Tomorrow’s HIV Prevention Trials of Vaccines and Antibodies

Abstract

Abstract Effective HIV prevention has the potential to change the landscape of HIV prevention trials. Low infection rates will make superiority studies necessarily large while non-inferiority trials will need some evidence that a counterfactual placebo group had a meaningful HIV infection rate in order to provide evidence of effective interventions. This paper explores these challenges in the context of immune related interventions of mAbs and vaccines. We discuss the issue of effect modification in the presence of PrEP, where subjects on PrEP may have less of a benefit of a mAb or (vaccine) than subjects off PrEP. We also discuss different methods of placebo infection rate imputation. We estimate infection risk as a function of mAb level (or vaccine induced immune response) in the mAb (or vaccine) arm and then extrapolate this infection risk to zero mAbs as a proxy for the placebo infection rate. Important aspects are the use of triangulation or multiple methods to impute the placebo infection rate, concern about extrapolation if few mAbs are close to zero, and the use of currently available data with placebo groups to rigorously evaluate the accuracy of imputation methods. We also discuss use of historical controls and some generalizations of the idea of (DMurray, J. 2019. “Regulatory Perspectives for Streamlining HIV Prevention Trials.” Statistical Communications in Infectious Diseases.) to use rectal gonorrhea rates to impute HIV infection rate. Generalizations include regression adjustment to calibrate for potential differences in baseline covariates for ongoing vs historical studies and the use of the gonorrhea, HIV relationship in a contemporaneous observational study. Examples of recent and ongoing trials of malaria chemoprophylaxis and HPV vaccines, where extremely effect prevention methods are available, are provided.