Host-cell proteins (HCPs), such as hamster phospholipase B-like 2 (PLBL2) and lysosomal phospholipase A2 (LPLA2), that co-purify with monoclonal antibodies (mAbs) are of significant concern because of their potential impacts on formulation stability and immunogenicity. However, such enzymes display highly variable and difficult-to-predict propensities to co-purify with different antibody therapeutics. Therefore, we have developed an integrated experimental and computational approach to address this challenging problem. Our approach involves generating large libraries of antibody mutants displayed on the surface of yeast, sorting them for high and low levels of interactions with host-cell enzymes, deep sequencing the enriched antibody libraries, and developing models for predicting mutations that suppress antibody binding to host-cell enzymes. We will discuss our findings of the most important molecular features that mediate interactions between antibodies and enzymes, and the implications of our findings for improving antibody drug development.

Experimental and computational analysis of monoclonal antibody interactions with host-cell proteins

ACS Spring 2022

The official archive and repository for all ACS Meetings & Expositions abstracts and select posters or presentations.