Comprehensive genomic alterations identification and potential shared neoantigens for non-small cell lung cancer immunotherapy.

Abstract

e20582 Background: Despite advances in therapies, the development of drug resistance in non-small cell lung cancer (NSCLC) poses a significant challenge. The emergence of neoantigen-based therapies offer a promising new frontier which enhances treatment efficacy and reduces associated costs. Methods: A total of 24 patients diagnosed with NSCLC were enrolled in this study. Primary tumor tissues and peripheral blood samples were collected from each patient. Paired whole exome and transcriptomic sequencing data were performed to identify somatic mutations and to quantify their respective expression levels. Somatic single nucleotide variants (SNVs) and small insertions or deletions (inDels) were detected using Strelka algorithm from whole exome sequencing (WES) data. Neoantigen prediction focused on human leukocyte antigens (HLA) type 1. HLA alleles were predicted using WES data encoding adjacent non-cancerous tissue samples, identifying four alleles that were present in at least 50% of the patients. Neoantigens were considered potentially immunogenic if their predicted median IC50 binding scores were ≤500nM while expressed transcripts per million (TPM) exceeds 1 in tumor samples. Results: Using systematic bioinformatics pipeline, we analyzed 24 NSCLC samples with matched tumor tissues and peripheral blood, and detected the 4677 mutations involving 2442 genes, including 2819 missense SNVs, 941 InDels, 229 gene fusion and 688 RNA alternative splicing events, 94.7% of which (4428 mutations involving in 2247 genes) were common with those reported in the TCGA-LUAD dataset. The average number of mutations was 194.8 in NSCLCs. We found that C>T/G>A transitions/ transversions were dominant. Missense mutations were the most frequent types of somatic mutation in the coding sequence regions. 98% of mutated genes were not detected simultaneously with DNA and RNA mutations. ADCK5 , CD151 and EIF3K were identified as potential new driver genes. Genomic variation profiling identified 2213 potential neoantigens with a median IC50 binding score of ≤500nM and a transcript expression level of >1 TPM. Among these, 41 were listed in databases TSNAdb, IEDB, and CTDatabase. A total of 14 genes harboring identical mutations were detected, with the identical mutations observed in over 50% of the patient cohort. Notably, mutations in 14 genes were either gene fusions or RNA alternative splicing events, and core antigenic sequences generated were found to be identical in the majority of patients. Of particular interest is the RPRD2 gene, where an RNA exitrons site was detected in all patient samples, along with the identification of the same neoantigen core sequence. Conclusions: Our analysis suggests that gene fusions and RNA splicing variations are more likely than point mutations to generate shared antigens, underscoring their importance for developing targeted immunotherapies.