CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

Results: A total 458 children attended the clinic; 244 (53%) were male; 22 (5%) 0-3 years, 192 (42%) 4-10 years and 244 (53%) 11-17 years. Median CD4 count was 713.5 cells/mm 3 (IQR 440-1088) and median time on ART 16 months (IQR 8-30). Changes to ART regimens were documented in 241/458 children (53%). ART regimen combinations were prescribed outside of recommended guidelines in 2/22 (9%) aged 0-3, 41/192 (21%) aged 4-10 and 19/100 (19%) aged 11-17. The majority (n=13, 59%) of infants were over-dosed on one drug of the ART regimen (Table 1). Among the young (4-10) and older children (11-17) 82 (46%) and 41 (18%) were either under- and /or over-dosed on one drug. Drugs commonly over-dosed in the older age groups were efavirenz and boosted lopinavir. Zidovudine was more frequently under- dosed and nevirapine was both frequently under- and over-dosed. 311/448 (69%) children were prescribed a fixed-dose combination (FDC). 77% (160) on FDC had all three drugs correctly dosed compared to those not on a FDC 62% (151) were correctly dosed. Conclusions: This study shows that ART regimens in children are often prescribed outside of national guidelines and at doses not adjusted to their weight. Suboptimal ART dosing might result in unwanted side effects, development of virological failure and potentially drug resistance. Training of health care workers should address appropriate dosing and simple aide-memoires may support accurate dosing.

847 Persistence of HIV Drug Resistance in Children Exposed to Nevirapine Prophylaxis Ruth M. Kanthula 1 ; Ingrid Beck 2 ; GiselaVan Dyk 3 ; Rachel Silverman 1 ; Scott Olson 1 ; Christen Salyer 2 ; Sharon Cassol 3 ;Theresa Rossouw 3 ; Lisa Frenkel 2 1 Univ of Washington, Seattle, WA, USA; 2 Seattle Children’s Rsr Inst, Seattle, WA, USA; 3 Univ of Pretoria, Pretoria, South Africa Background: Nevirapine (NVP) is administered to breastfeeding infants as prophylaxis for prevention of mother-to-child-transmission of HIV. Since drug-resistant viruses can compromise future antiretroviral therapy (ART) when prophylaxis fails, we studied HIV-infected, NVP-exposed children to determine the persistence of NVP resistance and correlate the detection of resistance with duration of exposure to NVP. Methods: HIV-infected, NVP-exposed children initiating ART were enrolled in a prospective observational study in Pretoria, South Africa. Dried blood spots were collected at enrollment and during monthly follow-up visits. Amplifiable HIV templates were quantified by real time PCR of gag , so that a target of 300 templates could be submitted to nested HIV pol PCR. NVP-resistance was assessed by an oligonucleotide ligation assay optimized for HIV-1 subtype C with the capability to quantify ≥2%mutant. The proportion of virus population comprised of mutant and wild-type viruses were quantified at four codons encoding high-level resistance to non-nucleoside reverse transcriptase inhibitors (NNRTIs). Results: A total of 88 HIV-infected children enrolled in the study from 2010-2013 at a median of 8 months (mo) of age (IQR:4-15). Children received single-dose (sd)-NVP at birth (n=40) or extended NVP prophylaxis (n=48) for a median of 6 weeks (IQR:4-9). At study enrollment, NVP-resistance was detected in 52% of children and among them, mutations were detected at Y181C (63%), K103N (43%), G190A (22%) and V106M (22%). Children with resistance had viral populations with high proportions of mutant variants (median 96% (IQR:82-100)). These children were also younger, with more recent NVP-exposure (median of 4 (IQR:2-7) vs. 14 (IQR:9-27) mo prior to enrollment, P<0.0001), and a greater proportion had received prolonged NVP-prophylaxis vs. sdNVP (60% vs. 43%; P=0.133). Of children with resistance at enrollment, 59%were followed longitudinally for a median of 13 mo (IQR:12-16) and resistant viruses persisted at a median concentration of 98% (IQR:77-100) to a median age of 21 mo (IQR:17-32). Conclusions: HIV-infected children who fail NVP-prophylaxis often have HIV populations consisting of NVP-resistant viruses, suggesting maternal transmission of resistant variants or NVP-selection at the time the HIV reservoir was established. The persistence of high mutant concentrations supports drug resistance testing in NVP-exposed children prior to NNRTI-ART in resource-limited communities. 848 Independent lineages of HIV-1 Multidrug Resistance in Children Failing Early ART Camille M. Lange 1 ; Stephane Hue 2 ; AvyViolari 3 ; Mark F. Cotton 4 ; Diana Gibb 5 ; Deenan Pillay 6 ; Ravindra K. Gupta 7 1 NIH, Bethesda, MD, USA; 2 London Sch of Hygiene & Trop Med, London, UK; 3 Univ of the Witwatersrand, Johannesburg, South Africa; 4 Stellenbosch Univ and Tygerberg Children’s Hosp, Cape Town, South Africa; 5 Med Rsr Council Clinical Trials Unit, London, UK; 6 Africa Cntr for Hlth and Pop Studies, Mtubatuba, South Africa; 7 Univ Coll London, London, UK Background: Early protease inhibitor (PI) based antiretroviral therapy (ART) is recommended in vertically infected children, bringing immediate clinical benefit and limiting the reservoir size, which may impact on future chances of cure. We previously reported linked multi-class drug resistance (MDR) in children that received nevirapine prophylaxis and failed early PI based ART (Lange et al, JAIDS 2015). Here we report evolutionary histories of drug resistant lineages in these children. Methods: Single genome sequencing was used to characterise genetically linked MDR in HIV-1 protease and reverse transcriptase from longitudinal plasma samples. Sequences were obtained from the pre-ART populations, during suppressive therapy, treatment failure, treatment interruption and re-started ART. The mean pairwise genetic distances (MPDs) between these populations were calculated. The evolution and population diversity of MDR lineages were analysed by Maximum Likelihood (ML) phylogenetic reconstruction. Results: 442 sequences were obtained from 2/10 children: n=249 from 7 time points in the first child; n=193 from 5 time points in the second child. Sampling depth was estimated as ≥ 5-10% of the viral population. ML trees showed multiple emergences of MDR on distinct intra-patient lineages. One child had a clonally expanded MDR lineage persisting for 85 weeks of ART. Accordingly, MPD at weeks 12, 40, 72 and 96 of ART were significantly lower than at baseline (0.6%, 0.5%, 0.2% and 0.3% vs 0.7% respectively; p<0.0001). In stark contrast the ML tree of the second child indicated temporal stepwise accumulation of resistance mutations. In this scenario the MPD at weeks 40 and 96 of ART and at week 60 of re-started ART were significantly higher than at baseline (0.3%, 0.4% and 0.3% vs 0.1% respectively; p<0.0001). MDR populations with PI resistance most frequently evolved from RT M184V mutated viral populations in both children. Conclusions: We report distinct patterns of multiple class HIV-1 drug resistance acquisition over time in children treated with early PI based ART: firstly clonal expansion from an MDR reservoir, and secondly through stepwise accumulation of major resistance mutations.

Poster Abstracts

355

CROI 2016