Development of nevirapine resistance in children exposed to the prevention of mother-to-child HIV-1 transmission programme in Maputo, Mozambique

RESEARCH ARTICLE De elopmen o Ne i apine Resis ance in Child en Exposed o he P e en ion o Mo he - o-Child HIV-1 T ansmission P og amme in Mapu o, Mozambique F ancisco An unes 1 *, Pe ei a Zindoga 2 , Pe pé ua Gomes 3‡ , O alho Augus o 4‡ , Isabel Mahumane 5‡ , Luís Veloso 6 , Emília Valadas 7 , Rica do Camacho 8‡ 1Ins i u o de Saúde Ambien al (ISAMB), Faculdade de Medicina da Uni e sidade de Lisboa, Lisboa, Po ugal, 2Depa amen o de Medicinas, Ins i u o Nacional de Saúde, Hospi al Cen al de Mapu o, Mapu o, Mozambique, 3Labo a ó io de Mic obiologia Clínica e Biologia Molecula , Se iço de Pa ologia Clínica, Cen o Hospi ala Lisboa Ociden al, Lisboa, Po ugal, 4Saúde da Comunidade, Faculdade de Medicina da Uni e sidade Edua do Mondlane, Mapu o, Mozambique, 5Pla a o mas Tecnológicas, Pla a o ma de Vi ologia Molecula , Ins i u o Nacional de Saúde, Mapu o, Mozambique, 6Clinical Da a Uni , Eu o ials Scien i ic Consul an s, Lisboa, Po ugal, 7Clínica Uni e si á ia de Doenças In ecciosas e Pa asi á ias, Faculdade de Medicina de Lisboa, Lisboa, Po ugal, 8Depa men o Mic obiology and Immunology, Rega Ins i u e o Medical Resea ch, Leu en, Belgium ‡PG and RC con ibu ed equally o his wok. OA and IM also con ibu ed equally o his wo k. * an [email protected] Abs ac Backg ound Single-dose ne i apine (sd-NVP) has been he main op ion o p e en ion o mo he - o-child ansmission (PMTCT) o HIV-1 in low- esou ce se ings. Howe e , sd-NVP can induce he selec ion o HIV-1 esis an mu a ions in mo he s and in an s. In Mozambique, he e a e lim- i ed da a ega ding he p o ile o NVP esis ance associa ed mu a ions (RAM) in he con ex o PMTCT. Objec i es To assess he p e alence and he ac o s associa ed wi h NVP RAM among child en bo n o HIV-1 in ec ed mo he s en olled in he PMTCT p og amme adop ed in Mozambique. Me hods One hund ed and i y se en child en aged 6 o 48 weeks we e sequen ially included (July 2011 o Ma ch 2012) a ou cen es in Mapu o. Geno yping o RAM was pe o med in sam- ples wi h HIV-1 RNA100 copies/μL (Vi oseq). Sequencing was pe o med wi h ABI 3100 (Applied Biosys ems). Logis ic eg ession modelling was unde aken o iden i y he ac o s associa ed wi h NVP RAM. PLOS ONE | DOI:10.1371/jou nal.pone.0131994 July 10, 2015 1/12 OPEN ACCESS Ci a ion: An unes F, Zindoga P, Gomes P, Augus o O, Mahumane I, Veloso L, e al. (2015) De elopmen o Ne i apine Resis ance in Child en Exposed o he P e en ion o Mo he - o-Child HIV-1 T ansmission P og amme in Mapu o, Mozambique. PLoS ONE 10 (7): e0131994. doi:10.1371/jou nal.pone.0131994 Edi o : So en Gan , Uni e si y o B i ish Columbia, CANADA Recei ed: Ma ch 30, 2015 Accep ed: June 9, 2015 Published: July 10, 2015 Copy igh : © 2015 An unes e al. This is an open access a icle dis ibu ed unde he e ms o he C ea i e Commons A ibu ion License, which pe mi s un es ic ed use, dis ibu ion, and ep oduc ion in any medium, p o ided he o iginal au ho and sou ce a e c edi ed. Da a A ailabili y S a emen : All ele an da a a e wi hin he pape and i s Suppo ing In o ma ion iles. Funding: This s udy was unded by he Associação pa a a In es igação e Desen ol imen o da Faculdade de Medicina (AIDFM). The unde had no ole in s udy design, da a collec ion and