Statistical issues in trials of preexposure prophylaxis

PURPOSE OF REVIEW We discuss selected statistical issues in the design and analysis of preexposure prophylaxis (PrEP) trials. The general principles may inform thinking for other interventions in HIV prevention. RECENT FINDINGS To date, four different designs have been used to determine the effectiveness of PrEP: randomized, double-blind, placebo-controlled; randomized, open-label, immediate or delayed access; nonrandomized comparison of HIV incidence according to the level of drug detected; comparison of the observed HIV incidence to the expected rate using historical control data. Open-label trials of PrEP, which assess public health effectiveness, complement the placebo-controlled trials which established the biological efficacy of TDF/FTC. Future trials of PrEP will be highly challenging to design since a no PrEP group is difficult to justify and the natural control regimen, TDF/FTC, is highly efficacious. SUMMARY Standard statistical paradigms for noninferiority trials should be reconsidered for evaluating alternative PrEP regimens.