Document Type: Research Articles
Department of Immunology, School of Basic Medical Sciences, Guangxi Medical University, Nanning, China.
Department of Otolaryngology and Head and Neck, The People’s Hospital of Guangxi Zhuang Autonomous Region, Nanning, China.
Department of Medical Oncology, Guangxi Medical University Cancer Hospital, Nanning, China.
Objective: This study aims to explore the roles of long non-coding RNAs (lncRNAs) for predicting survival in hepatocellular carcinoma (HCC) patients with cirrhosis. Methods: A set of lncRNAs differentially expressed between HCC patients with or without cirrhosis was identified using expression profiles of The Cancer Genome Atlas database, and these lncRNAs were screened for their risk scoring system to predict recurrence-free survival (RFS) or overall survival (OS). Predictive ability of risk scoring systems was confirmed using uni- and multivariate Cox analyses while adjusting for clinical features. Predictive lncRNAs were analyzed by functional enrichment analysis. Results: Our screen identified 22 lncRNAs that were upregulated in the presence of cirrhosis and 59 that were downregulated. To predict OS of HCC patients with cirrhosis, a risk scoring system was developed with four lncRNAs (LINC02086, LINC00880, LINC01549 and AC136475.3); to predict RFS in these patients, the risk scoring system contained five lncRNAs (SH3RF3-AS1, AC104117.3, AC136475.3, LINC00239 and MRPL23-AS1). All risk scoring systems were associated with an area under the receiver operating characteristic curve > 0.7. Based on uni- and multivariate Cox analyses, the risk scoring system could serve as a significant independent predictor for OS in HCC patients with cirrhosis. Functional enrichment analysis suggested that the lncRNAs in the risk scoring systems are involved primarily in the pathway of Wnt signal and cytokine-cytokine receptor interaction. Conclusion: Risk scoring systems based on lncRNAs can effectively predict OS of HCC patients with cirrhosis. The system should be further developed and validated in larger, preferably multi-site patient populations.