,2006,20:1117–1123.

40. MillsEJetal.Adherencetoantiretroviraltherapyinsub-SaharanAfricaandNorthAmerica:ameta-analysis.JAMA,2006,296:679–690.

41. FosterCetal.YoungpeopleintheUnitedKingdomandIrelandwithperinatallyacquiredHIV:thepediatriclegacyforadultservices.AIDSPatientCareandSTDs,2009,23:159–166.

42. TempleME,KoranyiKI,NahataMC.GastrostomytubeplacementinnonadherentHIV-infectedchildren.AnnalsofPharmacotherapy,2001,35:414–418.

43. KingJRetal.PharmacokineticsofantiretroviralsadministeredtoHIV-infectedchildrenviagastrostomytube.HIVClinicalTrials,2004,5:288–293.

44. NahiryaNtegePetal.TabletsaremoreacceptableandgivefewerproblemsthansyrupsamongyoungHIV-infectedchildreninresourcelimitedsettingsintheARROWtrial.2ndInternationalWorkshoponHIVPaediatrics,16–17July2010andXVIIIInternationalAIDSConference,18–23July2010,Vienna,Austria(http://pag.aids2010.org/Abstracts.apsx?AID=13055,accessed28June2012).

45. NyandikoWMetal.OutcomesofHIV-infectedorphanedandnon-orphanedchildrenonantiretroviraltherapyinwesternKenya.JournalofAcquiredImmuneDeficiencySyndromes,2006,43:418–425.

46. RintamakiLSetal.SocialstigmaconcernsandHIVmedicationadherence.AIDSPatientCareandSTDs,2006,20:359–368.

47. NachegaJBetal.AdherencetoantiretroviraltherapyinHIV-infectedadultsinSoweto,SouthAfrica.AIDSResearchandHumanRetroviruses,2004,20:1053–1056.

48. TamabcVetal.“It isnotthatI forget, it’s justthatIdon’twantotherpeopletoknow”:barrierstoandstrategiesforadherencetoantiretroviraltherapyamongHIVpatientsinnorthernVietNam.AIDSCare:PsychologicalandSocio-medicalAspectsofAIDS/HIV,2011,23:139–145.

49. CornellMetal.TemporalchangesinprogrammeoutcomesamongadultpatientsinitiatingantiretroviraltherapyacrossSouthAfrica,2002–2007.AIDS,2010,24:2263–2270.

50. OyugiJHetal.TreatmentinterruptionspredictresistanceinHIV-positiveindividualspurchasingfixed-dosecombinationantiretroviraltherapyinKampala,Uganda.AIDS,2007,21:965–971.

51. MarcellinFetal.DeterminantsofunplannedantiretroviraltreatmentinterruptionsamongpeoplelivingwithHIVinYaounde,Cameroon(EVALsurvey,ANRS12-116).TropicalMedicineandInternationalHealth,2008,13:1470–1478.

52. EholieSPetal.FieldadherencetohighlyactiveantiretroviraltherapyinHIV-infectedadultsinAbidjan,Côted’Ivoire.JournalofAcquiredImmuneDeficiencySyndromes,2007,45:355–358.

53. CraneJTetal.Thepriceofadherence:qualitativefindingsfromHIVpositiveindividualspurchasingfixed-dosecombinationgenericHIVantiretroviraltherapyinKampala,Uganda.AIDSBehavior,2006,10:437–442.

54. WeiserSetal.BarrierstoantiretroviraladherenceforpatientslivingwithHIVinfectionandAIDSinBotswana.JournalofAcquiredImmuneDeficiencySyndromes,2003,34:281–288.

55. KeiserOetal.Switchingtosecond-lineantiretroviraltherapyinresource-limitedsettings:comparisonofprogrammeswithandwithoutviralloadmonitoring.AIDS,2009,23:1867–1874.

56. PhillipsANetal.EffectontransmissionofHIV-1resistanceoftimingofimplementationofviralloadmonitoringtodetermineswitchesfromfirsttosecond-lineantiretroviralregimensinresource-limitedsettings.AIDS,2011,25:843–850.

57. FoxMPetal.Highratesofsurvival,immunereconstitution,andvirologicsuppressiononsecond-lineantiretroviraltherapyinSouthAfrica.JournalofAcquiredImmuneDeficiencySyndromes;2010,53:500–506.

58. Pujades-RodriguezMetal.Treatmentfailureandmortalityfactorsinpatientsreceivingsecond-lineHIVtherapyinresource-limitedcountries.JAMA,2010,304:303–312.

59. HamersRLetal.Cost-effectivenessoflaboratorymonitoringformanagementofHIVtreatmentinsub-SaharanAfrica:amodel-basedanalysis.AIDS,2012[Epubaheadofprint].

60. HosseinipourMCetal.Thepublichealthapproachtoidentifyantiretroviraltherapyfailure:high-levelnucleosidereversetranscriptaseinhibitorresistanceamongMalawiansfailingfirst-lineantiretroviraltherapy.AIDS,2009,23:1127–1134.

16

61. BennettDEetal.TheWorldHealthOrganization’sglobalstrategyforpreventionandassessmentofHIVdrugresistance.AntiviralTherapy,2008,13(Suppl.2):1–13.

62. BennettDEetal.RecommendationsforsurveillanceoftransmittedHIVdrugresistanceincountriesscalingupantiretroviraltreatment.AntiviralTherapy,2008,13(Suppl.2):25–36.

63. MyattM,BennettDE.AnovelsequentialsamplingtechniqueforthesurveillanceoftransmittedHIVdrugresistancebycross-sectionalsurveyforuseinlowresourcesettings.AntiviralTherapy,2008,13(Suppl.2):37–48.

64. JordanMRetal.WorldHealthOrganizationsurveys tomonitorHIVdrugresistancepreventionandassociated factors insentinelantiretroviraltreatmentsites.AntiviralTherapy,2008,13(Suppl.2):15–23.

65. BertagnolioSetal.WorldHealthOrganization/HIVRestNetDrugResistanceLaboratoryStrategy.AntiviralTherapy,2008,13(Suppl.2):49–57).

66. WHO–itspeopleandoffices[website].Geneva,WorldHealthOrganization,2012(http://www.who.int/about/structure/en/index.html,accessed28June2012).


17

H ighlyactiveantiretroviraltherapyhasbeenavailable

inmosthigh-incomecountriessinceitsintroduction

inthelate1990s.AnotherfeatureofHIVtreatment

programmesinhigh-incomecountriesisthewidespreaduse

ofdrugresistancegenotypingtosupportcasemanagement

and treatment monitoring. Despite important structural

andsocioeconomicdifferences, theirexperiencescanbe

informativeforlow-andmiddle-incomecountriesscaling

upaccesstoantiretroviraltherapy.

2.1 Drug resistance in ARV-naïve recently or chronically-infected populations

Availabledatasuggestthatbetween10%and17%ofARV-

naïve individuals inEurope, theUnitedStates,Japanand

Australiahavedrugresistancetoatleastoneantiretroviral

drug.

In Europe, a comprehensive review of 75 studies on

transmitted drug resistance published in 2009 covering

23209peoplefrom20countriesestimatedtheprevalence

oftransmitteddrugresistanceat10.9%.Drugresistance

mostfrequentlyinvolvedNRTI,withaprevalenceof7.4%.

The prevalence for NNRTI and PI was 3.4% and 2.9%,

respectively(1).1Levelsoftransmitteddrugresistanceseem

tohavedeclinedsignificantlyovertime,from11.5%between

1985 and 2003 to 7.7% between 2004 and 2009. This

reduction was largely caused by drops in the levels of

NRTIresistance,from8.0%to4.3%,andofPIresistance,

from3.3%to1.4%.Incontrast, theprevalenceofNNRTI

resistancechangedonlyslightlyoverthesameperiod,from

2.9%to3.2%.Astudyof25cohortsfromacrossEurope

1 Moststudiesdonotdifferentiatebetweenrecentlyorchronicallyinfectedindividuals.

2. HIV DRUG RESISTANCE IN HIGH-INCOME COUNTRIES

between1998and2009foundbroadlysimilarresults.Ina

groupof10056antiretroviraltherapy–naivepeople,954,or

9.5%,hadatleastonedrugresistancemutation(2).

In the United States, a study from the Center for AIDS

Research(NationalInstitutesofHealth)with14111people

covering the period from before 2003 through 2008

reported an overall genotypic resistance prevalence of

14.2%toatleastonedrug(8.3%NNRTI,8.2%NRTI,4.2%

PI)(3).InthestatesofWashingtonandColorado,theoverall

prevalenceofdrugresistancewas17%(11%NNRTI,6%

NRTI,3%PI)among506peoplewithrecentorestablished

HIVinfection (4),whileinSanFrancisco,16%of372people

diagnosedbetween2002and2009withacuteorearlyHIV

infectionhaddrugresistancetoatleastoneantiretroviral

drug(5).In2006,among2030newlydiagnosedindividuals

from10statesand1countyhealthdepartmentintheUnited

States, mutations associated with HIV drug resistance

werefoundin14.6%(NNRTI7.8%,NRTI5.6%,PI4.5%).

Abroaderreviewof45studiesconductedbetween1993

and2008(42intheUnitedStatesand3inCanada)found

that, among 8718 people, about 12.9% carried HIV drug

resistance.ResistancetoNRTI,at7.4%,wasobservedto

behighest, followedbyresistancetoNNRTIandPI,with

5.7%and3.2%,respectively(1).IncontrasttoEurope,the

review suggests that prevalence of HIV drug resistance

mayhave increasedinNorthAmericafrom11.6%before

2001to14.3%after2003,drivenlargelybytheincreasein

NNRTIresistance,from4.1%to8.3%,withNRTIresistance

decreasingfrom8.0%to6.4%.

In Japan, the prevalence of drug resistance mutations

in people newly diagnosed with HIV-1 infection doubled

from5.9%in2003to11.9%in2010(6,7),andtherelative

prevalenceofresistancebydrugclasschangedconsiderably

KEY FINDINGS

• Available data suggest that between 10% and 17% of ARV-naïve patients in Europe, United States, Japan and

Australiahavedrugresistancetoatleastoneantiretroviral.

• Withrespecttoacquireddrugresistance,evidenceindicatesthat(i)theproportionofpatientsachievingfullviral

suppressionhasincreasedovertime,thusminimizingtheemergenceofacquireddrugresistanceanditssubsequent

transmissionandthat(ii)resistancetoNRTIisthemostfrequentlyobservedtypeinpatientsfailingantiretroviral

therapy,followedbyNNRTIandPIresistance.

18

overthisperiod.Before2007,resistancetoNRTIwashigher

thantoNNRTIandPI,butsince2007resistancetoPIhas

becomemostprevalent,reaching4.9%in2010.Incontrast

toreportsfromotherhigh-incomecountries,transmitted

NNRTIresistanceseemstobelessfrequentinJapan.

InAustralia,researchconductedbetween1992and2001in

Sydneyinagroupof185recently-infectedindividualsfound

levelsoftransmitteddrugresistancetoreverse-transcriptase

inhibitorspeakinginthemid-1990s,droppingsignificantly

withtheintroductionofcombinationtherapyin1996andthen

reachingaplateauof10–15%duringtheyears1999–2001(8).Morerecently,anassessmentofdrugresistanceamong466

recentlyinfectedindividualsbetween1996and2007inthe

Victoria region foundanaverageannual transmitteddrug

resistanceprevalenceof16%,predominantlyassociatedwith

NRTIandNNRTI(9).Mutationsknowntocauseresistanceto

PIremaineduncommon.

2.2 Acquired drug resistance

Dataonacquireddrugresistanceinhigh-incomecountries

suggestthatNRTIresistanceisthemostfrequentformof

drugresistanceinpeopleforwhomantiretroviraltherapy

isfailing,followedbyNNRTIandPIresistance.

Among1988peoplefailingantiretroviraltherapybetween

2000 and 2004 from 15 European countries, 80.7%

had at least one drug resistance mutation (NRTI 75.5%,

NNRTI 48.5%, PI 35.8%). Predicted resistance to most

bPI was estimated at less than 25% (10). Similar results

wereobservedinanassessmentof16511drugresistance

genotypesfrom11492treatment-experiencedindividualsin

sevenEuropeancountriesbetween1999and2008:80.1%

had at least one drug resistance mutation (NRTI 67.2%,

NNRTI53.7%,PI32.4%),with17.2%showingresistanceto

threeclasses (11).Afteradjustingforconfoundingfactors,

peoplefailingtherapyinmorerecentcalendaryearsshowed

a decline in overall resistance to NRTI and PI but not to

NNRTI.

Thoughevidenceisstilllimitedandadditionalresearchis

needed,otherstudieshavealsoobservedthisdownward

trend in theprevalenceofacquireddrug resistance.This

findingisprobablyassociatedwiththeuseofimprovedfirst-

and second-line regimens with greater potential to fully

suppressviralreplication.Among5422individualsinBritish

Columbia,theincidenceofdrugresistanceinthosereceiving

antiretroviraltherapydroppedmorethan12-foldbetween

1996 and 2008, and viral suppression increased from

64.7%in2000to87.7%in2008(12).InanHIVoutpatient

study, the frequency of HIV resistance among people

receiving antiretroviral therapy for at least four months

with plasma viral load above 1000 copies/ml dropped

from 88% in 1999 to 79% in 2008, with a statistically

significant decline observed in the incidence of acquired

drugresistanceforPI.


19

REFERENCES

1. FrentzD,BoucherCA,vandeVijverDA.Temporalchangesintheepidemiologyoftransmissionofdrug-resistantHIV-1acrosstheworld.AIDSReviews,2012,14:17–27.

2. WittkopLetal.EffectoftransmitteddrugresistanceonvirologicalandimmunologicalresponsetoinitialcombinationantiretroviraltherapyforHIV(EuroCoord-CHAINjointproject):aEuropeanmulticohortstudy.LancetInfectiousDiseases,2011,11:363–371.

3. Poonetal.TransmitteddrugresistanceintheCFARnetworkofintegratedclinicalsystemscohort:prevalenceandeffectsonpre-therapyCD4andviralload.PLoSOne,2011,6:e21189.

4. Markovitz AR et al. Primary antiretroviral drug resistance in newly human immunodeficiency virus-diagnosed individuals testinganonymouslyandconfidentially.MicrobialDrugResistance,2011,17:283–289.

5. JainVetal.Transmitteddrugresistanceinpersonswithacute/earlyHIV-1inSanFrancisco,2002–2009.PLoSOne,2010,5:e15510.

6. Hattori J et al. Surveillance of drug resistance and phylodynamic network analysis of newly infected HIV/AIDS patients: Japan,2003to2010.19thConferenceonRetrovirusesandOpportunisticInfections,5–8March2012,Seattle,WA,USA(Paper729;http://www.retroconference.org/2012b/Abstracts/44128.htm,accessed28June2012).

7. HattoriJetal.Trendsintransmitteddrug-resistantHIV-1anddemographiccharacteristicsofnewlydiagnosedpatients:nationwidesurveillancefrom2003to2008inJapan.AntiviralResearch,2010,88:72–79.

8. AmmaranondPetal.NoincreaseinproteaseresistanceandadecreaseinreversetranscriptaseresistancemutationsinprimaryHIV-1infection:1992–2001.AIDS,2003,17:264–267.

9. RussellJSetal.PrevalenceoftransmittedHIVdrugresistancesincetheavailabilityofhighlyactiveantiretroviraltherapy.CommunicableDiseaseIntelligence,2009,33:216–220.

10. vandeVijverDAetal.HIV-1drug-resistancepatternsamongpatientsonfailingtreatmentinalargenumberofEuropeancountries.ActaDermatovenerologicaAlpina,Panonica,etAdriatica,2010,19:3–9.

11. ProsperiMCetal.DetectionofdrugresistancemutationsatlowplasmaHIV-1RNAloadinaEuropeanmulticentrecohortstudy.JournalofAntimicrobialChemotherapy,2011,66:1886–1896.

12. GillVSetal.ImprovedvirologicaloutcomesinBritishColumbiaconcomitantwithdecreasingincidenceofHIVtype1drugresistancedetection.ClinicalInfectiousDiseases,2010,50:98–105.

20

3.1 Overview

TransmittedHIVdrugresistanceoccurswhenpreviously

uninfected individuals are infected with drug-resistant

virus. The term transmitted HIV drug resistance is

appropriatelyappliedonlytoHIVdrugresistancedetected

inrecentlyinfectedindividualsbecause,overtimeandat

variablerates,mutationsmayreverttowild-type,become

archivedinviralDNAorfallbelowthesensitivitylevelof

standardgenotypingassays todetect them (1).However,

mostpublishedstudiesalsoincludeindividualswhomay

havebeeninfectedforaconsiderablylongertimeandare

consideredtobe“chronicallyinfected”.

To assess levels and trends of transmitted HIV drug

resistanceinrecentlyandchronicallyinfectedindividualsin

low-andmiddle-incomecountries,thepublishedliterature

was systematically reviewed, and the main findings are

presentedbelow.

3. TRANSMITTED HIV DRUG RESISTANCE IN LOW- AND MIDDLE-INCOME COUNTRIES

3.2 Literature review on drug resistance among ARV-naïve recently- or chronically-infected populations

A systematic literature search identified 126 articles,

spanning40countries,comprisingatotalof16650people

living with HIV (Table 3.1). Studies were considered if

they included untreated recently or chronically infected

individuals 15 years and older and had more than 10

genotypesavailable.Section3inAnnex1providesadditional

detailsonthemethodsused.Table1 inAnnex1 lists the

studies included. Geographically, most studies matching

the predefined selection criteria were in Africa, followed

byLatinAmericaandtheCaribbean,theWesternPacific

RegionandtheSouth-EastAsiaRegion.1

Many of the studies included in this meta-analysis were

performed using distinct methods and may differ with

respecttothepopulationstudied(suchasrecentorchronic

infections), the sampling frame (such as consecutive,

convenient or random selection from the general

population) and the laboratory methods used (such as

1 Datafromlow-andmiddle-incomecountriesintheEuropeanRegionandtheEasternMediterraneanRegionwereexcludedfromtheanalysisduetothepaucityofavailabledata.

KEY FINDINGS

1. Availabledatasuggestthattheestimatedprevalenceoftransmitteddrugresistanceincreasedbetween2003and

2010intheareassurveyed,althoughwithinexpectedlevels.

• A systematic review of published studies in ARV-naïve recently- or chronically-infected individuals (excluding

WHOsurveys)foundthatmorerecentsurveysreportedhigheraveragelevelsofHIVdrugresistance,reachingan

estimatedhighof6.6%(95%confidenceinterval5.1%–8.3%)ain2009.

• Pooled analysis of data from WHO surveys indicates that the estimated prevalence of transmitted HIV drug

resistancetoNNRTIincreasedbetween2004and2010.Thisestimatedincreasewasparticularlyapparentinthe

areassurveyedintheAfricanregion,wheretheprevalenceofNNRTIresistancereached3.4%(95%CI1.8%–5.2%)

in2009.

2. Data from WHO surveys suggest that greater coverage of antiretroviral therapy was associated with a higher

prevalenceoftransmitteddrugresistance,particularlytoNNRTI,althoughtheestimatedeffectondrugresistance

ofanincreaseinantiretroviraltherapycoverageremainedmodestintheareassurveyed.

a All confidence intervals quoted are at the 95% level.


21

driedbloodspotsorplasmaspecimensandthegenotyping

methods used). Thus, prevalence estimates may not be

nationallyorregionallyrepresentative.

Data from individual studies were abstracted and

aggregated by region and year and revealed that more

recentstudiesreportedhigherlevelsofHIVdrugresistance,

reaching a high of 6.6% (95% CI 5.1%-8.3%) in 2009.

Mostofthischangewasduetoanassociatedincreasein

theoverallprevalenceofmutationsconferringresistanceto

NNRTI.Noevidenceofincreasingresistanceovertimeto

NRTIorPIwasobserved(Table3.2).

In the Africa region, although overall prevalence levels

did not appear to vary significantly over time (Table 1 in

Annex2),amoredetailedanalysisbydrugclassshoweda

statisticallysignificantincreaseintheprevalenceofNNRTI

mutations.TheprevalenceofNNRTIresistancein2003was

1%(95%CI0.3%–2.1%)and6.4%(95%CI1.3%–17.5%)

in 2010 in the region. NRTI resistance varied little over

time. Reported resistance to PI was generally stable and

low,asthevastmajorityofpeoplelivingwithHIVinthis

regionreceivedanNNRTI-basedfirst-lineregimenduring

theperiodstudied.

Prevalence estimates for general or class-specific

mutationsdidnotvarysignificantly instudiesconducted

intheSouth-EastAsiaorWesternPacificregionsbetween

2003and2010.However,HIVdrugresistanceincreased

significantly in Latin America and the Caribbean, a fact

thatisprobablyassociated,amongotherreasons,withthe

earlier introduction and higher coverage of antiretroviral

therapyintheregion.Table1inAnnex2providesaregional

breakdownofresistanceprevalence.

3.3 WHO surveys to assess transmitted drug resistance

WHOrecommendsaminimum-resourcemethodtoassess

transmitted HIV drug resistance in specific geographical

areasofresource-limitedcountrieswheretransmittedHIV

drugresistanceislikelytobeseenfirst(suchasinurban

areas where antiretroviral therapy has been available for

atleastafewyears).IfHIVdrugresistancetransmission

islowinsuchareas,itisunlikelytobehigherelsewherein

thecountry.

The survey method for transmitted HIV drug resistance

samplesindividualsfrompopulationslikelytobeantiretroviral

drug–naiveandtohavebeenrecentlyinfected.Section4in

Annex1providesadditionalinformationonsurveymethods

for transmitted HIV drug resistance. This method is not

intendedtoprovideapointprevalenceestimatebutrather

toclassifytransmittedresistanceforeachdrugclassaslow

(prevalencelessthan5%),moderate(prevalencebetween

5%and15%)orhigh(prevalencemorethan15%).Surveys

arenotdesignedtobenationallyrepresentativeortoassess

trends over time. Instead, their main purpose is to alert

programmemanagersthatresistanceisbeingtransmittedin

specificgeographicalareasofacountryandthat,depending

on their results, wider policy action may be warranted.

Therefore,surveyresultscanbeinstrumental in informing

notonlytheselectionoffuturefirst-lineantiretroviraltherapy

regimensbutalsoinoptimizingapproachesforpreventing

mother-to-childtransmissionofHIVaswellasforpre-and

post-exposureprophylaxis.

3.3.1 OverviewBetween 2004 and 2010, 30 countries initiated 101

surveysusingtheWHO-recommendedmethodtoassess

transmitteddrugresistance.Datafrom82surveysfrom30

countriesweremadeavailabletoWHO.HIVdrugresistance

Number of surveys

2003 2004 2005 2006 2007 2008 2009 2010

Totalnumberofstudies 20 16 17 18 20 15 16 4

AfricanRegion 9 9 10 5 11 7 7 1

Western/Central 2 4 4 1 7 1 1 –

Southern 2 2 3 3 3 2 4 –

Eastern 5 3 3 1 1 4 2 1

South-EastAsiaRegion 3 – 2 5 2 1 – 2

WesternPacificRegion – 2 3 5 4 1 3 1

LatinAmericanandtheCaribbean 8 5 2 3 3 6 6 –

Totalnumberofcountriesrepresented 14 12 11 11 13 13 8 3

Numberofindividualsgenotyped 2281 1777 3568 1735 2572 3078 1503 136

Table 3.1 Studies included in the systematic review of drug resistance among ARV-naïve recently- or chronically-infected populations, by region and by year of survey, 2003–2010

% with at least one drug resistance mutation (95% confidence interval)

P-valuea2003 2004 2005 2006 2007 2008 2009 2010

Any 3.6(2.3–5.2)

4.5(2.3–7.3)

1.9(0.9–3.3)

2.5(1.2–4.1)

3.1(1.6–5.0)

4.9(3.6–6.3)

6.6(5.1–8.3)

2.1(0.1–5.8)

0.03

NRTI 2.0(0.9–3.4)

2.3(1.0–4.0)

0.7(0.1–1.5)

0.9(0.1–2.2)

1.2(0.4–2.4)

1.9(1.1–2.9)

2.0(0.8–3.5)

0.0(0.0–1.4)

0.46

NNRTI 0.9(0.2–2.0)

1.0(0.2–2.1)

1.1(0.4–2.0)

1.2(0.3–2.7)

1.2(0.5–2.2)

1.8(1.3–2.4)

3.3(2.3–4.4)

0.9(0.0–4.8)

<0.001

PI 0.3(0.0–1.0)

0.9(0.2–2.0)

0.0(0.0–0.1)

0.0(0.0–0.3)

0.2(0.0–0.6)

0.7(0.3–1.4)

0.9(0.2–1.9)

0.0(0.0–1.4)

0.48

a StatisticalmethodsaredescribedinSection3,Annex1.

Table 3.2 Estimated prevalence of HIV drug resistance among ARV–naive individuals from the published literature, 2003–2010

22

prevalencecouldbeclassifiedaslow,moderateorhighfor

atleastonedrugclassinasubsetof72surveysconducted

in26countries(Figures3.1and3.2,andTable3.3).1Section

2 in Annex 1 summarizes methods for sequence data

analysisandqualityassurance.

Table2inAnnex2providesindividualsurveyresults.Ten

surveys(notshown)hadinsufficientsamplesizestoallow

theirresultstobeclassifiedintoanyofthethreeprevalence

categories (low, moderate or high); nevertheless, their

1 Inthreesurveysthesamplesizewasinsufficienttoclassifyresistanceintooneofthethreecategories(<5%,5%–15%or>15%)butwassufficienttoclassifyresistanceasbeingabove5%(PhnomPenh,Cambodia,2008(NNRTI),KwaZulu-Natal,SouthAfrica,2008(NNRTIandNRTI)andKyiv,Ukraine,2009(NRTI).ThesesurveyswerethereforeconsideredashavingamoderateprevalenceofHIVdrugresistance(between5%and15%).

Figure 3.1 Countries (n=26) reporting results from WHO surveys of transmitted HIV drug resistance, 2004–2010

Country reporting results from WHO surveys of transmitted HIV drug resistance, 2004–2010

No data available or not participating in the surveys

Not applicable

Figure 3.2 Number of WHO transmitted HIV drug resistance surveys with classifiable results for at least one drug class, 2004–2010 (n = 72)

2

10 10 10

13

23

4

0

5

10

15

20

25

2004 2005 2006 2007 2008 2009 2010

Num

ber o

f sur

veys

Years of survey initiation

a

a Fourteensurveyswereimplementedin2010;however,onlyfourhadresultsavailableforanalysis.

patient-level data were included in the pooled analysis

presentedinsection3.3.3.2

Overall,91.7%of the72surveyswithclassifiable results

wereconductedbetween2005and2009.Geographically,

most(59.7%,or43/72)surveyswereimplementedinthe

AfricanRegion.

WHO recommends that surveys be repeated every

two years to detect signs of increasing transmission of

resistance. Of the 18 countries reporting from Africa, 10

conductedsurveysonly inoneyear,whereas8 repeated

them over time with variable frequency: 4 countries

implementeditintwodifferentyears(Botswana,Burkina

Faso, Kenya and Mozambique), three repeated it three

timesindifferentyears(Malawi,SwazilandandUganda)

andSouthAfricaconductedthesurveysannually.Table3in

Annex2listscountrieswithatleasttwosurveysrepeated

overtime.

Ofthe11countriescomprisingtheWHOSouth-EastAsia

Region,onlythree(India,IndonesiaandThailand)reported

results,foratotalofsixsurveys,allofwhichwereconducted

before2007.Thailandrepeatedthesurveytwiceindifferent

years,whereasIndiaandIndonesiaimplementedthesurvey

onlyonce.

2 ThesesurveyswereconductedinBotswana,Gaborone,2007;Burundi,Bujumbura,2007;Cambodia,multipleareas,2006;Cambodia,PhnomPenh,2006;CentralAfricanRepublic,Bangui,2007;Congo,Brazzaville,2006;Congo,PointeNoire,2006;IslamicRepublicofIran,multipleareas,2006;Mozambique,Maputo,2007;andSouthAfrica,WesternCape,2007.


23

Eighteen surveys were conducted in the Western Pacific

Regioninthreecountries(Cambodia,ChinaandVietNam),

mostlybetween2008and2009.China implemented 15

surveysbetween2007and2008inmultiplegeographical

areas,andVietNamperformedtwosurveys,oneinHanoi

(2006) and one in Ho Chi Minh City (2007). In the

EuropeanRegionandinLatinAmericaandtheCaribbean,

onlyonecountryineach,UkraineandMexico,respectively,

implementedsurveysaccordingtoWHOmethods.Most

ofthecountriesintheseregionshaveconcentratedorlow-

levelepidemics,andtheimplementationoftransmittedHIV

drugresistancesurveysusingcurrentmethods,designedto

beappliedinthecontextofgeneralizedHIVepidemics,was

particularlychallenging.1

Ofthe72surveys,41(56.9%)wereconductedinantenatal

caresitesamongpregnantwomen,andmostincludedonly

womenintheirfirstpregnancy–tominimizethelikelihood

of including women with previous exposure to regimens

forpreventingmother-to-childtransmission–andyounger

than25yearsofage–tominimizethelikelihoodofincluding

individualswithchronicinfectionandwithpriorexposure

toantiretrovirals.Twenty-eight(38.8%)wereconductedin

voluntarycounsellingandtestingsites,chieflyamongmen

andwomenyoungerthan25yearsofage.Onesurveywas

conductedamongsexworkers(Kampala,Uganda,2008),

one among people who inject drugs (Jakarta, Indonesia,

2006)andoneamongblooddonors(Bangkok,Thailand,

2005).

3.3.2 Classification of WHO surveys on transmitted HIV drug resistance

Table 3.4 summarizes the results of the 72 surveys that

could be classified as low, moderate or high prevalence

foratleastonedrugclass.Ofthe72surveys,52(72.2%)

had a low prevalence classification to all drug classes.

No survey was classified as having a high prevalence of

transmitted HIV drug resistance. However, 20 (27.8%)

hadamoderateprevalenceclassificationofresistanceto

oneormoreantiretroviraldrugclass(NRTIand/orNNRTI

and/or PI). Because of their important implications for

programme management and service delivery, surveys

showing moderate prevalence of drug resistance merit

particularattention.

