Abstract

3087 Background: While there has been much attention on mutation-associated neoantigens in tumors, there is less known about non-mutated tumor antigens that are shared across individuals. Understanding tumor-infiltrating T cell recognition of shared tumor antigens is important for understanding cancer immune recognition and escape, and may reveal novel targets for therapy. Methods: We have established a novel approach for discovering shared tumor antigens in human lung cancer. This approach involves identifying candidate T cell receptor (TCR) alpha/beta pairs that are predicted to exhibit specificity for shared tumor antigens in the context of a given human leukocyte antigen (HLA). We then screen the T cell receptor for binding to yeast display libraries of peptide-HLA. The Mark Davis lab at Stanford has previously developed an algorithm that groups T cell receptors into antigen specificity groups based on shared motifs within the TCR complementarity-determining region 3 (CDR3) sequences. Leveraging a dataset of over 700K CDR3 sequences from 178 HLA-typed non-small cell lung cancer (NSCLC) patients, we have found up to 4,300 antigen specificity groups after applying stringent cutoffs. We sequenced TCR alpha/beta pairs from 15 patients with lung adenocarcinoma (n = 4,705). Results: We identified an antigen specificity group enriched in tumor compared to adjacent uninvolved lungs. Antigen screening of the T cell receptor belonging to this specificity group using an A02 yeast display libraries led to the identification of a dominant peptide after four rounds of enrichment. We functionally validated that the peptide derived from the protein TMEM161A stimulated Jurkat cells expressing the TCR alpha/beta receptor of interest. We show that full-length TMEM161A protein is processed and presented into a peptide that stimulates primary T cells expressing the TCR alpha/beta receptor. We observe that a peptide from Epstein-Barr virus (EBV) protein LMP2 also stimulated the same TCR alpha/beta receptor. We have show that TMEM161A RNA and protein are overexpressed in human lung cancer compared to adjacent uninvolved lungs. Conclusions: We have demonstrated a novel approach toward antigen discovery and identified a shared tumor antigen TMEM161A in human lung cancer.

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call