CROI 2025 Abstract eBook

Abstract eBook

Poster Abstracts

1305 Comparing 2 mHealth Technologies to Support PrEP Use Among MSM and Transgender People in the US Albert Y. Liu 1 , David Glidden 2 , Juwann Moss 3 , Cat-Dancing Alleyne 1 , Janie Vinson 1 , Janet Myers 2 , Kimberly Koester 2 , Stefani Butts 4 , Rikki Montoya 5 , Tommy Sanchez 6 , Jorge Roman 5 , Maria Rybicki-Newman 6 , Susanne Doblecki-Lewis 7 , Hyman Scott 1 , Susan Buchbinder 1 1 San Francisco Department of Public Health, San Francisco, CA, USA, 2 University of California San Francisco, San Francisco, CA, USA, 3 Bridge HIV, San Francisco, CA, USA, 4 University of Miami, Miami, FL, USA, 5 San Francisco AIDS Foundation, San Francisco, CA, USA, 6 Whitman-Walker Health, Washington, DC, USA, 7 University of Miami Miller, Miami, FL, USA Background: PrEP adherence and persistence are suboptimal in real-world settings, with half to two thirds of PrEP users discontinuing within 6 months. We describe results of a comparative effectiveness trial of two mobile health technologies to support PrEP use among MSM and transgender people in the US. Methods: HIV-uninfected MSM and trans people were enrolled across 3 real world PrEP delivery sites (a sexual health clinic in San Francisco, a municipal STD clinic in Miami, and an LGBT health center in Washington DC). Participants were randomized 1:1 to receive PrEPmate, a bidirectional texting platform using personalized communication through weekly check-in messages or DOT Diary, a mobile app promoting self-management of PrEP use through a pill-taking/ sexual diary with feedback on PrEP protection levels (Figure). PrEP persistence was measured as having sufficient PrEP medication prescribed through 6 months, and PrEP adherence was measured by self-report at 6 months and having protective tenofovir diphosphate (TFV-DP) levels ≥700 fmol/punch in dried blood spots collected at 6 and 12 months (preliminary results available in a subset (n=61)). Effects of treatment assignment on PrEP outcomes were assessed using mixed effects logistic models. Results: From 7/2021-1/2024, we enrolled 296 participants. Mean age was 30.5 (IQR 26-37), 55% were Hispanic/Latinx, 16% African American; 91% were cisgender male, 9% identified as trans/non-binary; 31% were un-insured, 34% reported limited/marginal health literacy, and 39% were newly starting PrEP. The median number of sex partners in the past 3 months was 5; 55% reported recreational drug use; and 34% reported heavy alcohol use. PrEP persistence at 6 months was 77% in the PrEPmate vs. 78% in the DOT Diary arms (OR 0.92 95% CI 0.53-1.59, p=0.77). Median self-reported adherence was 90% in both arms (p=0.44). Protective TFV-DP was observed in 56% vs. 76% of samples in the PrEPmate vs. DOT Diary arms (p=0.17) (all on daily oral TDF/FTC). Acceptability of both tools was high, with >70% reporting they would use PrEPmate or DOT Diary in the future, and >75% would recommend them to friend. Conclusions: In this diverse, sexually active cohort, over three-quarters of participants were persistent on PrEP at 6 months, with similar rates across arms. Self-reported adherence was high in both groups, with a preliminary trend towards higher protective TFV-DP in the DOT Diary arm. These data provide evidence for these tools to support PrEP use in real-world settings.

