CROI 2016 Abstract eBook

Abstract Listing

Poster Abstracts

849 Transmitted Drug Resistance and First-Line ART Treatment Outcomes in Ugandan Children

Cissy Kityo 1 ; Ragna S. Boerma 2 ; Kim Sigaloff 2 ; Elizabeth Kaudha 1 ; Job Calis 3 ;Victor Musiime 4 ;Tamara Sonia Boender 2 ; Henry Mugerwa 1 ;Tobias Rinke deWit 2 ; Peter Mugyenyi 1 1 Joint Clinical Rsr Cntr, Kampala, Uganda; 2 Amsterdam Inst for Global Hlth and Develop, Amsterdam, Netherlands; 3 Academic Med Cntr, Amsterdam, Netherlands; 4 Makerere Univ Coll of Hlth Scis, Kampala, Uganda Background: Transmission of drug-resistant (TDR) HIV-1 can impair virologic response to antiretroviral therapy (ART) especially in children. This study evaluated the effect of TDR on virologic and acquired drug resistance (ADR) outcomes among children initiating first-line ART. Methods: Children ≤12 years initiating first-line ART were enrolled at 3 sites in Uganda between 2010 and 2011. Blood was taken at baseline and 6-monthly during 24 months of ART for later determination of viral load (VL) and pol genotypic testing if VL>1,000 copies/ml at JCRC Kampala laboratories. The 2014 IAS-USA mutation list and Stanford algorithm

were used to score drug resistance mutations (DRMs) and susceptibility. Patients were classified into two groups: fully active (no TDR or TDR without reduced susceptibility) and patients with TDR and partially active ART. Virological failure (VF) was defined as 2 consecutive VLs >1000 copies/ml, at least 6 months from ART initiation. Factors associated with VF and acquired drug resistance (ADR) were assessed in multivariate logistic regression analysis and evolution of resistance was described. Results: 317 children median age 4.9 years were enrolled on mainly NNRTI based regimens (91.2%). TDR mutations were detected in 47(16.9%) participants of whom 22(46.8%) initiated a partially active ART. 256(80.8%) participants were still on first line ART and in care at 24 months of follow-up of whom 32%(92/287) had VF. Children with TDR and partially active ART had significantly higher risk of VF (aOR:15.25, 95%-CI:3.77-61.7,p<0.001) and ADR (aOR:3.47, 95%-CI:1.31- 9.22,p<0.012). A single VL >1000 copies/ml at 6 months on treatment was strongly associated with VF (OR:22.09, 95%-CI:9.68-50.42,p<0.001) and ADR (OR:9.89, 95%-CI:5.16–18.94;p<0.001). Other factors associated with VF in the adjusted model were higher baseline VL (aOR:2.28,p<0.001) and WHO stage 2 compared to Stage 1 (aOR:10.3,p=0.022). Among the 66 children with prolonged viraemia, there was a high rate of acquisition of DRMs. Conclusions: In this pediatric cohort, TDR was found to be high and strongly associated with VF and ADR. Accumulation of DRMs was high and may jeopardize

future therapeutic options. Efforts are needed to incorporate affordable drug resistance testing in developing countries to prevent the use of suboptimal ART. In the context of universal access of VL monitoring, programmatic use of a 6 month VL on treatment would detect majority of future VFs early and allow for treatment adjustment to prevent ADR.

850 Drug Resistance Compromises Second-Line ART in Mozambican Children Failing First-Line

Paula M. Vaz 1 ;William C. Buck 2 ; Dulce A. Bila 3 ; Nilesh Bhatt 3 ; Kebba Jobarteh 2 ; Loide Cossa 3 ; Charity Alfredo 2 ; James Houston 4 ; Amina Sousa 3 ; ChunfuYang 4 1 Fundação Ariel Glaser, Maputo, Mozambique; 2 CDC, Maputo, Mozambique; 3 Inst Nacional de Saúde, Maputo, Mozambique; 4 CDC, Atlanta, GA, USA Background: Mozambique’s pediatric antiretroviral treatment (ART) program grew from 9,393 patients in 2007 to 41,400 in 2013. This rapid and ongoing expansion has made assurance of quality care a challenge and has occurred without viral load monitoring. In this context, we assessed the prevalence of virologic failure (VF) and HIV drug resistance mutations (DRM) in a cohort of ART-experienced children to understand implications for the current pediatric second-line regimen of LPV-r/3TC/ABC or TDF. Methods: A cross-sectional study was conducted at six clinics providing pediatric ART in Maputo. Children aged 1-14 years and active on ART for at least 12 months were enrolled from August 2013 to March 2014. Clinical and demographic information were collected and viral load was performed. Dried blood spots were prepared from samples with VF (>1000 copies/mL) for genotyping. DRM interpretation and drug susceptibility were assessed using HIVdb (Stanford v7.0). Statistical analysis was done with SAS (v9.2). Results: In total 713 children were enrolled; mean age of 102.9 months (95%CI: 81.0-124.7), mean time on ART of 60.3 months (95%CI: 38.8-81.2), 73.2% (95%CI: 43.5-90.6) were on d4T/3TC/NVP, 85.8% (95%CI: 75.2-92.3) had no immune suppression and 20.2% (95%CI: 15.0-26.6) had PMTCT exposure. VF was observed in 256 patients (35.9%) (95%CI: 26.9- 46.2%), and 96.9% (n=248) were successfully genotyped, with DRM found in 94.8% (n=235). High levels of NRTI and NNRTI mutations were observed with M184V (90.7%) and Y181C (49.6%) most common in each class. K65R and major PI DRMs occurred in 2.4% and 1.6% of sequences, respectively. Thymidine analog mutations (TAMs) were observed in 33.5% (n=83), all in children ≥5 years. TAM-2 was the main pathway observed (n=69; 83.1%) with a mixture of both pathways in 12 patients. Second-line ART had three, two, and one-drug efficacy in 5.4%, 82.4%, and 12.2% of patients with DR, respectively, age-stratified results in Fig. 1. Conclusions: VF was common (35.9%) in this large, treatment-experienced, older pediatric ART cohort, and was strongly associated with DRM (94.8%). Many of these children did not have immunologic failure and had likely been on failing regimens for some time, accumulating DRMs. These findings support the ongoing roll-out of viral load testing in Mozambique and have implications for pediatric second-line outcomes. Additional analysis of these data will look at differences in VF and DR by age group and associated risk factors.

Poster Abstracts

356

CROI 2016