Using genetic data to predict clinical outcomes in emergent antibody therapies against HIV.

Rapidly evolving pathogens such as tuberculosis, HIV, and influenza, pose a serious risk to global health. New interventions are in constant development, but just as quickly resistant traits emerge and become fixed in the pathogen population, rendering interventions ineffective. To frustrate the emergence of resistance, one can combine several interventions together. This combined approach underpins the success of modern HIV antiretroviral therapy and frontline tuberculosis treatment.

The principle behind combination therapy is clear: It is harder for a pathogen population to acquire resistant traits against multiple treatments than to acquire resistance to each treatment seperately. Can we quantify the clinical advantage of combination therapy? Using novel inference methods, we successfully predict patient outcome with emergent antibody treatments in both monoclonal and combination therapy trials. This analysis reveals the biology governing viral rebound and the features required for successful multivalent treatment. We can use this analysis to determine how many different antibodies would be needed to successfully prevent the emergence of HIV resistance. These results quantify both the promise and the limits of antibody-combination therapy against HIV.