analysis, decision o publish, o p epa a ion o he manusc ip . Compe ing In e es s: The co-au ho Luís Veloso is an employee o Eu o ials, Scien i ic Consul an s. This company ecei ed hono a ia om he unde o his p ojec , o suppo in medical w i ing ac i i ies. Resul s Se en y-nine child en had hei samples geno yped. Thei median age was 7.0 (3–12) mon hs and 92.4% ecei ed p ophylaxis wi h sd-NVP a bi h plus daily NVP. 35.4% o mo he s ecei ed an i e o i als (ARVs) o PMTCT. ARV RAM we e de ec ed in 43 (54.4%) o he child en. 45.6% o hese child en had a leas one NVP RAM. The mos common mu a ions associa ed wi h NVP esis ance we e K103N (n = 16) and Y181C (n = 15). NVP RAM was signi ican ly associa ed wi h mo he exposu e o PMTCT (c ude odds a io [OR] 30.3, 95% CI 4.93–186.34) and wi h mo he ’s CD4 coun <350 cells/mm 3 (c ude OR 3.08, 95% CI 1.02–9.32). In he mul i a iable analysis he mo he ’s exposu e o PMTCT was he only a iable signi ican ly associa ed wi h NVP RAM (adjus ed OR 48.65, 95% CI 9.33– 253.66). Conclusions We ound a high p e alence o NVP RAM among child en who we e exposed o he d ug egimen o PMTCT in Mozambique. The mo he s’exposu e o PMTCT signi ican ly inc eased he isk o NVP RAM. In oduc ion In 2010, an es ima ed 390 000 child en we e newly in ec ed wi h HIV, p ima ily due o mo he - o-child ansmission (MTCT) [1]. Ve ical ansmission, which can occu du ing p egnancy, labou , deli e y o b eas eeding, emains he main mode o HIV in ec ion in chil- d en [2]. The implemen a ion o p e en i e heal h policies such as p egnancy moni o ing and he adminis a ion o an i e o i al he apies (ART) led o a ma ked educ ion in MTCT a es in high-income coun ies [3]. Howe e , in limi ed- esou ce se ings, whe e ull access o ART has no been achie ed, MTCT a es emain ela i ely high [4]. Two o he s a egic goals o ‘The Global Plan owa ds he elimina ion o new HIV in ec- ions among child en by 2015 and keeping hei mo he s ali e’a e o achie e an ART co e age among p egnan women o 90% and o educe MTCT o a es <5%, in low- o medium- income coun ies wi h high es ima ed p e alence o p egnan women li ing wi h HIV [5]. In 2012, he co e age o ART p og ammes o p e en ion o MTCT (PMTCT) eached 65% ( ange: 57 o 70%), among he sub-Saha an A ican coun ies [4]. Mozambique is one o he nine coun ies in sub-Saha an A ica wi h a p e alence o HIV in ec ion abo e 10% [6]. In 2009, he co e age o ART in he capi al ci y o Mapu o eached 77%. The PMTCT o HIV is a p io i y o he Mozambican Go e nmen , which has imple- men ed a na ional scale up plan owa ds he elimina ion o MTCT up o 2015. Howe e , chil- d en’s access o PMTCT p og ammes in Mozambique is s ill limi ed (~59%) [7]. Single-dose ne i apine (sd-NVP), a non-nucleoside e e se ansc ip ase inhibi o (NNRTI), was adop ed by se e al A ican coun ies in he PMTCT due o i s good e icacy, low cos , ease o adminis a ion and long plasma hal -li e [8–10]. Howe e , sd-NVP can induce he selec ion o HIV-1 esis an mu a ions in mo he s and in an s. A ecen me a-analysis p o ided a pooled es ima e o NVP esis ance p e alence o 35.7% and 52.6% in women and child en who used sd-NVP, espec i ely [11]. Some au ho s sugges ha NVP-induced esis ance a e exposu e o MTCT p ophylac ic p og ammes may limi i ological esponse o subsequen use o NNRTIs in mo he s and in an s [3,12,13]. Ne i apine Resis ance in MTC