Almosttwothirds(60%or12of20)ofthesurveyswith

amoderateprevalenceclassificationreportedamoderate

levelofresistancetoNNRTI,50%(10of20)toNRTI,and

10%(2of20)toPI.

1 Othersurveysmayhavebeenconductedbutdatawerenotreportedormadeavailableforinclusioninthisanalysis.

Number of surveys

Geographicalregion 2004 2005 2006 2007 2008 2009 2010 Total

AfricanRegion 1 8 8 5 6 11 4 43

Eastern 3 2 1 1 6 1 14

Ethiopia 1 1

Kenya 1 1 2

Malawi 1 2 1 4

Mozambique 1 2 3

Uganda 1 1 1 3

UnitedRepublicofTanzania 1 1

Southern 1 4 3 3 3 4 3 21

Angola 1 1

Botswana 2 1 3

Lesotho 1 1

Namibia 1 1

SouthAfrica 1 2 1 2 2 2 2 12

Swaziland 1 1 1 3

Western/Central 1 3 1 2 1 8

BurkinaFaso 1 1 2

Cameroon 2 2

Chad 1 1

Côted’Ivoire 1 1

Ghana 1 1

Senegal 1 1

LatinAmericaandtheCaribbean 1 1

Mexico 1 1

EuropeanRegion 4 4

Ukraine 4 4

South-EastAsiaRegion 2 1 3 6

India 2 2

Indonesia 1 1

Thailand 2 1 3

WesternPacificRegion 1 2 7 8 18

Cambodia 1 1

China 1 6 8 15

VietNam 1 1 2

Overall 2 10 10 10 13 23 4 72

Table 3.3 Number of WHO surveys of transmitted HIV drug resistance with results classifiable for at least one drug class, by year of implementation and geographical region, 2004–2010

Category of transmitted HIV drug resistance Drug class N (%) of surveys

Lowprevalence(<5%) All 52(72.2%)

Moderateprevalence(5%–15%) Any 20(27.8%)

OnlyNNRTI 8(11.1%)

OnlyNRTI 7(9.7%)

OnlyPI 1(1.4%)

NRTIandNNRTI 3(4.2%)

NNRTIandPI 1(1.4%)

NRTIandPI 0(0%)

Highprevalence(>15%) Any 0(0%)

Totalnumberofsurveys 72

Table 3.4 Results of WHO transmitted HIV drug resistance surveys

24

Number (%) of surveys with moderate (5–15%) prevalence

YearTotal

surveysAnydrug

class NNRTI NRTI PI

2004–2006 22 4(18%) 1(5%) 3(14%) 0(0%)

2007–2010 50 16(32%) 11(22%) 7(14%) 2(4%)

a Mid-pointperiod.

CountryGeographical

area 2004 2005 2006 2007 2008 2009 2010

Botswana Francistown

Malawi Lilongwe NNRTI NNRTI

SouthAfrica Gauteng NRTI

SouthAfrica KwaZulu-NatalNNRTI+NRTI

NNRTINNRTI+NRTI

SwazilandManzini-Mbambanecorridor

Uganda Entebbe/Kampalab NRTI

China Beijing

China Hunan NNRTI

ChinaLiangshan(Sichuan)

China Shenzhen

a Green:lowprevalenceclassificationoftransmittedHIVdrugresistance;red:moderateprevalenceclassificationoftransmittedHIVdrugresistance.

b Entebbe/Kampalaareconsideredasbeingpartofthesamegeographicarea

Between 2004 and 2010, the proportion of surveys

reportingamoderateprevalenceoftransmittedHIVdrug

resistancetoatleastonedrugclassincreasedfrom18.2%

(4of22)in2004–2006to32%(16of50)in2007–2010

(Tables3.5and3.6).This increasewasmostlydrivenby

a considerable rise in the number of surveys reporting

moderateprevalenceofNNRTIresistance.Incontrast,the

frequencyofsurveysreportingamoderateprevalenceof

resistancetoNRTIremainedstable.

Geographically, the overall increase in the frequency of

surveyswithmoderateprevalenceappearstobecausedby

increasedreportsofmoderateprevalenceclassificationin

theAfricaRegion,wheretheproportionofsurveysreporting

moderateprevalencerosefrom17.6%(3of17)in2004–

2006to40.7%(11of27)in2007–2010.

Overall,65%(13of20)ofthesurveysshowingamoderate

prevalenceoftransmittedHIVdrugresistancetoanydrug

class were in the African Region, particularly in eastern

Africa(30%,6of20).Five(25%,5of20)wereconducted

in the Western Pacific Region, one in Latin America and

the Caribbean and another in the European Region. No

surveyfromtheSouth-EastAsiaRegionshowedresistance

between 5% and 15%. Table 4 in Annex 2 describes

the geographical distribution of surveys with moderate

classification.

Of the two surveys reporting moderate prevalence of

transmittedHIVdrugresistancetoPI,onewasinEastern

AfricaandoneintheWesternPacificRegion.Ofthethree

surveys with moderate prevalence of resistance for both

NNRTI and NRTI, two were from countries in southern

AfricaandoneinWestern/CentralAfrica.

Globally, in the 11 geographical areas in which surveys

were repeated over time (see Table 3.7) and therefore

allowedamoredetailedanalysis, fourreportedachange

fromlowtomoderateprevalence,signallinganincreasein

transmissionofdrug-resistantvirus(LilongweinMalawi,

Entebbe/Kampala in Uganda, KwaZulu-Natal in South

Africa, and Hunan in China). Two surveys conducted in

Beira(Mozambique)in2007and2009showedmoderate

levels of transmitted HIV drug resistance to NNRTI

and NRTI, respectively. In contrast, in five geographical

areas, successive surveys confirmed a low prevalence of

transmittedHIVdrugresistance.

Moderateprevalenceof transmittedHIVdrugresistance

wasreportedinGauteng(SouthAfrica)in2004,followed

byfiveconsecutivesurveysdocumenting lowprevalence

(Box3.1).

Year Total surveys

Number of surveys (% of annual total) with classification 5%–15%

for at least 1 drug class

2004 2 2(100%)

2005 10 0(0%)

2006 10 2(20%)

2007 10 2(20%)

2008 13 3(23%)

2009 23 9(39%)

2010 4 2(50%)

Total 72 20(28%)

Table 3.5 WHO surveys of transmitted HIV drug resistance with moderate prevalence classification (5%–15%), by year, 2004–2010

Table 3.6 Frequency of WHO surveys reporting moderate prevalence of transmitted HIV drug resistance, by period (before or after 2007)a

Table 3.7 WHO surveys of transmitted HIV drug resistance in selected areas, 2004–2010a


25

3.3.3 Pooled analysisToassesswhethertransmittedHIVdrugresistanceincreased

overtimeintheareassurveyed,sequencedatafromall82

surveys – a total of 3588 recently infected individuals –

werepooled.Figure3.3describestheregionaldistribution

of individuals included in the pooled analysis. Figure 3.4

providesabreakdownbythetypeofpopulationsurveyed.

In this sample, a statistically significant increase in the

prevalence of transmitted drug resistance toNNRTI was

Box 3.1 Assessing transmitted drug resistance in the provinces of KwaZulu-Natal and Gauteng, South Africa

South Africa initiated the roll-out of antiretroviral therapy nationally in 2004. In the province of KwaZulu-Natal, surveys to assess the prevalence of transmitted drug resistance were implemented in 2005 and repeated annually between 2007 and 2010 (Table 3.8). The prevalence of transmitted drug resistance was low in 2005 and 2007. However, the prevalence of transmitted HIV drug resistance to NNRTI was found to be moderate in 2008, 2009 and 2010. Similarly, the prevalence of transmitted HIV drug resistance to NRTI also increased to moderate in 2008 and 2010.

In Gauteng province, seven surveys of transmitted HIV drug resistance among women pregnant for the first time were conducted between 2004 and 2010. In this region, all surveys in all years showed low prevalence of resistance for all drug classes except for a moderate prevalence classification of NRTI resistance in 2004. While such result may represent a true moderate prevalence estimate, it may also have been caused by random misclassification error, as it was observed in the year when antiretroviral therapy was being rolled out and coverage was expected to be low. The survey may also have captured women infected with drug-resistant virus from partners participating in early clinical trials or whose virus had been exposed to drugs in other settings (private unregulated market). Nevertheless, subsequent surveys in Gauteng documented low prevalence of transmitted HIV drug resistance to all drug classes.

Antiretroviral therapy programme functioning should be investigated to address why moderate levels of transmitted HIV drug resistance were observed more frequently in KwaZulu-Natal compared with Gauteng. Specifically, factors known to be associated with HIV drug resistance such as loss to follow-up, retention, adherence, drug supply continuity, rates of population-level viral load suppression and prescribing practices should be assessed.

Table 3.8 Results of surveys to assess transmitted drug resistance in the provinces of KwaZulu-Natal and Gauteng (South Africa), 2004–20 10

Gauteng

Year NNRTI NRTI PI

2004 <5% 5-15% nc

2005 <5% <5% nc

2006 <5% <5% <5%

2007 <5% <5% <5%

2008 <5% <5% <5%

2009 <5% <5% <5%

2010 <5% <5% <5%

KwaZulu-Natal

Year NNRTI NRTI PI

2005 nc <5% nc

2007 <5% <5% <5%

2008 5-15% 5-15% <5%

2009 5-15% <5% <5%

2010 5-15% 5-15% <5%

nc=notclassifiable(insufficientspecimensavailabletoclassifytransmittedHIVdrugresistance)

Africa (61.7%)

Western Pacfic (23.4%)

South-East Asia (7.7%)

Europe (4.8%)

Eastern Mediterranean

(1.1%)

Latin America and the

Caribbean(1.3%)

Figure 3.3 Regional distribution of individuals from pooled analysis

observed between 2004 and 2010, particularly in the

areassurveyedintheAfricanRegion(Table3.9).Section9

in Annex 1 provides additional details on the statistical

methodsused.Section2inAnnex1summarizesmethods

forsequencedataanalysisandqualityassurance.

Figure 3.4 Populations surveyed (% of total number of individuals from pooled analysis)

Women in their first pregnancy(59.4%)

Voluntary counselling and testing

attendees(36.8%)

Sex workers (1.2%)

People who inject drugs(1.3%)

Blood donors(1.3%)

26

2004a 2005 2006 2007 2008 2009 2010aP-valueb

(adjustedforregion)

%(95%CI)

%(95%CI)

%(95%CI)

%(95%CI)

%(95%CI)

%(95%CI)

%(95%CI)

AnyDrug

AfricanRegion 10.0(2.8–23.7)

0.2(0.0–1.4)

0.6(0.0–2.4)

1.2(0.1–3.2)

1.8(0.1–4.8)

4.5(2.3–7.2)

2.8(0.1–7.7) 0.04

South-EastAsiaRegion — 0.7(0.0–4.8)

2.2(0.1–11.8)

1.0(0.2–3.8) — — — —

WesternPacificRegion — — 4.5(1.0–9.6)

4.4(1.1–9.4)

1.5(0.0–4.3)

2.4(0.6–4.8) — 0.41

LatinAmericaandtheCaribbean

8.5(2.4–20.4) — — — — — — —

EuropeRegion — — — — — 2.6(0.1–6.9) — —

EasternMediterraneanRegion — — 7.7(1.6–20.9) — — — — —

Overall 9.2(3.7–16.4)

0.3(0.2–1.4)

1.6(0.4–3.2)

1.6(0.5–3.1)

1.6(0.3–3.5)

3.4(2.1–5.1)

2.8(0.1–7.7) 0.06

NNRTI

AfricanRegion2.3

(0.1–12.0)0.0

(0.0–1.0)0.1

(0.0–0.9)0.0

(0.0–0.7)1.5

(0.1–3.9)3.4

(1.8–5.2)2.0

(0.2–5.0) <0.01


0.0(0.0–7.9)

0.3(0.0–2.6) — — — —


3.6(0.6–8.2)

0.5(0.2–2.1)

0.9(0.0–2.6) — 0.41


0.0(0.0–7.5) — — — — — — —

EuropeRegion — — — — — 0.8(0.3–3.4) — —


Overall0.7

(0.0–4.3)0.0

(0.0–0.8)0.2

(0.1–0.9)0.3

(0.0–1.3)0.9

(0.1–2.2)2.0

(1.1–3.2)2.0

(0.2–5.0) <0.01

NRTI


0.0(0.0–0.7)

0.1(0.0–1.1)

0.7(0.0–2.5)

0.5(0.2–2.0)

0.9(0.2–2.2)

0.6(0.0–3.7) 0.24


0.0(0.0–7.9)

0.0(0.0–1.5) — — — —


0.6(0.0–3.6)

0.3(0.0–1.8)

0.6(0.0–2.1) — 0.71


8.5(2.4–20.4) — — — — — — —

EuropeRegion — — — — — 1.2(0.2–4.6) — —


Overall 6.5(2.0–12.8)

0.0(0.0–0.8)

0.5(0.0–1.4)

0.4(0.0–1.4)

0.4(0.0–1.4)

0.9(0.3–1.7)

0.6(0.0–3.7) 0.37

PI


0.2(0.0–1.5)

0.0(0.0–0.5)

0.3(0.0–1.6)

0.0(0.0–0.8)

0.4(0.0–1.4)

0.0(0.0–1.1) 0.76


2.3(0.1–12.0)

0.3(0.0–2.5) — — — —


0.6(0.0–3.6)

0.2(0.0–1.6)

0.8(0.0–2.3) — 0.34


0.0(0.0–7.5) — — — — — — —

EuropeRegion — — — — — 0.2(0.0–2.4) — —


Overall 0.7(0.0–5.3)

0.1(0.0–1.1)

0.2(0.0–1.0)

0.3(0.0–1.3)

0.0(0.0–0.7)

0.5(0.0–1.2)

0.0(0.0–1.1) —

a Resultsmayhavebeenaffectedbythelimitedamountofdataavailableandshouldbeinterpretedcautiously(87specimensin2004and196in2010).b SomeP-valuescouldnotbecalculatedduetocollinearity,lackofdataand/orvariability.StatisticalmethodsaredescribedinSection9,Annex1.c Areassurveyedvariedconsiderablyamongcountriesandacrosstime.— Dataarenotavailableorapplicable.

Table 3.9 Estimates of transmitted HIV drug resistance by year of survey, region and antiretroviral therapy class (WHO transmitted HIV drug resistance surveys), 2004–2010c


27

Areassurveyedvariedconsiderably,amongcountriesand

acrosstime,sogeneralizationsmaynotbeappropriateor

applicable. Nevertheless, in a subset of six geographical

areasintheAfricanRegioninwhichtransmittedHIVdrug

resistancesurveyswererepeatedindifferentyears,three

areasreportedachangefromlowtomoderateprevalence

of transmitted HIVdrug resistance (Lilongwe inMalawi,

Entebbe/KampalainUgandaandKwaZulu-NatalinSouth

Africa),signallinganincreaseinthetransmissionofdrug-

resistantvirusintheseareas(Table3.7).

Fig. 3.5 depicts the relationship between the prevalence

of transmitted NNRTI drug resistance mutations and

antiretroviral therapy coverage. Antiretroviral therapy

coverage was defined as the number of people living

with HIV receiving antiretroviral drugs in the country in

which the survey was undertaken divided by the total

numberofpeoplelivingwithHIV.Controllingforregional

variability, available data indicate that higher levels of

antiretroviral therapy coverage are associated, though

modestly, with increased prevalence of transmitted drug

resistancetoNNRTI(p-valueadjustedforregion=0.039;

oddsratio=1.49,95%CI1.07–2.08).

Althoughsuchanincreaseremainedwithinexpectedlevels,

this finding is particularly critical in the context of the

expandeduseofantiretroviraldrugsnotonlyfortreatment

butalsoforthepreventionofHIVinfection.Unlesscarefully

monitoredandcontained,transmitteddrugresistancecan

potentially reduce the efficacy of standard antiretroviral

therapy–anissueaggravatedbythelimitedavailabilityof

alternativeantiretroviraldrugregimens.

3.3.4 Prevalence of HIV drug resistance mutationsFigure 3.6 shows the prevalence of drug resistance

mutationsacrossall82surveysperformedbetween2004

and 2010. For the description of the mutation list used,

seeSection5 inAnnex1.Overall,3.1%ofthe individuals

surveyedhadtransmitteddrugresistancetoat leastone

drugclass.Asexpected,NNRTImutations (2)weremost

commonlyobserved(1.6%),followedbyNRTI(1.3%)andPI

(0.7%).Only0.4%ofindividualshadbothNNRTIandNRTI

drug resistance mutations. Among individual mutations,

thoseatposition103(K103NorS)werethemostcommon

(0.8%), followed by D67N or G, K101E or P, Y181C, and

M184V,ranging from0.3%to0.4%.Onlyonesequence

carriedK65R,amutationconferringresistancetotenofovir.

Figure 3.5 Relationship between antiretroviral therapy coverage and prevalence of transmitted NNRTI drug resistance mutations

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

0% 5% 10% 15% 20% 25% 30% 35% 40%

Prev

alen

ce o

f NN

RTI r

esis

tanc

e m

utat

ions

:%

of g

enot

ypes


Eastern AfricaSouthern AfricaWestern/Central AfricaSouth-East AsiaWestern PacificOther

28

REFERENCES1. PingenMetal.Evolutionarypathwaysoftransmitteddrug-resistantHIV-1.JournalofAntimicrobialChemotherapy,2011,66:1467–1480.

2. BennettDEetal.DrugresistancemutationsforsurveillanceoftransmittedHIV-1drug-resistance:2009update.PLoSOne,2009,4:e4724.

0%

0.5%

1.0%

1.5%

2.0%

2.5%

3.0%

3.5%

Any m

utatio

n

Any N

NRTI

Any N

RTI

Any PI

NRTI an

d NNRTI

K103N

S

K101E

P

Y181 a

ny

G190 a

ny

Y188 a

ny

M184IV

D67GN

T215

any

K219 a

nyK70

RM41

LT6

9DL2

10W

Perc

enta

ge o

f gen

otyp

es

I

Red:percentageofindividualswithanydrugresistancemutationasdefinedbytheWHO2009SurveillanceHIVDrugResistancemutationlist(2).

Blue:anyNNRTImutation.Green:anyNRTImutation.

Orange:anyPImutation.

Lavender:bothNRTIandNNRTImutations.

DetailsofwhichmutationswereobservedmostcommonlyaredisplayedontherightinblueforNNRTIandingreenforNRTImutations.Alternativevariantsateachpositionarecombined:forexample,K103NSrepresentspeoplewithK103NorK103S;“any”designatesmultiplevariantsatthatposition(forexample,T215anyincludesD,F,I,N,SandY).Atotalof3381PRgenotypesand3539RTgenotypesareincluded;thetotalnumberofgenotypes(n=3588)wasusedasthedenominatorforcalculatingtheprevalenceof“anymutation”.

Figure 3.6 Prevalence of drug resistance mutations in individuals included in WHO transmitted HIV drug resistance surveys, 2004–2010


29

4. ACQUIRED DRUG RESISTANCE IN LOW- AND MIDDLE-INCOME COUNTRIES

KEY FINDINGS

1. SYSTEMATIC LITERATURE REVIEWAsystematicreviewofthepublishedliteratureindicatesthat,ineightlow-andmiddle-incomecountriesinAsiaand

sub-SaharanAfrica,60%ofthe573peoplefailingNNRTI-basedfirst-linetherapyafteramedianof12monthshad

resistancetoanyHIVdrugclass.Theremaining40%failedwithnoHIVdrugresistance,suggestingthatverypoor

adherenceorextendedtreatmentinterruptioncouldhaveplayedanimportantroleincausingvirologicalfailure.

2. WHO SURVEYS TO ASSESS ACQUIRED DRUG RESISTANCE

Resistance before initiation of first-line antiretroviral therapy • Fortysurveys,comprising6370people,wereperformedin12countriesbetween2006and2010usingastandardized

WHOprotocoltoassessacquireddrugresistance.Inapooledanalysisofpeopleinitiatingfirst-lineantiretroviral

therapy,prevalenceofHIVdrugresistancetoanydrugwas5%,rangingfrom4.8%in2007to6.8%in2010.Most

ofthisrisewasduetoanincreaseintheprevalenceofNNRTIdrugresistance,mainlyintheWHOAfricanRegion.

• In the clinics surveyed, higher national coverage of antiretroviral therapy was associated with slightly greater

prevalenceofresistancebeforeantiretroviral therapy initiation.Nevertheless, theoverallestimatedeffectofan

increaseinantiretroviraltherapycoverageondrugresistanceremainedmodestintheareassurveyed.

Resistance at 12 months among people failing antiretroviral therapy • Inasubsetof29surveyswith12-monthfollow-updataavailable,(i)5.1%ofthepeopleinitiatingtherapy–excluding

thosewhodiedorwhoweretransferredtootherfacilities–haddrugresistanceat12months,(ii)76.1%achieved

viralloadsuppressionandhadnoacquiredHIVdrugresistanceand(iii)18.8%hadpossibledrugresistance,asthey

wereeitherlosttofollow-upwithunknownoutcome,stoppedtreatmentorhadaviralloadabove1000copies/ml

andnoobserveddrugresistance.

• Ofthe29clinicsthatcontributed12-monthfollow-updata,31%failedtoachievetheWHO-recommendedtarget

ofhavingatleast70%ofpeoplewithviralloadsuppressionat12months.

• Amongpatientsaliveandreceivingantiretroviraltherapyat12months,9.4%experiencedtreatmentfailure.Ina

sub-setofthesepatientswithgenotypedataavailable,72.1%carriedHIVresistanttoanydrug(69.5%toNNRTI,

62.5%toNRTIand59.9%tobothNNRTIandNRTI).Theremaining27.9%failingtherapydidsoforreasonsnot

necessarilyrelatedtodrugresistance,suchastreatmentinterruption,and,intheabsenceofteststoidentifyHIV

drugresistance,wouldhavepotentiallybeenswitchedunnecessarilytocostliersecond-lineregimens.

30

4.1 Overview

Acquired HIV drug resistance occurs when resistance

mutations are acquired due to drug-selective pressure

in individuals receiving antiretroviral therapy. Acquired

HIVdrug resistancemayemergebecauseof suboptimal

adherence, treatment interruption, inadequate plasma

drugconcentrationsortheuseofsuboptimaldrugsordrug

combinations.

Some level of resistance is expected in populations on

antiretroviral therapy (1). In thiscontext,monitoringdrug

resistance at the population level is essential to identify

and implement measures to minimize the emergence of

drugresistance.

4.2 Literature review on acquired drug resistance in low- and middle-income countries

The published literature was systematically reviewed

to describe resistance among people failing first-line

antiretroviral therapy using NNRTI-based regimens after

12months in low-andmiddle-income countries. Studies

were considered if resistance data at a median duration

of12monthswereavailableforaminimumsamplesizeof

50peopleandincludedonlyindividualsolderthan15years

ofage.Section6inAnnex1providesmethodologicalnotes

ontheliteraturereviewprotocol.

AtotalofninestudiesfromtheWesternPacificandAfrican

regions were identified. Of these, four assessed patients

12monthsafterantiretroviraltherapyinitiation,2between

10and14monthsoftherapyinitiation,onebetween7and

18monthsandtwobetween6and27months.

Table4.1summarizesthenumberofstudiesreportingfirst-

lineNNRTItherapyfailures,byregion.Mostofthestudies

wereintheWHOAfricanRegion,contributinguniquedata

from6countries,andtwostudieswereconductedinthe

WesternPacificRegion.

Apooledanalysiscomprising573individualswithavailable

genotypes at failure from 9 studies in 8 countries was

performed and is presented in Table 4.2. Among the

peopleforwhomtherapyfailedat12months,anestimated

60% had drug resistance to any drug class (NRTI 55%,

NNRTI46%).Theremaining40%hadnodrugresistance,

mostlikelyduetoverypooradherenceand/ortreatment

interruption.Importantly,intheabsenceofteststoidentify

HIV drug resistance, people experiencing therapy failure

without drug resistance would have been switched to

second-lineregimensunnecessarily.

4.3 WHO surveys to assess acquired HIV drug resistance

In addition to monitoring transmitted drug resistance,

WHO recommends, as one of the key elements of its

global HIV drug resistance surveillance and monitoring

strategy,thesurveillanceofacquiredHIVdrugresistance

in populations receiving antiretroviral therapy (2). WHO

prospective surveys of acquired HIV drug resistance are

performed at select ART clinics and describe HIV drug

resistancepresentbeforeinitiationofantiretroviraltherapy.

Additionally,surveysestimatetheprevalenceofviralload

suppressionanddescribepatternsofHIVdrugresistance

inadultandpaediatric1populationsexperiencingvirological

failure12monthsafterinitiationoffirst-lineantiretroviral

therapy.Atenrolment,surveysincludebothantiretroviral

drug–naive and antiretroviral drug–exposed individuals.

1 OnlyMozambiquesurveyedchildren(aged13yearsoryounger).

Number of studies

Africa 7

Western/Central 4

Southern 1

Eastern 2

WesternPacific 2

Totalnumberofstudies 9

Totalcountriesrepresented 8

Totalnumberofpeoplemonitored 4248

Totalnumberofpeoplefailingwithgenotype 573

Table 4.1 Number of studies included in the systematic review of acquired drug resistance, by region

RegionPrevalence of HIV drug resistance (%) (95% CI)

Anydrugclass Africa 62(47–77)

WesternPacific 51(19–84)

Overall 60(47–72)

NRTI Africa 57(44–70)


Overall 55(42–67)

NNRTI Africa 47(25–69)


Overall 46(28–64)

Table 4.2 Pooled estimates of HIV drug resistance among people experiencing first-line NNRTI therapy failure at a median duration of 12 months with genotype available, by region and by class (95% confidence levels)


31

Section7inAnnex1providesmethodologicalnotesonthe

surveyprotocol.

Twelve-monthsurveyendpointsinclude:

• stillreceivingfirst-lineantiretroviraltherapy;

• switched to second-line antiretroviral therapy: a person

is classified as “switched to second-line antiretroviral

therapy”ifheorshechangedfromfirst-tosecond-line

antiretroviraltherapyregimenasaconsequenceoffirst-

linetreatmentfailureaccordingtonationalguidelines;

• losttofollow-up:apersonisclassifiedas“losttofollow-

up”ifheorshedidnotreturntotheclinicorpharmacy

forascheduledappointmentordrugpick-upformore

than90daysafterthelastmissedclinicalappointment

ordrugpick-upandtherewasnoinformationtoclassify

thepersoninoneoftheotherendpointcategoriessuch

asdeathortransferredout;

• died;

• stopped antiretroviral therapy: a person is classified as

having “stopped antiretroviral therapy” if he or she

ceased and did not restart antiretroviral therapy at 12

months,althoughheorsheremainedincareatthesite;

and

• documented transferred to another antiretroviral therapy

clinic:apersonisclassifiedashavingbeen“transferred

to anotherclinic” ifHIVcare was transferred froman

HIVdrugresistancesurveysitetoanyotheridentified

treatmentdeliverylocation.

TheWHO-recommendedclinic level target forviral load

suppression12monthsafterantiretroviraltherapyinitiation

isatleast70%(perprotocolanalysis,withlosstofollow-up

andstoppingtherapytreatedasfailure).

Importantly,clinicsamplingmaynothavebeenperformed

inwaystoensuretherepresentativenessofantiretroviral

therapyclinicsnationallyortoensurecomparabilityover

time.Therefore,nationaland/orregionalcomparisonsmay

notbeappropriateand/orapplicable.

4.3.1 Overview

Between 2006 and 2010, 82 monitoring surveys were

initiatedin22countries.Datafromatotalof40surveys

from12countrieswere included in this report.Thirty-six

surveyshadbaselinedataavailable,and29had12-month

endpoint information. Figure 4.1 and Table 4.3 show the

geographical distribution of the WHO acquired drug

resistancesurveys.

Overall, the vast majority (92.5%, or 37 of 40) of the

surveyswereconductedintheAfricanRegion(Table4.3

andFigure4.2).

Threecountries(Cameroon,IndonesiaandMozambique)

conducted the survey in only one antiretroviral therapy

clinic,fourcountries(Burundi,India,KenyaandSwaziland)

surveyedtwoclinicsandfivecountries(Malawi,Nigeria,

South Africa, Zambia and Zimbabwe) implemented the

surveyinmultipleclinics.

4

11

13

1

3

8

0

2

4

6

8

10

12

14

Num

ber o

f sur

veys

Year of survey initiation

2006 2007 2008 2009 2010

Only baseline survey results available

Surveys with 12-month endpoint data available

Figure 4.1 Distribution of WHO surveys of acquired HIV drug resistance, by year of survey initiation, 2006–2010

2006 2007 2008 2009 2010 Total

AfricanRegion 4 10 11 4 8 37

EasternAfrica 4 4 5 13

Burundi 2 2

Kenya 1 1 2

Malawi 4a 4 8

Mozambiqueb 1 1

Western/Central 3 1 4

Cameroon 1 1

Nigeria 3 3

SouthernAfrica 6 3 3 8 20

SouthAfrica 3 3

Swaziland 2 2

Zambia 3 3

Zimbabwec 1 3 8 12

South-EastAsia 1 2 3

India 1 1 2

Indonesia 1 1

Total 4 11 13 4 8 40

a FoursurveysperformedinMalawiin2006usedacross-sectionalanalysisofpeoplereceivingantiretroviraltherapyfor12months;thusbaselinedemographicandgenotypicdataareunavailable.

b Paediatricsurveyconductedamongpeopleaged13yearsoryounger.c SurveysinitiatedinZimbabwein2009(threesurveys)and2010(eightsurveys)werein

progressandhadonlybaselinegenotypeinformationavailable.

