Background
Broadly neutralizing antibodies (bnAbs) are among the most promising agents in HIV-1 treatment and prophylaxis. With numerous clinical trials across the world seeking to evaluate application of potent bnAbs, the issue of viral escape and resistance is pressing and important. However, thus far, our understanding of bnAb resistance pathways among diverse HIV-1 clades remains limited. Our study addresses these limitations with the development of medium-high throughput assays that allow the simultaneous analysis of multiple HIV-1 strains for their ability to escape bnAb neutralization.
Methods
We developed cell-culture assays and computational and experimental validation pipelines, for the selection and identification of bnAb resistance mutations. We have conducted an unprecedented number (over 7,000) of resistance-selection experiments across 20 strains representing HIV-1 global diversity against 2 bnAbs in clinical development; the CD4-binding site antibody 3BNC117 and the V3-glycan dependent antibody 10-1074. We isolated hundreds of viruses that replicated in the presence of a bnAb and each virus isolate was phenotyped and sequenced. Individual mutations predicted to confer resistance were tested via pseudotype neutralization assays against the bnAb used for selection and additional bnAbs.
Results
We found substantial variability among HIV-1 isolates in their propensity to escape bnAbs, as well as in the identity of mutations, and mechanisms, that conferred bnAb resistance. Our data reveal that bnAb escape is readily achieved with single amino acid substitutions, for the majority of HIV-1 strains, but resistance mutations for a single antibody can be numerous and highly dependent on the strain context. Unconventional mechanisms of escape were also uncovered for some strains. Some of the resistance mutations were highly specific to the selecting bnAb while others were able to confer resistance to at least one other bnAb targeting an overlapping epitope.
Conclusions
Our study identifies previously unappreciated pathways of bnAb escape and reveals that the genetic barrier to bnAb escape is generally low. These findings have direct implications for bnAb therapeutic application; the great clinical promise of bnAbs risks being compromised by injudicious deployment. Nevertheless, approaches such as those documented in our study can identify combinations of bnAbs with a higher barrier to escape that could guide strategies to minimize the development of resistance.