HIV drug resistance in a community‐randomized trial of universal testing and treatment: HPTN 071 (PopART)

Abstract

Introduction Universal HIV testing and treatment (UTT) has individual and public health benefits. HPTN 071 (PopART), a community‐randomized trial in Zambia and South Africa, demonstrated that UTT decreased HIV incidence. This endpoint was assessed in a cohort of >48,000 randomly selected adults in the study communities. We evaluated the impact of UTT on HIV drug resistance in this cohort and compared other resistance‐related outcomes in participants with recent versus non‐recent HIV infection. Methods Two years after the start of HPTN 071 (2016–2017), 6259 participants were HIV positive and 1902 were viremic (viral load >400 copies/ml). HIV genotyping and antiretroviral (ARV) drug testing were performed for viremic participants in three groups: seroconverters (infected <1 year), non‐seroconverters (infected >1 year, random subset) and participants with unknown duration of infection (random subset). A two‐stage cluster‐based approach was used to assess the impact of the study intervention on drug resistance. Treatment failure was defined as being viremic with ARV drugs detected. Participants were classified as ARV naïve based on self‐report and ARV drug testing. Results Genotyping results were obtained for 758 participants (143 seroconverters; 534 non‐seroconverters; and 81 unknown duration of infection). The estimated prevalence of resistance in the study communities was 37% for all viremic persons and 11% for all HIV‐positive persons. There was no association between UTT and drug resistance. Resistance was detected in 14.0% of seroconverters and 40.8% of non‐seroconverters (non‐nucleoside reverse transcriptase inhibitor resistance: 14.0% and 39.9%; nucleoside/nucleotide reverse transcriptase inhibitor resistance: 0.7% and 15.5%; protease inhibitor resistance: 0% and 1.9%; multi‐class resistance: 0.7% and 16.1%, respectively). ARV drugs were detected in 2/139 (1.4%) of seroconverters and 94/534 (17.6%) of non‐seroconverters tested. These participants were classified as failing ART; 88 (93.6%) of the non‐seroconverters failing ART had resistance. Mutations used for surveillance of transmitted drug resistance were detected in 10.5% of seroconverters and 15.1% of non‐seroconverters who were ARV naive. Conclusions UTT was not associated with an increase in drug resistance in this cohort. Higher rates of drug resistance and multi‐class resistance were observed in non‐seroconverters compared to seroconverters.

Demographics of people who transmit HIV-1 in Zambia: a molecular epidemiology analysis in the HPTN-071 PopART study

“…Once again, there was no evidence of a difference in the distribution of mutations with respect to the trial arm of the recipient (chi-squared test p=0•321). For comparison, 79•7%, 11% and 9•3% of recipients had virus that was sensitive, low-level, or high-level resistant, in line with a previously published study on resistance 22 .…”

Demographics of sources of HIV-1 transmission in Zambia: a molecular epidemiology analysis in the HPTN 071 PopART study

“…For comparison, of 285 recipients with non-missing data, 227 (209–242; 79·7%), 31 (21–46; 11·0%), and 27 (16–40; 9·3%) recipients had virus that was respectively sensitive, low-level, or high-level resistant, in line with a previously published study on resistance. 23 …”

Population‐level analysis of natural control of HIV infection in Zambia and South Africa: HPTN 071 (PopART)

“…For this reason, our estimated prevalence of 3.1% should be considered a minimum estimate. Second, HPTN 071 included a disproportionate number of women and was conducted in Zambia and South Africa, where almost all infections are caused by HIV subtype C. In a prior study, we performed HIV genotyping and subtyping for a subset of 758 participants in HPTN 071; 95.9% of those participants had subtype C HIV infection [35]. These factors may limit the generalizability of our findings to other cohorts and to regions where other subtypes predominate.…”