Table 4.3 Distribution of WHO surveys of acquired HIV drug resistance by location and year, 2006–2010

32

Figure 4.2 Geographical distribution of countries (n=12) reporting results from WHO surveys of acquired HIV drug resistance, 2006–2010

Country reporting results from WHO surveys of acquired HIV drug resistance, 2006–2010

No data available or not participating in the surveys

Not applicable

WHOrecommendsthatsurveysberepeatedatthesame

select clinics at regular intervals to monitor programme

performance.WhereasKenya,SouthAfricaandZimbabwe

implementedsurveysinmultipleyearsatdifferentclinics,

only Malawi surveyed the same clinics twice: four sites

in 2006 and again in 2008 (Box 4.1). In Malawi, survey

resultsandcomplementaryoperationalresearchfindings

havebeenusedtostrengthenhealthinformationsystems,

leadingtomoreaccurateidentificationofdeathsandthose

patientstransferredtootherfacilities.

Mostsurveys(77.5%,or31of40)wereperformedinurban

areas, whereas 17.5% (7 of 40) and 2% (1 of 40) were

implemented in rural and semiurban areas, respectively.

Half of the participating clinics were public (20 of 40),

32.5%wereprivate(13of40)andaminoritywerepublic

Box 4.1 Improving clinic performance in Malawi

Malawi implemented WHO surveys of HIV drug resistance in four sites in 2006 and repeated them at the same four sites in 2008. Each of the four clinics was located in a different region of the country. All were large sites in urban areas, two were public sites and two were public sites receiving external technical support. The WHO target for clinic-level HIV drug resistance prevention (as measured by viral load suppression 12 months after antiretroviral therapy initiation) is 70% or higher. In 2006, both clinic 1 and clinic 2 fell short of the target, with HIV drug resistance prevention estimates of 60% and 68% respectively. In 2008, both clinics surpassed the target at 85% and 75%, respectively. The improvement in survey results was largely driven by a reduction in the prevalence of possible HIV drug resistance and, in particular, fewer people being classified as lost to follow-up 12 months after antiretroviral therapy initiation. Clinic 1 succeeded in decreasing the proportion of patients lost to follow-up by strengthening its health information systems, leading to more accurate identification of deaths and of those patients who had been transferred out to other facilities.

Percentage of patients meeting indicated endpoint

2006 2008

Clinic1

Clinic2

Clinic3

Clinic4

Clinic1

Clinic2

Clinic3

Clinic4

HIVdrugresistanceprevention 60.0 67.6 79.1 83.1 85.0 74.8 73.3 83.2

PossibleHIVdrugresistance 37.5 27.6 18.2 13.6 9.2 20.4 24.2 10.1

HIVdrugresistancedetected 2.5 4.8 2.7 3.4 5.8 4.9 2.5 6.7

Losttofollow-up 16.2 19.6 13.8 4.4 5.9 13.3 17.3 8.0

Death 11.5 8.4 11.7 11.3 10.5 8.0 6.7 7.3

Stop 3.4 0 0 0 0 0 0 0

withexternalsupport(17.5%,7of40).Mostofthesurveys

with available data were conducted between 2007 and

2008.Thisisduetotheprospectivenatureofthesurvey

method, which requires up to 12 months to fully enrol

patientsinthecohortandanadditionalyeartoreachthe

requisite 12-month observation endpoint. As such, most

ofthesurveysperformedin2009and2010didnothave

availabledatareadyforinclusioninthisreport.

4.3.2 Drug resistance before initiation of first-line antiretroviral therapy (survey baseline)

Inapooledanalysisof6370peopleenrolledin40surveys

of acquired drug resistance between 2007 and 2010,

596from4surveyshadnobaselinegenotypeavailable

because no specimen was obtained; of the remaining

5774peoplein36surveys,680hadnobaselinegenotype

becauseofPCRamplificationfailure.


33

In total, 5.0% of the people with available baseline

genotypes had one or more mutations in any drug class

beforetherapyinitiation(forthedefinitionofthemutation

list,seeSection5inAnnex1).

Of 5066 people with baseline reverse-transcriptase

(RT) genotypes, 228 (4.5%) had one or more mutation

associatedwithresistancetoNRTIorNNRTI(3.7%NNRTI,

1.4%NRTI,0.6%bothNNRTIandNRTI),and28of5068

withproteasegenotypes(0.6%)hadoneormoremutation

associatedwithresistancetoPI(Figure4.4).

Geographically,theprevalenceofNNRTIorNRTImutations

atbaselinewas4.3% insurveysconducted in theWHO

African Region (3.6% in the Eastern Africa subregion,

5.1% in the Southern Africa subregion and 2.5% in the

Western/CentralAfricasubregion)andreached6.3%inthe

surveysconductedintheSouth-EastAsiaRegion.

Section 2 in Annex 1 summarizes methods used for

sequencedataanalysisandqualityassurance.

Figure 4.3 depicts the disposition of people in acquired

drugresistancesurveys fromenrolment to the12-month

endpoint,focusingonsurveyresultsfrompeopleinitiating

first-lineantiretroviraltherapy.

Themeanprevalenceofresistancemutationsatbaseline

was4.8%(95%CI3.8%-6.0%) in2007,3.9%(95%CI

3.0%-4.9%)in2008,4.6%(95%CI2.2%-7.8%)in2009

N Pa2ents enrolled 6370

N Pa2ents at Baseline from Completed Surveys

N Pa2ents Alive and on First Line ART at 12 months

N Pa2ents with VL Data at 12 months

VL > 1000 cp/ml

Lost to Follow-‐up

Transferred Out

Deaths

Stopped ART

Switched to second-‐line ART

Unclassifiable

VL < 1000 cp/ml

VL unclassifiable

HIVDR No HIVDR

No specimen

VL >1000 cp/ml, genotype failed

RT Genotypes

5066

RT SDRM 228

Pts in Uncompleted Surveys No specimen 596

No genotype 680

Genotypic Results Available at Baseline

5094

PR Genotypes

5068

PR SDRM 28

PR:proteaseregionoftheHIV-1.RT:reverse-transcriptaseregionofHIV-1.SDRM:denotestheuseofthe2009WHOsurveillancedrugresistancemutationslistindataanalysis.VL:viralload.FoursurveysperformedinMalawiin2006(n=596)usedacross-sectionalanalysisofpeoplereceivingantiretroviraltherapyfor12months;hence,nobaselinedemographicandgenotypedatawereunavailable.Pts:Patients.N:number.Cp=copies.

Figure 4.3 Flow diagram of individuals enrolled in WHO surveys of acquired HIV drug resistance: from baseline to 12-month endpoints

34

0

1%

2%

3%

4%

5%

6%

NRTI and N

NRTI mutatio

ns

Any mutatio

nsRT m

utations

NNRTI mutatio

nsNRTI m

utations

PI mutatio

ns

K103N

S

Y181C

G190A

S

K101E

V106A

M

Y188C

HL

M18

4IVT21

5DFIS

Y

M41

L

K219E

N

K70R

L210

W

D67N

Perc

enta

ge o

f bas

elin

e ge

noty

pes

Mutationsweredefinedusingthe2009WHOsurveillancedrugresistancemutationslist.

Figure 4.4 Prevalence of HIV drug resistance mutation at baseline in WHO acquired HIV drug resistance surveys

and reached 6.8% (95% CI 4.8%-9.0%) in 2010 (Table

4.4). Among the sites surveyed in the African Region,

baselineNNRTIresistancerosefrom3.4%(95%CI2.4%-

4.5%)to5.4%(95%CI3.7%-7.4%)inthesameperiod,

astatisticallysignificantincrease(p-value=0.03),afact

thatmayberelatedtopreviousantiretroviraldrugexposure

(prevention of mother-to-child transmission, previous

antiretroviral therapy) or to transmitted drug resistance.

Table 5 in Annex 2 shows the estimated prevalence of

baselineresistancebyregionandbydrugclass.Section9

in Annex 1 provides additional details on the statistical

methodsused.

Figure4.5depictstherelationshipbetweentheprevalence

ofHIVdrugresistancemutationsamongpeopleinitiating

treatmentandantiretroviraltherapycoverage,definedas

thenumberofpeoplelivingwithHIVreceivingantiretroviral

drugsdividedbythetotalnumberofpeoplelivingwithHIV

inthecountrywherethesurveywasundertaken.Inclinics

surveyed, the prevalence of resistance mutations was

positivelycorrelatedwithcoverageofantiretroviraltherapy

(p-valueadjusted for region=0.025;odds ratioper 10%

ART increase 1.38, 95% CI 1.09–1.75). Nevertheless, the

overallestimatedeffectondrugresistanceofanincrease

in antiretroviral therapy coverage remained modest,

suggesting that treatment was expanded in the areas

surveyedwithouttriggeringunexpectedincreases inHIV

drugresistance.Section9 inAnnex1providesadditional

detailsonthestatisticalmethodsused.

% (95% CI)

P-valuea2007 2008 2009 2010

Any 4.8(3.8–6.0) 3.9(3.0–4.9) 4.6(2.2–7.8) 6.8(4.8–9.0) 0.06

NRTI 1.2(0.7–2.0) 1.3(0.8–2.0) 1.1(0.3–2.2) 1.0(0.3–2.1) 0.70

NNRTI 3.7(2.5–4.9) 2.4(1.6–3.3) 3.3(1.8–5.1) 5.5(3.8–7.4) 0.06

PI 0.3(0.0–0.7) 0.4(0.1–0.8) 0.5(0.0–1.7) 0.0(0.0–0.4) 0.97

a Statisticalmethodsaredescribedinsection9,Annex1.

Table 4.4 Prevalence of HIV drug resistance at baseline in WHO acquired HIV drug resistance surveys (n=36), by year of surveys and drug class, 2007–2010

Box 4.2 Relationship between previous exposure to antiretroviral drugs and detection of resistance-associated mutations at baseline

A subset of individuals enrolled in WHO surveys of acquired HIV drug resistance responded to a questionnaire about previous exposure to antiretroviral drugs for the purpose of characterizing the relationship between previous drug exposure and HIV drug resistance at baseline. Overall, 3464 people had both information about prior exposure and a RT genotype result; 286 (8.3%) reported previous antiretroviral drug exposure, 44 of whom (15.4% of the 286 reporting prior exposure) had one or more RT resistance mutations at baseline. In contrast, 3178 (89.7%) reported no previous antiretroviral drug exposure, 124 of whom (3.9% of the 3178 reporting no prior exposure) had one or more resistance mutations in RT at baseline. This suggests that people reporting prior exposure to antiretrovirals are more likely to carry HIV drug resistance at baseline (p-value < 0.001, Fisher exact test).1

1 Table6inAnnex2detailsindividualsurveyresultsbyantiretroviraltherapyclinic.


35

4.3.3 Acquired drug resistance among people failing first-line antiretroviral therapy at 12 months

Ofthe6370peopleenrolled,4764completedthesurvey

andhadendpointdataavailableforanalysis.1Ofthese,3475

werealiveandreceivingfirst-lineantiretroviraltherapyafter

12 months. Seven switched to second-line regimens, 13

stoppedtherapy,294transferredcaretoanotherclinic,599

werelosttofollow-up,362diedand14hadunclassifiable

surveyendpoints(Table4.5).Table7andFigure2inAnnex

1 Elevensurveysinitiatedin2009and2010inZimbabwewith1606peopleenrolledwerestillongoingasofthewritingofthisreport,andonlybaselinedatawereavailable.

Figure 4.5 Relationship between antiretroviral therapy coverage and prevalence of NNRTI drug resistance mutations at ART initiation

0%

2%

4%

6%

8%

10%

12%

14%

16%

18%

20%

0% 5% 10% 15% 20% 25% 30% 35%

Prev

alen

ce o

f NN

RTI r

esis

tanc

e m

utat

ions

: %

of g

enot

ypes


Eastern AfricaSouthern AfricaWestern/Central AfricaSouth-East Asia


2provideclinic-leveldataonthenumberandproportion

ofpeoplelosttofollow-up,stoppingantiretroviraltherapy,

transferringout,dyingandswitchingclinics.

WHOsurveysofacquiredHIVdrugresistancehavethree

survey outcomes: HIV drug resistance prevention (viral

load<1000copies/ml),HIVdrugresistance2andpossible

HIVdrugresistance(includedinthiscategoryarepeople

lost to follow-up, individuals who stopped antiretroviral

therapy, for whom drug resistance cannot be assessed,

2 HIVdrugresistancedefinedaslow,moderateorhighinterpretationusingStanfordHIVdrugresistancealgorithm

RegionPeople

enrolled

People on first-line ART at 12 months

n (%)

Lost to follow-upn (%)

Stopped ARTn (%)

Transferred outn (%)

Deathsn (%)

Switched to second-line

ARTn (%)

Unclassifiablen (%)

AfricanRegion 4365 3211(73.6%) 541(12.4%) 11(0.3%) 268(6.1%) 315(7.2%) 7(0.2%) 12(0.3%)

EasternAfrica 2023 1494(73.9%) 189(9.3%) 6(0.3%) 176(8.7%) 148(7.3%) 0(0%) 10(0.5%)

SouthernAfrica 1710 1314(76.8%) 168(9.8%) 5(0.3%) 80(4.7%) 136(8%) 5(0.3%) 2(0.1%)

Western/CentralAfrica 632 403(63.8%) 184(29.1%) 0(0%) 12(1.9%) 31(4.9%) 2(0.3%) 0(0%)

South-EastAsia 399 264(66.2%) 58(14.5%) 2(0.5%) 26(6.5%) 47(11.8%) 0(0%) 2(0.5%)

Overall 4764 3475(72.9%) 599(12.6%) 13(0.3%) 294(6.2%) 362(7.6%) 7(0.1%) 14(0.3%)

Table 4.5 Endpoints of WHO acquired HIV drug resistance surveys (n=25) with both baseline and endpoint data available

36

RegionHIV drug resistance prevention (% of

people initiating therapya)

Any HIV drug resistance at endpointb

Possible HIV drug resistance(% of people initiating

therapya)% of people initiating

therapya% of people genotyped at

treatment failure

AfricanRegion 76.6% 4.7% 69.5% 18.8%

Eastern 79.4% 4.3% 63.7% 16.4%

Southern 80.3% 4.7% 73.3% 15.0%

Western/Central 59.9% 6.0% 74.5% 34.1%

South-EastAsia 71.4% 8.9% 93.3% 19.7%

Overall 76.1% 5.1% 72.1% 18.8%

a Excludespeoplewhodiedorwhoweretransferredtoanotherantiretroviraltherapyfacility.b HIVdrugresistancedefinedasadrugresistancepredictionoflow,intermediateorhighlevelusingtheStanfordHIVdatabasealgorithm.Alternatively,ifcalculatedbasedonthenumberofsurveillance

drugresistancemutationsatendpoint,subregional,regionalandoverallproportionsremainidentical.

Table 4.6 Outcomes of the HIV drug resistance surveys at endpoints

Outcomes: HIVDR preven2on

HIVDR HIVDR Possible

N Pa2ents enrolled 6370

N Pa2ents at Baseline from Completed Surveys 4764

N Pa2ents Alive and on First Line ART at 12 months 3475

N Pa2ents with VL Data at 12 months 3219

VL > 1000 cp/ml 301

Lost to Follow-‐up 599

Transferred Out 294

Deaths 362

Stopped ART 13

Unclassifiable 14

VL < 1000 cp/ml 2918

VL unclassifiable 35

HIVDR 192 No HIVDR 75

No specimen 220

VL >1000 cp/ml, genotype failed 34

RT Genotypes

5066

RT SDRM 228

Pts in Uncompleted Surveys 1606 No specimen 596

No genotype 680

Genotypic Results Available at Baseline

5094

PR Genotypes

5068

PR SDRM 28

Switched to second-‐line ART 7

HIVDR 2

No VL 4

VL >1000 cp/ml, genotype not done 1

PR:proteaseregionofHIV-1.RT:reverse-transcriptaseregionofHIV-1.VL:viralload.Pts:patients.N:number.Cp:copies.

Figure 4.6 Flow diagram of acquired HIV drug resistance survey: 12-month endpoints and outcomes


37

and people with viral load greater than 1000 copies/ml

12monthsafter therapy initiationbutnodrugresistance

mutationsdetected).

Table 4.6 summarizes survey outcomes by region and

subregion. Section 8 in Annex 1 provides a detailed

explanationofeachsurveyoutcomeandTable8inAnnex

2providesclinic-specificoutcomeresults.

4.3.3.1 Drugresistanceinpatientsfailingtherapyat

12months

In total, 194 people – 5.1% of those initiating therapy,

excludingpatientswhodiedorwhoweretransferredout

to other facilities – had drug resistance at 12 months.1

Prevalence of drug resistance at 12 months varied

considerablyamongclinics,from0.6%inonesiteinSouth

Africa(2007)to9.7%inaclinicinIndia(2008).

Among those patients failing therapy, the prevalence of

drugresistancewas72.1%,rangingfrom25%inoneclinic

inBurundi(2007)to100%inaclinicinKenya(2008),one

clinicinNigeria(2008),oneclinicinMozambique(2007),

twoclinicsinMalawi(onein2006andonein2008)and

oneclinicinIndonesia(2008).Thisimpliesthatalmosta

thirdofthepeoplewerefailingtherapyforreasonsother

thandrugresistance.Althoughseveralfactorsmaybeat

play, people with viral loads exceeding 1000 copies/ml

butwithoutdrugresistancearelikelytohaveexperienced

treatment interruptionorhavehadverypooradherence.

Resistance to NNRTI and NRTI was, respectively, 69.5%

and 62.5% among people failing therapy with genotype

dataavailable.Table4.7summarizesHIVdrugresistance

resultsatsurveyendpointbydrugclassandbyregion.Table

9inAnnex2summarizestheclinic-levelresultsfortheHIV

drugresistance).

Overall,only15%(36of229)ofpeoplefailingARTwith

matching baseline-endpoint genotypes had virus with

RT inhibitor resistance before antiretroviral therapy

initiation. This implies that treatment failure among the

remaining 85% was probably not associated with pre-

existingresistance,althoughsomemayhavehadresistant

virusespresentatlevelsbelowthesensitivityofstandard

genotypingassays.

4.3.3.2 Drugresistanceprevention

In total, 76.1% of people initiating treatment achieved

viral loadsuppressiononastandardfirst-lineregimenat

12months.

1 ThisincludestwopatientswithHIVdrugresistantvirusatthetimeofswitchingtosecond-linetherapypriorto12months.

RegionNumber of

patients Any NRTI Any NNRTI

NRTI and NNRTI Any druga

AfricanRegion 239 59.8% 66.9% 57.3% 69.5%

Eastern 102 52.9% 61.8% 51.0% 63.7%

Southern 90 64.4% 68.9% 60.0% 73.3%

Western/Central 47 66.0% 74.5% 66.0% 74.5%

South-EastAsia 30 83.3% 90.0% 80.0% 93.3%

Overall 269 62.5% 69.5% 59.9% 72.1%

a AnydrugincludesNRTI,NNRTIandPI.TheresultsforPIsarenotshown.PIdrugresistancewasonlyobservedfornelfinavir,resultingfromthepresenceofmultiplepolymorphicmutations,especiallyinsubtypesC,G,andCRF02_AG.NelfinavirresistancewasobservedinninespecimenswithoutNRTIorNNRTIresistance.Nodrugresistancewaspredictedforanyritonavir-boostedPI.

Table 4.7 HIV drug resistance results among people failing therapy at 12 months, by region and drug class

In the subset of people who were alive and receiving

antiretroviraltherapy12monthsaftertreatmentinitiation

and had available viral load data, 90.6% (2918 of 3219)

achievedviralloadsuppression.Nevertheless,attheclinic

level,31.0%(9of29)ofsitesdidnotachievetheWHO-

suggested target of having at least 70% of people with

viral load suppression 12 months after therapy initiation.

Moreover,anadditional27.6%(8of29)ofclinicsclustered

justabovethetarget(70–80%).Infouroftheclinics,less

than60%ofpatientsachievedviralloadsuppression.Poor

performanceleadstomajorconsequencesintermofcost

and probably adversely affects morbidity and mortality

outcomes. These results are particularly concerning in

countries in which multiple clinics reported consistently

under-performingresults(Figure4,annex2.).Table10in

Annex2summarizestheclinic-levelresultsfortheHIVdrug

resistancepreventionoutcome.

4.3.3.3 Possibledrugresistance

Atotalof722(18.8%)patientswereclassifiedashaving

possible HIV drug resistance (75 with viral load greater

than 1000 copies/ml at 12 months and no resistance,

13whostoppedantiretroviraltherapy,599whowerelost

tofollow-up,34withviralloadgreaterthan1000copies/

mlat12monthsbutwithspecimensfailingtoamplifyand

1withviralloadgreaterthan1000copies/mlatswitchbut

failingtoamplifyPCRproducts).

Although only 5.1% of people initiating therapy had HIV

drugresistanceat12months,thelevelofpossibleHIVdrug

resistance,whichfactorsintheunknownoutcomesassociated

withpeople lost to follow-uporwhostoppedantiretroviral

therapy,wasmuchgreater,at 18.8%.This impliesthat the

prevalenceofHIVdrugresistancecouldbeconsiderablyhigher

thansuggestedbydirectmeasuresofHIVdrugresistance.

PossibleHIVdrugresistancerangedwidelyfrom4.1%in

one clinic in Zimbabwe in 2008 to 46.2% in a clinic in

Swazilandin2008.Table11inAnnex2summarizesclinic-

levelresultsforpossibleHIVdrugresistanceoutcome.

38

Inthisanalysis,highlevelsofpossibledrugresistancewere

mostlydrivenbythesubstantialproportionsofpeoplewho

werelosttofollow-uporwhostoppedantiretroviraltherapy.

Figure3inAnnex2providesclinic-leveldataonpossible

HIVdrugresistance.WHOearlywarningindicatorguidance

recommends that no more than 20% of patients should

be lost to follow-up12monthsafter treatment initiation.

Of the 29 surveys conducted between 2006 and 2010,

17% (5 of 29) did not meet this target. Of note, almost

onethird(29.1%)ofthepeopleinitiallyenrolledinsurveys

conductedintheWestern/CentralAfricasubregionwere

lost to follow-up at 12 months, considerably higher than

theaveragesinotherregionsandsubregions.Theobserved

rates of lost to follow-up and possible drug resistance

suggest theneed tostrengthendefaulter tracingand re-

engagement mechanisms as well as health information

systems.Exceptionally,atbothsitessurveyedinBurundi,

possibleHIVdrugresistancewasmostlycausedbypeople

withviralloadgreaterthan1000copies/mlandnoHIVdrug

resistanceongenotyping,suggestingpatientsarelikelyto

haveexperiencedtreatmentinterruptionorhavehadvery

pooradherence.

4.3.3.4 PrevalenceandpatternsofHIVdrugresistance

amongpeopleexperiencingtreatmentfailure

12monthsafterinitiation

Themajority(87%)ofthepeoplebeingtreatedreceived

athymidineanalogue–containingregimen,andarelatively

small proportion were on tenofovir-containing regimens

(about 12%).Amongthe269 individuals failingfirst-line

antiretroviral therapy with a genotype result available,

38.7% retained susceptibility to both 3TC/FTC and

tenofovir,46.8%hadareductioninsusceptibilityto3TC

onlyandnonehadtenofovirresistanceonly.Only14.5%had

reducedsusceptibilitytobothtenofovirand3TC.

Box 4.3 Possible HIV drug resistance: why is it important?

People categorised as having possible drug resistance are most likely to have experienced treatment interruption and/or have had very poor adherence. Treatment interruptions of NNRTI-based regimens of 48 hours or longer are associated with the selection of NNRTI drug resistance and increased risk of virological failure (3,4). The fact that no HIV drug resistance was observed in 27.9% of patients failing ART at 12 months in WHO surveys may be accounted for by the fact that HIV drug resistance may have been present but predominantly reverted to drug-sensitive wild-type virus. Moreover, standard population-based sequencing (standard commercial and laboratory assays) only detects drug resistance if it is present at about 10–20% of the virus population (5). Notably, HIV drug resistance present as minority variants may pass undetected, persisting for months or years after treatment (6–8) and may re-emerge in the viral population after treatment is reinitiated, impacting treatment outcomes adversely (9).

Correlation between regimen and HIV drug resistance

outcomewasnotpossibleduetothelackofpatient-level

data.Thusitwasnotfeasibletodeterminewhetherreduced

susceptibilitytotenofovirresultedfromtheuseoftenofovir

or stavudine among the people experiencing treatment

failure.

Figure 4.7 shows HIV drug resistance among patients

with therapy failure 12 months after initiation, by drug

and drug class. Overall, these data suggest that, if

populations experiencing first-line antiretroviral therapy

failurewereswitchedtosecond-lineregimenssoonafter

initial virological failure, the virus would retain at least

partial susceptibility to currently recommended second-

lineNRTIcomponents,thusmaximizingtheirresponseto

boosted PI-based second-line antiretroviral therapy. This

assessment issupportedbytheresultsof recentstudies

oftheresponsetosecond-linetherapyinlow-andmiddle-

incomecountries(10–13).

Figure 4.8 shows the prevalence of HIV drug resistance

mutations among people experiencing treatment failure

at12months.CommonlyobservedNRTImutationswere

M184V (58.7%), K65R (10.4%), D67N (7.1%), K70R

(6.7%), multiple variants at T215 (5.6%) and multiple

variantsatK219(4.8%).

One or more thymidine analogue resistance-associated

mutations(TAM)were identified in 15.6%of thepeople

being treated. Table 4.8 presents the distribution of

endpointgenotypes(n=269)withrespecttothenumber

ofTAMdetectedandwhethertheTAMpatternresembled

thatseenforTAMpathways1or2.Onesequencehada

T215STmixtureandcouldnotbeassignedtoaparticular

pathway.TAMsaredefinedas:M41L,D67N,K70EorR,

L210W,anymutationatT215andanymutationatK219.

Onlynineof thepeople(3.3%)hadthreeormoreTAM,

conferringhigh-levelresistancetoNRTI.CommonNNRTI

mutations included K101E (9.3%), K103N or S (29%),

V106AorM(10.4%),Y181C,IorV(29.4%),Y188C,Hor

L(6.7%)andG190AorS(17.5%).

Table 12 in Annex 2 describes the distribution of HIV

subtypesobservedbycountry.Table13inAnnex2provides

details of drug resistance among people experiencing

treatment failure at 12 months, by antiretroviral therapy

clinicandgeographicalregion.Table14inAnnex2provides

detailsaboutregionalandsite-specificprevalenceofmajor

resistance-associatedmutations.


39

Number of TAMs

Number of people (%)

TAM pathway 1a

(n)

TAM pathway 2b

(n)

TAM pathway

undefined (n)

0 227(84.6%)

1 23(8.2%) 3 19 1

2 10(3.7%) 3 7 0

3 3(1.1%) 0 3 0

4 5(1.9%) 1 4 0

5 1(0.4%) 0 1 0

Total 269 7 34 1

a Pathway1wasassignedifanyofthefollowingwaspresent:M41L,L210WorT215Y.b Pathway2wasassignedifanyofthefollowingwerepresent:D67N,K70EorR,any

mutationatK219orT215F.IncaseswheretherewasoverlapofTAM1andTAM2mutations,theaminoacidatposition215(ForY)wasusedtomakethedetermination.

Table 4.8 Prevalence of thymidine analogue resistance-associated mutations (TAM), by pattern

Any drug

Any RTI

NRTI and N

NRTI

Any PIa NVP

EFV RPVETR

3TC/FT

CABC ddI

d4T TDFAZT

Perc

enta

ge o

f end

poin

t gen

otyp

es

0% 10% 20% 30% 40% 50% 60% 70% 80% 90%

100%

Any NNRTI

Any NRTI

a PIdrugresistancewasonlyobservedfornelfinavir,resultingfromthepresenceofmultiplepolymorphicmutations,especiallyinsubtypesC,GandCRF02_AG.NelfinavirresistancewasneverobservedinspecimenswithoutpredictedNRTIorNNRTIresistance.Nodrugresistancewaspredictedforanyritonavir-boostedPI.DetailedmethodologicalnoteareavailableinSection5,annex1.

Figure 4.7 HIV drug resistance among people experiencing treatment failure at 12 months, by drug and drug class

Perc

enta

ge o

f end

poin

t gen

otyp

es

0%

10% 20% 30% 40% 50% 60% 70% 80%

RT mutatio

ns

NNRTI mutatio

ns

Any PI m

utations

K103N

S

Y181C

IV

G190A

ES

V106A

MK10

1E

Y188C

HL

M230L

P225H

M184IV K65

R

D67GN

NRTI mutatio

ns

NRTI and N

NRTI mutatio

ns

K70ER

T215IF

NSY

K219E

NQM41

LL7

4IV


Figure 4.8 Prevalence of HIV drug resistance–associated mutations among people experiencing treatment failure at 12 months

40

REFERENCES1. PingenMetal.Evolutionarypathwaysoftransmitteddrug-resistantHIV-1.JournalofAntimicrobialChemotherapy,2011,66:1467–1480.


1. RichmanDDetal.TheprevalenceofantiretroviraldrugresistanceintheUnitedStates.AIDS,2004,18:1393–1401.

2. JordanMRetal.WorldHealthOrganizationsurveys tomonitorHIVdrugresistancepreventionandassociated factors insentinelantiretroviraltreatmentsites.AntiviralTherapy,2008,13(Suppl.2):15–23.

3. Parienti JJ et al. Predictors of virologic failure and resistance in HIV-infected patients treated with nevirapine- or efavirenz-basedantiretroviraltherapy.ClinicalInfectiousDiseases,2004,38:1311–1316.

4. OyugiJHetal.TreatmentinterruptionspredictresistanceinHIV-positiveindividualspurchasingfixed-dosecombinationantiretroviraltherapyinKampala,Uganda.AIDS,2007,21:965–971.

5. HalvasEKetal.Blinded,multicentercomparisonofmethodstodetectadrug-resistantmutantofhumanimmunodeficiencyvirustype1atlowfrequency.JournalofClinicalMicrobiology,2006,44:2612–2614.

6. Palmer S et al. Selection and persistence of non-nucleoside reverse transcriptase inhibitor-resistant HIV-1 in patients starting andstoppingnon-nucleosidetherapy.AIDS,2006,20:701–710.

7. HanceAJetal.Changesinhumanimmunodeficiencyvirustype1populationsaftertreatmentinterruptioninpatientsfailingantiretroviraltherapy.JournalofVirology,2001,75:6410–6417.

8. LecossierDetal.DetectionofminoritypopulationsofHIV-1expressingtheK103Nresistancemutationinpatientsfailingnevirapine.JournalofAcquiredImmuneDeficiencySyndromes,2005,38:37–42.

