Abstract A26: Towards the identification of novel tumor antigens in human lung cancer

Abstract

Abstract Immunotherapy has changed the treatment paradigm for patients with lung cancer and other solid tumors, but little is known about the antigen specificities of the T cell residing in tumors. In order to identify novel shared tumor antigens, we have previously developed an algorithm that allows the inference of T-cell specificities based on shared motifs within the T-cell receptor (TCR) CDR3 sequences. Leveraging with a dataset of 789K CDR3 sequences from 178 HLA-typed non-small cell lung cancer (NSCLC) patients, we have found up to 40K motifs after applying stringent cutoffs. Publicly available CDR3 sequences from various tetramer databases assisted with robust assignments of unique T-cell clones from the NSCLC cohorts to known specificities. For example, we have identified and validated multiple tumor-infiltrating T-cell clones that are specific to viral antigens. Furthermore, the availability of HLA genotype results for our lung cancer cohorts facilitated prediction of motifs that are potentially “restricted” to a given HLA allele or supertype. Together with single cell-sorted T-cell clones from our Stanford NSCLC cohort (n = 5,448), we have narrowed down to 60 tumor-specific CDR3 motifs (wherein at least one Stanford sequence could be found) that are potentially restricted to the A02 allele. Within these top A02-enriched motifs, 49 candidates TCR alpha/beta pairs were identified. In an effort to find novel tumor antigens, we prioritized four top T-cell candidates and screened A02 yeast display libraries with distinct peptide length (8-11mer). Four rounds of repeated screens led to the enrichment of a few dominant mimotopes with similar sequences. Database search identified a near-perfect match to a peptide 9mer from the TMEM161A locus that is predicted to bind HLA-A02 with high affinity. Functionally, the synthetic TMEM161A peptide 9mer stimulated Jurkat cells expressing the specific TCR alpha/beta chains as well as the top mimotopes. In conclusion, our results highlight the robustness of our current pipeline for the identification of novel shared tumor antigens. The ongoing work involves further characterizing the functional state of these specific T cells within the tumor microenvironment. Citation Format: Shin-Heng Chiou, Diane Tseng, Xinbo Yang, Alexandre Reuben, Alana McSween, Julie Wilhelmy, Chunlin Wang, Joseph Shrager, Jianjun (Jay) Zhang, K. Christopher Garcia, Crystal Mackall, Mark M Davis. Towards the identification of novel tumor antigens in human lung cancer [abstract]. In: Proceedings of the AACR Special Conference on Tumor Immunology and Immunotherapy; 2019 Nov 17-20; Boston, MA. Philadelphia (PA): AACR; Cancer Immunol Res 2020;8(3 Suppl):Abstract nr A26.