A multifaceted approach for identification, validation, and immunogenicity of naturally processed and in silico-predicted highly conserved SARS-CoV-2 peptides.

SARS-CoV-2 remains a major global public health concern. Antibody waning and immune escape variant emergence necessitate the development of next generation vaccines that induce cross-reactive durable immune responses. T cell responses to SARS-CoV-2 demonstrate higher conservation, antigenic breadth, and longevity than antibody responses. Therefore, we sought to identify pathogen-derived T cell epitopes for a potential peptide-based vaccine. We pursued an approach leveraging: 1) liquid chromatography and tandem mass spectrometry (LC-MS/MS)-based identification of peptides from ancestral SARS-CoV-2-infected cell lines, 2) epitope prediction algorithms, and 3) overlapping peptide libraries. From this strategy, we identified 380 unique SARS-CoV-2-derived peptide sequences, including 53 antigenic HLA class I and class II peptides from multiple structural and non-structural/accessory viral proteins. These peptide sequences were highly conserved across variants of concern/interest (VoC/VoIs), and are estimated to achieve coverage of >96% of the world population. Our findings validate this discovery pipeline for peptide identification and immunogenicity testing, and are a crucial step toward the development of a next-generation multi-epitope SARS-CoV-2 peptide vaccine, and a novel vaccine platform methodology.