9. Li JZ et al. Low frequency HIV-1 drug resistance mutations and risk of NNRTI-based antiretroviral treatment failure. JAMA, 2011,305:1327–1335.

10. HosseinipourMCetal.Thepublichealthapproachtoidentifyantiretroviraltherapyfailure:high-levelnucleosidereversetranscriptaseinhibitorresistanceamongMalawiansfailingfirst-lineantiretroviraltherapy.AIDS,2009,23:1127–1134.

11. BartlettJAetal.Lopinavir/ritonavirmonotherapyaftervirologicfailureoffirst-lineantiretroviraltherapyinresource-limitedsettings.AIDS,2012,[Epubaheadofprint].

12. HosseinipourMCetal.Second-linetreatment in theMalawiantiretroviralprogramme:highearlymortality,butgoodoutcomes insurvivors,despiteextensivedrugresistanceatbaseline.HIVMedicine,2010,11:510–518.

13. SigaloffKCEetal.Second-lineantiretroviraltreatmentsuccessfullyre-suppressesdrug-resistantHIV-1afterfirst-linefailure:prospectivecohortinsub-SaharanAfrica.JournalofInfectiousDiseases,2012,205:1739–1744.


41

5.1 Overview1

Inthefaceofslowlyincreasingdrugresistancetrends,and

thegrowinguseofantiretroviraltherapyforbothtreatment

andprevention,effortsmustberedoubledtoensurethatthe

emergenceofdrugresistanceisadequatelymonitoredand

minimized.Severalantiretroviraltreatmentprogrammeand

sitefactorshavebeenshown(seeChapter1)tobeclosely

associated with the emergence and transmission of HIV

drugresistance,includingthequalityofcare,adherenceto

antiretroviraltherapyandclinicandprogrammefunctioning

(1,2).

Whereas genotyping is expensive and complex, the

monitoringofsuchfactorsiscomparativelyinexpensiveand

canbesuccessfullyusedtotimelyidentifygapsinservice

deliverysothatcorrectiveactioncanbetakentominimize

the emergence of HIV drug resistance. In 2004, WHO

1 ThissectionreliesextensivelyonBennettetal(4).

5. EARLY WARNING INDICATORS

KEY MESSAGES

• Earlywarning indicatorsofHIVdrugresistancemonitor factorsat individualclinicsknowntocreatesituations

favourabletotheemergenceofHIVdrugresistance.Thetimelyidentificationofclinicswithsuboptimalperformance

helpstotargetappropriateinterventionsthatcanpotentiallyreducetheriskofHIVdrugresistanceemergingand

optimizecare.Since2004,earlywarningindicatorshavebeenmonitoredat2017antiretroviraltherapyclinicsin

50countriesassessing131686people.

• Overall,75%ofclinicsmonitoredmetthetargetof100%ofpatientsreceivingprescriptionsforantiretroviraltherapy

inaccordancewithnationalorWHOguidelines.Whereas74%ofclinicssurveyedinAfricaand80%inAsiamet

thistarget,only46%achieveditinLatinAmericaandCaribbean.

• Withrespecttopatientslosttofollow-upat12months(earlywarningindicator2),overall69%ofclinicsmonitored

mettheWHO-recommendedtarget,rangingfrom59%inAfricato75%inAsiaand85%inLatinAmericaand

theCaribbean.Sixty-sevenpercentofclinicsmettherecommendedlevelforretentiononfirst-lineantiretroviral

therapyat12months(earlywarningindicator3).

• Seventeen per cent of reporting clinics achieved WHO’s recommended target for on-time drug pick-up (early

warningindicator4),and58%metWHO’srecommendedtargetforon-timeappointmentkeeping(earlywarning

indicatorassessing5).Withrespecttodrugsupplycontinuity(earlywarningindicator6),only65%ofreporting

clinicsprovidedacontinuoussupplyofantiretroviraldrugduringa12-monthperiod.

• Althoughthesmallnumberofreportingsitesprecludesregionalorglobalgeneralizations,reporteddataidentified

importantgapsinservicedeliveryandprogrammeperformance,particularlyinprocurementandsupplysystems,

patientadherenceandclinicretention.

developedasetofeightHIVdrugresistanceearlywarning

indicatorstomonitorthesefactors,eachassociatedwitha

recommendedtargetforclinic-levelmonitoring.

Since 2004, more than 50 countries have monitored

one or more early warning indicators at select clinics.

AlthoughWHOrecommendsthatearlywarningindicators

bemonitoredannuallyatallantiretroviral therapyclinics

within a country or at a large number of representative

clinics, most countries have monitored early warning

indicators inaconvenientsampleofsites.Therefore,the

dataobtainedarenotnationallyrepresentativeandpreclude

the assessment of regional/global trends. Nevertheless,

reportsdocumentedimportantgapsinservicedeliveryand

programmeperformance.

Table 5.1 summarizes the results from cohorts of people

initiatingantiretroviraltherapybetween2004and2009,

assessing 131 686 people at 2107 clinics since 2004,

42

comprising: African Region, 907 clinics in 25 countries;

Asia(WesternPacificRegionandSouth-EastAsiaRegion

combined):1048clinicsin6countries;LatinAmericaand

the Caribbean: 148 clinics in 18 countries; and European

Region:4clinicsin1country.

Earlywarningindicators1,2and3(prescribingpractices,loss

tofollow-upandretentiononfirst-lineantiretroviraltherapy

at12months,respectively)werethethreeindicatorsmost

frequentlymonitored.Despitetheirimportantrelationship

toHIVdrugresistance,aminorityofclinicsreportedearly

warning indicators 4 and 5, and the reporting of early

warningindicator6wasintermediate.Thefrequencywith

which early warning indicators 1–6 were reported was

probably associated with the ease of data abstraction.

Earlywarningindicator7(adherenceassessedthroughpill

count; rarely implemented in programme practice) was

monitoredinonlytwocountries(lessthan1%ofclinics)and

wasexcludedfromtheanalysis.Veryfewclinicsreported

on early warning indicator 8 because of limited routine

useofviralloadtestingforclinicalmonitoringpurposes.In

thefuture,asviralloadtestingbecomesmoreaccessible,

reportingof ratesofviral loadsuppression isanticipated

toincrease.

The percentage of adult clinics meeting WHO-

recommendedtargetsvariedconsiderablybyearlywarning

indicatorandregion(Table5.1).

Available data indicate that, overall, 75% of clinics

monitored met the target of 100% of the service users

receiving prescriptions for antiretroviral therapy in

accordancewithnationalorWHOguidelines(earlywarning

indicator 1). Whereas 74% and 80% of clinics in Africa

andAsia,respectively,metthistarget,only46%achieved

itinLatinAmericaandtheCaribbean.Thismayberelated

to the greater use of more individualized approaches to

antiretroviraltherapyinLatinAmericaandtheCaribbean

andtheclassificationoffirst-lineregimensthatcontained

PIortenofoviras“inappropriate”whennotrecommended

bynationalguidelines,eventhoughtheywouldnotunduly

haveselectedforHIVdrugresistance.

Withrespecttoearlywarningindicator2(losstofollow-up

at12months),69%ofclinicsmettheWHO-recommended

target,rangingfrom59%intheAfricanRegionto75%in

Asia and 85% in Latin America and the Caribbean. No

directcomparisonscanbemadesincethecountriesand

clinicssurveyedwerenotthesame,butthisresultisbroadly

consistentwiththerelativelyhigherlevelsoflosstofollow-

upobservedinsomeofthesitesmonitoredintheAfrican

Regioninthecontextofsurveysofacquireddrugresistance

(Chapter4).

Sixty-sevenpercentoftheclinicsmettherecommended

level forearlywarning indicator3(retentiononfirst-line

antiretroviral therapy), with regional averages ranging

Indicator

Early warning indicator 1: Prescribingpractices

Early warning indicator 2:

Losstofollow-up


Retentiononfirst-line

antiretroviraltherapyat12months


On-timeantiretroviraldrugpick-up


On-timeappointment

keeping

Early warning indicator 6: Antiretroviraldrugsupplycontinuity


Viralloadsuppressionat

12months

Target 100% ≤20% ≥70% ≥90% ≥80% 100% ≥70%

AfricanRegion(allyears)

Numberofclinics 907 794 863 321 309 537 24

%ofclinicsmeetingrecommendedlevel 74% 59% 61% 15% 43% 63% 96%

Asia(allyears)

Numberofclinics 1048 1043 1045 10 1037 100 —

%ofclinicsmeetingrecommendedlevel 80% 75% 72% 0% 64% 89% —

LatinAmericaandtheCaribbean(allyears)



Total(allregions,allyears)



Thesitessurveyedreflecthealthsystemsthatarehighlyheterogeneousinstructureandfunding,andsuchdifferencesmayhaveinfluencedearlywarningindicatorfindings.Inaddition,country-specificdataheavilyinfluencedregionalandglobaldata;forexample,Thailandmonitoredaconsiderablylargernumberofclinicsthananyothercountry(902of2107adultclinicsincludedintheanalysisand296of331paediatricclinics).Moreover,clinicsamplingmaynothavebeenperformedinwaystoensuretherepresentativenessofantiretroviraltherapyclinicsnationally.Nationaland/orregionalcomparisonsmaythereforenotalwaysbeappropriateorapplicable.Earlywarningindicator7(adherenceassessedthroughpillcount)wasexcludedfromtheanalysissinceitwasmonitoredinonlytwocountries(lessthan1%ofclinics).— Datanotavailableorapplicable.

Table 5.1 Number of clinics monitored and percentage of clinics achieving recommended targets by early warning indicator and region by adult cohorts, 2004–2009


43

between60%and70%.Improvingretentiononfirst-line

antiretroviraltherapyat12monthsisessential,sincemany

countries and clinics have only one second-line regimen

availableandnosalvagealternatives.Thus,itisnecessary

tooptimizeadherencetofirst-lineantiretroviraltherapyand

minimizeinappropriateswitchingtosecond-lineregimens

duringthefirst12monthstoenhancethelong-termsuccess

ofpopulation-basedantiretroviraltherapy.

Although fewclinicsmonitoredviral loadsuppressionat

12 months, among those that did, 85% met the WHO-

recommendedtarget.

Seventeen per cent of reporting clinics achieved WHO’s

recommendedlevelforearlywarningindicator4(on-time

drug pick-up), and 58% achieved WHO’s recommended

targetforearlywarningindicator5(on-timeappointment

keeping).Withrespecttoearlywarning indicator6,only

65% of reporting clinics provided a continuous supply

of antiretroviral drugs during a 12-month period, ranging

from51%to89%indifferentregions.Althoughthesmall

numberofreportingsitesprecludesgeneralizingtheserates

tospecificregions,availabledataindicatethatprocurement

and supply distribution remain as important programme

challenges.

5.2 Revised early warning targets and indicators

In 2012, after a critical review of the available medical

literatureandthemultiplechallengesobservedwithdata

collection and reporting, early warning indicators were

simplified and harmonized with other monitoring and

evaluation frameworksandprocesses, including thoseof

theUnitedNationsSpecialSessiononHIV/AIDSandthe

UnitedStatesPresident’sEmergencyPlanforAIDSRelief.

Thenumberofcoreindicators(Table5.2)hasbeenreduced

to four: on-time pill pick-up, dispensing practices, drug

supply continuity and clinic retention at 12 months. A

fifth indicator, viral load suppression at 12 months, is

recommended and should be monitored at sites where

viral load testing is routinely performed 12 months after

therapyinitiation.

Therevisedsetofindicatorsisanticipatedtorequireless

dataabstraction,facilitatingwideruptakeandreporting.

Recommended targets have been adjusted to take into

account new scientific evidence on optimal programme

managementandperformance.Monitoringofearlywarning

indicatorsisnowbasedonascorecardapproach(Table5.2)

tofacilitatetheinterpretationofprogrammedata.Scorecards

produce three classifications: red (poor performance,

belowthedesiredlevel),amber(fairperformance,notyet

at the desired level) and green (excellent performance,

achievingthedesiredlevel).Scorecardingalsoallowsfora

greyclassificationifclinicsdonotmonitoraspecificearly

warningindicatorandawhiteclassificationifan indicator

isnot reported inaspecificyear followingpredetermined

nationalconvention(3).

Earlywarningindicatorsprovidecrucialinformationonthe

performanceoftreatmentclinicsandcanbeinstrumental

in prioritizing actions andallocating resources forclinics

mostinneed.Aggregatingearlywarningindicatorresults

fromarepresentativesampleorallclinicswithinacountry

can highlight broader programmatic issues hampering

the achievement of desired outcomes so that treatment

outcomes can be maximized and the emergence of HIV

drugresistancecanbeminimized.

Table 5.2 Revised set of early warning indicators and WHO-recommended targets (2012)

HIV drug resistance early warning indicator scorecard

Earlywarningindicator Status Target

1.On-timepillpick-up Red/amber/green

Red:<80%Amber:80–90%Green:>90%

2.Retentionincarea Red/amber/green/white

Red:<75%retainedafter12monthsofantiretroviraltherapyAmber:75–85%retainedafter12monthsofantiretroviraltherapyGreen:>85%retainedafter12monthsofantiretroviraltherapy

3.Pharmacystock-outs Red/green

Red:<100%ofa12-monthperiodwithnostock-outsGreen:100%ofa12-monthperiodwithnostock-outs

4.Dispensingpractices Red/green

Red:>0%dispensingofmono-ordualtherapyGreen:0%dispensingofmono-ordualtherapy

5.Viralloadsuppressionb

Red/amber/green

Red:<70%viralloadsuppressionafter12monthsofantiretroviraltherapyAmber:70–85%viralloadsuppressionafter12monthsofantiretroviraltherapyGreen:>85%viralloadsuppressionafter12monthsofantiretroviraltherapy

Red:poorperformance,belowthedesiredlevel.Amber:fairperformance,notyetatthedesiredlevelbutprogressingtowardsthedesired

level.Green:excellentperformance,achievingthedesiredlevel.Grey:datanotavailable.White:retentionindicatornotreportedinaspecificyearfollowingapredeterminednational

convention.a Retentionindicatorisidenticaltothefollowingindicators:UNGASSno.24;UnitedStates

Presidents’EmergencyPlanforAIDSReliefno.T1.3.D;andGlobalFundtoFightAIDS,TuberculosisandMalariaimpactno.HIV-I3retentionindicator(whichisonlymonitoredandreportedbiannually).

b Thetargetsforviralsuppressionforchildren<2yearsoldhavebeenmodifiedasfollows: Red:<60%viralloadsuppressionafter12monthsofantiretroviraltherapy. Amber:60–70%viralloadsuppressionafter12monthsofantiretroviraltherapy. Green:>70%viralloadsuppressionafter12monthsofantiretroviraltherapy.

44

REFERENCES1. BennettDEetal.TheWorldHealthOrganization’sglobalstrategyforpreventionandassessmentofHIVdrugresistance.Antiviral

Therapy,2008,13(Suppl.2):1–13.

2. JordanMR.AssessmentsofHIVdrugresistancemutationsinresource-limitedsettings.ClinicalInfectiousDiseases,2011,52:1058–1060.

3. HIVdrugresistance[website].Geneva,WorldHealthOrganization,2012(http://www.who.int/hiv/drugresistance,accessed28June2012).

4. BennettDEetal.HIVDrugResistanceEarlyWarningIndicatorsinCohortsofIndividualsStartingAntiretroviralTherapybetween2004and2009:WorldHealthOrganizationGlobalReportfrom50Countries.ClinicalInfectiousDiseases,2012:54(Suppl4).S280-S289.


45

In December 2003, less than 400 000 people received

antiretroviraltherapyinlow-andmiddle-incomecountries,

representinglessthan7%oftheestimatednumberofpeople

inneed.Communitieswerebeingravagedbytheepidemic,

lifeexpectancywasfallingprecipitouslyinmanycountries

andtheeconomicandsocialgainsachievedovertheprevious

decadeswerebeing reversed.Given thesecircumstances,

rapidlyexpandingaccesstoantiretroviral therapywasnot

onlyanethical imperativetowardsthoseaffectedbuthad

alsobecomeaglobalsecurityneed.Nevertheless,despitethe

urgentneedforaction,concernexistedabouthowdelivering

alifelonginterventioninsettingswithlimitedresourcesand

infrastructuremightaffecttheemergenceandtransmission

ofdrug-resistantHIV.

Since2003,coverageofantiretroviraltherapyhasgrown

dramaticallyand,asofDecember2011,morethan8million

people were receiving antiretroviral therapy in low- and

middle-income countries. Based on data from published

studiesandontheresultsofsurveysconductedfollowing

standardized WHO methods, this report reveals three

majorconclusions.First,withtheexpansionoftreatment

achieved over the last eight years, there are signals of

increasingprevalenceoftransmittedHIVdrugresistance

amongrecently-infectedpopulationsintheareassurveyed,

particularly to NNRTI. However, though increasing,

transmittedHIVdrug resistancehasnotoccurredat the

high levels some hadpredicted asaconsequence of the

rapidscale-upofantiretroviraltherapy.Assuch,currently-

recommended first-line regimens should lead to viral

suppression for most individuals initiating antiretroviral

therapy.

Second, with respect to acquired drug resistance, WHO

surveysindicatethat,ifpeopleareswitchedtosecond-line

regimenssoonaftervirologicalfailure,standardsecond-line

treatment combinations are likely to be effective for the

majorityofpatientsfailingfirst-linetherapy.

Third, drug resistance surveillance provides important

informationontheeffectivenessofARTprogrammesand

services.MonitoringofARTprogrammefunctioningthrough

WHOHIVdrugresistanceearlywarningindicatorsin50

countrieshashighlightedtheexistenceofimportantgapsin

servicedeliveryandprogrammeperformance,particularly

withrespecttoprocurementandsupplysystems,adherence

andclinicretention.

6. CONCLUSIONS

The results presented in this report are not intended

to be representative of the countries from which they

were reported and should not be generalized beyond

the populations surveyed. However, findings should be

interpreted as an alert to programme managers that

resistancetransmissionandacquisitionareoccurringand

thatwiderpolicyactionmaybewarranted.

Transmitted drug resistance

Overall, transmitted resistance is estimated to have

increasedintheareasandpopulationssurveyed,andthis

patternappearstohavebeendrivenbyincreasedresistance

to the NNRTI class. An increase in transmitted drug

resistancewasparticularlyapparentinsomeoftheareas

surveyedintheAfricanRegion,afactthatmaybepartly

explained by the relative abundance of data from these

areas. Such an increase in transmitted resistance is not

unexpectedandprobablyreflectstheconsiderableprogress

achieved by many low- and middle-income countries in

expandingaccesstoantiretroviraldrugs.

AvailableHIVdrugresistancedatasuggestthatcurrently

recommended first-line antiretroviral therapy regimens

are effective for most people initiating treatment. As

antiretroviraltherapycontinuestoberolledout,however,

increasedratesoftransmitteddrugresistancemayoccur,

androbustsurveillancesystemsmustbeinplacetodetect

potential future increases in a timely manner. Moreover,

focused efforts are needed to identify levels and trends

amongspecificpopulationsathigherriskofHIVinfection,

such as menwhohavesexwithmen,people who inject

drugsandsexworkers,amongwhomHIVprevalencetends

tobeconsiderablygreaterthanbackgroundlevels.

Reports from surveys showing moderate levels of

transmitted resistance deserve particular attention.

Surveillanceoftransmittedresistanceshouldberepeated

intheseareastoconfirmtheresultsandbeexpandedto

additionalregions.Inaddition,antiretroviraltherapyclinic

andprogrammefactorsinareasreportingmoderatelevels

of transmitted drug resistance should be investigated to

assesstheirpotentialcontributionstotheemergenceand

transmissionofdrug-resistantHIV.

If levels higher than 15% of transmitted drug resistance

are detected, it is recommended that full-scale national

surveillanceofHIVdrugresistanceinpopulationsinitiating

46

antiretroviraltherapybeperformedimmediatelytoidentify

any changes needed to ensure the effectiveness of first-

line antiretroviral therapy. Moreover, an analysis of drug

resistance among women living with HIV should be

conductedtoinformtheselectionofregimensforpreventing

mother-to-childtransmission.Thesesurveysshouldprovide

a point prevalence estimate of HIV drug resistance and

trigger public health action based on cost–effectiveness

thresholds. Importantly,atpresentnochanges tocurrent

treatmentorprophylacticguidelinesarewarrantedbased

onthedatapresented.

Acquired drug resistance

DatafrompublishedstudiesandWHOsurveysinlow-and

middle-income countries indicate that, after 12 months

of antiretroviral therapy, between 82% and 91% of the

peopleassessedachievedviralloadsuppression(treatment

success). Among those experiencing therapy failure,

between60%and70%haddrugresistance,implyingthat

theremaining30%–40%experiencedtherapy failure for

otherreasons,suchasverylowadherenceorlongtreatment

interruptions,and, intheabsenceofHIVdrugresistance

testing, could potentially have been switched to costlier

second-line regimens unnecessarily. Notably, of the 304

peopleinWHOsurveysfailingtherapyinthefirst12months

after treatment initiation,only7switchedtosecond-line

antiretroviraltherapy.Thismaybeduetothelimitedability

todetectearlyfailureusingclinicalorimmunologicalmeans

or difficulty in accessing second-line treatment. It also

illustratesthepotentialofroutineviralloadmonitoring.

IntheareasassessedbyWHOsurveys,theprevalenceof

HIVdrugresistanceinpopulationsinitiatingantiretroviral

therapywasrelativelylow(5%).Theresistanceprofileof

the people experiencing treatment failure at 12 months

suggeststhat,iftheywereswitchedtosecond-linetherapy

atthisspecifictime,mostwouldlikelyrespondtocurrently

recommended,boostedPI-basedsecond-lineantiretroviral

regimens.

At18.8%,theprevalenceofpossibleHIVdrugresistance

observed in WHO surveys is concerning and merits

attention. Although the causes of possible HIV drug

resistancevariedfromclinictoclinic,theyweregenerally

associatedwithhighratesoflosstofollow-upobservedin

someofthesitessurveyed,especiallyinWestern/Central

Africa,suggestingtheneedtostrengthenmechanismsto

traceandre-engagedefaultersincare.

Poor retention rates in many clinics are also concerning.

Given the relationship between treatment interruption

and HIV drug resistance, observed retention rates are

concerning,especiallyasantiretroviral therapycontinues

tobescaledup,andclinicswillfacethedoublechallenge

of successfully managing a growing number of patients

forlonger.

Early warning indicators

MonitoringofHIVdrugresistanceearlywarningindicators

is an important component of global and national

strategiestominimizetheemergenceofpreventableHIV

drug resistance.Monitoringearlywarning indicatorscan

identify weaknesses at the antiretroviral therapy clinic

and programme levels that may result in suboptimal

treatment or treatment interruption, potentially causing

HIV drug resistance to emerge. Early warning indicators

analyseroutinelycollecteddatathroughadrugresistance

lens. As such, they are the first line in preventing HIV

drugresistance.

Monitoring early warning indicators also identifies

successfulclinicsthatcouldserveasbestpracticemodels

for other clinics. Between 2004 and 2009, 50 countries

monitoredoneormoreearlywarningindicatorsatselect

clinics. Although no global trends or conclusions can be

assessed, such experiences have shown that important

gaps in service delivery and programme performance

affect a considerable proportion of clinics delivering

antiretroviral therapy, particularly with respect to the

fragilityofprocurementandsupplysystemsandinadequate

adherenceandclinicretention.

Giventhelimitednumberofantiretroviraldrugsavailable

inmanylow-andmiddle-incomecountries,includingthe

absence of third-line or salvage regimens, and the cost

andtoxicityofsecond-linedrugs,thedurationoftimeon

effective fully suppressive first-line antiretroviral therapy

regimens must be maximized. Moreover, as viral load

monitoringandindividualHIVdrugresistancegenotyping

are often unavailable, successful antiretroviral therapy

programmesshouldstrivetoexceedrecommendedtargets

assessedthoughearlywarningindicatormonitoring.

Inaddition,asincreasingnumbersofpeopleareplacedon

second-lineantiretroviraltherapy,developingstrategiesfor

surveillanceofdrugresistancetosecond-lineandsalvage

regimensmaybenecessary.


47

WHOrecommendsthatsurveysberepeatedregularlyto

detectsignalsofincreasingtransmissionofresistanceand

toassesstheimprovementofprogrammesinminimizing

the emergence of acquired resistance. However, few

countries have repeated surveys, and many have never

engaged in surveillance activities. It is essential that

HIV drug resistance surveillance activities be perceived

and integrated as critical components of the monitoring

and evaluation framework of treatment programmes. In

addition,whilecostmaybeperceivedasabarrier,HIVdrug

resistance surveillance activities represent only a small

fractionoftheglobalinvestmentintheHIVresponse.

Robust programme monitoring, including surveillance of

transmitted andacquiredHIV drug resistance, is vital to

ensurethatadecadeofdecliningHIV-relatedmorbidityand

mortalityisnotreversed.WHO,throughitspartnernetwork

of collaborating institutions, is committed to monitoring

HIV drug resistance globally and to advocate for scaling

up routine surveillance using standardized methods and

increasedmobilizationofnationalandinternationalfunds

tosupportHIVdrugresistancesurveillance.

48

ANNEX 1. METHODOLOGICAL NOTES

Section 1. Regional and subregional country groupings1

Central Africa: Cameroon;CentralAfricanRepublic;Chad;Congo;DemocraticRepublicoftheCongo;EquatorialGuinea;Gabon;SaoTome

andPrincipe

Eastern Africa: Burundi;Comoros;Djibouti;Eritrea;Ethiopia;Kenya;Madagascar;Malawi;Mauritius;Mozambique;Rwanda;Seychelles;Somalia;

Sudan;Uganda;UnitedRepublicofTanzania

Southern Africa: Angola;Botswana;Lesotho;Namibia;SouthAfrica;Swaziland;Zambia;Zimbabwe

Western Africa: Benin;BurkinaFaso;CapeVerde;Côted’Ivoire;Gambia;Ghana;Guinea;Guinea-Bissau;Liberia;Mali;Mauritania;Niger;Nigeria;

Senegal;SierraLeone;Togo

South-East Asia: Bangladesh;Bhutan;DemocraticPeople’sRepublicofKorea;India;Indonesia;Maldives;Myanmar;Nepal;SriLanka;Thailand;

Timor-Leste

Western Pacific: Australia;BruneiDarussalam;Cambodia;China;CookIslands;Fiji;Japan;Kiribati;LaoPeople’sDemocraticRepublic;Malaysia;

MarshallIslands;Micronesia(FederatedStatesof);Mongolia;Nauru;NewZealand;Niue;Palau;PapuaNewGuinea;Philippines;

RepublicofKorea;Samoa;Singapore;SolomonIslands;Tonga;Tuvalu;Vanuatu;VietNam

Section 2. WHO Sequence data analysis and quality assurance for surveys to assess transmitted and acquired drug resistance

Genotypingofprotease(PR)andreversetranscriptase(RT)wasperformedinlaboratorieswithintheWHOLaboratoryNetwork,

mostlyusingin-housemethodsbasedonRT-PCRofRNAextractedfromplasmaordriedbloodspots,followedbystandard

bulksequencingtechniques.Insomecases,commercialkits(TruGeneorViroSeq)wereused.Memberlaboratoriesundergo

anintensiveinspectionandreviewprocessandparticipateinannualexternalproficiencytesting(1).

NucleotidesequenceswereanalysedusingtheCalibratedPopulationResistance(CPR,version5)programontheStanford

HIVDatabasewebsite(http://cpr-v.stanford.edu/cpr/servlet/CPR),andthefollowingparametersandthresholdswereused

forsequencerejection:(1)aminoacidsequenceidenticaltothesubtypeBconsensus,i.e.mostlikelyalabstraincontaminant;

(2)anyinsertionsnotnearPRaminoacidposition38orRTaminoacidposition69;(3)anydeletionsnotnearRTposition69,

atthelastcodonsequencedinRTorpastRTposition300;(3)anystopcodonsnotpresentasmixturesunlesslocatedafter

RTposition300;(4)anyframeshiftsresultinginmorethanthreeconsecutivemutations;(5)morethan20atypicalmutations;

(6)missingPRsequencebetweenposition46and90orbetweenRTposition41and190;and(7)morethantwoambiguous

aminoacids(X’s)inPRorRTbeforeposition300oranyatadrugresistancemutationsite.

AnalysiswasperformedusingMEGA5.05(http://www.megasoftware.net)byconstructingneighbour-joiningtreesandgenetic

distancematricesfromtrimmedsequences(PRpositions1-99andRTpositions1-250)with1000bootstrapiterationsand

missingdataandgapshandledbypairwisedeletion.Forsurveysoftransmitteddrugresistance,whereitisnotexpectedto

observetwohighlyrelatedsequences,onememberofanypairwithgeneticdistanceof0or1(thatis,0oronly1nucleotide

1 SubregionalcountrygroupingforAfricaisavailableatwww.unicef.org/wcaro/WCARO_SOAC08_Fig011.pdf(accessed11July2012).


49

differencebetweenthe2PR-RTsequences)wasrejected.ForsurveysofacquiredDR,expectedbaseline-endpointpairs(based

onpatientIDcodes)wereconfirmed,oriffoundnottoclusterontheneighbour-joiningtree,wererejected.Insomecases,

sequenceswererelabelledwhenphylogeneticanalysisindicatedthataspecimenhadbeenmislabelled.

Section 3. Methods of the literature review on drug resistance in ARV-naive recently- or chronically-infected populations in low- and middle-income countries

English-languagearticlesfromPubMed,EMBASEandmajorconferenceabstractsweresearchedfortheperiod1January2003

to31July2011.Studieswereconsiderediftheyincludeduntreatedrecentlyorchronicallyinfectedindividualsolderthan15years

andhadmorethan10specimenssuccessfullygenotyped.Thegeographical focuswas limitedto low-andmiddle-income

countriesfromAsia,sub-SaharanAfrica(eastern,southern,western/central)andLatinAmericaandtheCaribbean.Studies

wereexcludediftheyonlyreportedresistanceinthecontextofpreventingmother-to-childtransmissionorusedsequencing

methodsotherthanstandardbulksequencing,suchasgenomesequencing,allele-specificPCRandultra-deepsequencing.

