CROI 2025 Abstract eBook

Abstract eBook

Poster Abstracts

1347 PrEP Use Trajectories and HIV Incidence Among PrEP Users in Brazil: Findings From the ImPrEP Study Carolina Coutinho 1 , Beatriz Grinsztejn 1 , Brenda Hoagland 1 , Marcelo Cunha 1 , Iuri Leite 1 , Ronaldo Ismério 1 , Mayara Secco Torres da Silva 1 , Maria Cristina Abbade 2 , Alessandro S. Farias 3 , Jose Valdez Madruga 4 , Marcos Benedetti 5 , Cristina Pimenta 5 , Thiago Torres 1 , Valdilea Gonçalves Veloso 1 , for the ImPrEP CAB Brasil Study Group 1 Instituto Nacional de Infectologia Evandro Chagas, Rio de Janeiro, Brazil, 2 Secretaria Municipal de Saúde de São Paulo, São Paulo, Brazil, 3 Centro Especializado em Diagnóstico, Assistência e Pesquisa (CEDAP), Salvador, Brazil, 4 Centro de Referência e Treinamento DST/AIDS-SP, Sao Paulo, Brazil, 5 Oswaldo Cruz Foundation, Rio de Janeiro, Brazil Background: New HIV cases continue to rise in Latin America, especially among young men who have sex with men (MSM) and transgender women (TGW). Understanding long-term oral PrEP use is crucial for improving adherence and retention to services. This study explored PrEP use patterns over time in Brazil, identifying distinct trajectories and associated factors. Methods: ImPrEP was a prospective, single-arm, multicenter oral PrEP study conducted in Brazil, Mexico, and Peru . Eligible MSM/TGW (aged ≥18, HIV-negative, risk behaviors in past 6 months). For this analysis, we included Brazilian participants followed from Feb 2018 to Jul 2024. Group-based trajectory models identified adherence patterns using MPR ≥ 0.6 (possessing at least 4 PrEP pills/week) as reference for adequate adherence. A multinomial model assessed factors associated to belonging to each group. Discontinuation was defined as not receiving PrEP for at least six months and not returning until the end of follow-up. Results: Of 3,928 enrolled, 3,780 returned for at least one follow-up visit and were included in this analysis. Median follow-up duration was 2.97 years (IQR:1.30–5.13years); 11,797.56 person-years. We identified 3 distinct trajectories: G1 (60%) with consistent probability of having adequate adherence during follow-up, G2 (36%) with gradual decline in adequate adherence, while G3 (4%) with rapid decline (Figure). Higher odds of being in G2 and G3 compared to G1 were younger age (G2: aOR=1.50, 95%CI:1.26–1.79; G3: aOR=1.83, 95%CI:1.18–2.85), transactional sex (G2: aOR=1.44, 95%CI:1.10–1.88; G3: aOR=1.82, 95%CI:0.96–3.44), and stimulant use (G2: aOR=1.24, 95%CI:1.04– 1.48). Overall, 66% discontinued PrEP during follow-up, and HIV incidence was higher among them compared to those on PrEP (1.67 [95%CI:1.34-2.06] vs. 0.26 [95% CI:0.18-0.36]). HIV incidence rate was 4.22 times (95%CI;1.66 14.09; p=0.002) higher in G3 than in G1 (G1: 0.18 [95%CI:0.11-0.33]; G2: 0.29 [95%CI:0.17-0.48]; G3: 0.97 [95%CI: 0.40-2.32]). Conclusions: We identified 3 distinct oral PrEP adherence trajectories in Brazil, with those with rapid adherence decline being of younger age, engaging in transactional sex, and using stimulant drugs. Higher HIV incidence among those who discontinued PrEP underlines the importance of retention in PrEP services. Long-acting PrEP technologies may help address the unique challenges faced by these subgroups, enhancing PrEP retention and overall effectiveness.

