Applying immunopeptidomics and machine learning to improve neoantigen prediction for therapeutic and diagnostic use
Neoantigens are increasingly critical in immuno-oncology as therapeutic targets for neoantigen-based personalized cancer vaccines (PCVs) and as potential biomarkers for immunotherapy response. However, optimizing technologies for identifying neoepitopes that are more likely to provoke an immune response remains an important challenge. Current major histocompatibility complex (MHC) presentation prediction algorithms are primarily trained using in vitro MHC binding data, which does not encompass certain important factors for neoantigen presentation such as proteasomal cleavage and transport. Recent advances in immuno-anity purification and mass spectrometry technology make it possible to identify processed cell surface MHC bound peptides in an in vivo setting, providing the opportunity for the development of improved neoantigen prediction pipelines.