WHOtransmittedHIVdrugresistancesurveyresultspublishedbycountryauthorswereexcludedfromthisreview.Individuals

whowerenewlydiagnosedathealthfacilitiesorthoseeligibletoinitiateantiretroviraltherapywereclassifiedaschronically

infected.Recentlyinfectedindividualsweredefinedthroughepidemiologicalsurrogatecriteriaforrecentinfection,through

serialantibodytestingorthroughadetunedantibodyalgorithm.

Thestudieswereassessedaccordingtomid-pointyearof recruitmentandbyregion.Heterogeneitybetweenstudieswas

examinedbypoolingstudiesusingrandom-effectsmeta-analysesandassessingtheI2statistic.Owingtothefactthatthe

proportionofindividualswithadrugresistancemutationwasverylow,wewereunabletousethestandardnormalapproximation

tothebinomialdistributiontoperformthesemeta-analyses.Instead,wetransformedtheindividualstudiesusingaFreeman-

Tukey–typearcsinesquareroottransformation:y=arcsine[√(r/(n+1)]+arcsine[√(r+1)/(n+1)],withavarianceof1/(n+1);

whereristhenumberofindividualswithamutation,andnisthenumberofindividualsgenotyped.

TheI2statisticwasassessedonthesetransformedproportionsbeforebacktransformationforestimationofpooledprevalences.

Pooledestimatesof theprevalenceofdrugclass-specificmutations(NRTI,NNRTIandPI)by regionandover timewere

calculated.StatisticalanalysiswasperformedinStataversion11.2(StataCorp,USA).

Meta-regressionswereperformedbyusingmixed logistic regressionmodels.Specifically thesemodels includedafixed

effecttoaccountfordifferencesbetweenWHOregionsandrandomeffectsatthestudyleveltoaccountforbetween-study

heterogeneitywithinregion.

Manyofthestudiesincludedinthismeta-analysiswereperformedusingdistinctmethodsandmaydifferwithrespecttothe

populationstudied(suchasrecentorchronicinfections),thesamplingframe(suchasconsecutive,convenient,orrandom

selectionfromgeneralpopulation)andthelaboratorymethodsused(suchasdriedbloodspotsorplasmasamplesorgenotyping

methodsused).Individualstudiesmayalsohavebeeninfluencedbyregionalfactorssuchasantiretroviraltherapycoverage

andavailability,variationinHIVsubtypes,qualityofcareattheindividualsitesandantiretroviraltherapyprogrammes,country

incomelevelsandthestructureororganizationofhealthservices.Assuch,prevalenceestimatesmaynotbenationallyor

regionallyrepresentative.

Moreover,studies reported resistancedataaccording toanyof the internationally recognized lists,andvariations inhow

mutationsaredefinedmayhaveinfluencedindividualstudyresultsand,hence,aggregateanalyses.Thismayparticularlybethe

caseforestimatesofPIresistance.Stratificationofthedatasetbyclassesandregionsmayhavereducedthestatisticalpower

todetectregion-specifictrendsovertime.

50

Study Region Country

Mid-point year of

recruitment

deMadeirosetal. LatinAmerica Brazil 2003

Cardosoetal. LatinAmerica Brazil 2003

Vergneetal. Western/CentralAfrica BurkinaFaso 2003

Vessiereetal. Western/CentralAfrica Cameroon 2003

Perezetal. LatinAmerica Cuba 2003

Kassauetal. EasternAfrica Ethiopia 2003

Lloydetal. LatinAmerica Honduras 2003

Balakrishnanetal. South-EastAsia India 2003

Deshpandeetal. South-EastAsia India 2003

Escoto-Delgadilloetal. LatinAmerica Mexico 2003

Bartoloetal. EasternAfrica Mozambique 2003

Bellocchietal. EasternAfrica Mozambique 2003

Perreiraetal. EasternAfrica Mozambique 2003

Lamaetal. LatinAmerica Peru 2003

Lamaetal. LatinAmerica Peru 2003

Bessongetal. SouthernAfrica SouthAfrica 2003

Jacobsetal. SouthernAfrica SouthAfrica 2003

Chonwattanaetal. South-EastAsia Thailand 2003

Galluzzoetal. EasternAfrica Uganda 2003

Bouchardetal. LatinAmerica Venezuela 2003

FerreiradaSilvaetal. SouthernAfrica Angola 2004

Dilerniaetal. LatinAmerica Argentina 2004

Dilerniaetal. LatinAmerica Argentina 2004

Gonsalezetal. LatinAmerica Brazil 2004

Rodriguesetal LatinAmerica Brazil 2004

Lyetal. WesternPacific Cambodia 2004

Soaresetal. Western/CentralAfrica Cameroon 2004

Ndembietal. Western/CentralAfrica Cameroon 2004

Koizumietal. Western/CentralAfrica Cameroon 2004

Zhangetal. WesternPacific China 2004

Tonietal. Western/CentralAfrica Coted’Ivoire 2004

Nafisaetal. EasternAfrica Kenya 2004

Vianietal. LatinAmerica Mexico 2004

Lahuertaetal. EasternAfrica Mozambique 2004

Lyagobaetal. EasternAfrica Uganda 2004

Lyagobaetal. SouthernAfrica Zimbabwe 2004

Petronietal. LatinAmerica Argentina 2005

Tebitetal. Western/CentralAfrica BurkinaFaso 2005

Marechaletal. Western/CentralAfrica CAR 2005

Zhongetal. WesternPacific China 2005

Liuetal. WesternPacific China 2005

Liaoetal. WesternPacific China 2005

Lihanaetal. EasternAfrica Kenya 2005

Deracheetal. Western/CentralAfrica Mali 2005

Ahumada-Ruizetal. LatinAmerica Panama 2005

Diop-Ndiayeetal Western/CentralAfrica Senegal 2005

McIntyreetal. SouthernAfrica SouthAfrica 2005

Orrelletal. SouthernAfrica SouthAfrica 2005

Barthetal. SouthernAfrica SouthAfrica 2005


Mid-point year of

recruitment

Moshaetal. EasternAfrica Tanzania 2005

Nyombietal. EasternAfrica Tanzania 2005

Apisarnthanaraketal. South-EastAsia Thailand 2005

Lallemantetal. South-EastAsia Thailand 2005

Ferreiraetal. LatinAmerica Brazil 2006

Oliveiraetal. Western/CentralAfrica CapeVerde 2006

Liuetal. WesternPacific China 2006

Hanetal. WesternPacific China 2006

Zhangetal. WesternPacific China 2006

Tuetal. WesternPacific China 2006

Murilloetal. LatinAmerica Honduras 2006

Kandathiletal. South-EastAsia India 2006

Kamotoetal. EasternAfrica Malawi 2006

Huangetal. SouthernAfrica SouthAfrica 2006

vanZyletal. SouthernAfrica SouthAfrica 2006

Maphalalaetal. SouthernAfrica Swaziland 2006

Apisarnthanaraketal. South-EastAsia Thailand 2006

Sirivichayakuletal. South-EastAsia Thailand 2006

Sirivichayakuletal. South-EastAsia Thailand 2006

Auwanitetal. South-EastAsia Thailand 2006

Rangeletal. LatinAmerica Venezuela 2006

ThaoVuetal. WesternPacific Vietnam 2006

Pandoetal. LatinAmerica Argentina 2007

Bussmannetal. SouthernAfrica Botswana 2007

Sprinzetal. LatinAmerica Brazil 2007

DesaFilhoetal. LatinAmerica Brazil 2007

Nouhinetal. WesternPacific Cambodia 2007

Aghokengetal. Western/CentralAfrica Cameroon 2007

Burdaetal. Western/CentralAfrica Cameroon 2007



ChunfuYangetal. WesternPacific China 2007

Chinetal. WesternPacific China 2007

Djokoetal. Western/CentralAfrica DRC 2007

Chaturburjetal. South-EastAsia India 2007

Lalletal. South-EastAsia India 2007

Agwaleetal. Western/CentralAfrica Nigeria 2007

Yaotseetal. Western/CentralAfrica Togo 2007

Leeetal. EasternAfrica Uganda 2007

Ishizakietal. WesternPacific Vietnam 2007

Tshabalalaetal SouthernAfrica Zimbabwe 2007

Zijenahetal. SouthernAfrica Zimbabwe 2007


Inocencioetal. LatinAmerica Brazil 2008


Nzeyimanaetal. EasternAfrica Burundi 2008

DiazGranadosetal. LatinAmerica Columbia 2008

Rajeshetal. South-EastAsia India 2008

Priceetal. EasternAfrica Kenya 2008

Forthepurposeoftheanalysis,eachstudyprovidingdataforbothchronicandrecentlyinfectedindividualswasconsidered

astwoseparatestudies.

Table 1. Studies included in the literature review of HIV drug resistance among ARV-naive recently- or chronically-infected populations


51


Mid-point year of

recruitment

Haidaraetal. Western/CentralAfrica Mali 2008

Avila-Riosetal. LatinAmerica Mexico 2008

Priceetal. EasternAfrica Rwanda 2008

Bessongetal. SouthernAfrica SouthAfrica 2008

Priceetal. EasternAfrica Uganda 2008

Castilloetal. LatinAmerica Venezuela 2008

Phanetal. WesternPacific Vietnam 2008

Priceetal. SouthernAfrica Zambia 2008

Castelbrancoetal. SouthernAfrica Angola 2009

Arrudaetal. LatinAmerica Brazil 2009

Ferreiraetal. LatinAmerica Brazil 2009

Carvalhoetal. LatinAmerica Brazil 2009

BacelarAciolilinsetal. LatinAmerica Brazil 2009

Soaresetal. LatinAmerica Brazil 2009

Grafetal. LatinAmerica Brazil 2009


Mid-point year of

recruitment

Diakiteetal. Western/CentralAfrica Guinea-Canakry 2009

LihanaRetal. EasternAfrica Kenya 2009

Kamotoetal. EasternAfrica Malawi 2009

Mavhanduetal. SouthernAfrica SouthAfrica 2009

Parboosingetal. SouthernAfrica SouthAfrica 2009

Bontelletal. WesternPacific Vietnam 2009

Deanetal. WesternPacific Vietnam 2009

Ishizakietal. WesternPacific Vietnam 2009

Tshabalalaetal SouthernAfrica Zimbabwe 2009

Lietal. WesternPacific China 2010

Neogietal. South-EastAsia India 2010

Thoratetal. South-EastAsia India 2010

Nazziwaetal. EasternAfrica Uganda 2010

Section 4. Methodological notes on the design and interpretation of WHO transmitted HIV drug resistance surveys

Surveystomonitortransmitteddrugresistancesampleindividualsfrompopulationslikelytobeantiretroviraldrug–naiveand

tohavebeenrecentlyinfected,inthiscaseindividualsyoungerthan25yearsofageand,inthecaseofwomen,onlythose

withnopreviouspregnanciesorpregnantforthefirsttime.Whereavailable,evidenceofrecentinfectionorseroconversion

byavalidlaboratorytestorevidenceofaCD4countexceeding500cellspermm3mayalsobeusedtodetermineeligibility.

ConsecutiveHIV-positivespecimensfromeligibleindividualsdiagnosedatsitesofferingservicesrelatedtoantenatalcare,

voluntarycounsellingandtesting,sexuallytransmittedinfectionsorpreventingmother-to-childtransmissionmaybeused.In

settingswheretheHIVepidemicisdrivenbyaparticularmodeoftransmission,HIVdrugresistancetransmissionsurveyscan

targetaseparatesubpopulation(suchassexworkersorpeoplewhoinjectdrugs).

Briefly,theWHOHIVdrugresistancesurveymethodsamplesasmallnumber(n≤47)ofeligibleindividualsconsecutively

encounteredatspecificsiteswithinanareaduringalimitedtimeperiod.Thismethodisnotintendedtoestimatethepoint

prevalenceoftransmittedHIVdrugresistancebutratherusestruncatedsequentialsamplingtoclassifytransmittedresistance

foreachdrugclassaslow(prevalencelowerthan5%),moderate(prevalencebetween5%and15%)orhigh(prevalencehigher

than15%)(2).Surveyresultsarenotintendedtoberepresentativeofthecountriesfromwhichtheywerereportedandshould

notbegeneralizedbeyondthepopulationssurveyed.

BecauseHIVserosurveystoestimateHIVprevalenceinspecificareasarealreadyinplaceinmostlow-andmiddle-income

countries(3),WHOrecommendsusingeligibleremnantspecimensfromthesesurveyswherepossible.Thesurveyonlyintends

tocollectepidemiologicalinformationthatisroutinelyavailablefrommedicalrecords.Driedbloodspotsarethemostcommonly

usedspecimentype,andfewsurveyshaveusedplasmaorserum.

TheresultswereconsideredifsurveyswereconductedaccordingtoWHO-recommendedmethodsandiftheysatisfiedthe

followingfourcriteria:(1)thesurveyprotocoland/orreportwasmadeavailabletoWHO;(2)HIVdrugresistancegenotyping

testingwasperformedinaWHO-designatedlaboratory;(3)theindividualsequencedatawerequalityassuredbyWHO;and

(4)whenrequested,patient-levelepidemiologicalinformationwasmadeavailabletoWHOforadditionalqualityassuranceof

thedata.Surveysconductedbefore2007thathadHIVdrugresistancegenotypingconductedinanon-designatedlaboratory

(atthetimewhentheWHOlaboratorynetworkwasnotatitsfullcapacity)wereincludedinthisreportonlyifqualityassurance

oftherawsequencedataandphylogeneticanalysisconductedbyWHOoradesignatedlaboratorywasconsideredsatisfactory.

Section2inthisannexprovidesadditionaldetailsongenotypingandqualityassurance.

Individualsequencesnotpassing thequalityassuranceassessmentconductedbyWHOwereexcluded fromtheanalysis

providedinthisreportand,consequently,someofthesurveyresultspresentedhereinmaydifferslightlyfromresultsfromthe

samesurveyspublishedelsewhere.

52

WHOhasupdatedandrefinedthemethodsandsuggestedpublichealthandprogrammaticactionsassociatedwithsurveysto

assesstransmittedHIVdrugresistance.Itisnowrecommendedthatdifferentsitetypeswithinadefinedgeographicalregion

maybecombinedifonesitetypeisanticipatedtoprovideaninsufficientsamplesizetomakeaprevalenceclassificationduring

amaximumperiodof12monthsofspecimencollection.Inaddition,alternativesurveyinclusioncriteriaarebeingconsideredto

facilitateimplementationinlow-prevalencesettingswherestandardcriteriadonotpermitsurveyimplementation.

Section 5. Measuring and classifying HIV drug resistance: the WHO HIV drug resistance surveillance mutations list and the Stanford HIV resistance database

Mutationsoccurrandomly,andmanyareharmless.Infact,mostmutationsplaceHIVatadisadvantagebyreducingtheviral

“fitness”andslowingitsabilitytoinfectcells.However,severalmutationscanactuallygiveHIVasurvivaladvantagewhenHIV

medicationsareused,becausethesemutationscanblockdrugsfromworkingagainsttheHIVenzymestheyaredesignedto

target.

HIV isalsopolymorphic.Aposition inanHIVgenome iscalledpolymorphic if it isdifferent fromwhat isobserved ina

standardlaboratoryreferencestrainofthevirus.Thesenucleotidedifferences(polymorphisms)arecommonlyseeninthe

viruspopulationsof infected individuals.Generally,polymorphismshaveno impacton replicationcapacityandmayeven

causevariantstoreplicatelesswell.However,polymorphismsinthepresenceofothermajorHIVdrugresistancemutations

maymakethevirusabletobetterreplicateinthepresenceofdrugsthatwouldotherwisenormallysuppresstheirreplication.

TheWHOsurveillancedrugresistancemutations(SDRM)listpublishedin2007andupdatedin2009consistsofmajordrug

resistancemutationsselectedforbyantiretroviralusebutexcludesmutationsconsideredtobepolymorphicbasedontheir

prevalenceinuntreatedsubjects(4).Athresholdprevalenceof0.5%hasbeenusedtodefineamutationasbeingpolymorphic.

Assuch,anassessmentofHIVdrugresistancebasedonthesurveillancedrugresistancemutationlistidentifiesthepresence

orabsenceofmajordrugresistancemutations.

In2012,publisheddatafromstudiesofuntreatedsubjectswerereanalysedusingtheStanfordHIVresistancedatabase.Using

theseupdateddata,mutationsatposition46inprotease(M46IorL)werefoundtohavethehighestprevalenceofallthePI

surveillancedrugresistancemutations(0.21%and0.26%,respectively).Basedonarevisedthresholdof0.2%,M46IandL

havebeenremovedfromthelistofmutationsusedtoanalysetheWHOsurveydatainthisreport.Thiseffectivelyincreases

thespecificityoftheanalysisalthoughatthepotentialexpenseofreducedsensitivity.Byreducingtheprevalencethresholdto

differentiateamutationfromapolymorphism,theproportionoffalse-positivesislikelytobereducedandthepositivepredictive

valueofthedetectionofPIresistanceislikelytoincreaseaccordingly.ForthepurposeofanalysisofbaselineHIVdrugresistance,

the2009WHOsurveillancemutationslist,excludingmutationsM46IandL,isusedtoidentifymutationsinbaselinesequences.

ThesurveillancedrugresistancemutationlistwasdevelopedspecificallytohelpidentifyHIVwithevidenceofpriordrugexposure

andtoavoidconsideringnaturallyoccurringpolymorphismsasrepresentingtransmitteddrugresistance.Assuch,thesurveillance

drugresistancemutationlistwasusedforthepurposesoftheanalysisoftransmitteddrugresistance(seeChapter3)and

resistancebeforeantiretroviraltherapyinitiation(baselineofthesurveyofacquireddrugresistance.seeChapter4).

Nevertheless,somepolymorphismsareknowntocontributetoreduceddrugsusceptibility.Whendrugresistancetoantiretroviral

drugsneeds tobepredicted,anymutations, includingpolymorphicmutations,known tocontribute tosusceptibilityare

consideredanddataareinterpretedusingascoringsystem,oralgorithm(5),suchastheoneavailableontheStanfordHIV

databasewebsite(http://sierra2.stanford.edu/sierra/servlet/JSierra).TheendpointsofWHOsurveysofacquiredHIVdrug

resistanceareanalysedwithinthisframework,andapredictedresistanceclassificationoflow,moderateorhighisconsidered

asresistant(seeChapter4).

Insomecases,adrugresistanceinterpretationusingthisalgorithmmayresultinavirusbeingclassifiedashavinglow-level

resistanceintheabsenceofamutationincludedinthesurveillancedrugresistancemutationlist.Conversely,avirusmayhave

oneormoremutationsaspartofthesurveillancedrugresistancemutationlistwithouthavingasufficientnumberofmutations

toresultinalow-leveldrugresistanceinterpretation.Thismeansthat,whentakingintoaccountbothmajordrugresistance


53

mutationsandtheeffectthatpolymorphisms,ifpresent,mayhaveontheoverallsusceptibilityofadrugtoaparticularHIV

virus,itisnotuncommontoseeanabsenceofPImutationsbasedonthesurveillancedrugresistancemutationlistbutsome

levelofpredictedPIresistance(thisisparticularlytruefornelfinavirandcertainunboostedPI).

ThispatternwasobservedamongpeopleinitiatingantiretroviraltherapyinWHOacquiredHIVdrugresistancesurveys.Only28

people(0.6%)hadanysurveillancedrugresistancemutationrelatedtoPI(Figure4.4),but611people(12%)hadatleastlow-

levelpredictedresistancetoaPI,nearlyalways(in607people)asaresultoflow-levelpredictedresistancetonelfinavir,based

onthepresenceofmultiplenaturallyoccurringpolymorphicmutationssuchasL10IorF,K20IandT74S(Figure1inAnnex2).1

Similarly,while187people(3.7%)hadatleastoneNNRTIsurveillancedrugresistancemutation(Figure4.4),290people(5.7%)

hadatleastlow-levelpredictedresistancetoanNNRTI(Figure1inAnnex2),nearlyalwaysasaresultofloworintermediate-

levelpredictedresistancetonevirapine,basedonthepresenceofpolymorphicmutationssuchasA98G,K103RandV179D,

E138A,F227LandY318F.OnepersonhadarareY181Smutation,whichleadstoaninterpretationoflow-levelresistanceto

multipleNNRTI,butthismutationisnotonthecurrentsurveillancedrugresistancemutationlist.

Section 6. Methods of the literature review on acquired drug resistance in low- and middle-income countries

PubMed,EMBASEandtheScienceCitationIndexweresearchedforprospectiveorcross-sectionalstudiesfortheperiodbetween

1January1994and31December2011.Thegeographicalfocuswasrestrictedtolow-andmiddle-incomecountriesfromAsia

(South-EastAsiaandWesternPacificRegions),sub-SaharanAfrica(eastern,southern,western/central),LatinAmericaand

theCaribbean.Studieswereincludediftheyreportedsequencedataonatleast50genotypesinpeoplefailingNNRTI-based

first-lineantiretroviraltherapyatamediandurationoftherapyof12months.

Studieswereexcluded if theyonly reported resistance in thecontextofpreventingmother-to-child transmissionorused

sequencingmethodsotherthanstandardbulksequencing,suchasgenomesequencing,allele-specificPCRandultra-deep

sequencing.AnypublishedWHOHIVdrugresistancesurveyswerealsoexcluded.

Dataontheclinicalcharacteristicsofpopulation,historyofantiretroviraltherapyexposureandvirologicalresponseswere

abstracted.Definitionsofvirologicalfailure,theproportionsassessedforresistanceandtheresultingresistantgenotypeswere

recorded.Studyauthorswerecontactedforfurtherinformationifnecessary.Mutationsweredefinedaccordingtointernationally

acceptedlists.

Studieswereeithercohortstudies,wherepeopleinitiatingNNRTI-based

antiretroviraltherapywerefollowedup,orcross-sectional,whereparticipants

wereassessedforfailureat12monthorless.Thevariablecommontoeach

ofthesereportswastheproportionofpeoplecarryingresistantvirusamong

peoplefailingantiretroviraltherapy(numberofpeoplegenotypedatfailure

withresistancedividedbythenumberexperiencingantiretroviraltherapy

failurewithgenotypeavailable).Definitionsoftreatmentfailureincluded

clinical,immunologicaland/orvirologicalmeasures.

Informationondurationoftherapywasderivedatthestudylevelasmedian

duration, so theactualdurationof therapy for individuals isdistributed

aroundthemedian.Thisimpliesthatstudiesmayincludepeoplewhohavebeenontherapyformore(orless)than12months.

Forinstance,astudyreportingamediandurationoftherapyof12monthshasupto50%ofobservationsabovethemedian,

thereforepotentiallyincludingpeoplewhomayhavebeenreceivingtherapyformorethan12months.Inaddition,resistance

attreatmentfailuremayhavebeenrelatedtotheresistancealreadypresentatbaseline,andparticipantsrecruitedintothese

studiesmaynotberepresentativeofthegeneralpopulationwithHIVonantiretroviraltherapy.

1 Fourpeoplehaddrugresistancepredictedtoaritonavir-boostedPIotherthannelfinavir:atazanavir/r(I50L),fosamprenavir/r(multiplepolymorphisms),indinavir/r(V82M)ortipranavir/r(multiplepolymorphisms).


Ndembietal2010 EasternAfrica Uganda

Ramadhanietal2007 EasternAfrica Tanzania

Kouanfacketal2009 West/CentralAfrica Cameroon

Messouetal2011 West/CentralAfrica IvoryCoast

Dagnraetal2011 West/CentralAfrica Togo

Aghokengetal2011 West/CentralAfrica Cameroon

Garridoetal2008 SouthernAfrica Angola

Zolfoetal2011 WesternPacific Cambodia

Ruanetal2010 WesternPacific China

54

Section 7. Methodological notes on the design and interpretation of WHO acquired HIV drug resistance surveys

The researchprotocol stipulates that,ateachantiretroviral therapyclinicbeingsurveyed,acohortofabout 130people

initiatingfirst-lineantiretroviral therapybeenrolled.Drug resistancegenotyping isperformedbefore treatment initiation

(baseline)foreveryone,andeveryoneisthenfollowedfor12months.Consecutiveindividualsinitiatingfirst-lineantiretroviral

therapyattheselectedsiteareeligibletoparticipateinthesurvey,regardlessofpreviousexposuretoantiretroviraldrugsfor

preventingmother-to-childtransmissionorotherreasons.Baselinespecimensarecollectedwithinonemonthbeforeinitiation

ofantiretroviraltherapy.Aspatientswhodiedorwhoweretransferredtootherfacilitiesarenotincludedintheanalysis,an

effectivesurveysamplesizeof96patientswithclassifiableendpointsprovidesa95%confidenceintervalof+/-10%forthe

proportionwithHIVDRprevention,regardlessofthecumulativeincidenceofviralsuppression.HIVisquantified(viralload)

at12monthsforpeoplemaintainedonfirst-linetreatmentoratthetimeofswitchtosecond-lineantiretroviraltherapyfor

peopleexperiencingtherapyfailurebefore12months.Amongpeoplewithviralloadexceeding1000copies/ml,genotypingis

performedtocharacterizedrugresistancemutationsusingpopulation-basesequencing.Additionalrelevantdemographicand

epidemiologicalinformationisgathered,includingpreviousexposuretoantiretroviraldrugs(forpreventingmother-to-child

transmissionorpreviousantiretroviraltherapy).

AlthoughWHOprospectivesurveysofacquiredHIVdrugresistancehaveprovideddetailedsite-specificinformation,their

uptakehadbeenlimitedbecauseoftheirprospectivenature,whichrequireduptotwoyearsoffollow-uptoassessHIVdrug

resistanceoutcomes,andtherelativelylargesamplesizerequired.Toaddressthischallenge,anewcross-sectionalmethodhas

beendevelopedtoassessacquiredHIVdrugresistanceatrepresentativeantiretroviraltherapyclinicsamongpeopleforwhom

treatmentisfailingwithdetectablevirus.ThisnewmethodusesLotQualityAssuranceSamplingtoclassifytheratesofviral

loadsuppression12–15and24–36monthsafterinitiatingantiretroviraltherapyinadultpopulationsandinpaediatricpopulations

receivingantiretroviraltherapyfor12ormoremonths.Surveysaredesignedtobeimplementedroutinelyatrepresentativesites

inacountry.AlthoughWHOcontinuestosupportprospectivesurveysofacquiredHIVdrugresistance,thecross-sectional

methodisanticipatedtobemoreeasilyimplementedandprovidemoretimelyandmorenationallyrepresentativeresults.This

newcross-sectionalmethodiscurrentlybeingpilotedinNamibia.

Section 8. The three outcomes of WHO acquired HIV drug resistance surveys: prevented, detected and possible HIV drug resistance

WHOsurveysofacquiredHIVdrugresistancehavethreesurveyoutcomes:HIVdrugresistanceprevention,HIVdrugresistance

andpossibleHIVdrugresistance.BecausedeathduringthefirstyearoftreatmentisunlikelytobeattributabletoHIVdrug

resistanceandbecauseHIVdrugresistanceoutcomesforpeopletransferringtootherclinicscannotbeusedtoassessthe

functioningof thesentinelsites,peoplewith thesesurveyendpointsarenot included in thecalculationsof theestimated

prevalenceofHIVdrugresistancepreventionat12months.

1)HIVdrugresistanceprevention:

HIVdrugresistanceisconsideredtohavebeenpreventedif,12monthsafterantiretroviraltherapyinitiationoratthetimeof

theswitchtosecond-linetherapy,peoplehavesuppressedviralloads,definedashavinglessthan1000copies/ml.Athreshold

of1000copies/mlwaschosenbecauseofsensitivityandreproducibilityofstandardcommercialgenotypingassayscommonly

availableinlow-andmiddle-incomecountriesatthetimetheprotocolwasdeveloped.ThelevelofHIVdrugresistanceprevention

inacohortisassessedasfollows:

Numerator:peoplewithviralloadlessthan1000copies/ml12monthsafterantiretroviraltherapyinitiationoratthetimeof

switchtosecond-linetherapy

Denominator:peoplereceivingfirst-lineantiretroviraltherapyat12monthswithclassifiableviralloadresults+peopleswitchingto

second-lineantiretroviraltherapywithclassifiableviralloadresult+peoplelosttofollow-up+peoplewhostoppedantiretroviral

therapyduringthesurvey.


55

2)HIVdrugresistance:

HIVdrugresistance isconsideredtohaveoccurred ifdrugresistance isdetectedbygenotyping inpeoplewithviral loads

greaterthan1000copies/ml12monthsafterantiretroviraltherapyinitiationoratthetimeofswitchtosecond-linetherapy.

TheprevalenceofdetectedHIVdrugresistanceinacohortisassessedasfollows:

2a)HIVdrugresistance(asa%ofthepeopleinitiatingtherapy):

Numerator:peoplewithaviralloadgreaterthan1000copies/ml12monthsafterantiretroviraltherapyinitiationoratswitchto

second-linetherapywithHIVdrugresistance

Denominator:peoplereceivingfirst-lineantiretroviraltherapyat12monthswithclassifiableviralloadresults+peopleswitchingto


therapyduringthesurvey

2b)HIVdrugresistance(asa%ofpeoplefailingtherapywithgenotypingresultsavailable):

Numerator:peoplewithviralloadgreaterthan1000copies/ml12monthsafterantiretroviraltherapyinitiationoratswitchto

second-lineantiretroviraltherapyandHIVdrugresistance

Denominator:peoplewithviralloadgreaterthan1000copies/mlwithgenotypingavailable12monthsafterantiretroviraltherapy

initiationoratthetimeofswitchingtosecond-linetherapy

3)PossibleHIVdrugresistance:

HIVdrugresistanceisconsideredtobepossibleamongpeoplewho(i)stoppedantiretroviraltherapyduringthesurveyperiod,

(ii)werelosttofollow-up,(iii)hadaviralloadgreaterthan1000copies/mlbutfailedtohaveasuccessfulgenotypingassay

and(iv)hadviralloadsgreaterthan1000copies/mlandnodetectedHIVdrugresistance12monthsafterantiretroviraltherapy

initiationoratthetimeofswitchingtosecond-lineantiretroviraltherapy:

Numerator:peoplewithviralloadgreaterthan1000copies/mlandnodetectedHIVdrugresistanceat12-monthsurveyendpoint

(onantiretroviraltherapyat12monthsandatswitch)+peoplewhostoppedantiretroviraltherapy+peoplelosttofollow-up

+peoplewithunclassifiableviralloadat12-monthsurveyendpoint(onantiretroviraltherapyat12monthsandatthetimeof

switchingtosecond-linetherapy)+peoplewithviralloadgreatergreaterthan1000copies/mlbutfailedtohaveasuccessful

genotypingassay.