HIV-1 T ansmission PLOS ONE | DOI:10.1371/jou nal.pone.0131994 July 10, 2015 2/12 This does no al e he au ho 's adhe ence o all he PLOS ONE policies on sha ing da a and ma e ials. Besides he s anda d medical w i ing ac i i ies, LV has gi en ele an con ibu ion o he da a analysis and in e p e a ion, decision-making p ocess, e iew and inal app o al o he manusc ip . Conside ing ha LV mee s he au ho ship c i e ia s ipula ed by ICMJE, all o he au ho s ag ee o include him as co-au ho o his manusc ip . The o he au ho s ha e decla ed ha no compe ing in e es s exis . The p ophylac ic egimen adop ed by Mozambique ollows he Op ion A p econized by he Wo ld Heal h O ganiza ion (WHO) 2009 Guidelines [14]. In in an s, his op ion consis o sd- NVP a bi h plus NVP p ophylaxis up o he end o b eas eeding. Mo he s ecei e zido udine (AZT) an epa um, sd-NVP a he onse o labou and wice daily AZT + lami udine (3TC) o 7 days pos pa um (sd-NVP and AZT + 3TC can be omi ed i AZT an epa um did no exceed 4 weeks). Su eys conduc ed in Mapu o in 2007 and 2009 e ealed a es o ansmi ed d ug esis ance o all ART classes below 5% [15]. Thus a , he e is limi ed knowledge ega ding he p e alence o NVP i al esis ance mu a- ions in child en bo n o HIV-1-in ec ed mo he s and who we e exposed o ma e nal o child p og ams o PMTCT in Mozambique. This knowledge is essen ial o ailo ARV s a egies in his popula ion and maximize he e ec i eness o ARV egimens in p e en ing MTCT. In ou wo k, we es ima ed he p e alence o i al esis ance o NVP among child en exposed o he PMTCT p og amme adop ed in Mozambique. We also explo ed he mos impo an ac o s associa ed wi h NVP esis ance. Me hods Se ing and popula ion This c oss-sec ional s udy was ca ied ou be ween July 2011 and Ma ch 2012. Child en aged 6 o 48 weeks included in he Mozambican PMTCT p og am and bo n o HIV-1 in ec ed mo h- e s we e sequen ially en olled in he s udy acco ding o hei scheduled appoin men a ou paedia ic HIV cen es in Mapu o’s egion. The s udy included mo he s who needed ART o ea hei own heal h o mo he s who jus needed ARVs o educe he isk o MTCT o HIV. The PMTCT p o ocol ollowed in Mozambique a he ime o his s udy was based on WHO Rapid Ad ice—Use o an i e o i al d ugs o ea ing p egnan women and p e en ing HIV in ec ion in in an s– e sion 2 [14]. HIV-1 in ec ed p egnan women who we e eligible o ART (p ima ily h ough CD4 sc eening) s a ed his ea men , i espec i e o he ges a ional age, and con inued h oughou p egnancy, deli e y and he ea e . The i s -line ART egimen included AZT + 3TC + NVP. The ma e nal ARV p ophylaxis consis ed o an epa um daily AZT as ea ly as 14 weeks o ges a ion, sd-NVP a onse o labou and wice daily AZT + 3TC o 7 days pos pa um. The p ophylaxis in b eas eeding child en consis ed o sd-NVP a bi h and hen daily adminis a ion o NVP un il one week a e all exposu e o b eas milk ended. Non-b eas eed- ing child en ecei ed sd-NVP a bi h and hen daily adminis a ion o NVP o AZT un il 4 o 6 weeks o age. HIV-in ec ed in an s ecei ed one o he ollowing ART egimens: AZT o d4T (s a udine) + 3TC + NVP o AZT o d4T + 3TC + LPV/ (lopina i / i ona i ). Child en wi h- ou CD4 coun a ailable and ha had ecei ed ART ou o he PMTCT p og amme we e excluded om he s udy. W i en in o med consen was ob ained om all child en’s pa en s o legal gua dians. The s udy was app o ed by Mozambique’s