not be preferred or feasible. In the SEARCH Dynamic Choice Prevention (DCP) studies, participants at risk for HIV could choose between oral PrEP, PEP, CAB-LA, and the DVR. We describe characteristics of ring users, reasons for choosing the DVR, and adherence patterns in the first 3 months of ring use. Methods: The SEARCH DCP intervention offers choice of biomedical prevention options with flexibility to switch between options over time guided by changing HIV risk and preference. The DVR was added as an option in 2 ongoing studies in rural Uganda and Kenya1) oral PrEP, PEP, CAB LA, DVR (NCT 05549726) and 2) oral PrEP, PEP, DVR (NCT 04810650). Participants were seen in the clinic at DVR initiation and every 12 weeks during follow-up, with HIV testing, and monthly reminder phone calls. Participants had the option to return to the clinic monthly for ring insertion until comfortable with self-care. Results: Of 1560 women in the intervention arms of the two studies, 34 (2%) chose and used the Dapivirine ring during follow-up. 21(62%) were from Uganda, and 7 (21%) were aged 15-24. Prior to selecting the ring, 18 (53%) had ever used PrEP, 2 (6%) had ever used PEP, 6 (18%) had ever used CAB-LA, and 15 (44%) had never used any biomedical prevention option. At initiation, 21 (62%) chose the DVR due to experienced or anticipated pill burden, 8 (23%) initiated the ring because it is discrete, 4 (12%) preferred CAB-LA but were ineligible, and 1(3%) chose the ring because they would forget to take pills. Following the baseline visit, 30/34(88%) women self-inserted the ring at week 4. The remaining 5 (14%) had the ring inserted by a health care provider. Of 25 who reached at least 3 months since ring initiation, 21 (84%) changed their rings on schedule, 4 discontinued rings due to menorrhagia (1), change in risk (1), or travel outside the community (2). Conclusions: Among women offered prevention options in a Dynamic Choice model that included oral PrEP, PEP, and CAB LA, adding the Dapivirine Ring as an option was feasible and a preferred option in 2% of women to date. The Ring provided an alternative to some women who had not chosen any of the other biomedical prevention options previously offered. 1307 Machine Learning to Predict Persons With HIV at Risk of Falling Out of Care Ravi Goyal, Gordon Honerkamp-Smith, Alan Wells, Susan J. Little, Thomas Martin University of California San Diego, La Jolla, CA, USA Background: Effective antiretroviral therapy to maintain durable viral suppression is key to ending the HIV epidemic in the United States. We evaluated the ability of machine learning (ML) algorithms to predict people with HIV (PWH) who are at risk of falling out of care. Methods: This was a retrospective study among all PWH in San Diego County. The study used mandatorily reported de-identified public health HIV data between 2017 and 2022. The data was partitioned temporally into two training (training and refinement datasets) and one validation dataset to accurately predict performance for predicting future observations. The outcome of interest was developing unsuppressed viral load (>200 copies/ml) during a year-long prediction window. Variables included socio-demographic data, baseline HIV infection information including stage and transmission risk, and longitudinal HIV viral load and CD4 data. HIV viral load testing patterns and trends were incorporated into the analysis. A random forest model was used to generate outcome predictions for the overall population and was stratified by race. Mediation analyses were undertaken to assess underlying causality. Results: A total of 18,919 PWH were included in the analysis from which 13.1%, 13.0% and 12.1% had unsuppressed viral load during the observation window in the training, refinement and validation datasets respectively. The model had an area under the receiver operating characteristic (AUROC) curve of 82.2 (95%CI 79.3-85.0), a sensitivity of 33.8% (95%CI 28.6-39.0), and specificity of 96.9% (95%CI 95.7-97.2) corresponding to a positive predictive value of 55.7% (95%CI 48.7-62.8) and negative predictive value of 91.7% (95%CI 90.6-92.8). The AUROC was similar across race/ethnicities. Prior viral load testing pattern/trends such as linear trend, proportion of unsuppressed viral loads, and slope of viral load were identified as the most important variables by the model; however, they partially acted as mediators of underlying demographic (e.g., race) and HIV infection risk (e.g., injection drug use). Relative importance of variables and outcome prediction by race are shown in Figure 1 . Conclusions: ML algorithms using mandatory reported public health HIV data can predict which PWH will have future unsuppressed viral load and would be easily generalizable to other U.S. jurisdictions. Future work will assess its

Poster Abstracts

The figure, table, or graphic for this abstract has been removed.

1306 Characteristics of Dapivirine Ring Users in the SEARCH Dynamic Choice HIV Prevention Studies Elijah Kakande 1 , Nicole Sutter 2 , Laura Balzer 3 , Douglas Black 2 , Jane Kabami 1 , James Ayieko 4 , Gabriel Chamie 2 , Helen Sunday 1 , Marilyn Nyabuti 5 , Colette Aoko 4 , Catherine A. Koss 2 , Maggie Czarnogorski 6 , Maya Petersen 3 , Moses Kamya 7 , Diane V. Havlir 2 , for the SEARCH Collaboration 1 Infectious Diseases Research Collaboration, Kampala, Uganda, 2 University of California San Francisco, San Francisco, CA, USA, 3 University of California Berkeley, Berkeley, CA, USA, 4 Kenya Medical Research Institute, Kilifi, Kenya, 5 KEMRI Kenya, Nairobi, Kenya, 6 ViiV Healthcare, Brentford, UK, 7 Makerere University College of Health Sciences, Kampala, Uganda Background: The Dapivirine ring (DVR) is effective for HIV prevention among women at risk for HIV, and an alternative when other biomedical options may

CROI 2025 434