Denominator:peopleonfirst-lineantiretroviraltherapyat12monthswithclassifiableviralloadresults+peopleswitchingto


therapyduringthesurvey.

Section 9: Methods for statistical analyses

Anexploratoryanalysiswasperformedandpooledproportionsofthenumberofindividualswithdrugresistancemutations

weredetermined.Astheproportionofindividualswithadrugresistancemutationwaslow,itwasnotpossibletorelyonthe

standardnormalapproximationtothebinomialdistributiontoestimatepooledproportions.Instead,individualstudieswere

transformedusingaFreeman-Tukey–typearcsinesquareroottransformation:y=arcsine[√(r/(n+1)]+arcsine[√(r+1)/(n+1)],

withavarianceof1/(n+1);whereristhenumberofindividualswithamutation,andnisthenumberofindividualsgenotyped.

Usingthisprocedure,confidenceintervalsforindividualstudiesdonotneedtobesymmetriconthenaturalscaleanditisstill

possibletocalculateconfidenceintervalswhentherearezeromutations.(Reference:Miller,J.J.TheInverseoftheFreeman-Tukey

DoubleArcsineTransformation.TheAmericanStatistician,AmericanStatisticalAssociation,1978,32,p.138).

56

Randomeffectsmeta-analyseswereperformedonthetransformedproportionsusingDerSimonian-Lairdweightingbefore

back-transformationofthepooledproportions.HeterogeneitywasassessedusingtheI2statisticfrommeta-analysesofthe

transformedproportions.Pooledproportionsusingthismethodweretypicallylowerthanwhatasimplepoolingofdataacross

studieswouldsuggest.Thisisinlinewith(i)thelowmutationratesobservedintheavailabledatasetand(ii)withthereduced

variability,andthereforeincreasedprecision,ofestimatedmutationratesinstudieswherelevelsarecloseto0or100%compared

towhenlevelsarecloseto50%withanequalnumberofindividualsgenotyped.

Insomesurveysonlypartialsequencedatawereavailable(PRorRTregionsonly),sothatwhencalculatingtheprevalenceof“any

drugresistancemutation”,anaverageofthetotalnumberofgenotypeswithPRandRTsequenceswasusedasthedenominator.

StatisticalanalyseswereconductedinStataversion11.2(StataCorp,Texas),includingtheuseofthemetanpackageformeta

analysisandthegllammpackageformixedlogisticregressionmodels.

Analysis of change in levels of transmitted drug resistance by calendar yearTodeterminewhetherprevalenceoftransmittedresistanceincreasedovertime,datafromallsurveyswerepooledaccording

toregionandsub-region,andyearofimplementation.Toexplorethesignificanceoftheobservedvariationovertime,meta-

regressionwasperformedbyusingamixedlogisticregressionmodel.Specificallythesemodelsincludedafixedeffecttoaccount

fordifferencesbetweenWHOregions,andrandomeffectsatthestudyleveltoaccountforbetween-studyheterogeneitywithin

eachregion.Modelsusingrandomeffectsattheregionalandcountrylevelswerealsoexplored,withoutsignificantchanges

intheoutcome.

Analysis of change in levels of HIV drug resistance by ART coverage Meta-regressionswereperformedbyusingmixedlogisticregressionmodelsincludingafixedeffecttoaccountfordifferences

betweenWHOregions,andrandomeffectsatthestudyleveltoaccountforbetweenstudyheterogeneitywithineachregion.

TotestfortheimportanceofARTcoverageoryear,likelihoodratiotestswereusedtocomparemodelswithandwithouta

lineartermforARTcoverageandbaselineyear.Asthemodelsarelogisticregressionmodels,coefficientsfromthemodelare

logoddsratios,andsoarelinearonthelogoddsscale,butnotonthenaturalscale.Duetothelowprevalenceofmutations,

theoddsratio–whichistheexponentialofthecoefficientsfromthelogisticregressionmodel–isapproximatelyequaltothe

ratioofthemutationratesper1unitincreaseintheexplanatoryvariable.Therefore,foramutationrateof1%,anoddsratioof

1.4%representsanincreasetoapproximately1.4%.

REFERENCES1. ParkinN,BremerJ,BertagnolioS.GenotypingexternalqualityassuranceintheWorldHealthOrganizationHIVDrugResistanceLaboratory

Networkduring2007–2010.ClinicalInfectiousDiseases,2012,54(Suppl.4):S266.

2. MyattM,BennettDE.AnovelsequentialsamplingtechniqueforthesurveillanceoftransmittedHIVdrugresistancebycross-sectionalsurveyforuseinlowresourcesettings.AntiviralTherapy,2008,13(Suppl.2):37–48.

3. UNAIDS/WHOWorkingGrouponGlobalHIV/AIDSandSTISurveillance.GuidelinesforconductingHIVsentinelserosurveysamongpregnantwomenandothergroups.Geneva,UNAIDS,2003(http://data.unaids.org/Publications/IRC-pub06/jc954-anc-serosurveys_guidelines_en.pdf?preview=true,accessed28June2012).


5. LiuTF,ShaferRW.WebresourcesforHIVtype1genotypic-resistancetestinterpretation.ClinicalInfectiousDiseases,2006,42:1608–1618.


57

AN

NEX

2. S

UPP

LEM

ENTA

L TA

BLE

S A

ND

FIG

URE

S

WHO

Reg

ion

p-va

luea

2003

2004

2005

2006

2007

2008

2009

2010

Any drug class

Afric

a3.5

(1.6

–5.8

)4.

2(1.

1–8.

8)2.2

(0.8

–4.0

)0.

7(0

.0–2

.3)2.5

(0.8

–4.9)

3.6(2

.2–5.4

)4.

8(3

.1–6.

7)6.

4(1.

3–17.

5)NS

Sout

h-Ea

stA

sia1.6

(0.4

–3.4

)—

1.0(0

.6–4

.0)

1.3(0

.1–3.5

)3.3

(2.8

–18.9)

2.9(0

.6–8

.2)—

1.0(0

.0–5

.8)

NS

Wes

tern

Pac

ific

—9.

8(0

.9–2

5)1.8

(0.2–

4.5)

3.3(1

.0–6

.6)

1.9(0

.6–3

.7)4.

0(1.

6–8.

2)5.

6(3

.2–8.

7)0.

0(0

.0–16

.1)NS

Latin

Am

erica

and

the

Carib

bean

4.7

(2.3–

7.6)

3.1(1

.7–4.

9)7.7

(2.1–

18.5

)7.6

(4.8

–10.9)

8.1(4

.0–13

.4)

6.3

(5.3–

7.5)

9.8

(7.5

–12.4

)—

0.01

Over

all3.6

(2.3–

5.2)

4.5

(2.3–

7.3)

1.9(0

.9–3

.3)2.5

(1.2–

4.1)

3.1(1

.6–5

.0)

4.9

(3.6

–6.3)

6.6

(5.1–

8.3)

2.1(0

.1–5.

8)0.

03

NRTI

Afric

a1.5

(0.4

–3.1)

2.0(0

.4–4

.6)

0.8

(0.0

–2.2)

0.2

(0.0

–0.3)

1.0(0

.0–2

.7)1.1

(0.3–

2.2)

0.7

(0.1–

1.7)

0.0

(0.0

–7.5

)NS

Sout

h-Ea

stA

sia0.

3(0

.0–1.

4)—

0.9

(0.2–

3.3)

1.0(0

.0–2

.8)

2.5(0

.0–14

.2)1.9

(0.2–

6.8)

—0.

0(0

.0–2

.6)

NS

Wes

tern

Pac

ific

—6.

3(0

.1–18

.8)

0.7

(0.0

–2.0

)1.5

(0.1–

3.9)

0.7

(0.0

–2.1)

1.7(0

.4–5

.0)

2.6(0

.7–10

.1)0.

0(0

.0–16

.1)NS

Latin

Am

erica

and

the

Carib

bean

4.0

(1.9–

6.6)

1.4(0

.3–2.9

)1.9

(0.0

–10.3)

3.6(1

.2–7.1

)2.8

(0.9

–5.6

)3.1

(1.8

–4.6

)4.1

(2.0

–6.8

)—

NS

Over

all2.0

(0.9

–3.4

)2.3

(1.0

–4.0

)0.

7(0

.1–1.5

)0.

9(0

.1–2.2

)1.2

(0.4

–2.4

)1.9

(1.1–

2.9)

2.0(0

.8–3

.5)

0.0

(0.0

–1.4)

NS

NNRTI

Afric

a1.0

(0.3–

2.1)

0.6

(0.0

–2.1)

1.1(0

.4–2

.2)0.

2(0

.0–0

.3)0.

7(0

.1–1.5

)2.1

(1.0

–3.4

)3.3

(1.6

–5.5

)6.

4(1.

3–17.

5)0.

01

Sout

h-Ea

stA

sia0.

1(0.

0–1.0

)—

0.7

(0.1–

4.6)

1.1(0

.1–2.9

)0.

0(0

.0–2

.6)

1.0(0

.0–5

.2)—

0.0

(0.0

–2.6

)NS

Wes

tern

Pac

ific

—4.

5(0

.7–22

)0.

9(0

.2–3.4

)2.2

(0.8

–4.1)

1.5(0

.4–3

.1)2.9

(0.9

–6.6

)2.8

(0.5

–6.5

)0.

0(0

.0–16

.1)NS

Latin

Am

erica

and

the

Carib

bean

1.2(0

.1–4.

2)0.

8(0

.1–1.9

)5.

8(1.

2–15

.9)4.1

(2.1–

6.7)

4.5

(2.1–

7.7)

2.4(1

.1–4.

0)3.4

(2.0

–5.2)

—0.

06

Over

all0.

9(0

.2–2.0

)1.0

(0.2–

2.1)

1.1(0

.4–2

.0)

1.2(0

.3–2.7

)1.2

(0.5

–2.2)

1.8(1

.3–2.4

)3.3

(2.3–

4.4)

0.9

(0.0

–4.8

)<0

.001

PI

Afric

a0.

1(0.

0–0.

8)0.

8(0

.0–2

.5)

0.0

(0.0

–0.1)

0.2

(0.0

–1.7)

0.1(

0.0–

0.5)

0.4

(0.0

–1.3)

0.1(

0.0–

0.8)

0.0

(0.0

–7.5

)NS

Sout

h-Ea

stA

sia0.

1(0.

0–1.2

)—

0.0

(0.0

–0.3)

0.0

(0.0

–0.4

)0.

9(0

.0–5

.2)0.

0(0

.0–3

.5)

—0.

0(0

.0–2

.6)

—

Wes

tern

Pac

ific

—1.5

(0.0

–4.1)

0.2

(0.0

–0.8

)0.

0(0

.0–0

.4)

0.3

(0.0

–0.2)

1.2(0

.1–4.1

)0.

5(0

.0–2

.4)

0.0

(0.0

–16.1)

NS

Latin

Am

erica

and

the

Carib

bean

0.7

(0.0

–2.1)

0.9

(0.0

–2.8

)0.

0(0

.0–6

.8)

0.4

(0.2–

1.7)

1.5(0

.3–3.2

)1.2

(0.7–

1.8)

2.9(1

.6–4

.6)

—NS

Over

all0.

3(0

.0–1.

0)0.

9(0

.2–2.0

)0.

0(0

.0–0

.1)0.

0(0

.0–0

.3)0.

2(0

.0–0

.6)

0.7

(0.3–

1.4)

0.9

(0.2–

1.9)

0.0

(0.0

–1.4)

NS

aSt

atist

icalm

etho

dsar

ede

scrib

edin

Sec

tion

9,A

nnex

1.NS

:not

stat

istica

llysi

gnifi

cant

.—

Dat

anot

avail

able

ora

pplic

able

.

Tabl

e 1 P

reva

lenc

e of

HIV

dru

g re

sista

nce

amon

g an

tiret

rovi

ral t

hera

py–n

aive

indi

vidu

als f

rom

the

publ

ished

lite

ratu

re, b

y ye

ar, r

egio

n an

d cl

ass o

f ant

iretr

ovira

l dru

g (%

with

at l

east

one

dru

g re

sista

nce

mut

atio

n (9

5% co

nfide

nce

inte

rval

)), 2

003–

2010

58

WHO

Reg

ion

Subr

egio

nCo

untr

yGe

ogra

phic

al a

rea

Year

of i

mpl

emen

tatio

nPo

pula

tion

NNRT

INR

TIPI

Afric

aEa

ster

nEt

hiop

iaAd

disA

baba

2005

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aEa

ster

nKe

nya

Nairo

bi20

05Pr

egna

ntw

omen

<5%

<5%

<5%

Afric

aEa

ster

nKe

nya

Mom

basa

2009

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s5–

15%

<5%

5–15

%

Afric

aEa

ster

nM

alaw

iLil

ongw

e20

06Pr

egna

ntw

omen

<5%

<5%

<5%

Afric

aEa

ster

nM

alaw

iBl

anty

re20

09Pr

egna

ntw

omen

<5%

<5%

<5%

Afric

aEa

ster

nM

alaw

iLil

ongw

e20

09Pr

egna

ntw

omen

5–15

%<5

%<5

%

Afric

aEa

ster

nM

alaw

iLil

ongw

e20

10Pr

egna

ntw

omen

5–15

%<5

%<5

%

Afric

aEa

ster

nM

ozam

biqu

eBe

ira20

07Pr

egna

ntw

omen

<5%

5–15

%<5

%

Afric

aEa

ster

nM

ozam

biqu

eBe

ira20

09Pr

egna

ntw

omen

5–15

%<5

%<5

%

Afric

aEa

ster

nM

ozam

biqu

eM

aput

o20

09Pr

egna

ntw

omen

<5%

<5%

<5%

Afric

aEa

ster

nUn

ited

Repu

blic

ofTa

nzan

iaDa

resS

alaam

2005

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aEa

ster

nUg

anda

Ente

bbe

2006

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aEa

ster

nUg

anda

Kam

pala

2008

Sexw

orke

rsNC

<5%

<5%

Afric

aEa

ster

nUg

anda

Kam

pala

2009

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s<5

%5–

15%

<5%

Afric

aSo

uthe

rnAn

gola

Luan

da20

09Pr

egna

ntw

omen

<5%

<5%

<5%

Afric

aSo

uthe

rnBo

tsw

ana

Fran

cisto

wn

2005

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aSo

uthe

rnBo

tsw

ana

Gabo

rone

2005

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aSo

uthe

rnBo

tsw

ana

Fran

cisto

wn

2007

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aSo

uthe

rnLe

soth

oM

aser

u20

09Pr

egna

ntw

omen

NC<5

%<5

%

Afric

aSo

uthe

rnNa

mib

iaW

indh

oek

2006

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aSo

uthe

rnSo

uth

Afric

aGa

uten

g20

04Pr

egna

ntw

omen

<5%

5–15

%NC

Afric

aSo

uthe

rnSo

uth

Afric

aGa

uten

g20

05Pr

egna

ntw

omen

<5%

<5%

NC

Afric

aSo

uthe

rnSo

uth

Afric

aKw

aZul

u-Na

tal

2005

Preg

nant

wom

enNC

<5%

NC

Afric

aSo

uthe

rnSo

uth

Afric

aGa

uten

g20

06Pr

egna

ntw

omen

<5%

<5%

<5%

Afric

aSo

uthe

rnSo

uth

Afric

aGa

uten

g20

07Pr

egna

ntw

omen

<5%

<5%

<5%

Afric

aSo

uthe

rnSo

uth

Afric

aKw

aZul

u-Na

tal

2007

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aSo

uthe

rnSo

uth

Afric

aGa

uten

g20

08Pr

egna

ntw

omen

<5%

<5%

<5%

Afric

aSo

uthe

rnSo

uth

Afric

aKw

aZul

u-Na

tala

2008

Preg

nant

wom

en5–

15%

5–15

%<5

%

Afric

aSo

uthe

rnSo

uth

Afric

aGa

uten

g20

09Pr

egna

ntw

omen

<5%

<5%

<5%

Afric

aSo

uthe

rnSo

uth

Afric

aKw

aZul

u-Na

tal

2009

Preg

nant

wom

en5–

15%

<5%

<5%

Afric

aSo

uthe

rnSo

uth

Afric

aGa

uten

g20

10Pr

egna

ntw

omen

<5%

<5%

<5%

Afric

aSo

uthe

rnSo

uth

Afric

aKw

aZul

u-Na

tal

2010

Preg

nant

wom

en5–

15%

5–15

%<5

%

Tabl

e 2

WHO

surv

eys o

f tra

nsm

itted

HIV

dru

g re

sista

nce

with

resu

lts cl

assifi

able

for a

t lea

st o

ne d

rug

clas

s


59

WHO

Reg

ion

Subr

egio

nCo

untr

yGe

ogra

phic

al a

rea

Year

of i

mpl

emen

tatio

nPo

pula

tion

NNRT

INR

TIPI

Afric

aSo

uthe

rnSw

azila

ndM

anzin

i-Mba

bane

2006

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aSo

uthe

rnSw

azila

ndM

anzin

i-Mba

bane

2008

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aSo

uthe

rnSw

azila

ndSh

iselw

eni

2010

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aW

este

rn/c

entra

lBu

rkin

aFas

oBo

boD

ioul

asso

2005

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aW

este

rn/c

entra

lBu

rkin

aFas

oOu

agad

ougo

u20

09Pr

egna

ntw

omen

5–15

%5–

15%

<5%

Afric

aW

este

rn/c

entra

lCa

mer

oon

Doua

la20

06Pr

egna

ntw

omen

<5%

5–15

%<5

%

Afric

aW

este

rn/c

entra

lCa

mer

oon

Yaou

ndé

2006

Preg

nant

wom

en5–

15%

<5%

<5%

Afric

aW

este

rn/c

entra

lCh

adN’

Djam

ena

2006

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aW

este

rn/c

entra

lCo

ted

'Ivoi

reAb

idjan

2008

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aW

este

rn/c

entra

lGh

ana

Man

yaK

robo

2008

Preg

nant

wom

en<5

%<5

%<5

%

Afric

aW

este

rn/c

entra

lSe

nega

lDa

kar

2007

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s<5

%<5

%<5

%

Amer

icas

Mex

icoM

exico

City

2004

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s<5

%5–

15%

<5%

Euro

pean

Ukra

ine

Done

tsk

2009

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s<5

%<5

%<5

%

Euro

pean

Ukra

ine

Kher

son

2009

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

sNC

<5%

NC

Euro

pean

Ukra

ine

Kyiva

2009

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

sNC

5–15

%NC

Euro

pean

Ukra

ine

Odes

sa20

09Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

<5%

<5%

<5%

Sout

h-Ea

stA

siaIn

dia

Kakin

ada

2007

Preg

nant

wom

en<5

%<5

%<5

%

Sout

h-Ea

stA

siaIn

dia

Mum

bai

2007

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s<5

%<5

%NC

Sout

h-Ea

stA

siaIn

done

siaJa

karta

2006

Peop

lew

hoin

ject

dru

gs<5

%<5

%<5

%

Sout

h-Ea

stA

siaTh

ailan

dBa

ngko

k20

05Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

<5%

NC<5

%

Sout

h-Ea

stA

siaTh

ailan

dBa

ngko

k20

05Bl

ood

dono

rs<5

%<5

%<5

%

Sout

h-Ea

stA

siaTh

ailan

dCh

iang

Mai

2007

Preg

nant

wom

en<5

%<5

%<5

%

Wes

tern

Pac

ific

Cam

bodi

aPh

nom

Pen

ha20

08Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

5–15

%NC

<5%

Wes

tern

Pac

ific

Chin

aHu

nan

2007

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s<5

%<5

%<5

%

Wes

tern

Pac

ific

Chin

aBe

ijing

2008

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s<5

%<5

%<5

%

Wes

tern

Pac

ific

Chin

aSh

enzh

en(G

uang

dong

)20

08Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

<5%

<5%

<5%

Wes

tern

Pac

ific

Chin

aLia

ngsh

an(S

ichua

n)20

08Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

<5%

<5%

<5%

Wes

tern

Pac

ific

Chin

aUr

umqi

City

(Xin

jiang

)20

08Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

<5%

NC5–

15%

Wes

tern

Pac

ific

Chin

aYi

liCity

(Xin

jiang

)20

08Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

<5%

<5%

<5%

Wes

tern

Pac

ific

Chin

aKu

nmin

g(Y

unna

n)20

08Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

<5%

<5%

<5%

Wes

tern

Pac

ific

Chin

aBe

ijing

2009

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s<5

%<5

%<5

%

Wes

tern

Pac

ific

Chin

aGu

izhou

2009

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s<5

%<5

%<5

%

Wes

tern

Pac

ific

Chin

aHu

nan

2009

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s5–

15%

<5%

<5%

Tabl

e 2

WHO

surv

eys o

f tra

nsm

itted

HIV

dru

g re

sista

nce

with

resu

lts cl

assifi

able

for a

t lea

st o

ne d

rug

clas

s (co

nt.)

60

WHO

Reg

ion

Subr

egio

nCo

untr

yGe

ogra

phic

al a

rea

Year

of i

mpl

emen

tatio

nPo

pula

tion

NNRT

INR

TIPI

Wes

tern

Pac

ific

Chin

aJia

ngsu

2009

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s<5

%<5

%<5

%

Wes

tern

Pac

ific

Chin

aSh

enzh

en(G

uang

dong

)20

09Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

<5%

<5%

<5%

Wes

tern

Pac

ific

Chin

aLia

ngsh

an(S

ichua

n)20

09Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

<5%

<5%

<5%

Wes

tern

Pac

ific

Chin

aDe

hong

(Yun

nan)

2009

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s<5

%<5

%<5

%

Wes

tern

Pac

ific

Chin

aZh

ejian

g20

09Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

<5%

5–15

%<5

%

Wes

tern

Pac

ific

Viet

Nam

Hano

i20

06Vo

lunt

aryc

ouns

ellin

gan

dte

stin

gat

tend

ees

<5%

<5%

<5%

Wes

tern

Pac

ific

Viet

Nam

HoC

hiM

inh

City

2007

Volu

ntar

ycou

nsel

ling

and

test

ing

atte

ndee

s5–

15%

<5%

<5%

Over

allto

tals

Notc

lassifi

able

53

6

<5%

5559

64

5–15

%12

102

>15%

00

0

Tota

l72

7272

aM

oder

ate

class

ifica

tions

des

igna

ted

altho

ugh

spec

imen

num

berd

oesn

otp

erm

itdi

stin

ctio

nbe

twee

nm

oder

ate

and

high

.NC

:not

clas

sifiab

le.

Tabl

e 2

WHO

surv

eys o

f tra

nsm

itted

HIV

dru

g re

sista

nce

with

resu

lts cl

assifi

able

for a

t lea

st o

ne d

rug

clas

s (co

nt.)


61

Year

of I

mpl

emen

tatio

n

Regi

onCo

untry

2004

2005

2006

2007

2008

2009

2010

Afric

aBo

tsw

ana

Fran

cisto

wn;

Ga

boro

neFr

ancis

tow

n

Afric

aBu

rkin

aFas

oBo

boD

ioul

asso

Ouag

adou

gou

(NNR

TI+N

RTI)

Afric

aKe

nya

Nairo

biM

omba

sa(N

NRTI

+PI)

Afric

aM

alaw

iLil

ongw

eLil

ongw

e(N

NRTI

);Bl

anty

reLil

ongw

e(N

NRTI

)

Afric

aM

ozam

biqu

eBe

ira(N

RTI)

Beira

(NNR

TI);

Map

uto

Afric

aSo

uth

Afric

aGa

uten

g(N

RTI)

Gaut

eng;

Kw

aZul

u-Na

tal

Gaut

eng

Gaut

eng;

KwaZ

ulu-

Nata

l

Gaut

eng;

Kw

aZul

u-Na

tal

(NNR

TI+

NRT

I)

Gaut

eng;

Kw

aZul

u-Na

tal

(NNR

TI)

Gaut

eng;

Kw

aZul

u-Na

tal(

NNRT

I+NR

TI)

Afric

aSw

azila

ndM

anzin

i-Mba

bane

Man

zini-M

baba

neSh

iselw

eni

Afric

aUg

anda

bEn

tebb

eKa

mpa

laKa

mpa

la(N

RTI)

Sout

hEa

stA

siaTh

ailan

dBa

ngko

k(2)

Chian

gM

ai

Wes

tern

Pac

ific

Chin

aHu

nan

Beijin

g;K

unm

ing;

Lian

gsha

n;

Shen

zhen

;Ur

umqi

city

(PI);

Yi

licity

Beijin

g;D

ehon

g;G

uizh

ou;

Huna

n(N

NRTI

);Jia

ngsu

;Lia

ngsh

an;

Shen

zhen

;Zh

ejian

g(N

RTI)

Wes

tern

Pac

ific

Viet

Nam

Hano

iHo

Chi

Min

hCi

ty(N

NRTI

)

aAr

easw

ithm

oder

ate

class

ifica

tion

oftr

ansm

itted

dru

gre

sista

nce

are

inre

d.

bEn

tebb

ean

dKa

mpa

lain

Uga

ndaa

reco

nsid

ered

asth

esa

me

geog

raph

icar

ea.

Tabl

e 3

Coun

trie

s with

at l

east

two

WHO

tran

smitt

ed d

rug

resis

tanc

e su

rvey

s rep

eate

d ov

er ti

me,

by

year

of s

urve

y, 2

004–

2010

a

62

Afric

aLa

tin A

mer

ica

and

the

Carib

bean

Euro

peSo

uth-

East

Asi

aW

este

rn P

acifi

cDr

ug cl

ass

num

ber o

f sur

veys

with

m

oder

ate

clas

sific

atio

nEa

ster

nSo

uthe

rnW

este

rn/C

entra

lAf

rica

tota

l

Any

206

(30%

)a4

(20%

)3

(15%

)13

(65%

)1(

5%)

1(5%

)0

(0%

)5

(25%

)

NNRT

I12

4(3

3.3%

)3

(25%

)2

(16.7%

)9

(75%

)0

(0%

)0

(0%

)0

(0%

)3

(25%

)

NRTI

102

(20%

)3

(30%

)2

(20%

)7

(70%

)1(

10%

)1(

10%

)0

(0%

)1(

10%

)

PI2

1(50

%)

0(0

%)

0(0

%)

1(50

%)

0(0

%)

0(0

%)

0(0

%)

1(50

%)

NNRT

Iand

NRT

I3

0(0

%)

2(6

6.7%

)1(

33.3%

)3

(100%

)0

(0%

)0

(0%

)0

(0%

)0

(0%

)

Tota

lnum

bero

fsu

rvey

s72

1421

843

14

618

aPe

rcen

tage

sare

out

oft

hen

umbe

rfor

eac

hdr

ugcl

assc

ateg

ory.

Tabl

e 4

Tran

smitt

ed d

rug

resis

tanc

e su

rvey

s with

mod

erat

e cl

assifi

catio

n (b

etw

een

5% a

nd 15

%) d

rug

resis

tanc

e


63

Year

of I

mpl

emen

tatio

n

WHO

Reg

ion

2007

2008

2009

2010

p-va

luea

Any drug class

Afric

a4.

6(3

.6–5

.8)

3.7(2

.8–4

.8)

4.6

(2.2–

7.8)

6.8

(4.8

–9.0

)0.

04

Sout

h-Ea

stA

sia7.9

(4.0

–13.7)

5.4(2

.9–8

.6)

——

0.34

Over

all4.

8(3

.8–6

.0)

3.9(3

.0–4

.9)4.

6(2

.2–7.8

)6.

8(4

.8–9

.0)

0.06

NRTI

Afric

a1.1

(0.6

–1.7)

1.0(0

.5–1.

6)1.1

(0.3–

2.2)

1.0(0

.3–2.0

)0.

75

Sout

h-Ea

stA

sia3.6

(1.2–

8.2)

4.3

(2.0

–7.2)

——

0.74

Over

all1.2

(0.7–

2.0)

1.3(0

.8–2

.0)

1.1(0

.3–2.2

)1.0

(0.3–

2.0)

0.70

NNRTI

Afric

a3.4

(2.4

–4.5

)2.3

(1.4

–3.3)

3.3(1

.8–5

.0)

5.4(3

.7–7.4

)0.

03

Sout

h-Ea

stA

sia7.9

(4.0

–13.7)

3.0(1

.2–5.

6)—

——

Over

all3.7

(2.5

–4.9)

2.4(1

.6–3

.3)3.3

(1.8

–5.0

)5.4

(3.7–

7.4)

0.06

PI

Afric

a0.

3(0

.1–0.

8)0.

5(0

.1–0.

9)0.

5(0

.1–1.7

)0.

0(0

.0–0

.4)

0.82

Sout

h-Ea

stA

sia0.

0(0

.0–2

.6)

0.0

(0.0

–0.7)

——

—

Over

all0.

3(0

.0–0

.7)0.

4(0

.1–0.

8)0.

5(0

.1–1.7

)0.

0(0

.0–0

.4)

0.97

aSt

atist

icalm

etho

dsar

ede

scrib

edin

Sec

tion

9,A

nnex

1.—

Dat

aare

not

avail

able

ora

pplic

able

.

