Baseline variables associated with subsequent HIV seroconversion among gay, bisexual and other men who have sex with men and transgender women: a prospective, multicenter PrEP demonstration study (ImPrEP)

Background: Ongoing implementation of HIV pre-exposure prophylaxis (PrEP) in Latin America should consider lessons learned from implementation projects such as ImPrEP (Brazil/Mexico/Peru, 2018–2021). In this analysis we assessed the effect of variables ascertained in early appointments on HIV seroconversion among ImPrEP participants. Methods: ImPrEP enrolled HIV-negative men who have sex with men and transgender women (MSM/TGW) aged 18+ years reporting recent condomless anal sex, anal sex with HIV-positive partners, transactional sex, or sexually transmitted infections (STI). Participants received a 30-day PrEP supply; at the 30-day visit and quarterly thereafter they completed behavioural assessments, underwent HIV testing, and received 3-month PrEP supplies if HIV-negative. PrEP adherence was measured using the medication possession ratio (MPR) at the 30-day visit. We used Cox's proportional hazards regression to examine the effect of our sociodemographic, behavioural, STI, and early PrEP care engagement variables of interest on time to HIV seroconversion. Findings: Compared to participants in Brazil, the hazard ratio for HIV seroconversion was higher among those in Peru (HR = 7.91, 95% CI: 4.74–13.20). Compared to participants aged ≥35 years, the HR for HIV seroconversion was higher for those aged 18–24 (aHR = 4.84, 95% CI: 2.55–9.17 and 25–34 (aHR = 2.43, 95% CI: 1.21–4.91). HIV seroconversion was also associated with transgender identity (aHR = 2.28, 95% CI: 1.12–4.66), transactional sex (aHR = 1.88, 95% CI: 1.18–2.99), receptive condomless anal sex (aHR = 2.42, 95% CI: 1.42–4.12), STI diagnosis (aHR = 1.93, 95% CI: 1.25–2.99), and a MPR < 0.6 (aHR = 2.64, 95% CI: 1.52–4.60). Interpretation: While moderate-high, HIV incidence among ImPrEP participants represented a considerable reduction from figures observed among MSM/TGW not using PrEP/PEP. Interventions to improve PrEP adherence are needed among new Latin American PrEP users, especially if baseline factors associated with seroconversion are present. Long-acting injectable PrEP can also become useful for this population. Funding: This study was funded by UNITAID.