Na ional Heal h Bioe hics Commi ee and by he Mozambican Minis y o Heal h. Da a and specimen collec ion Socio-demog aphic and clinical da a we e ob ained om exis ing medical eco ds. The mo he - ela ed da a included he WHO clinical s age o disease, ype o PMTCT app oach (ART o p ophylaxis), p ophylac ic scheme and du a ion o exposu e o ARVs. The mo he ’s i al load was no assessed a he ime o labou (is no ou ine p ac ice). The child- ela ed a i- ables we e gende , age, deli e y mode, bi h weigh , eeding p ac ice, CD4 coun , p ophylac ic egimen and HIV-1 RNA le els. Ne i apine Resis ance in MTC HIV-1 T ansmission PLOS ONE | DOI:10.1371/jou nal.pone.0131994 July 10, 2015 3/12 Two samples o pe iphe al blood (5 ml each) we e collec ed om each child by enipunc- u e. One sample was used o ou ine analysis and he o he was shipped wi hin 5 hou s o he Immunology Labo a o y o Heal h Na ional Ins i u e in Mozambique. The e, 200 μL we e used o de e mine he CD4 coun by low cy ome y using FACSCalibu (Bec on-Dickinson, San Jose, Cali o nia, USA); he emaining sample was cen i uged, plasma was isola ed and s o ed a -80°C. This sample was shipped o Hospi al Egas Moniz in Lisbon wi hin 12 mon hs a e i s collec ion, o geno yping o HIV-1 RNA and de ec ion o ARV esis ance associa ed mu a- ions (RAM). The shipmen was pe o med in leak-p oo con aine s speci ic o ozen samples and con aining d y ice. Tempe a u e was checked a a i al o labo a o y and no anomalies we e de ec ed. HIV-1 RNA geno yping was pe o med by means o a comme cially a ailable ki (Vi oSeq HIV-1 geno yping sys em; Cele a Diagnos ics/Abbo Molecula Diagnos ics, Rome, I aly) based on RNA ex ac ion, e o ansc ip ion, DNA ampli ica ion, and sequencing by an au oma ed sequence (ABI 3100; Applied Biosys ems, Fos e Ci y, CA) [16]. Geno yping o ARV RAM was only ca ied ou in samples wi h HIV-1 RNA le els 100 copies/μL, using he S an o d Resis ance Da abase (HIVdb e sion 6.05) [17]. A NVP RAM was conside ed i a leas one o he ollowing gene ic changes in e e se an- sc ip ase was de ec ed: A98G, K101E, K101H, K101P, K103N, V106A, V106M, V108I, Y181C, Y181V, Y188L and G190A. HIV sub ypes we e de e mined by phylogene ic analysis o pol egion sequences, using he Rega Sub yping Tool 2.0 [18]. S a is ical Analysis The sample size was calcula ed based on he popula ion o Mapu o, 1.209.993 pe sons, and a a e o ansmission o HIV d ug esis ance o all ele an ARVs and ARV classes o 4.0%, obse ed in Malawi [19], a neighbou ing coun y o Mozambique wi h simila cha ac e is ics (low income, high illi e acy a es, poo sani a ion and high a e o sexual ac i i y). A sample o 120 child en allowed he es ima ion o p e alence o NVP RAM wi h a 95% con idence in e al (CI) and a ma gin o e o o 3.5%. O e all, 157 child en we e en olled o accoun o specimen losses abo e 20%. Quan i a i e a iables we e summa ised as median, minimum and maximum and quali a- i e a iables we e summa ised as absolu e equency. CIs we e compu ed o he mos p e a- len mu a ions (>5 cases). Odds a io (OR) and 95%CI we e used o explo e he associa ion be ween independen a iables and NVP RAM. The bi a iable analysis was conduc ed o all mo he and child cha ac e is ics po en ially associa ed wi h NVP RAM. All a iables wi h a p alue 0.20 in he bi a iable analysis, as well as child’s gende and age, we e included in he