Tabl

e 5

Estim

ated

pre

vale

nce

of m

utat

ions

in W

HO a

cqui

red

drug

resis

tanc

e su

rvey

s at b

asel

ine

by re

gion

and

by

drug

clas

s % (9

5% co

nfide

nce

inte

rval

s), 2

007–

2010

64

Surv

ey

Num

ber o

f peo

ple

with

gen

otyp

es

adm

inis

tere

d a

ques

tionn

aire

on

prev

ious

exp

osur

e to

AR

Vs (A

)

Num

ber o

f peo

ple

repo

rtin

g pr

evio

us

expo

sure

to A

RVs

(incl

udin

g fo

r PM

TCT)

(B)

Num

ber o

f peo

ple

repo

rtin

g pr

evio

us

expo

sure

to A

RVs

with

RT

SDRM

at

base

line

(C)

Num

ber o

f peo

ple

repo

rtin

g pr

evio

us

expo

sure

to A

RVs

with

out R

T SD

RM a

t ba

selin

e (D

)

Num

ber o

f peo

ple

repo

rtin

g no

pre

viou

s ex

posu

re to

ARV

s w

ith R

T SD

RM a

t ba

selin

e (E

)

Num

ber o

f peo

ple

repo

rtin

g no

pre

viou

s ex

posu

re to

ARV

s w

ithou

t RT

SDRM

at

base

line

(F)

% re

port

ing

prev

ious

ex

posu

re to

ARV

s w

ith R

T SD

RM a

t ba

selin

e

(C/(

C+D)

)

% re

port

ing

no

prev

ious

exp

osur

e to

AR

Vs w

ith R

T SD

RM

at b

asel

ine

(E

/(E+

F))

Keny

a-12

007

210

92

76

195

22.2%

3.0%

Keny

a-2

2008

206

60

67

193

0.0%

3.5%

Sout

hAf

rica-

1200

721

14

40

819

910

0.0%

3.9%

Sout

hAf

rica-

220

0719

619

613

217

531

.6%

1.1%

Sout

hAf

rica-

320

0718

310

37

616

730

.0%

3.5%

Zam

bia-

1200

721

94

13

720

825

.0%

3.3%

Zam

bia-

220

0720

810

46

719

140

.0%

3.5%

Zam

bia-

320

0710

44

04

694

0.0%

6.0%

Zim

babw

e-12

008

206

143

114

188

21.4

%2.1

%

Zim

babw

e-2

2009

126

170

174

105

0.0%

3.7%

Zim

babw

e-3

2009

124

71

64

11314

.3%3.4

%

Zim

babw

e-4

2009

133

272

258

987.4

%7.5

%

Zim

babw

e-5

2010

243

12

120

33.3%

4.8%

Zim

babw

e-6

2010

488

26

535

25.0

%12

.5%

Zim

babw

e-7

2010

149

793

762

683.8

%2.9

%

Zim

babw

e-8

2010

11210

010

795

0.0%

6.9%

Zim

babw

e-9

2010

11010

010

991

0.0%

9.0%

Zim

babw

e-10

201

010

10

12

70.

0%22

.2%

Zim

babw

e-11

2010

132

191

186

107

5.3%

5.3%

Zim

babw

e-12

201

012

96

24

5118

33.3%

4.1%

Cam

eroo

n-12

009

141

20

23

136

0.0%

2.2%

Nige

ria-3

200

819

52

11

418

950

.0%

2.1%

Indi

a-12

007

140

115

66

123

45.5

%4.

7%

Indi

a-2

2008

148

43

15

139

75.0

%3.5

%

3464

286

4424

212

430

5415

.4%

3.9%

RT:r

ever

setr

ansc

ripta

se.S

DRM

:sur

veilla

nce

drug

resis

tanc

em

utat

ion.

Tabl

e 6

Prio

r exp

osur

e to

ant

iretr

ovira

l dru

gs a

nd d

etec

tion

of re

sista

nce

mut

atio

ns a

t bas

elin

e, b

y in

divi

dual

WHO

HIV

dru

g re

sista

nce

surv

ey


65

Figu

re 1

Pred

icte

d HI

V dr

ug re

sista

nce

in p

eopl

e in

itiat

ing

antir

etro

vira

l the

rapy

(WHO

HIV

dru

g re

sista

nce

surv

ey b

asel

ine

spec

imen

s)a

Percentage of baseline genotypes

0%

2%

4%

6%

8%

10%

12%

18%

20%

14%

16%

Any drug

NRTI and N

NRTI

Any PI

NVP

EFV

RPV

ETR

3TC/FT

C

ABC

ddI

d4T

TDF

AZT

Any NNRTI

Any NRTI

aDe

taile

dm

etho

dolo

gica

lnot

esar

eav

ailab

lein

Sec

tion

5,A

nnex

1.

66

WHO

Reg

ion

Subr

egio

nSu

rvey

Num

ber

of p

eopl

e en

rolle

d

Peop

le re

ceiv

ing

first

-line

ART

at

12 m

onth

s n

Lost

to fo

llow

-up

n (%

)St

oppe

d AR

Tn

(%)

Tran

sfer

red

out

n (%

)De

aths

n

(%)

Switc

h to

seco

nd li

ne

n (%

)Un

clas

sifia

ble

n

(%)

Afric

aEa

ster

nBu

rund

i–12

007

125

1141(

0.8%

)—

1(0.

8%)

9(7

.2%)

——

Afric

aEa

ster

nBu

rund

i–22

007

136

126

3(2

.2%)

—3

(2.2%

)3

(2.2%

)—

1(0.

7%)

Afric

aEa

ster

nKe

nya–

1200

722

118

117

(7.7%

)—

8(3

.6%

)15

(6.8

%)

——

Afric

aEa

ster

nKe

nya–

220

0822

320

510

(4.5

%)

1(0.

4%)

5(2

.2%)

2(0

.9%

)—

—

Afric

aEa

ster

nM

alaw

i–12

006

148

7424

(16.

2%)

5(3

.4%

)28

(18.

9%)

17(1

1.5%

)—

—

Afric

aEa

ster

nM

alaw

i–22

006

143

8928

(19.

6%)

—14

(9.8

%)

12(8

.4%

)—

—

Afric

aEa

ster

nM

alaw

i–32

006

145

9520

(13.8

%)

—13

(9%

)17

(11.7

%)

——

Afric

aEa

ster

nM

alaw

i–42

006

160

737

(4.4

%)

—62

(38.

8%)

18(1

1.3%

)—

—

Afric

aEa

ster

nM

alaw

i–12

008

153

1129

(5.9

%)

—13

(8.5

%)

16(1

0.5%

)—

3(2

%)

Afric

aEa

ster

nM

alaw

i–22

008

150

108

20(1

3.3%

)—

8(5

.3%)

12(8

%)

—2

(1.3%

)

Afric

aEa

ster

nM

alaw

i–32

008

150

99)

26(1

7.3%

)—

14(9

.3%)

10(6

.7%)

—1(

0.7%

)

Afric

aEa

ster

nM

alaw

i–42

008

150

11712

(8%

)—

7(4

.7%)

11(7

.3%)

—3

(2%

)

Afric

aEa

ster

nM

ozam

biqu

e–12

007

11910

112

(10.

1%)

——

6(5

%)

——

East

ern

Afric

a20

2314

9418

9(9

.3%)

6(0

.3%)

176

(8.7%

)14

8(7

.3%)

0(0

%)

10(0

.5%

)

Afric

aSo

uthe

rnSo

uth

Afric

a–12

007

224

176

13(5

.8%

)—

11(4

.9%

)23

(10.

3%)

1(0.

4%)

—

Afric

aSo

uthe

rnSo

uth

Afric

a–2

2007

205

169

12(5

.9%

)1(

0.5%

)8

(3.9

%)

14(6

.8%

)1(

0.5%

)—

Afric

aSo

uthe

rnSo

uth

Afric

a–3

2007

208

166

18(8

.7%)

—12

(5.8

%)

11(5

.3%)

1(0.

5%)

—

Afric

aSo

uthe

rnSw

azila

nd–1

2008

133

6927

(20.

3%)

—20

(15%

)15

(11.3

%)

—2

(1.5%

)

Afric

aSo

uthe

rnSw

azila

nd–2

200

812

769

48(3

7.8%

)—

8(6

.3%)

2(1.

6%)

——

Afric

aSo

uthe

rnZa

mbi

a–12

007

239

167

30(1

2.6%

)—

13(5

.4%

)27

(11.3

%)

2(0

.8%

)—

Afric

aSo

uthe

rnZa

mbi

a–2

2007

228

197

6(2

.6%

)4

(1.8%

)4

(1.8%

)17

(7.5

%)

——

Afric

aSo

uthe

rnZa

mbi

a–3

2007

11685

10(8

.6%

)—

1(0.

9%)

20(1

7.2%

)—

—

Afric

aSo

uthe

rnZi

mba

bwe–

1200

823

021

64

(1.7%

)—

3(1.

3%)

7(3

%)

——

Sout

hern

Afri

ca17

1013

1416

8(9

.8%

)5

(0.3%

)80

(4.7%

)13

6(8

%)

5(0

.3%)

2(0

.1%)

Afric

aW

este

rn/C

entra

lCa

mer

oon–

1200

914

176

47(3

3.3%

)-

5(3

.5%

)13

(9.2%

)-

-

Afric

aW

este

rn/C

entra

lNi

geria

–120

0814

090

43(3

0.7%

)-

2(1.

4%)

5(3

.6%

)-

-

Afric

aW

este

rn/C

entra

lNi

geria

–22

008

143

8454

(37.8

%)

-1(

0.7%

)4

(2.8

%)

--

Afric

aW

este

rn/C

entra

lNi

geria

–32

008

208

153

40(1

9.2%

)-

4(1.

9%)

9(4

.3%)

2(1%

)-

Wes

tern

/Cen

tralA

frica

1710

1314

168

(9.8

%)

5(0

.3%)

80(4

.7%)

136

(8%

)5

(0.3%

)2

(0.1%

)

Afric

a43

6532

1154

1(12

.4%

)11

(0.3%

)26

8(6

.1%)

315

(7.2%

)7

(0.2%

)12

(0.3%

)

Sout

h-Ea

stA

siaIn

dia–

1200

714

188

28(1

9.9%

)—

11(7

.8%

)13

(9.2%

)—

1(0.

7%)

Sout

h-Ea

stA

siaIn

dia–

220

0814

810

419

(12.8

%)

1(0.

7%)

10(6

.8%

)14

(9.5

%)

——

Sout

h-Ea

stA

siaIn

done

sia–1

2008

11072

11(10

%)

1(0.

9%)

5(4

.5%

)20

(18.

2%)

—1(

0.9%

)

Sout

h-Ea

stA

sia39

926

458

(14.

5%)

2(0

.5%

)26

(6.5

%)

47(1

1.8%

)0

(0%

)2

(0.5

%)

Over

all47

6434

7559

9(12

.6%

)13

(0.3%

)29

4(6

.2%)

362

(7.6

%)

7(0

.1%)

14(0

.3%)

Tabl

e 7

WHO

Acq

uire

d HI

V dr

ug re

sista

nce

surv

ey e

ndpo

ints


67

0%

10%

20

%

30%

40

%

50%

60

%

70%

80

%

90%

10

0%

Burundi-1 20

07

Burundi-2 20

07

Kenya-1

2007

Kenya-2

2008

Malawi-1 20

06

Malawi-2 20

06

Malawi-3 20

06

Malawi-4 20

06

Malawi-1 20

08

Malawi-2 20

08

Malawi-3 20

08

Malawi-4 20

08

Mozambique-1

2007

Eastern Afri

ca

South Africa-1

2007

South Africa-2

2007

South Africa-3

2007

Swaziland-1

2008

Swaziland-2

2008

Zambia-1 20

07

Zambia-2 20

07

Zambia-3 20

07

Zimbabwe-1

2008

Southern Africa

Cameroon-1 20

09

Nigeria-1

2008

Nigeria-2

2008

Nigeria-3

2008

Weste

rn/Central A

frica India-1

2007

India-2 20

08

Indonesia-1

2008

South-East Asia

Ove

rall Un

clas

sifia

ble

Sw

itch

De

aths

Tr

ansf

er o

ut

Stop

ped

antir

etro

vira

l the

rapy

Lo

st to

follo

w-u

p

Peop

le re

ceiv

ing

first

-line

ant

iretro

vira

l the

rapy

at 1

2 m

onth

s

Figu

re 2

WHO

acq

uire

d dr

ug re

sista

nce

surv

ey e

ndpo

ints

, by

clin

ic su

rvey

ed

68

WHO

Reg

ion

Subr

egio

nSu

rvey

Urba

n/ru

ral

Site

type

Peop

le w

ith R

T SD

RM

at A

RT in

itiat

ion

(%)

HIV

drug

resi

stan

ce

prev

entio

n (%

of p

eopl

e in

itiat

ing

ARTb )

Any

HIV

drug

resi

stan

ce a

t end

poin

ta

HIV

drug

resi

stan

ce p

ossi

ble

(% o

f peo

ple

initi

atin

g AR

b )(%

of p

eopl

e in

itiat

ing

ARTb )

(% o

f peo

ple

faili

ng

with

gen

otyp

es)

Afric

aEa

ster

nBu

rund

i–12

007

Urba

nPr

ivate

1.8%

84.4

%2.8

%25

.0%

12.8

%

Afric

aEa

ster

nBu

rund

i–22

007

Urba

nPu

blic

0.8%

80.6

%4.

6%27

.8%

14.8

%

Afric

aEa

ster

nKe

nya–

1200

7Ur

ban

Publ

icw

ithe

xter

nal

supp

ort

3.3%

82.6

%2.6

%41

.7%14

.7%

Afric

aEa

ster

nKe

nya–

220

08Ur

ban

Priva

te2.9

%86

.1%4.1

%10

0.0%

9.8%

Afric

aEa

ster

nM

alaw

i–12

006

Urba

nPu

blic

with

ext

erna

lsu

ppor

t—

60.0

%2.5

%66

.7%37

.5%

Afric

aEa

ster

nM

alaw

i–22

006

Urba

nPu

blic

—67

.6%

4.8%

83.3%

27.6

%

Afric

aEa

ster

nM

alaw

i–32

006

Urba

nPu

blic

—79

.1%2.7

%10

0.0%

18.2%

Afric

aEa

ster

nM

alaw

i–42

006

Urba

nPu

blic

with

ext

erna

lsu

ppor

t—

83.1%

3.4%

66.7%

13.6

%

Afric

aEa

ster

nM

alaw

i–12

008

Urba

nPu

blic

with

ext

erna

lsu

ppor

t6.

7%85

.0%

5.8%

77.8

%9.

2%

Afric

aEa

ster

nM

alaw

i–22

008

Urba

nPu

blic

3.5%

74.8

%4.

9%83

.3%20

.4%

Afric

aEa

ster

nM

alaw

i–32

008

Urba

nPu

blic

2.9%

73.3%

2.5%

60.0

%24

.2%

Afric

aEa

ster

nM

alaw

i–42

008

Urba

nPu

blic

with

ext

erna

lsu

ppor

t4.

0%83

.2%6.

7%10

0.0%

10.1%

Afric

aEa

ster

nM

ozam

biqu

e–12

007

Urba

nPu

blic

7.0%

79.8

%8.

3%10

0.0%

11.9%

East

ern

Afric

a3.6

%79

.4%

4.3%

63.7%

16.4

%

Afric

aSo

uthe

rnSo

uth

Afric

a–12

007

Rura

lPr

ivate

5.7%

85.3%

3.8%

58.3%

10.9

%

Afric

aSo

uthe

rnSo

uth

Afric

a–2

2007

Urba

nPr

ivate

4.1%

82.9

%6.1

%71

.4%

11.0%

Afric

aSo

uthe

rnSo

uth

Afric

a–3

2007

Urba

nPu

blic

with

ext

erna

lsu

ppor

t4.

9%86

.0%

0.6%

50.0

%13

.5%

Afric

aSo

uthe

rnSw

azila

nd–1

2008

Rura

lPr

ivate

3.2%

61.8

%6.

7%85

.7%31

.5%

Afric

aSo

uthe

rnSw

azila

nd–2

200

8Ur

ban

Publ

ic2.4

%50

.0%

3.8%

80.0

%46

.2%

Afric

aSo

uthe

rnZa

mbi

a–12

007

Urba

nPr

ivate

3.7%

78.1%

2.7%

62.5

%19

.3%

Afric

aSo

uthe

rnZa

mbi

a–2

2007

Urba

nPr

ivate

5.3%

85.2%

7.4%

83.3%

7.4%

Afric

aSo

uthe

rnZa

mbi

a–3

2007

Urba

nPr

ivate

5.8%

77.9

%7.0

%66

.7%15

.1%

Afric

aSo

uthe

rnZi

mba

bwe–

1200

8Ur

ban

Priva

te3.4

%90

.4%

5.5%

80.0

%4.1

%

Afric

aSo

uthe

rnZi

mba

bwe-

220

09Ru

ral

Priva

te3.0

%—

——

—

Afric

aSo

uthe

rnZi

mba

bwe-

320

09Ur

ban

Publ

ic4.

0%—

——

—

Afric

aSo

uthe

rnZi

mba

bwe-

420

09Ru

ral

Publ

ic7.5

%—

——

—

Tabl

e 8

WHO

HIV

dru

g re

sista

nce

surv

ey o

utco

mes

(HIV

dru

g re

sista

nce;

HIV

dru

g re

sista

nce

prev

entio

n; H

IV d

rug

resis

tanc

e po

ssib

le),

by in

divi

dual

ant

iretr

ovira

l the

rapy

clin

ic


69

WHO

Reg

ion

Subr

egio

nSu

rvey

Urba

n/ru

ral

Site

type

Peop

le w

ith R

T SD

RM

at A

RT in

itiat

ion

(%)

HIV

drug

resi

stan

ce

prev

entio

n (%

of p

eopl

e in

itiat

ing

ARTb )

Any

HIV

drug

resi

stan

ce a

t end

poin

ta

HIV

drug

resi

stan

ce p

ossi

ble

(% o

f peo

ple

initi

atin

g AR

b )(%

of p

eopl

e in

itiat

ing

ARTb )

(% o

f peo

ple

faili

ng

with

gen

otyp

es)

Afric

aSo

uthe

rnZi

mba

bwe-

520

10Ur

ban

Publ

ic7.1

%—

——

—

Afric

aSo

uthe

rnZi

mba

bwe-

620

10Pe

ri-Ur

ban

Priva

te12

.5%

——

——

Afric

aSo

uthe

rnZi

mba

bwe-

720

10Ru

ral

Priva

te4.

6%—

——

—

Afric

aSo

uthe

rnZi

mba

bwe-

820

10Ur

ban

Publ

ic6.1

%—

——

—

Afric

aSo

uthe

rnZi

mba

bwe-

920

10Ru

ral

Priva

te8.

8%—

——

—

Afric

aSo

uthe

rnZi

mba

bwe-

102

010

Rura

lPu

blic

20.0

%—

——

—

Afric

aSo

uthe

rnZi

mba

bwe-

1120

10Ru

ral

Publ

ic6.

0%—

——

—

Afric

aSo

uthe

rnZi

mba

bwe-

122

010

Urba

nPu

blic

5.0%

——

——

Sout

hern

Afri

ca5.1

%80

.3%4.

7%73

.3%15

.0%

Afric

aW

este

rn/C

entra

lCa

mer

oon–

1200

9Ur

ban

Publ

ic2.2

%56

.1%3.3

%66

.7%40

.7%

Afric

aW

este

rn/C

entra

lNi

geria

–120

08Ur

ban

Publ

ic0.

7%59

.4%

4.5%

54.5

%36

.1%

Afric

aW

este

rn/C

entra

lNi

geria

–22

008

Urba

nPu

blic

4.5%

51.1%

9.5%

100.

0%39

.4%

Afric

aW

este

rn/C

entra

lNi

geria

–32

008

Urba

nPu

blic

with

ext

erna

lsu

ppor

t2.6

%69

.1%6.

3%70

.6%

24.6

%

Wes

tern

/Cen

tralA

frica

2.5%

59.9

%6.

0%74

.5%

34.1%

Afric

a4.

3%76

.5%

4.7%

69.5

%18

.8%

Sout

h-Ea

stA

sia—

Indi

a–12

007

Urba

nPu

blic

7.9%

67.0

%7.8

%90

.0%

25.2%

Sout

h-Ea

stA

sia—

Indi

a–2

2008

Urba

nPu

blic

5.4%

73.4

%9.

7%92

.3%16

.9%

Sout

h-Ea

stA

sia—

Indo

nesia

–120

08Ur

ban

Publ

ic5.

5%75

.0%

9.2%

100.

0%15

.8%

Sout

h-Ea

stA

sia6.

3%71

.4%

8.9%

93.3%

19.7%

Over

all4.

5%76

.1%5.1

%72

.1%18

.8%

RT:r

ever

setr

ansc

ripta

se.S

DRM

:sur

veilla

nce

drug

resis

tanc

em

utat

ion.

NA:

not

avail

able

.aH

IVd

rug

resis

tanc

ede

fined

asa

drug

resis

tanc

epr

edict

ion

oflo

w,in

term

ediat

eor

hig

hle

velu

sing

the

Stan

ford

HIV

dat

abas

ealg

orith

m.A

ltern

ative

ly,if

calcu

lated

bas

edo

nth

enu

mbe

rof

surv

eilla

nce

drug

resis

tanc

em

utat

ions

ate

ndpo

int,

subr

egio

nal,r

egio

nala

ndo

vera

llpro

porti

onsr

emain

iden

tical.

bE

xclu

desp

eopl

ew

hod

ied

orw

how

ere

trans

ferre

dto

anot

hera

ntire

trovir

alth

erap

yfac

ility.

—D

ataa

ren

otav

ailab

leo

rapp

licab

le.

aHI

Vdr

ugre

sista

nce

defin

edas

adr

ugre

sista

nce

pred

ictio

nof

low,

inte

rmed

iate

orh

igh

leve

lusin

gth

eSt

anfo

rdH

IVd

atab

ase

algor

ithm

.Alte

rnat

ively,

ifca

lculat

edb

ased

on

the

num

bero

fsur

veilla

nce

drug

resis

tanc

em

utat

ions

ate

ndpo

int,

subr

egio

nal,r

egio

nala

ndo

vera

llpr

opor

tions

rem

ainid

entic

al.

bEx

clude

speo

ple

who

die

dor

who

wer

etra

nsfe

rred

toan

othe

rant

iretro

viral

ther

apyf

acilit

y.

Tabl

e 8

WHO

HIV

dru

g re

sista

nce

surv

ey o

utco

mes

(HIV

dru

g re

sista

nce;

HIV

dru

g re

sista

nce

prev

entio

n; H

IV d

rug

resis

tanc

e po

ssib

le),

by in

divi

dual

ant

iretr

ovira

l the

rapy

clin

ic (c

ont.)

70

WHO

Reg

ion

Subr

egio

nSu

rvey

Peop

le

enro

lled

(A

)

Peop

le re

ceiv

ing

first

-line

ART

at

12 m

onth

s with

vi

ral l

oad

resu

lts

clas

sifia

ble

(B)

Peop

le sw

itchi

ng

with

vira

l loa

d cl

assi

fiabl

e (C

)

Lost

to

follo

w-u

p (D

)St

oppe

d AR

T (E

)

Deno

min

ator

fo

r cal

cula

ting

outc

omes

(F

: B+C

+D+E

)

Peop

le w

ith

vira

l loa

d >1

000

copi

es/m

l and

ge

noty

ped

at 12

m

onth

s or s

witc

h (G

)

Num

erat

or

outc

ome

2: p

eopl

e w

ith a

ny H

IVDR

at

12 m

onth

s or

switc

h (H

)

Peop

le w

ith

any

HIVD

R at

12

mon

ths o

r sw

itch

amon

g th

ose

geno

type

d (H

/G)

Out

com

e 2:

HIV

DR

at 12

mon

ths o

r sw

itch

(H/F

)

Afric

aEa

ster

nBu

rund

i–12

007

125

108

01

010

912

325

.0%

2.8%

Afric

aEa

ster

nBu

rund

i–22

007

136

105

03

010

818

527

.8%

4.6%

Afric

aEa

ster

nKe

nya–

1200

722

117

30

170

190

125

41.7%

2.6%

Afric

aEa

ster

nKe

nya–

220

0822

318

30

101

194

88

100.

0%4.1

%

Afric

aEa

ster

nM

alaw

i–12

006

148

510

245

803

266

.7%2.5

%

Afric

aEa

ster

nM

alaw

i–22

006

143

770

280

105

65

83.3%

4.8%

Afric

aEa

ster

nM

alaw

i–32

006

145

900

200

1103

310

0.0%

2.7%

Afric

aEa

ster

nM

alaw

i–42

006

160

520

70

593

266

.7%3.4

%

Afric

aEa

ster

nM

alaw

i–12

008

153

1110

90

120

97

77.8

%5.

8%

Afric

aEa

ster

nM

alaw

i–22

008

150

830

200

103

65

83.3%

4.9%

Afric

aEa

ster

nM

alaw

i–32

008

150

940

260

120

53

60.0

%2.5

%

Afric

aEa

ster

nM

alaw

i–42

008

150

107

012

0119

88

100.

0%6.

7%

Afric

aEa

ster

nM

ozam

biqu

e–12

007

11997

012

010

99

910

0.0%

8.3%

East

ern

Afric

a20

2313

310

189

615

2610

265

63.7%

4.3%

Afric

aSo

uthe

rnSo

uth

Afric

a–12

007

224

170

113

018

412

758

.3%3.8

%

Afric

aSo

uthe

rnSo

uth

Afric

a–2

2007

205

150

112

116

414

1071

.4%

6.1%

Afric

aSo

uthe

rnSo

uth

Afric

a–3

2007

208

153

018

017

12

150

.0%

0.6%

Afric

aSo

uthe

rnSw

azila

nd–1

2008

133

620

270

897

685

.7%6.

7%

Afric

aSo

uthe

rnSw

azila

nd–2

200

812

758

048

010

65

480

.0%

3.8%

Afric

aSo

uthe

rnZa

mbi

a–12

007

239

157

030

018

78

562

.5%

2.7%

Afric

aSo

uthe

rnZa

mbi

a–2

2007

228

193

06

420

318

1583

.3%7.4

%

Afric

aSo

uthe

rnZa

mbi

a–3

2007

11676

010

086

96

66.7%

7.0%

Afric

aSo

uthe

rnZi

mba

bwe–

1200

823

021

50

40

219

1512

80.0

%5.

5%

Sout

hern

Afri

ca17

1012

342

168

514

0990

6673

.3%4.

7%

Afric

aW

este

rn/C

entra

lCa

mer

oon–

1200

914

176

047

012

36

466

.7%3.3

%

Afric

aW

este

rn/C

entra

lNi

geria

–120

0814

090

043

013

311

654

.5%

4.5%

Afric

aW

este

rn/C

entra

lNi

geria

–22

008

143

830

540

137

1313

100.

0%9.

5%

Afric

aW

este

rn/C

entra

lNi

geria

–32

008

208

150

140

019

117

1270

.6%

6.3%

Wes

tern

/Cen

tralA

frica

632

399

118

40

584

4735

74.5

%6.

0%

Afric

a43

6529

643

541

1135

1923

916

669

.5%

4.7%

Sout

h-Ea

stA

siaIn

dia–

1200

714

187

028

0115

109

90.0

%7.8

%

Sout

h-Ea

stA

siaIn

dia–

220

0814

810

40

191

124

1312

92.3%

9.7%

Sout

h-Ea

stA

siaIn

done

sia–1

2008

11064

011

176

77

100.

0%9.

2%

Sout

h-Ea

stA

sia39

925

50

582

315

3028

93.3%

8.9%

Over

all47

6432

193

599

1338

3426

919

472

.1%5.1

%

Tabl

e 9

WHO

acq

uire

d dr

ug re

sista

nce

surv

ey: “

HIV

drug

resis

tanc

e” o

utco

me,

by

indi

vidu

al a

ntire

trov

iral t

hera

py cl

inic


71

WHO

regi

onSu

breg

ion

Surv

eyPe

ople

enr

olle

d (A

)

Peop

le o

n fir

st-li

ne A

RT a

t 12

mon

ths

with

vira

l loa

d re

sults

clas

sifia

ble

(B

)

Patie

nts s

witc

hing

with

vi

ral l

oad

clas

sifia

ble

(C

)

Lost

to

follo

w-u

p

(D)

Stop

ART

(E

)

Deno

min

ator

fo

r cal

cula

ting

outc

omes

(F

: B+C

+D+E

)

Num

erat

or: p

atie

nts

with

vira

l loa

d <1

000

copi

es/m

l at

12 m

onth

s (G)

Out

com

e 1:

HIV

drug

resi

stan

ce

prev

entio

n (G

/F)

Afric

aEa

ster

nBu

rund

i–12

007

125

108

01

010

992

84.4

%

Afric

aEa

ster

nBu

rund

i–22

007

136

105

03

010

887

80.6

%

Afric

aEa

ster

nKe

nya–

1200

722

117

30

170

190

157

82.6

%

Afric

aEa

ster

nKe

nya–

220

0822

318

30

101

194

167

86.1%

Afric

aEa

ster

nM

alaw

i–12

006

148

510

245

8048

60.0

%

Afric

aEa

ster

nM

alaw

i–22

006

143

770

280

105

7167

.6%

Afric

aEa

ster

nM

alaw

i–32

006

145

900

200

11087

79.1%

Afric

aEa

ster

nM

alaw

i–42

006

160

520

70

5949

83.1%

Afric

aEa

ster

nM

alaw

i–12

008

153

1110

90

120

102

85.0

%

Afric

aEa

ster

nM

alaw

i–22

008

150

830

200

103

7774

.8%

Afric

aEa

ster

nM

alaw

i–32

008

150

940

260

120

8873

.3%

Afric

aEa

ster

nM

alaw

i–42

008

150

107

012

0119

9983

.2%

Afric

aEa

ster

nM

ozam

biqu

e–12

007

11997

012

010

987

79.8

%

East

ern

Afric

a20

2313

310

189

615

2612

1179

.4%

Afric

aSo

uthe

rnSo

uth

Afric

a–12

007

224

170

113

018

415

785

.3%

Afric

aSo

uthe

rnSo

uth

Afric

a–2

2007

205

150

112

116

413

682

.9%

Afric

aSo

uthe

rnSo

uth

Afric

a–3

2007

208

153

018

017

114

786

.0%

Afric

aSo

uthe

rnSw

azila

nd–1

2008

133

620

270

8955

61.8

%

Afric

aSo

uthe

rnSw

azila

nd–2

200

812

758

048

010

653

50.0

%

Afric

aSo

uthe

rnZa

mbi

a–12

007

239

157

030

018

714

678

.1%

Afric

aSo

uthe

rnZa

mbi

a–2

2007

228

193

06

420

317

385

.2%

Afric

aSo

uthe

rnZa

mbi

a–3

2007

11676

010

086

6777

.9%

Afric

aSo

uthe

rnZi

mba

bwe–

1200

823

021

50

40

219

198

90.4

%

Sout

hern

Afri

ca17

1012

342

168

514

09113

280

.3%

Afric

aW

este

rn/C

entra

lCa

mer

oon–

1200

914

176

047

012

369

56.1%

Afric

aW

este

rn/C

entra

lNi

geria

–120

0814

090

043

013

379

59.4

%

Afric

aW

este

rn/C

entra

lNi

geria

–22

008

143

830

540

137

7051

.1%

Afric

aW

este

rn/C

entra

lNi

geria

–32

008

208

150

140

019

113

269

.1%

Wes

tern

/Cen

tralA

frica

632

399

118

40

584

350

59.9

%

Afric

a43

6529

643

541

1135

1926

9376

.5%

Sout

h-Ea

stA

siaIn

dia–

1200

714

187

028

0115

7767

.0%

Sout

h-Ea

stA

siaIn

dia–

220

0814

810

40

191

124

9173

.4%

Sout

h-Ea

stA

siaIn

done

sia–1

2008

11064

011

176

5775

.0%

Sout

h-Ea

stA

sia39

925

50

582

315

225

71.4

%

Over

all47

6432

193

599

1338

3429

1876

.1%

Tabl

e 10

WHO

acq

uire

d dr

ug re

sista

nce

surv

ey: “

HIV

drug

resis

tanc

e pr

even

tion”

out

com

e, b

y in

divi

dual

ant

iretr

ovira

l the

rapy

clin

ic

72

WHO

Reg

ion

Subr

egio

nSu

rvey

Peop

le e

nrol

led

(A)

Peop

le re

ceiv

ing

first

-line

an

tiret

rovi

ral t

hera

py a

t 12

mon

ths w

ith v

iral l

oad

resu

lts cl

assi

fiabl

e (B

)

Peop

le

switc

hing

w

ith v

iral l

oad

clas

sifia

ble

(C)

Lost

to

follo

w-u

p (D

)St

oppe

d AR

T (E

)

Deno

min

ator

fo

r out

com

e 3

(F: B

+C+D

+E)

Vira

l loa

d >1

000

copi

es/

ml a

nd w

ild

type

(G

)

Vira

l loa

d >1

000

copi

es/m

l, ge

noty

pe fa

il (H

)

Num

erat

or o

f ou

tcom

e 3

(I: D

+E+G

+H)

Out

com

e 3:

HIV

dr

ug re

sist

ance

po

ssib

le

(I/F

)

Afric

aEa

ster

nBu

rund

i–12

007

125

108

01

010

99

414

12.8

%

Afric

aEa

ster

nBu

rund

i–22

007

136

105

03

010

813

016

14.8

%

Afric

aEa

ster

nKe

nya–

1200

722

117

30

170

190

74

2814

.7%

Afric

aEa

ster

nKe

nya–

220

0822

318

30

101

194

08

199.

