Abstract Body

PrEP with TDF/FTC is highly effective for preventing HIV in diverse populations. Among MSM, stimulant use and condomless anal sex with multiple partners (CAS-MP) are known risk factors for HIV acquisition. CAS-MP has been associated with increased PrEP adherence, but less is known about the association with stimulant use. We hypothesized that stimulant use would modify the effect of CAS-MP, decreasing PrEP adherence.

We performed a secondary analysis of PATH-PrEP, a two-site open label study evaluating PrEP for MSM in Los Angeles, California. TDF/FTC was offered to 296 MSM for 48 weeks between April 2014 and July 2016. Adherence levels to TDF/FTC were assessed via TFV-DP levels in dried blood spots (DBS) and TFV in plasma samples at 4, 8, 12, 24, 36 and 48 weeks. All participants received standardized adherence counseling, which was escalated if TFV plasma levels were below the limit of quantitation (<10 ng/mL). Risk behavior was assessed through a computerized survey instrument at all visits. We modeled an interaction between stimulant use (methamphetamine, ecstasy, or cocaine) and CAS-MP (2 or more partners) over time on protective TFV-DP levels in DBS (>700 fmol/punch). We included individuals with at least one follow-up visit (n=283). We used a generalized linear mixed model, controlling for age, ethnicity, enrollment site, education, income, and sex work. Predicted probabilities were estimated to plot the interaction and 95% CIs (figure).

Median age was 33 (IQR 28-42), 51% were White, 28% Latino and 10% Black. At baseline, 61% reported CAS-MP and 34% reported stimulant use in the prior 30 days. Over all observed time points, 80% of samples had protective levels. At week 4, those reporting CAS-MP without stimulant use had higher odds of protective levels (AOR 2.4, 95% CI 1.3-4.6) than non-stimulant and non-CAS-MP (reference). Those with CAS-MP and stimulant use had decreased odds of protective levels (AOR 0.2, 95% CI 0.1-0.7). Over time, those reporting CAS-MP without stimulant use did not have a significant change in odds of protective levels (AOR 1.00, 95% CI 0.96-1.03), while those who reported CAS-MP and stimulant use had increased odds of protective levels (AOR 1.07 per week, 95% CI 1.01-1.13).

Stimulant use moderated the effect of CAS-MP on adherence over time, increasing the odds of protective levels of PrEP, contrary to our initial hypothesis. Stimulant use should not be a deterrent to prescribe PrEP to high-risk individuals engaging in CAS-MP.