Preferences for oral and injectable PrEP among qualitative sub-study participants in HPTN 084

Background HPTN 084 compared the safety and efficacy of long-acting injectable cabotegravir (CAB) to daily oral TDF/FTC for prevention of HIV-1 in uninfected African women. Like a similar trial in MSM/TGW (HPTN 083), the trial was stopped early for efficacy, expediting the need to consider introduction strategies for different populations. We examine survey and qualitative data from a four-country sub-study to examine oral and injectable PrEP acceptability and considerations for CAB access among African women. Methods Participants completed baseline and follow-up surveys on HIV risk perception, sexual behavior. product acceptability and adherence during the blinded trial. Additionally, up to two in-depth interviews each with 73 sub-study participants explored product use and trial-related experiences, during the blinded and unblinded study periods. Using survey data, we classified participants as: engaged in female sex work (FSW), having multiple non-transactional partners, or monogamous. A study statistician identified participants’ assigned study arm. We followed a thematic analysis process to read transcripts, develop a codebook and apply codes in NVivo to transcripts with intermittent intercoder reliability checks; using Excel matrices to explore differences across risk categories and study arms. Findings Participants overwhelmingly preferred injections to pills, appreciating the ease, convenience, and privacy of a long-acting formulation. Many participants described challenges with contraceptive and/or study pill adherence, impeded by late night work, unexpected travel, or heavy drinking. Women in the TDF/FTC arm were more likely to describe side effects, compared to those in the CAB arm. Pain also varied widely by study arm. When considering post-trial access to CAB, limited PrEP knowledge, cost and concerns around stigma and poor service quality were potential access barriers. Conclusion Women’s desire for privacy and ease of use outweighed injectable concerns, resulting in a strong preference for CAB. Cost and accessibility will need to be addressed by implementation programs.