A unified peptide array platform for antibody epitope binning, mapping, specificity and predictive off-target binding

Abstract Therapeutic antibody efficacy is largely determined by the target epitope. In addition, off-target binding can result in unanticipated side-effects. Therefore, characterization of the epitope and binding specificity are critical in antibody discovery. Epitope binning provides low-resolution of an antibody epitope and is typically performed as a cross-blocking assay to group antibodies into overlapping or non-overlapping bins. Epitope mapping identifies the epitope with high resolution but requires low throughput methods. In addition to binning and mapping, there is a need for a scalable and predictive approach to reveal off-target binding early in antibody discovery to reduce the risk of in vivo side effects. Peptide microarrays are an information-rich platform for antibody characterization. However, the potential of peptide microarrays in early-stage antibody discovery has not been realized because they are not produced at the scale, quality and format needed for reliable high-throughput antibody characterization. A unified, peptide library platform for high-resolution antibody epitope binning, mapping and predictive off-target binding characterization is described here. This platform uses highly scalable array synthesis and photolithography to synthesize more than 3 million addressable peptides. These arrays conform to a microplate format and each synthesis is qualified with mass spectrometry. Using this platform, a scalable approach to early-stage epitope and specificity characterization, with prediction of off-target interaction(s), is demonstrated using a panel of anti-HER2 monoclonal antibodies. This study highlights the prospect of this platform to improve antibody discovery productivity by generating epitope and specificity information much earlier with potentially hundreds of antibody clones.