A comparison of F/TAF and F/TDF as HIV pre-exposure prophylaxis in a predominantly cisgender women population in East and South Africa: a randomised, factorial, non-inferiority trial

Background: F/TAF was shown to be noninferior to F/TDF as pre-exposure prophylaxis (PrEP) in men, but approval was not extended to cisgender women. We report the results of PrEPVacc, in which a predominantly female population was randomly allocated to receive daily oral F/TDF or F/TAF for ∼6 months within a HIV-1 prophylactic vaccine trial. Setting: Four study sites in 3 African countries (Uganda, Tanzania, South Africa). Methods: The 2 regimens were compared by the averted infections ratio (AIR)—the proportion of infections averted by F/TAF relative to F/TDF. The counterfactual HIV incidence, an essential component of this metric, was derived from a preceding registration cohort. Dried blood spots (DBS) were collected at regular time points for later assessment of tenofovir diphosphate levels in selected subpopulations. Results: 1380 participants (697 F/TDF, 683 F/TAF) were included in the primary analysis (total follow-up 709.2 person-years); 87% were cisgender women. Three HIV infections (0.86/100 person-years) occurred in the F/TAF group versus 2 in the F/TDF group (0.56/100 person-years). The counterfactual HIV incidence was estimated to be 2.59/100 person-years (90% CI: 1.86 to 3.52), giving an AIR of 0.85 (90% CI: 0.31 to 1.66). Based on the week 8 DBS sample, only an estimated 14% of participants were classified as taking 2–3 tablets per week and 9% ≥4 tablets per week. Conclusions: Despite similar HIV incidence rates, the noninferiority of F/TAF was not demonstrated, probably because of low statistical power primarily driven by low adherence. However, there is compelling evidence from multiple studies supporting the efficacy of F/TAF as PrEP regardless of sex.