Abstract Body

HIV incidence is high among young black MSM (YBMSM) in the US, and effective implementation of pre-exposure prophylaxis (PrEP) has great potential to reduce new infections. Scale up of PrEP is ongoing in this key population; yet, we continue to observe high HIV incidence (interim estimate 6%/year) in our cohort of YBMSM with access to PrEP services. A better understanding of patterns of PrEP persistence and discontinuation among YBMSM is needed.

The EleMENt study is an observational cohort examining relationships between substance use and HIV risk behavior among HIV-negative YBMSM (n=299) aged 18-29 in Atlanta, GA. All participants were offered optional PrEP at each study visit over the 24-month follow-up. Clinical visits, labs, transportation, and navigation services for manufacturer assistance plans (MAP) to obtain no/low cost TDF/FTC were provided by the study. For initiators, we recorded time on and off PrEP based on frequent study surveys, prescription records, dates of MAP approvals, counseling notes, and other participant contacts. PrEP discontinuation events were defined as a ≥2 week lapse in PrEP use. Time to first PrEP discontinuation was assessed with the Kaplan-Meier method, with a Cox proportional hazard model used to identify factors associated with discontinuation.

After 483 person-years of follow up, 42% (125/299) of YBMSM initiated PrEP through the EleMENt program. Overall, PrEP initiators were ‘on PrEP’ for 69% of possible person-time after initiation. 63% (79/125) discontinued PrEP at least once during study follow-up, and 68% of discontinuers (54/79) subsequently restarted PrEP. 22% (27/125) discontinued two or more times. The median time to first PrEP discontinuation was 219 days (95% CI 181-280). In a multivariable model, marijuana use (adjusted hazard ratio [aHR] 2.07, 95% CI 1.24-3.47), age <22 years (aHR 3.63, 95% CI 1.95-6.74) and having fewer than 3 sex partners (aHR 2.16, 95% CI 1.30-3.58) were associated with PrEP discontinuation.

Persistent PrEP coverage in this cohort of YBMSM was suboptimal and discontinuations, including multiple discontinuations, were common despite additional support services available through the study. Interventions to support PrEP persistence, especially for younger and substance using YBMSM, will be necessary to achieve full effectiveness of PrEP. For the future, regimens that do not require adherence to a daily medication could help facilitate PrEP persistence in this key population.