mul i a iable logis ic eg ession analysis. Adjus ed ORs and 95% CIs we e es ima ed o all independen p edic o s in he inal selec ed logis ic model. All P alues we e wo-sided and he le el o signi icance was se a p<0.05. S a is ical analysis was pe o med using he S a is ical Package o he Social Sciences, e - sion 19.0 (SPSS Inc., Chicago, IL, USA). Resul s Gene al cha ac e is ics o he eligible popula ion O e all, 157 child en me he eligibili y c i e ia. The median age o he child en was 7.0 mon hs (1–12 mon hs) and 90 (57.3%) we e male. Mos child en (75.8%; 119 o 157) had CD4 pe cen - age >20%. Se en y-one pe cen o mo he s (111 o 157) we e clinical s age 3 o HIV in ec ion and 127 (80.9%) we e ecei ing ART. Ne i apine Resis ance in MTC HIV-1 T ansmission PLOS ONE | DOI:10.1371/jou nal.pone.0131994 July 10, 2015 4/12 Se en y-nine o he 157 child en (50.3%) had hei sample sequenced o geno ypic esis- ance. The easons o no pe o ming esis ance es s we e he absence o HIV-1 RNA esul s (n = 22) and unde ec able HIV-1 RNA le els <50 copies/μL (n = 56). Table 1 summa ises he cha ac e is ics o child en wi h esis ance es esul s as well as he mo he s’ a iables o in e es . The median age o child en was 7 mon hs (3–12 mon hs) and 49 o 79 (62.0%) we e male. The as majo i y o child en (87.3%; 69 o 79) we e b eas ed and CD4 pe cen age was >20% in 48 o 79 child en (60.8%). ARV p ophylaxis consis ed o sd-NVP a bi h plus daily NVP in 92.4% o he child en. Almos all he child en had HIV sub ype C, 93.6% (74 o 79). The emaining sub ypes we e A/C, A/D, C/B, C/D and B (da a no shown). No s a is ically signi ican di e ences we e ound o any o he a iables o in e es be ween he 79 child en who had he samples sequenced and he 78 child en who did no . Table 1. Mo he and child socio-demog aphic and clinical cha ac e is ics among he popula ion wi h esis ance es esul s. Child en (n = 79) Age (mon hs) 7.0 3–12 Male 49 62.0% Deli e y mode aginal 66 83.5% caese ian 13 16.5% Bi h weigh 2.500 g 43 54.4% <2.500 g 36 45.6% Feeding p ac ice o mula eeding 10 12.7% exclusi e b eas eeding 69 87.3% CD4 pe cen age 25.00 2–55% <10% 10 12.7% 10–20% 21 26.6% >20% 48 60.8% ARV p ophylaxis AZT 6 7.6% sd-NVP + daily NVP 73 92.4% Mo he (n = 79) Clinical s age o HIV in ec ion (WHO) 1 and 2 13 16.5% 3 and 4 66 83.5% MTCT app oach † ART 51 64.6% P ophylaxis 28 35.4% Du a ion o exposu e o ARVs (mon hs) 8.00 0–38 CD4 coun (cells/mm 3 ) 540.00 108–1120 Da a a e numbe (%), median ( ange). AZT, zido udine; sd-NVP, single-dose ne i apine; MTCT, mo he - o-child ansmission; ARVs, an i e o i als; ART, an i e o i al he apy; mm 3 , cubic millili e. † The mo he ’sfi s -line ART egimen included AZT + 3TC + NVP. The ma e nal ARV p ophylaxis consis ed o an epa um daily AZT as ea ly as 14 weeks o ges a ion, sd-NVP a onse o labou and wice daily AZT + 3TC o 7 days pos pa um. doi:10.1371/jou nal.pone.0131994. 