Abstract PO006: Computational identification of female-biased breast cancer candidate genes correlated with tumor-infiltrating immune cell types

Abstract

Listen

Translate article

Abstract Biological differences between sexes are prevalent in many aspects, including in their reactions to tumor invasion. It is important to identify the prognostic biomarkers in order to better understand the interactions between the host immune system and tumors in different sexes. By examining the correlation between female-biased genes identified from The Cancer Genome Atlas (TCGA) Breast Invasive Carcinoma (BRCA) RNA-seq dataset and tumor-infiltrating immune cell types, we can better understand how female-biased genes exhibit their host immune reactions to cancer antigens. We searched for BRCA candidate genes from significant clusters reported by a recent study against published literature on sex-biased genes expressed in breast tissue, and a total list of 114 female-biased genes in BRCA was compiled. We employed TIMER2.0, a function visualization tool of tumor-infiltrating immune cells for TCGA dataset, to study the association between immune infiltration and gene expression for our selected female-biased genes in the context of CD8 T-cell and B-cell types. Both cell types give similar correlation trends. Indeed, the majority of female-biased genes (45 out of 58) show positive correlation between their expression and immune cell abundance level in both cell types, which could imply strong host immune reaction to female-biased cancer antigens. More than half of the genes (26 out of 45) showed positive correlation in both cell types for exhibiting higher expressions in tumor samples than in normal samples. Specifically, two genes, RUNX3 and CXCR6, have high positive correlation in both immune cell types and higher expressions in tumor samples than normal samples, suggesting that these genes could act as antigens on tumor cells in breast cancer. To pinpoint additional genes that are putatively involved in responding to tumor immune infiltration by CD8 T-cells and B-cells, we ran the DAVID functional annotation tool to identify enriched Gene Ontology (GO) terms and pathways for our selected 45 genes showing positive correlation in both cell types. The major GO terms in significantly enriched functional annotation clusters are “membrane,” “cell membrane,” “receptor,” “domain: SH2,” and “transcription from RNA polymerase II promoter.” The genes sharing these GO terms have been previously shown to be involved in breast cancer prognostic process (e.g. CXCR6, KIT, RUNX3) and also immune cell regulation (e.g. CD8B, DAPP1, LY75). Our result indicates that many identified female-biased genes having positive associations with CD8 T-cell and B-cell abundance levels could serve as alternative therapeutic markers which may guide researchers to develop effective cancer treatment vaccines and design immune checkpoint blockade drugs. Citation Format: Eric Li, Yongsheng Bai. Computational identification of female-biased breast cancer candidate genes correlated with tumor-infiltrating immune cell types [abstract]. In: Abstracts: AACR Virtual Special Conference: Tumor Immunology and Immunotherapy; 2020 Oct 19-20. Philadelphia (PA): AACR; Cancer Immunol Res 2021;9(2 Suppl):Abstract nr PO006.