Abstract

Abstract Tumor evolution is dictated by the genetic, epigenetic and transcriptional landscape of the ‘cell-of-origin’ or the tumor initiating cell. While the genomic landscape of cancers has been extensively profiled, efforts to identify the cell-of-origin in individual tumors have been largely restricted to cell type specific activation of oncogenes in genetically engineered mouse models and correlative studies on gene expression patterns between normal tissue and tumor. Prior work from our lab has shown that highly expressed genes within a cell exhibit an overall increase in mutation rates as a consequence of transcription-coupled damage (TCD) which overwhelms transcription-coupled repair (TCR) relative to genes with low or medium level expressions. We therefore hypothesized that tumor cells retain a TCD-mutational signature representative of their cell-of-origin, enabling us to identify the cell-of-origin from the mutational landscape of human tumors.To test this hypothesis, we explored tumors originating from the lung which comprise comprise of distinct histological subtypes: small cell lung cancer (SCLC) and non-small cell lung cancer (NSCLC) that is further categorized into lung adenocarcinoma (LUAD), squamous cell lung cancer (LUSC) and large cell carcinoma (LCC). To identify lung epithelial cell type signatures, we first analyzed a publicly available single cell transcriptomic atlas of the normal human lung of 9,407 cells. For each of the 14 epithelial cell types defined in the lung atlas, we curated a gene signature of its respective unique and highly expressed genes. We next studied the somatic genetic variants in a dataset of 488 samples, which consisted of 230 LUAD tumors and 178 LUSC tumors from TCGA as well as 80 SCLC tumors from a previously published dataset. For each of the 488 tumors, we generated TCD-scores as a function of mutation frequency for each cell-type specific gene signature.Unsupervised hierarchical clustering on the 14 TCD-scores corresponding to the 14 epithelial cell types for each tumor, revealed heterogeneous patterns of cell-type specific mutational signatures. More specifically, LUAD samples were heterogeneously enriched for high AT2-specific, club-specific, ciliated-specific TCD-scores, while LUSC tumors were enriched for basal-specific and ciliated-specific TCD-scores. Together these data suggest that a tumor’s mutational profile can serve as a metric for the identification of the cell-of-origin. Further work will test the extension in different cancer types in other organs or a pan-cancer analysis. The identification of cell-of-origin for cancer subtypes would enable a better understanding of the tumor biology, and ultimately may lead to preventive therapies for individuals at high risk of developing cancer. Citation Format: Ayushi S. Patel, Itai Yanai. Cell-type specific mutational signatures as a proxy for cell-of-origin identification in lung cancer histological subtypes [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2022; 2022 Apr 8-13. Philadelphia (PA): AACR; Cancer Res 2022;82(12_Suppl):Abstract nr 1696.

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