001 Ne i apine Resis ance in MTC HIV-1 T ansmission PLOS ONE | DOI:10.1371/jou nal.pone.0131994 July 10, 2015 5/12 The mo he s we e p edominan ly clinical s age 3 o 4 (83.5%; 66 o 79), 28 (35.4%) ecei ed daily AZT as ea ly a 14 weeks o ges a ion, sd-NVP a onse o labou and wice daily AZT + 3TC o 7 days pos pa um, and 51 (64.6%) we e on i s -line ART egimen. The median du a- ion o mo he ’s exposu e o ARVs was 8 mon hs (0–38 mon hs) (Table 1). NVP esis ance associa ed mu a ions (RAM) ARV RAM we e de ec ed in 43 o he 79 samples ha we e sequenced (54.4%; 95% CI 43.5– 65.4). Thi y-six samples (45.6%; 95% CI 34.6–56.6) had a leas one NNTRI RAM, 33 samples (41.8%; 95% CI 30.9–52.65) had NRTI RAM and se en samples (8.9%; 95% CI 2.6–15.1) had PI RAM. O e all, 74 NNTRI RAM, 54 NRTI RAM and eigh PI RAM we e obse ed. The NNRTI RAM mos equen ly ep esen ed in his ARV class we e K103N (n = 16, 21.6%, 95% CI 12.2– 31.0) and Y181C (n = 15, 20.3%, 95% CI 11.1–29.4) and G190A (n = 10, 13.5%, 95% CI 5.7– 21.3). M184V (n = 27, 50.0%, 95% CI 36.7–63.3), T69N and D67N (n = 4, 7.4%) we e he mos common NRTI RAM among his he apeu ic class. PI esis ance mu a ions we e M46I (n = 5, 62.5%) and Q58E (n = 1, 12.5%), L90M (n = 2, 25.0%) (Fig 1). All he samples wi h NNRTI RAM had a leas one NVP RAM. O e all, 66 NVP RAM we e ound. O hese, he mos equen mu a ions we e K103N (n = 16, 24.2%, 95% CI 13.4–34.6) and Y181C (n = 15, 22.7%, 95% CI 12.6–32.8), and G190A (n = 10, 15.2%, 95% CI 6.5–23.8). The e we e also mul iple occu ences o V108I (n = 6), A98G (n = 6), K101E (n = 4), V106A (n = 4), K101P (n = 1). The mu a ions K101H, V106M, Y181V and Y188L we e ound in only one sample each (Fig 2). Fac o s associa ed wi h geno ypic esis ance o NVP In he bi a iable analysis, NVP RAM was signi ican ly associa ed wi h mo he ’s exposu e o p ophylaxis ARV (c ude OR 30.3, 95% CI 4.93–186.34, P<0.0001) and wi h mo he ’s CD4 coun <350 cells/mm 3 (c ude OR 3.08, 95% CI 1.02–9.32, P<0.04). No s a is ically signi ican associa ion was ound be ween NVP RAM and child- ela ed a iables, mo he HIV clinical Fig 1. P o ile o d ug esis ance associa ed mu a ions by ARV class. Pe cen ages we e calcula ed based on he o al numbe o RAM ound o each ARV class (NNTRI = 74, NRTI = 54 and PI = 8). doi:10.1371/jou nal.pone.0131994.g001 Ne i apine Resis ance in MTC HIV-1 T ansmission PLOS ONE | DOI:10.1371/jou nal.pone.0131994 July 10, 2015 6/12 s age and du a ion o mo he ’s exposu e o ART (Table 2). In he mul i a iable op imiza ion logis ic eg ession, he mo he ’s exposu e o p ophylaxis ARV was he only a iable ha emained signi ican ly associa ed wi h NVP RAM (adjus ed OR 48.65, 95% CI 9.33–253.66, P<0.001). See S1 Table. S udy Da ase . Discussion Ou s udy aimed o es ima e he p e alence o NVP RAM in child en bo n o HIV-1 in ec ed mo he s and who we e exposed o he PMTCT p og amme p econized in Mozambique. The p e alence o NVP RAM ound in ou sample was app oxima ely 46%. This a e is wi hin he b oad ange o p e alence a es o NVP RAM ound among se e al sub-Saha an A ican coun- ies, which a ies om 27%, in Sou h A ica, o 76% in Malawi [20–22]. In 2010, a p ospec i e s udy wi h 740 Mozambican child en who had ecei ed sd-NVP p ophylaxis e ealed a high a e o NVP esis ance (87%) among he g oup wi h es ablished in u e o in ec ion and a a e o 38% when in ec ion occu ed in he pe ipa um pe iod [23]. O no ice, almos all o he child en had sub ype C HIV in ec ion. This inding is in line wi h a es p e iously obse ed in Mozambique. Se e al s udies showed ha bo h women and in an s wi h sub ype C HIV a e mo e likely o de elop NVP esis ance a e exposu e o sd- NVP han hose in ec ed wi h o he HIV sub ypes [24,25]. We ound ha he mo he s who ook AZT and/o sd-NVP o p o ide MTCT p ophylaxis o hei in an , ins ead o combina ion ART and wi h