Abstract PO-066: Computational identification of key pathway and differentially-expressed gene signatures in ovarian cancer stem cells

Abstract

Abstract Many studies have demonstrated the causativeness of cancer stem cells (CSCs), which are able to self-renew, metastasize, differentiate, and resist chemotherapy. Although multiple cell surface markers have been used to identify and isolate cancer stem cells, their uncertainty makes these surface markers unreliable in identifying heterogeneous cancer stem cells. Some transcription factors have been shown to be overexpressed in CSCs residing in multiple types of cancers, and some pathways have been found to regulate CSC populations. However, since some CSCs do not express all of the identified genes, it is important to establish a panel of genes and pathways that can consistently distinguish CSCs. To determine a panel of consensus genes or pathways that can be used to recognize CSCs, we attempted to identify differentially expressed genes and enriched pathways in CSCs, which we call cancer stem cell signature. In this study, we employed two ovarian cancer RNA-seq data (GSE33874 and GSE82304) from the Gene Expression Omnibus (GEO) and considered genes with significant fold changes (|logFC|≥1). We used ComPath to identify pathways that are enriched in each of our selected RNA-seq datasets, then we developed a customized python script to reveal pathways commonly enriched in both datasets. We subsequently applied ComPath’s mapping catalog to elucidate equivalent or hierarchical relationships across common pathways from three different databases (KEGG, Reactome, and WikiPathways). In total, we identified 241 commonly upregulated genes and 96 commonly downregulated genes. We also present 30 equivalent relationships and 98 hierarchical relationships between pathways. Our results confirm several pathways that are reported to be involved in CSC regulation, such as the Wnt signaling pathway and the Hedgehog pathway. Furthermore, we also identified multiple pathways that have been previously associated with poor prognosis or high risk of ovarian cancer, such as the Prolactin signaling pathway and the coagulation cascade, indicating that these pathways may also be involved in the maintenance of ovarian cancer stem cells. We will continue to investigate ovarian CSC gene and pathway signatures by involving more RNA-seq data and pathway analysis databases. Citation Format: Renata H. Fu, Yongsheng Bai, Qi-En Wang. Computational identification of key pathway and differentially-expressed gene signatures in ovarian cancer stem cells [abstract]. In: Proceedings of the AACR Virtual Special Conference on Tumor Heterogeneity: From Single Cells to Clinical Impact; 2020 Sep 17-18. Philadelphia (PA): AACR; Cancer Res 2020;80(21 Suppl):Abstract nr PO-066.