Shimin Chen1, Enfa Zhao2. 1. Department of Gastroenterology, Traditional Chinese Medical Hospital of Taihe Country, Taihe, 236600, China. Electronic address: asmchen@sina.cn. 2. Department of Structural Heart Disease, The First Affiliated Hospital of Xi'an Jiaotong University, Xi'an, 710061, China.
Abstract
OBJECTIVE: Epithelial-to-mesenchymal transition (EMT) is an essential biological process of cancer progression associated with increased metastatic potential and initiation. Herein, we aimed to develop and validate a robust EMT-related prognostic signature that could predict the prognosis of patients with hepatocellular carcinoma (HCC). METHODS: Messenger RNA expression matrix and clinicopathological data were retrieved from The Cancer Genome Atlas (TCGA) and identified differentially expressed genes (DEGs) between HCC tissues and adjacent non-tumor samples. Univariate Cox regression analysis, least absolute shrinkage and selection operator (LASSO) Cox regression and multivariate Cox regression analysis were performed to establish a prognosis signature. Kaplan-Meier survival curve, time-dependent receiver operating characteristic (ROC), multivariate Cox regression analysis, nomogram, C-index, and decision curve analysis (DCA) were performed to investigate the prognostic performance of the signature. The prognostic performance of the new signature was further validated in an independent external cohort. A support vector machine (SVM) approach was performed to evaluate the diagnostic value of the identified genes. RESULTS: A seven-gene signature was formulated to classify patients into high-risk and low-risk groups with discrepant overall survival (OS) in two cohorts (all P < 0.0001), and the former illustrated shorter survival time than the latter even stratified by various groups. The new signature has presented an excellent performance for predicting survival prognosis. Multivariate analysis showed that the signature was an independent risk factor for HCC. The SVM classifier based on the seven genes presented an excellent diagnostic power in differentiating early HCC and normal tissues. Gene Set Enrichment Analyses (GSEA) demonstrated multiple biological processes and pathways to provide novel insights into the development of HCC. CONCLUSION: We established and validated a prognostic signature based on EMT-related genes with good predictive value for HCC survival. The diagnostic performance of the signature had been demonstrated to accurately distinguish early HCC from control individuals.
OBJECTIVE: Epithelial-to-mesenchymal transition (EMT) is an essential biological process of cancer progression associated with increased metastatic potential and initiation. Herein, we aimed to develop and validate a robust EMT-related prognostic signature that could predict the prognosis of patients with hepatocellular carcinoma (HCC). METHODS: Messenger RNA expression matrix and clinicopathological data were retrieved from The Cancer Genome Atlas (TCGA) and identified differentially expressed genes (DEGs) between HCC tissues and adjacent non-tumor samples. Univariate Cox regression analysis, least absolute shrinkage and selection operator (LASSO) Cox regression and multivariate Cox regression analysis were performed to establish a prognosis signature. Kaplan-Meier survival curve, time-dependent receiver operating characteristic (ROC), multivariate Cox regression analysis, nomogram, C-index, and decision curve analysis (DCA) were performed to investigate the prognostic performance of the signature. The prognostic performance of the new signature was further validated in an independent external cohort. A support vector machine (SVM) approach was performed to evaluate the diagnostic value of the identified genes. RESULTS: A seven-gene signature was formulated to classify patients into high-risk and low-risk groups with discrepant overall survival (OS) in two cohorts (all P < 0.0001), and the former illustrated shorter survival time than the latter even stratified by various groups. The new signature has presented an excellent performance for predicting survival prognosis. Multivariate analysis showed that the signature was an independent risk factor for HCC. The SVM classifier based on the seven genes presented an excellent diagnostic power in differentiating early HCC and normal tissues. Gene Set Enrichment Analyses (GSEA) demonstrated multiple biological processes and pathways to provide novel insights into the development of HCC. CONCLUSION: We established and validated a prognostic signature based on EMT-related genes with good predictive value for HCC survival. The diagnostic performance of the signature had been demonstrated to accurately distinguish early HCC from control individuals.