a low CD4 coun had a s a is ically signi - ican associa ion wi h NVP RAM in he bi a iable analysis. Child en whose mo he s had CD4 coun <350 cells/mm 3 , had a highe isk o NVP esis ance han hose child en whose mo he s Fig 2. P o ile o NVP esis ance associa ed mu a ions (RAM). Pe cen ages we e calcula ed based on he o al numbe o NVP RAM (n = 66). K103N and Y181C we e he mos equen NVP RAM (24.2% and 22.7% o he samples, espec i ely). doi:10.1371/jou nal.pone.0131994.g002 Ne i apine Resis ance in MTC HIV-1 T ansmission PLOS ONE | DOI:10.1371/jou nal.pone.0131994 July 10, 2015 7/12 had CD4 coun 350 cells/mm 3 (OR = 3.08). The low CD4 coun obse ed in mo he s could be a ibu able o i ological ailu e and selec ion o esis ance, which, in u n, could be ans- mi ed o he child. The majo i y o mo he s in ou sample (84%) we e clinical s age 3 o 4 o HIV in ec ion. In he mul i a iable logis ic model he mo he s ha ook an i e o i als o p ophylaxis hei e us/in an signi ican ly inc eased isk o NVP esis ance (OR = 48.65) among child en Table 2. Logis ic eg ession p edic ing NVP RAM adjus ed o independen a iables o in e es . NVP RAM (n = 36) Wi hou NVP RAM (n = 43) C ude OR (95% CI) P- alue Adjus ed OR (95% CI)* P- alue Child gende Female 12 40.0% 18 60.0% e e ence Male 24 48.0% 25 51.0% 1.44 (0.57–3.65) 0.4396 0.88 (0.25–3.12) 0.853 Child age (mon hs) <6 9 36.0% 16 64.0% e e ence 6–12 27 50.0% 27 50.0% 1.78 (0.66–4.79) 0.2482 1.85 (0.50–6.87) 0.356 mean age ±SD 7.80 ±2.89 6.67 ±2.58 0.061 Deli e y mode Vaginal 7 53.9% 6 46.2% e e ence Caese ian 29 43.9% 37 56.1% 0.67 (0.20–2.24) 0.5148 Bi h weigh (g) <2500 18 50.0% 18 50.0% 1.39 (0.57–3.42) 0.4690 2500 18 41.9% 25 58.1% e e ence Feeding p ac ice Exclusi e b eas eeding 33 47.8% 36 52.2% e e ence In an o mula 3 30.0% 7 70.0% 0.47 (0.11–2.00) 0.2932 P ophylaxis wi h NVP in child en 32 45.7% 40 57.1% 0.60 (0.12–2.92) 0.5222 Mo he exposu e o ARV ART 11 21.6% 40 78.4% e e ence MTCT P ophylaxis 25 89.3% 3 10.7% 30.3 (4.93– 186.34) <0.0001 48.65 (9.33–253.66) <0.001 Mo he CD4 coun (cells/mm 3 ) <350 12 66.7% 6 33.3% 3.08 (1.02–9.32) 0.0408 1.75 (0.37–8.26) 0.477 350 24 39.3% 37 60.7% e e ence mean CD4 coun 504.44 ±283.14 619.23 ±214.18 0.043 Du a ion o mo he ’s exposu e o ARVs (mon hs) 0.31 (0.08–1.31) 0.111 <12 8 33.3% 16 66.7% e e ence 12 28 50.9% 27 49.1% 2.07 (0.75–5.75) 0.1517 Mo he WHO s aging o HIV in ec ion 1 and 2 6 46.2% 7 53.9% e e ence 3 and 4 30 45.5% 36 54.6% 0.97 (0.29–3.23) 0.9633 Da a a e numbe (%); mean (s anda d de ia ion). *The ollowing independen a iables we e included in he mul iple logis ic adjus ed eg ession analysis: child gende , child age, mo he exposu e o ARV, mo he CD4 coun , ma e nal du a ion o exposu e o ARVs. Values in bold deno e s a is ical significance. ART, an i e o i al he apy; CI, Confidence In e al; OR, Odds Ra io; NVP RAM, ne i apine esis ance associa ed mu a ions; MTCT, mo he - o-child- ansmission; mm 3 , cubic millili e. doi:10.1371/jou nal.pone.0131994. 