8%

Afric

aEa

ster

nM

alaw

i–12

006

148

510

245

801

030

37.5

%

Afric

aEa

ster

nM

alaw

i–22

006

143

770

280

105

10

2927

.6%

Afric

aEa

ster

nM

alaw

i–32

006

145

900

200

1100

020

18.2%

Afric

aEa

ster

nM

alaw

i–42

006

160

520

70

591

08

13.6

%

Afric

aEa

ster

nM

alaw

i–12

008

153

1110

90

120

20

119.

2%

Afric

aEa

ster

nM

alaw

i–22

008

150

830

200

103

10

2120

.4%

Afric

aEa

ster

nM

alaw

i–32

008

150

940

260

120

21

2924

.2%

Afric

aEa

ster

nM

alaw

i–42

008

150

107

012

0119

00

1210

.1%

Afric

aEa

ster

nM

ozam

biqu

e–12

007

11997

012

010

90

113

11.9%

East

ern

Afric

a20

2313

310

189

615

2637

1825

016

.4%

Afric

aSo

uthe

rnSo

uth

Afric

a–12

007

224

170

113

018

45

220

10.9

%

Afric

aSo

uthe

rnSo

uth

Afric

a–2

2007

205

150

112

116

44

018

11.0%

Afric

aSo

uthe

rnSo

uth

Afric

a–3

2007

208

153

018

017

11

423

13.5

%

Afric

aSo

uthe

rnSw

azila

nd–1

2008

133

620

270

891

028

31.5

%

Afric

aSo

uthe

rnSw

azila

nd–2

200

812

758

048

010

61

049

46.2%

Afric

aSo

uthe

rnZa

mbi

a–12

007

239

157

030

018

73

336

19.3%

Afric

aSo

uthe

rnZa

mbi

a–2

2007

228

193

06

420

33

215

7.4%

Afric

aSo

uthe

rnZa

mbi

a–3

2007

11676

010

086

30

1315

.1%

Afric

aSo

uthe

rnZi

mba

bwe–

1200

823

021

50

40

219

32

94.1

%

Sout

hern

Afri

ca17

1012

342

168

514

0924

1321

115

.0%

Afric

aW

este

rn/C

entra

lCa

mer

oon–

1200

914

176

047

012

32

150

40.7%

Afric

aW

este

rn/C

entra

lNi

geria

–120

0814

090

043

013

35

048

36.1%

Afric

aW

este

rn/C

entra

lNi

geria

–22

008

143

830

540

137

00

5439

.4%

Afric

aW

este

rn/C

entra

lNi

geria

–32

008

208

150

140

019

15

247

24.6

%

Wes

tern

/Cen

tralA

frica

632

399

118

40

584

123

199

34.1%

Afric

a43

6529

643

541

1135

1973

3466

018

.8%

Sout

h-Ea

stA

siaIn

dia–

1200

714

187

028

0115

10

2925

.2%

Sout

h-Ea

stA

siaIn

dia–

220

0814

810

40

191

124

10

2116

.9%

Sout

h-Ea

stA

siaIn

done

sia–1

2008

11064

011

176

00

1215

.8%

Sout

h-Ea

stA

sia39

925

50

582

315

20

6219

.7%

Over

all47

6432

193

599

1338

3475

3472

218

.8%

Tabl

e 11

WHO

acq

uire

d dr

ug re

sista

nce

surv

ey: “

HIV

drug

resis

tanc

e po

ssib

le” o

utco

me,

by

indi

vidu

al a

ntire

trov

iral t

hera

py cl

inic


73

Coun

try

and

regi

onA

CCR

F01

CRF0

2CR

F06

CRF1

6D

GO

ther

aTo

tal

Buru

ndi

3220

0—

——

—4

12

239

Keny

a27

754

——

—26

565

—41

8

Mala

wi

155

8—

——

—3

3—

565

Moz

ambi

que

—99

——

——

1—

—10

0

East

ern

Afric

a31

091

1—

——

2664

92

1322

Sout

hAf

rica

257

8—

1—

12

—6

590

Swaz

iland

—24

9—

——

1—

——

250

Zam

bia

351

9—

3—

—1

32

531

Zim

babw

e3

1374

——

——

11

—13

79

Sout

hern

Afri

ca8

2720

—4

—2

44

827

50

Cam

eroo

n20

——

89—

—7

516

137

Nige

ria11

3—

235

26—

517

74

461

Wes

tern

/Cen

tralA

frica

313

—32

426

—12

182

2059

8

Afric

a34

936

34—

328

2628

8019

530

4670

Indi

a1

285

——

——

——

—28

6

Indo

nesia

—1

108

1—

——

——

110

Sout

h-Ea

stA

sia1

286

108

1—

——

——

396

Over

all35

039

2010

832

926

2880

195

3050

66

aOt

hers

ubty

pesw

ithfe

wer

than

10sp

ecim

ense

ach:

B,C

RF09

,CRF

11,C

RF13

,F,H

and

K.

Tabl

e 12

Dist

ribut

ion

of su

btyp

es b

y co

untr

y am

ong

peop

le in

itiat

ing

antir

etro

vira

l the

rapy

in W

HO a

cqui

red

drug

resis

tanc

e su

rvey

s

74

WHO

Reg

ion

Subr

egio

nSu

rvey

nAZ

Td4

TFT

C/3T

CTD

FAB

Cdd

IAn

y NR

TINV

PEF

VET

RRP

VAn

y NN

RTI

NRTI

and

NNR

TIAn

y PI

Any

drug

Afric

aEa

ster

nBu

rund

i–12

007

12—

—16

.7%—

16.7%

—16

.7%25

.0%

25.0

%16

.7%16

.7%25

.0%

16.7%

8.3%

25.0

%

Afric

aEa

ster

nBu

rund

i–22

007

185.

6%5.

6%27

.8%

—27

.8%

11.1%

27.8

%27

.8%

27.8

%11.

1%16

.7%27

.8%

27.8

%5.

6%27

.8%

Afric

aEa

ster

nKe

nya–

1200

78

——

75.0

%—

75.0

%12

.5%

75.0

%10

0.0%

100.

0%25

.0%

25.0

%10

0.0%

75.0

%—

100.

0%

Afric

aEa

ster

nKe

nya–

220

0812

8.3%

8.3%

25.0

%—

33.3%

16.7%

33.3%

33.3%

33.3%

8.3%

8.3%

33.3%

25.0

%—

41.7%

Afric

aEa

ster

nM

alaw

i–12

006

3—

—66

.7%—

66.7%

33.3%

66.7%

66.7%

66.7%

33.3%

33.3%

66.7%

66.7%

—66

.7%

Afric

aEa

ster

nM

alaw

i–22

006

6—

—33

.3%—

33.3%

16.7%

33.3%

83.3%

83.3%

83.3%

83.3%

83.3%

33.3%

—83

.3%

Afric

aEa

ster

nM

alaw

i–32

006

3—

33.3%

100.

0%33

.3%10

0.0%

33.3%

100.

0%10

0.0%

100.

0%66

.7%66

.7%10

0.0%

100.

0%—

100.

0%

Afric

aEa

ster

nM

alaw

i–42

006

3—

—66

.7%—

66.7%

—66

.7%66

.7%66

.7%66

.7%66

.7%66

.7%66

.7%—

66.7%

Afric

aEa

ster

nM

alaw

i–12

008

911.

1%11.

1%77

.8%

11.1%

77.8

%33

.3%77

.8%

66.7%

66.7%

44.4

%44

.4%

66.7%

66.7%

—77

.8%

Afric

aEa

ster

nM

alaw

i–22

008

6—

—50

.0%

—50

.0%

—50

.0%

83.3%

83.3%

33.3%

33.3%

83.3%

50.0

%—

83.3%

Afric

aEa

ster

nM

alaw

i–32

008

5—

20.0

%60

.0%

20.0

%60

.0%

20.0

%60

.0%

60.0

%60

.0%

——

60.0

%60

.0%

20.0

%60

.0%

Afric

aEa

ster

nM

alaw

i–42

008

8—

25.0

%75

.0%

25.0

%75

.0%

62.5

%75

.0%

100.

0%10

0.0%

75.0

%75

.0%

100.

0%75

.0%

—10

0.0%

Afric

aEa

ster

nM

ozam

biqu

e–12

007

911.

1%22

.2%10

0.0%

22.2%

100.

0%44

.4%

100.

0%10

0.0%

100.

0%88

.9%

88.9

%10

0.0%

100.

0%—

100.

0%

East

ern

Afric

a10

23.9

%8.

8%52

.0%

6.9%

52.9

%20

.6%

52.9

%61

.8%

61.8

%36

.3%37

.3%61

.8%

51.0

%2.9

%63

.7%

Afric

aSo

uthe

rnSo

uth

Afric

a–12

007

128.

3%16

.7%58

.3%16

.7%58

.3%25

.0%

58.3%

58.3%

58.3%

25.0

%25

.0%

58.3%

58.3%

41.7%

58.3%

Afric

aSo

uthe

rnSo

uth

Afric

a–2

2007

14—

21.4

%57

.1%14

.3%57

.1%35

.7%57

.1%71

.4%

71.4

%14

.3%21

.4%

71.4

%57

.1%14

.3%71

.4%

Afric

aSo

uthe

rnSo

uth

Afric

a–3

2007

2—

—50

.0%

—50

.0%

50.0

%50

.0%

50.0

%50

.0%

——

50.0

%50

.0%

—50

.0%

Afric

aSo

uthe

rnSw

azila

nd–1

2008

714

.3%28

.6%

71.4

%28

.6%

71.4

%28

.6%

71.4

%85

.7%85

.7%—

—85

.7%71

.4%

14.3%

85.7%

Afric

aSo

uthe

rnSw

azila

nd–2

200

85

60.0

%60

.0%

80.0

%40

.0%

80.0

%80

.0%

80.0

%80

.0%

80.0

%40

.0%

40.0

%80

.0%

80.0

%—

80.0

%

Afric

aSo

uthe

rnZa

mbi

a–12

007

8—

50.0

%50

.0%

37.5

%50

.0%

50.0

%50

.0%

62.5

%62

.5%

50.0

%62

.5%

62.5

%50

.0%

12.5

%62

.5%

Afric

aSo

uthe

rnZa

mbi

a–2

2007

185.

6%33

.3%72

.2%27

.8%

72.2%

44.4

%72

.2%83

.3%83

.3%72

.2%72

.2%83

.3%72

.2%27

.8%

83.3%

Afric

aSo

uthe

rnZa

mbi

a–3

2007

9—

22.2%

55.6

%22

.2%55

.6%

22.2%

55.6

%55

.6%

55.6

%33

.3%33

.3%55

.6%

44.4

%33

.3%66

.7%

Afric

aSo

uthe

rnZi

mba

bwe–

1200

815

26.7%

20.0

%73

.3%13

.3%73

.3%26

.7%73

.3%60

.0%

60.0

%40

.0%

46.7%

60.0

%53

.3%13

.3%80

.0%

Sout

hern

Afri

ca90

11.1%

27.8

%64

.4%

22.2%

64.4

%36

.7%64

.4%

68.9

%68

.9%

36.7%

40.0

%68

.9%

60.0

%21

.1%73

.3%

Afric

aW

este

rn/C

entra

lCa

mer

oon–

1200

96

——

66.7%

—66

.7%33

.3%66

.7%66

.7%66

.7%33

.3%33

.3%66

.7%66

.7%—

66.7%

Afric

aW

este

rn/C

entra

lNi

geria

–120

0811

9.1%

18.2%

36.4

%—

36.4

%18

.2%45

.5%

54.5

%54

.5%

27.3%

27.3%

54.5

%45

.5%

72.7%

54.5

%

Afric

aW

este

rn/C

entra

lNi

geria

–22

008

1323

.1%38

.5%

84.6

%30

.8%

84.6

%46

.2%84

.6%

100.

0%10

0.0%

76.9

%76

.9%

100.

0%84

.6%

7.7%

100.

0%

Afric

aW

este

rn/C

entra

lNi

geria

–32

008

175.9

%11.

8%64

.7%11.

8%64

.7%29

.4%

64.7%

70.6

%70

.6%

64.7%

64.7%

70.6

%64

.7%5.9

%70

.6%

Wes

tern

/Cen

tralA

frica

4710

.6%

19.1%

63.8

%12

.8%

63.8

%31

.9%

66.0

%74

.5%

74.5

%55

.3%55

.3%74

.5%

66.0

%21

.3%74

.5%

Afric

a23

97.9

%18

.0%

59.0

%13

.8%

59.4

%28

.9%

59.8

%66

.9%

66.9

%40

.2%41

.8%

66.9

%57

.3%13

.4%

69.5

%

Sout

h-Ea

stA

siaIn

dia–

1200

710

30.0

%30

.0%

80.0

%20

.0%

80.0

%30

.0%

90.0

%80

.0%

80.0

%40

.0%

50.0

%80

.0%

80.0

%—

90.0

%

Sout

h-Ea

stA

siaIn

dia–

220

0813

23.1%

23.1%

69.2%

15.4

%69

.2%30

.8%

69.2%

92.3%

92.3%

53.8

%53

.8%

92.3%

69.2%

7.7%

92.3%

Sout

h-Ea

stA

siaIn

done

sia–1

2008

742

.9%

57.1%

100.

0%28

.6%

100.

0%57

.1%10

0.0%

100.

0%10

0.0%

85.7%

100.

0%10

0.0%

100.

0%—

100.

0%

Sout

h-Ea

stA

sia30

30.0

%33

.3%80

.0%

20.0

%80

.0%

36.7%

83.3%

90.0

%90

.0%

56.7%

63.3%

90.0

%80

.0%

3.3%

93.3%

Over

all26

910

.4%

19.7%

61.3%

14.5

%61

.7%29

.7%62

.5%

69.5

%69

.5%

42.0

%44

.2%69

.5%

59.9

%12

.3%72

.1%

Tabl

e 13

Pre

dict

ed d

rug

resis

tanc

e am

ong

peop

le e

xper

ienc

ing

failu

re o

f firs

t-lin

e an

tiret

rovi

ral t

hera

py a

t 12

mon

ths,

by in

divi

dual

clin

ic


75

WHO

Re

gion

Subr

egio

nSu

rvey

nK6

5RD6

7NK7

0RM

184I

VT2

15

any

K219

an

y

Any

NRTI

dr

ug

resi

stan

ce

mut

atio

nAn

y TA

MK1

01E

K103

NSV1

06AM

Y181

any

Y188

an

yG1

90

any

Any

NNRT

I dr

ug

resi

stan

ce

mut

atio

n

NRTI

an

d NN

RTI

Any

drug

re

sist

ance

m

utat

ion

Afric

aEa

ster

nBu

rund

i–12

007

12—

——

16.7%

——

16.7%

——

16.7%

—16

.7%—

—25

.0%

16.7%

25.0

%

Afric

aEa

ster

nBu

rund

i–22

007

18—

5.6%

5.6%

27.8

%—

—27

.8%

5.6%

5.6%

11.1%

5.6%

5.6%

5.6%

5.6%

27.8

%27

.8%

27.8

%

Afric

aEa

ster

nKe

nya–

1200

712

——

—25

.0%

——

33.3%

8.3%

—33

.3%—

——

—33

.3%25

.0%

41.7%

Afric

aEa

ster

nKe

nya–

220

088

——

—75

.0%

——

75.0

%12

.5%

—62

.5%

12.5

%25

.0%

—12

.5%

100.

0%75

.0%

100.

0%

Afric

aEa

ster

nM

alaw

i–12

006

3—

——

66.7%

——

66.7%

33.3%

——

33.3%

33.3%

—66

.7%66

.7%66

.7%66

.7%

Afric

aEa

ster

nM

alaw

i–22

006

6—

——

33.3%

——

33.3%

—16

.7%16

.7%16

.7%66

.7%16

.7%16

.7%83

.3%33

.3%83

.3%

Afric

aEa

ster

nM

alaw

i–32

006

333

.3%—

—10

0.0%

——

100.

0%—

—33

.3%—

66.7%

—33

.3%10

0.0%

100.

0%10

0.0%

Afric

aEa

ster

nM

alaw

i–42

006

3—

——

66.7%

——

66.7%

——

——

66.7%

——

66.7%

66.7%

66.7%

Afric

aEa

ster

nM

alaw

i–12

008

911.

1%—

—77

.8%

11.1%

11.1%

77.8

%11.

1%11.

1%33

.3%—

44.4

%—

11.1%

66.7%

66.7%

77.8

%

Afric

aEa

ster

nM

alaw

i–22

008

6—

——

50.0

%—

—50

.0%

——

50.0

%—

33.3%

——

83.3%

50.0

%83

.3%

Afric

aEa

ster

nM

alaw

i–32

008

520

.0%

——

60.0

%—

—60

.0%

——

60.0

%20

.0%

——

—60

.0%

60.0

%60

.0%

Afric

aEa

ster

nM

alaw

i–42

008

825

.0%

37.5

%—

62.5

%—

—75

.0%

37.5

%—

50.0

%12

.5%

75.0

%12

.5%

12.5

%10

0.0%

75.0

%10

0.0%

Afric

aEa

ster

nM

ozam

biqu

e–12

007

922

.2%—

11.1%

88.9

%—

11.1%

100.

0%22

.2%11.

1%22

.2%22

.2%77

.8%

11.1%

22.2%

100.

0%10

0.0%

100.

0%

East

ern

Afric

a10

26.

9%3.9

%2.0

%50

.0%

1.0%

2.0%

52.9

%9.

8%3.9

%29

.4%

7.8%

32.4

%3.9

%9.

8%61

.8%

51.0

%63

.7%

Afric

aSo

uthe

rnSo

uth

Afric

a–12

007

1216

.7%8.

3%—

41.7%

8.3%

8.3%

58.3%

16.7%

8.3%

33.3%

16.7%

16.7%

—16

.7%58

.3%58

.3%58

.3%

Afric

aSo

uthe

rnSo

uth

Afric

a–2

2007

1414

.3%—

—57

.1%—

—57

.1%—

7.1%

28.6

%35

.7%7.1

%7.1

%28

.6%

71.4

%57

.1%71

.4%

Afric

aSo

uthe

rnSo

uth

Afric

a–3

2007

2—

——

50.0

%—

—50

.0%

——

—50

.0%

——

—50

.0%

50.0

%50

.0%

Afric

aSo

uthe

rnSw

azila

nd–1

2008

714

.3%14

.3%14

.3%71

.4%

14.3%

14.3%

71.4

%14

.3%—

57.1%

14.3%

—28

.6%

—85

.7%71

.4%

85.7%

Afric

aSo

uthe

rnSw

azila

nd–2

200

85

—20

.0%

20.0

%80

.0%

60.0

%20

.0%

80.0

%60

.0%

—60

.0%

—40

.0%

—20

.0%

80.0

%80

.0%

80.0

%

Afric

aSo

uthe

rnZa

mbi

a–12

007

837

.5%

——

25.0

%—

—50

.0%

—12

.5%

25.0

%12

.5%

37.5

%12

.5%

12.5

%62

.5%

50.0

%62

.5%

Afric

aSo

uthe

rnZa

mbi

a–2

2007

1827

.8%

—5.

6%72

.2%—

5.6%

72.2%

11.1%

22.2%

27.8

%16

.7%44

.4%

—33

.3%83

.3%72

.2%83

.3%

Afric

aSo

uthe

rnZa

mbi

a–3

2007

922

.2%—

—55

.6%

——

55.6

%—

—11.

1%33

.3%11.

1%—

33.3%

55.6

%44

.4%

66.7%

Afric

aSo

uthe

rnZi

mba

bwe–

1200

815

6.7%

20.0

%26

.7%73

.3%6.

7%6.

7%73

.3%33

.3%26

.7%6.

7%6.

7%6.

7%20

.0%

26.7%

60.0

%53

.3%80

.0%

Sout

hern

Afri

ca90

17.8%

6.7%

7.8%

60.0

%6.

7%5.

6%64

.4%

14.4

%12

.2%26

.7%18

.9%

20.0

%7.8

%23

.3%68

.9%

60.0

%73

.3%

Afric

aW

este

rn/

Cent

ral

Cam

eroo

n–12

009

6—

——

66.7%

16.7%

—66

.7%16

.7%16

.7%50

.0%

——

16.7%

16.7%

66.7%

66.7%

66.7%

Afric

aW

este

rn/

Cent

ral

Nige

ria–1

2008

11—

—9.1

%36

.4%

—18

.2%45

.5%

18.2%

—27

.3%—

27.3%

—9.1

%54

.5%

45.5

%54

.5%

Afric

aW

este

rn/

Cent

ral

Nige

ria–2

200

813

23.1%

15.4

%23

.1%84

.6%

7.7%

7.7%

84.6

%30

.8%

15.4

%30

.8%

7.7%

46.2%

7.7%

23.1%

100.

0%84

.6%

100.

0%

Afric

aW

este

rn/

Cent

ral

Nige

ria–3

200

817

5.9%

5.9%

—64

.7%5.9

%—

64.7%

11.8%

17.6%

17.6%

—41

.2%5.9

%17.

6%64

.7%64

.7%64

.7%

Wes

tern

/Cen

tralA

frica

478.

5%6.

4%8.

5%63

.8%

6.4%

6.4%

66.0

%19

.1%12

.8%

27.7%

2.1%

34.0

%6.

4%17.

0%72

.3%66

.0%

72.3%

Afric

a23

911.

3%5.4

%5.4

%56

.5%

4.2%

4.2%

59.8

%13

.4%

8.8%

28.0

%10

.9%

28.0

%5.9

%16

.3%66

.5%

57.3%

69.0

%

Tabl

e 14

Pre

vale

nce

of d

rug

resis

tanc

e–as

soci

ated

mut

atio

ns a

t tre

atm

ent f

ailu

re (1

2 m

onth

s), b

y in

divi

dual

clin

ic

76

WHO

Re

gion

Subr

egio

nSu

rvey

nK6

5RD6

7NK7

0RM

184I

VT2

15

any

K219

an

y

Any

NRTI

dr

ug

resi

stan

ce

mut

atio

nAn

y TA

MK1

01E

K103

NSV1

06AM

Y181

any

Y188

an

yG1

90

any

Any

NNRT

I dr

ug

resi

stan

ce

mut

atio

n

NRTI

an

d NN

RTI

Any

drug

re

sist

ance

m

utat

ion

Sout

h–Ea

stA

siaIn

dia–

1200

710

—20

.0%

10.0

%80

.0%

20.0

%10

.0%

90.0

%40

.0%

20.0

%30

.0%

—20

.0%

30.0

%30

.0%

80.0

%80

.0%

90.0

%

Sout

h-Ea

stA

siaIn

dia–

220

0813

—23

.1%15

.4%

69.2%

15.4

%7.7

%69

.2%23

.1%15

.4%

53.8

%15

.4%

38.5

%—

30.8

%92

.3%69

.2%92

.3%

Sout

h-Ea

stA

siaIn

done

sia–1

2008

714

.3%14

.3%28

.6%

85.7%

14.3%

14.3%

100.

0%42

.9%

—14

.3%—

71.4

%14

.3%14

.3%10

0.0%

100.

0%10

0.0%

Sout

h-Ea

stA

sia30

3.3%

20.0

%16

.7%76

.7%16

.7%10

.0%

83.3%

33.3%

13.3%

36.7%

6.7%

40.0

%13

.3%26

.7%90

.0%

80.0

%93

.3%

Tota

l26

910

.4%

7.1%

6.7%

58.7%

5.6%

4.8%

62.5

%15

.6%

9.3%

29.0

%10

.4%

29.4

%6.

7%17.

5%69

.1%59

.9%

71.7%

Note

:Spe

cimen

swith

PR

only

geno

type

sava

ilabl

ear

eex

clude

dfo

rthe

tabl

e.

Tabl

e 14

Pre

vale

nce

of d

rug

resis

tanc

e–as

soci

ated

mut

atio

ns a

t tre

atm

ent f

ailu

re (1

2 m

onth

s), b

y in

divi

dual

clin

ic (c

ont.)


77

0%

10%

20

%

30%

40

%

50%

60

%

70%

80

%

90%

10

0%

Burundi-1 20

07

Burundi-2 20

07

Kenya-1

2007

Kenya-2

2008

Malawi-1 20

06

Malawi-2 20

06

Malawi-3 20

06

Malawi-4 20

06

Malawi-1 20

08

Malawi-2 20

08

Malawi-3 20

08

Malawi-4 20

08

Mozambique-1

2007

Eastern Afri

ca

South Africa-1

2007

South Africa-2

2007

South Africa-3

2007

Swaziland-1

2008

Swaziland-2

2008

Zambia-1 20

07

Zambia-2 20

07

Zambia-3 20

07

Zimbabwe-1

2008

Southern Africa

Cameroon-1 20

09

Nigeria-1

2008

Nigeria-2

2008

Nigeria-3

2008

Weste

rn/Central A

frica India-1

2007

India-2 20

08

Indonesia-1

2008

South-East Asia

Ove

rall VL

> 1

000

c/m

l, G

T fa

il

VL >

100

0 c/

ml a

nd w

ild ty

pe

Stop

ped

antir

etro

vira

l the

rapy

Lo

st to

follo

w-u

p

Percentage of HIVDR possible

Figu

re 3

Bre

akdo

wn

of p

ossib

le H

IV d

rug

resis

tanc

e ou

tcom

e, b

y in

divi

dual

clin

ic

78

0%

10%

20

%

30%

40

%

50%

60

%

70%

80

%

90%

10

0%

Burundi-1 20

07

Burundi-2 20

07

Kenya-1

2007

Kenya-2

2008

Malawi-1 20

06

Malawi-2 20

06

Malawi-3 20

06

Malawi-4 20

06

Malawi-1 20

08

Malawi-2 20

08

Malawi-3 20

08

Malawi-4 20

08

Mozambique-1

2007

Eastern Afri

ca

South Africa-1

2007

South Africa-2

2007

South Africa-3

2007

Swaziland-1

2008

Swaziland-2

2008

Zambia-1 20

07

Zambia-2 20

07

Zambia-3 20

07

Zimbabwe-1

2008

Southern Africa

Cameroon-1 20

09

Nigeria-1

2008

Nigeria-2

2008

Nigeria-3

2008

Weste

rn/Central A

frica India-1

2007

India-2 20

08

Indonesia-1

2008

South-East Asia

Ove

rall

HIV

DR

poss

ible

A

ny H

IVD

R de

tect

ed

HIV

DR

prev

entio

n

Figu

re 4

The

thre

e ou

tcom

es o

f sur

veys

of a

cqui

red

HIV

drug

resis

tanc

e, b

y an

tiret

rovi

ral t

hera

py cl

inic

HIV/AIDS Programme

For more information, contact:

World Health OrganizationDepartment of HIV/AIDS20, avenue Appia1211 Geneva 27Switzerland

E-mail: [email protected]

http://www.who.int/hiv/en/

ISBN 978 92 4 150393 8


v.2/HIVDR 2003-2010- 23 July - WHO1.Anti-HIV agents – therapeutic use. 2.Drug resistance, Viral...

Documents

Transcript of v.2/HIVDR 2003-2010- 23 July - WHO1.Anti-HIV agents – therapeutic use. 2.Drug resistance, Viral...