Optimizing clinical dosing of combination broadly neutralizing antibodies for HIV prevention

Abstract

Broadly neutralizing antibodies are promising agents to prevent HIV infection and achieve HIV remission without antiretroviral therapy (ART). As learned from effective ART, HIV viral diversity necessitates combination antibody cocktails. Here, we demonstrate how to optimally choose the ratio within combinations based on the constraint of a total dose size. Optimization in terms of prevention efficacy outcome requires a model that synthesizes 1) antibody pharmacokinetics (PK), 2) a mapping between concentration and neutralization against a genetically diverse pathogen (e.g., distributions of viral IC50 or IC80), 3) a protection correlate to translate in vitro potency to clinical protection, and 4) an in vivo interaction model for how drugs work together. We find that there is not a general solution, and the optimal dose ratio likely will be different if antibodies cooperate versus if both products must be simultaneously present. Optimization requires trade-offs between potency and longevity; using an in silico case-study, we show a cocktail can outperform a bi-specific antibody (a single drug with 2 merged antibodies) with superior potency but worse longevity. In another practical case study, we perform a triple antibody optimization of VRC07, 3BNC117, and 10-1074 bNAb variants using empirical PK and a pre-clinical correlate of protection derived from animal challenge studies. Here, a 2:1:1 dose emphasizing VRC07 would optimally balance protection while achieving practical dosing and given conservative judgements about prior data. Our approach can be immediately applied to optimize the next generation of combination antibody prevention and cure studies.