002 Ne i apine Resis ance in MTC HIV-1 T ansmission PLOS ONE | DOI:10.1371/jou nal.pone.0131994 July 10, 2015 8/12 compa ing o mo he s who ecei ed ART. This inding co obo a es he esul s obse ed in Mapu o by Vaz e al [26], subs an ia ing he e idence ha sd-NVP-based p ophylac ic egi- mens lead o a ma ked inc ease in he isk o esis ance o his ARV and o he subsequen use o NNRTIs [3,12,27,28]. In ou s udy, he isk o i al esis ance o NVP was nume ically highe among child en whose mo he s we e exposed o ARVs o mo e han 12 mon hs bu his inding had no s a is- ical signi icance (OR = 2.07). In ou sample, a high p opo ion o child en (87.3%) we e b eas ed. An analysis o b eas milk samples om mo he s o ci y o Bei a in Mozambique who we e exposed o sd-NVP, showed a pe sis ence o NVP- esis an HIV-1 up o 8 mon hs pos pa um [29]. A ound 48% o he child en in ou s udy ha we e b eas ed had NVP esis ance, which is abo e he a e o NVP esis ance de ec ed in Bei a, Mozambique (37.5%) [30] and below he a e ound in Malawi (66%) [21]. In he bi a iable analysis we ound no s a is ically signi ican associa ion be ween b eas eeding and an inc eased isk o NVP RAM. Thi y-six o he 79 child en ha we e success ully geno yped in ou s udy exhibi ed a leas one mu a ion con e ing esis ance o NNRTIs. The mos equen e e se ansc ip ase mu a- ions we e Y181C, K103N and G190A. Simila esis ance pa e ns ha e been ound in o he s udies [31–33]. PI RAM we e de ec ed in a small sample o child en, which e lec s he low use o PI in his popula ion. Limi a ions The ac ha we did no know he mo he s ART his o y and hei ARV esis ance p o ile con- s i u es one majo limi a ion o ou s udy. As a esul , we could no disc imina e be ween esis- ance mu a ions ha we e due o ART o PMTCT and he ones ha we e selec ed in he child. WHO endo ses he egula su eillance o ARV esis an HIV in pa ien s who ecei e ART o he i s ime. Se e al s udies conduc ed in sub-Saha an A ican coun ies ound a es o p i- ma y esis ance a ound 5% [34]. Fu he mo e, we we e unable o iden i y i mo he s we e p e- iously exposed o NVP-con aining egimens and, he e o e, assess he impac o his po en ial con ounde o NVP esis ance de elopmen . We ha e no in es iga ed he pe sis ence o de ec able NVP esis ance in ou sample. E i- dence shows ha child en’s exposu e o sd-NVP is associa ed wi h NVP- esis an mu a ions ha can pe sis o a yea o mo e [35–37]. This pe sis ence seems o be mo e ma ked a e exposu e o ex ended NVP p ophylaxis [38,39]. Among HIV-in ec ed Ugandan in an s wi h NVP esis ance a 6 weeks o age, NVP- esis an HIV a ian s we e s ill de ec ed a 6 mon hs in all se en in an s who ecei ed he ex ended NVP egimen, compa ed o only one o six in an s who ecei ed sd-NVP alone [38]. A la ge sample would ha e p o ided mo e obus indings ega ding he ou comes o in e - es . Some o he specimens we e no sequenced due o poo quali y o delay in i s shipmen . Mo eo e , he a ailable sequencing assays we e no able o de ec esis ance o samples wi h HIV RNA le els <100 copies/μL. Wi h he apid scale up o ART in Mozambique, mo e han 16.000 child en ha e been exposed o ARVs up o 2011. Despi e g adually imp o ing access o ART, he p opo ion o child en cu en ly ea ed in he coun y is conside ably small. As o Decembe 2010 only 23% o HIV-in ec ed child en aged <15 yea s old in need o ART we e ecei ing his he apy [26, 33]. The iden i ica ion o he ac o s mos associa ed wi h i ological esponse and p e en ion o esis ance o ARVs is c ucial o an app op ia e managemen o ART p og ammes, he op imi- za ion o he use o esou ces and a be e de ini ion o ea men s a egies. This knowledge is Ne i apine Resis ance in MTC HIV-1 T ansmission PLOS ONE | DOI:10.1371/jou nal.pone.0131994 July 10, 2015 9/12