Identification of stemness-related LncRNA signature for predicting the prognosis and therapy response in colorectal cancer

Abstract

Abstract Background: Cancer stem cells (CSC) carry out a vital responsibility throughout the entire progress of colorectal cancer (CRC), and fulfil an essential biological function. However, lncRNAs participate in regulating CRC stem cells (CCSCs) and correlate strongly with the patients' prognosis. Therefore, it is crucial to identify the CCRC-related lncRNAs in CRC. Methods: We identified CCRCs-related lncRNAs through the Cell marker and TCGA databases. And the CCSC-related lncRNAs model was constructed by the differential, cox survival , and lasso regression analysis. Combining the GEO dataset, we determined the prognostic value by Kaplan-Meier analysis, univariate and multivariate cox survival analysis. Moreover, principal component analysis (PCA), clinical characterization, nomogram, gene mutation, gene set enrichment analysis (GSEA), immune microenvironment (TME), chemotherapy, intergroup differential gene, and protein-protein interaction (PPI) analysis were conducted to analyze the risk model. Furthermore, the core genes in the sub-module were comprehensively characterized. Results: In this research, abnormally expressed, prognostic and CSC-related lncRNAs were firstly identified. Through the lasso regression model, we obtained a robust risk signature consisting of 4 CCSC-related lncRNAs (ZEB1-AS1，LINC00174，FENDRR and ALMS1-IT1). Then, the risk model was confirmed applicable in both TCGA and GEO cohorts. Further verification, the signature can be verified as a independent prognostic factor for CRC. Based on the CCSC-related lncRNA model, the high- and low-risk groups exhibited different stemness statuses, including gene expression, mutation status, signaling pathways, TME and chemotherapy response. The HOX family and HOX4 were centrally located in the PPI interaction and had an influential contribution in CRC. Conclusions: We established a 4 CCSC-related lncRNA signature with a promising prognosis. And the signature can appropriately estimate the gene mutation, TME, and chemotherapy outcomes for CRC patients. Furthermore, the CCSC-related lncRNAs and HOX4 can serve as noble biomarkers and promote the management of therapy clinically.