Search

Objective: To identify predictors of HIV acquisition in Botswana., Design: We applied machine learning approaches to identify HIV risk predictors using existing data from a large, well-characterized HIV incidence cohort., Methods: We applied machine learning (randomForestSRC) to analyze data from a large population-based HIV incidence cohort enrolled in a cluster-randomized HIV prevention trial in 30 communities across Botswana. We sought to identify the most important risk factors for HIV acquisition, starting with 110 potential predictors., Results: During a median 29-month follow-up of 8,551 HIV-negative adults, 147 (1.7%) acquired HIV. Our machine learning analysis found that for females, the most important variables for predicting HIV acquisition were the use of injectable hormonal contraception, frequency of sex in the prior 3 months with the most recent partner and residing in a community with HIV prevalence of 29% or higher. For the small proportion (0.3%) of females who had all three risk factors, their estimated probability of acquiring HIV during 29 months of follow-up was 34% (approximate annual incidence of 14%). For males, non-long-term relationships with the most recent partner and community HIV prevalence of 34% or higher were the most important HIV risk predictors. The 6% of males who had both risk factors had a 5.1% probability of acquiring HIV during the follow-up period (approximate annual incidence of 2.1%)., Conclusions: Machine learning approaches allowed us to analyze a large number of variables to efficiently identify key factors strongly predictive of HIV risk. These factors could help target HIV prevention interventions in Botswana., Clinical Trials Registration: NCT01965470., (Copyright © 2024 Wolters Kluwer Health, Inc. All rights reserved.)

Background: Risk for nondiagnostic and false-positive HIV testing has not been quantified for neonates., Methods: From April 2015 to July 2018, we screened HIV-exposed infants in Botswana less than 96 h from birth by qualitative DNA PCR. Repeat blood draws for DNA and RNA PCR testing occurred for initial positive and indeterminate results to establish final diagnosis. We compared screening DNA PCR cycle threshold values with final HIV status of the child., Results: Of 10 622 HIV-exposed infants, 10 549 (99.3%) had no HIV DNA detected (negative), 42 (0.4%) had HIV DNA detected (positive), and 31 (0.3%) tested indeterminate at first HIV screen. Repeat testing identified 2 (5.0%) of 40 positive screens (2 declined additional testing) as false positives and confirmed 2 (6.5%) of 31 indeterminate screens as infected. Median cycle threshold value at screening was 28.1 (IQR 19.8--34.8) for children with final positive status, and 35.5 (IQR 32.8--41.4) for indeterminates who were ultimately negative. Six (15%) of 40 infants with final positive status had cycle threshold value greater than 33 at first screen, whereas 3 (9.7%) of 31 indeterminates with final negative status had cycle threshold value 33 or less at first screen. This threshold resulted in a negative predictive value of 82% and a positive predictive value of 92% for a single screen., Conclusion: Although a DNA PCR cycle threshold value of 33 was predictive of the final HIV status in newborns, overlap occurred for true positives, false positives, and initial indeterminates. Testing additional samples should be standard practice for positive and indeterminate HIV DNA PCR tests in the first week of life.

Background: Scale-up of antiretroviral therapy (ART) and introduction of treat-all strategy necessitates population-level monitoring of acquired HIV drug resistance (ADR) and pretreatment drug resistance (PDR) mutations., Methods: Blood samples were collected from 4973 HIV-positive individuals residing in 30 communities across Botswana who participated in the Botswana Combination Prevention Project (BCPP) in 2013-2018. HIV sequences were obtained by long-range HIV genotyping. Major drug-resistance mutations (DRMs) and surveillance drug resistance mutations (SDRMs) associated with nucleoside reverse transcriptase inhibitors (NRTI) and nonnucleoside reverse transcriptase inhibitors (NNRTI) were analyzed according to the Stanford University HIV Drug Resistance Database. Viral sequences were screened for G-to-A hypermutations. A threshold of 2% was used for hypermutation adjustment. Viral suppression was considered at HIV-1 RNA load ≤400 copies/ml., Results: Among 4973 participants with HIV-1C sequences, ART data were available for 4927 (99%) including 3858 (78%) on ART. Among those on ART, 3435 had viral load data and 3297 (96%) were virologically suppressed. Among 1069 (22%) HIV-infected individuals not on ART, we found NRTI-associated and NNRTI-associated SDRMs were found in 1.5% (95% confidence interval [CI] 1.0-2.5%) and 2.9% (95% CI 2.0-4.2%), respectively. Of the 138 (4%) of individuals who had detectable HIV-1 RNA, we found NRTI-associated and NNRTI-associated drug resistance mutations in 16% (95% CI 10-25%) and 33% (95% CI 25-42%), respectively., Conclusion: We found a low prevalence of NRTI-associated and NNRTI-associated PDR-resistance mutations among residents of rural and peri-urban communities across Botswana. However, individuals on ART with detectable virus had ADR NRTI and NNRTI mutations above 15%.

Background: National initiatives offering non-nucleoside reverse transcriptase inhibitor (NNRTI)-based combination antiretroviral therapy (cART) have expanded in sub-Saharan Africa. The Tshepo study is the first clinical trial evaluating the long-term efficacy and tolerability of efavirenz versus nevirapine-based cART among adults in Botswana., Methods: A 3-year randomized study (n = 650) using a 3 x 2 x 2 factorial design comparing efficacy and tolerability among: (i) zidovudine/lamivudine versus zidovudine/didanosine versus stavudine/lamivudine; (ii) efavirenz versus nevirapine; and (iii) community-based supervision versus standard adherence strategies. This paper focuses on comparison (ii)., Results: There was no significant difference by assigned NNRTI in time to virological failure with resistance (log-rank P = 0.14), nevirapine versus efavirenz [risk ratio (RR) 1.54, 95% CI 0.86-2.70]. Rates of virological failure with resistance were 9.6% nevirapine-treated (95% CI 6.8-13.5) versus 6.6% efavirenz-treated (95% CI 4.2-10.0) at 3 years. Women receiving nevirapine-based cART trended towards higher virological failure rates when compared with efavirenz-treated women, Holm-corrected (log-rank P = 0.072), nevirapine versus efavirenz (RR 2.22, 95% CI 0.94-5.00). A total of 139 patients had 176 treatment-modifying toxicities, with a shorter time to event in nevirapine-treated versus efavirenz-treated patients (RR 1.85, 1.20-2.86; log-rank P = 0.0002)., Conclusion: Tshepo-treated patients had excellent overall immunological and virological outcomes, and no significant differences were observed by randomized NNRTI comparison. Nevirapine-treated women trended towards higher virological failure with resistance compared with efavirenz-treated women. Nevirapine-treated adults had higher treatment modifying toxicity rates when compared with those receiving efavirenz. Nevirapine-based cART can continue to be offered to women in sub-Saharan Africa if patient education concerning toxicity is emphasized, routine safety monitoring chemistries are performed and the potential risk of efavirenz-related teratogenicity is considered.