T cell receptor repertoire analysis of tumor-infiltrating T cells in non-small cell lung cancer [tet_sorted_PBLT_cells]

We analyzed 778,938 T cell receptor (TCR) CDR3b sequences from 178 non-small cell lung cancer (NSCLC) patients using GLIPH2 (Grouping of Lymphocyte Interactions with Paratope Hotspots 2). We identified over 66,000 TCR shared specificity groups, of which over 400 were enriched in tumor vs. adjacent lung. We focused on one tumor-enriched specificity group and used a yeast display library to identify the peptide epitopes. These include peptides from TMEM161A, which is an overexpressed antigen in NSCLC, and homologous peptides from Epstein-Barr virus (EBV) and E. coli, aligning with recent reports of virus-specific T cells in tumor infiltrates. Overall, we analyzed a very large dataset of TCR sequences from NSCLC patients and found thousands of specificities shared by three or more patients. This creates a database of TCR shared specificity groups in NSCLC, and also a template for mining and understanding the specificity of infiltrating T cells for any cancer. Single-cell RNA-Seq data collected with FACS sorted T cells from PBL of HLA-A*02+, healthy donors and HLA-A*02+ lung cancer patients.