and completed a questionnaire in English or Setswana. At follow-up, those who reported initiating PrEP provided blood samples (finger prick) for dried blood spots (DBS) to assess PrEP adherence and persistence. PrEP readiness was measured via self-report and PrEP uptake was assessed via medical record data. Results: A repeated-measures mixed model indicated a significant effect of treatment-by-time interaction on PrEP readiness, χ²(3, 9) = 30.08, p = 0.004. Logistic regressions, accounting for clustering, revealed that there was not a statistically significant effect of the treatment on PrEP uptake – although those in the S&D + PrEP/SRH (OR = 2.50; CI = 1.47, 4.25) and S&D + PrEP/SRH + YW (OR = 2.94; CI 1.75, 4.95) arms were more likely to initiate PrEP than those in the control arm. AGYW in the PrEP/SRH + YW arm (OR = 2.74, CI = 1.26, 5.94) were more likely than those in the control arm to be adherent. AGYW in the S&D + PrEP/SRH (OR = 1.89, CI = 1.09, 3.22), PrEP/SRH + YW (OR = 2.32, CI = 1.99, 2.71) and the S&D + PrEP/SRH + YW (OR = 1.82, CI = 1.31, 2.52) arms were more likely to persist on PrEP than those in the control arm. Conclusions: Our study and its findings offers some important evidence of the impact of a stigma reduction training and woman-focused intervention that can improve PrEP uptake, adherence, and persistence among AGYW. 1346 Impact of CAB-LA Versus Oral PrEP in Reducing HIV Incidence Among MSM in Thailand: A Modelling Study David Van de Vijver 1 , Artit Wongsa 2 , Jeroen van Kampen 1 , Haoyi Wang 3 , Kai J. Jonas 3 , Rena Janamnuaysook 2 , Stephanie Popping 4 , Nittaya Phanuphak 2 1 Erasmus University Medical Center, Rotterdam, Netherlands, 2 Institute of HIV Research and Innovation, Bangkok, Thailand, 3 Maastricht University, Maastricht, Netherlands, 4 Academic Medical Center, Amsterdam, Netherlands Background: Long acting preexposure prophylaxis (PrEP) can address periods of partial adherence among individuals using oral PrEP. Using PrEP dispensing records, we determined the proportion of partially adherent men-who-have sex-with-men (MSM) in Thailand who return within ten weeks and could benefit from cabotegravir as long-acting PrEP (CAB-LA). We then used a mathematical HIV transmission model to study the epidemiological impact of CAB-LA in Thai MSM. Methods: PrEP dispensing records were analyzed from 12 key population-led clinics in Thailand (2019-2022). We included individuals whose first visit was at least one year before the last recorded visit in 2022. For those who received three or more PrEP prescriptions, we calculated the ratio of days covered by PrEP (assuming daily use) to the total days between prescriptions. A ratio >75% indicated high adherence to oral PrEP (effectiveness of 93%), and a ratio <75% was classified as partial adherence (effectiveness 60%, Jourdain Lancet Public Health 2022). Individuals who returned within 10 weeks were assumed protected by CAB-LA (34% higher effectiveness). Based on literature, we modeled the tail phase of CAB-LA, estimating reduced effectiveness of 0-26%, 49-76%, and 82-95% during the first, second, and third 12-week periods following the last CAB-LA injection. An HIV transmission model was calibrated to the Thai MSM HIV epidemic, including a discrete choice experiment showing that 14% prefer CAB-LA. As the number of PrEP users is unknown, we estimated this population to range between 10,000 and 100,000 in 2024-2034. Results: A number of 34,843 PrEP prescriptions were dispensed to 7,257 MSM, with common discontinuations (45% did not return within six months after start). Among individuals filling at least 3 prescriptions, 52.6% had periods of partial adherence to oral PrEP that could not be covered by CAB-LA, 14.0% had periods of partial adherence to oral PrEP that could be covered by CAB-LA, and 33.0% were consistently protected by both PrEP modalities. Compared to a scenario when only oral PrEP is available, adding CAB-LA will avert between 323 (0.7%, 10,000 users in 2024-2034) and 1316 (4.4%, 100,000) new infections in the next decade. The incidence of HIV in 2034 is expected to decrease in the next decade by 1.3% and 9.1% for 10,000 and 100,000 PrEP users, respectively. Conclusions: Although discontinuation of PrEP is a challenge, offering both CAB-LA and oral PrEP as HIV prevention options can further reduce HIV incidence.

Poster Abstracts

CROI 2025 449