Identification of an Autophagy-Related Gene Signature that Can Improve Prognosis of Hepatocellular Carcinoma Patients

Abstract

Background: Autophagy is a programmed cell degradation mechanism that has been associated with several physiological and pathophysiological processes, including malignancy. Improper induction of autophagy has been proposed to play a pivotal role in the progression of hepatocellular carcinoma (HCC).Methods: Univariate Cox regression analysis of overall survival (OS) was performed to identify risk-associated autophagy-related genes (ARGs) in HCC data set from The Cancer Genome Atlas (TCGA). Multivariate cox regression was then performed to develop a risk prediction model for the prognosis of 370 HCC patients. The multi-target receiver operating characteristic (ROC) curve was used to determine the model’s accuracy. Besides, the relationship between drug sensitivity and ARGs expression was also examined. Results: A total of 62 differentially expressed ARGs were identified in HCC patients. Univariate and multivariate regression identified five risk-associated ARGs ( HDAC1, RHEB , ATIC, SPNS1 and SQSTM1 ) that were correlated with OS in HCC patients. Of importance, the risk-associated ARGs were independent risk factor s in the multivariate risk model including clinical parameters such as malignant stage (HR=1.433, 95% CI=1.293-1.589, P<0.001). In addition, the area under curve for the prognostic risk model was 0.747, which indicates the high accuracy of the model in prediction of HCC outcomes. Interestingly, the risk-associated ARGs were also correlated with drug sensitivity in HCC cell lines.Conclusions: We developed a novel prognostic risk model by integrating the molecular signature and clinical parameters of HCC, which can effectively predict the outcomes of HCC patients.