Construction of prognostic model of autophagy-related LncRNA in hepatocellular carcinoma / 西安交通大学学报(医学版)

ABSTRACT

【Objective】 To construct and verify a model for predicting the prognosis of liver cancer patients with autophagy-related long non-coding RNA (LncRNA) expression. 【Methods】 LncRNA expression spectrum, mRNA expression spectrum and clinical characteristics of 374 patients with liver cancer were obtained from the Cancer Genome Atlas (TCGA) Database. The expression levels of autophagy-related genes (from the Human Autophagy Database) were extracted from the mRNA expression profile, and the LncRNA was screened to obtain the highly relevant autophagy-related genes by correlation analysis with the LncRNA expression profile. Differential expression analysis and single factor analysis were conducted between the expressions of the above established autophagy-related LncRNA in 374 cancer tissues and 50 normal tissues to screen autophagy LncRNAs which were differentially expressed and was associated with a significant liver cancer prognosis. The above screened LncRNA were included in the risk of Cox proportional model selection and the prognosis of patients with liver cancer prediction model was established. The risk index (risk score) was calculated according to the forecast model and patients could be divided into high-risk and low-risk groups. Kaplan-Meier survival analysis, Cox regression analysis and receiver operating characteristic curve (ROC) were conducted to evaluate the prognostic value of the model. Gene set enrichment analysis (GSEA) was carried out to analyze the signal pathway of gene enrichment in patients with a high or a low risk. 【Results】 We screened 582 autophagy LncRNAs (R≥0.6, P<0.001) to obtain 23 LncRNAs which were differentially expressed and associated with a significant liver cancer prognosis. A prognostic model consisting of 6 LncRNAs, namely, MKLN1-AS, AC012360.3, AC145207.5, AL513320.1, AC099850.3 and AL049840.2 were constructed. KM survival analysis showed that the survival time of the high-risk group was significantly lower than that of the low-risk group (median survival time 6.937 years and 2.323 years, P<0.001). Cox regression analysis showed that the survival time of patients in the high-risk group was significantly lower than that in the low-risk group (HR=3.773, 95% CI 2.252-6.322, P<0.001); the area under the time dependent ROC curve for 1, 2, 3, 4, 5 and 6-year overall survival was 0.760, 0.729, 0.731, 0.722, 0.696 and 0.685. GSEA analysis showed that the genes in the high-risk group were mainly concentrated in cell cycle, autophagy, tumor, ubiquitin-mediated proteolysis, and RNA degradation. 【Conclusion】 The prognostic model composed of MKLN1-AS, AC012360.3, AC145207.5, AL513320.1, AC099850.3 and AL049840.2 can be used to predict the prognosis of liver cancer patients, which is expected to guide clinical treatment.