Developing and Validating an Autophagy Gene-Set-Based Prognostic Signature in Hepatocellular Carcinoma Patients

Abstract

Background: Hepatocellular carcinoma (HCC), a major type of cancer affecting the lives of people across the world, is one of many diseases whose advancement is thought to be influenced by autophagy dysfunction. Here, the prognostic significance of autophagy in HCC will be investigated. Methods: The Cancer Genome Atlas (TCGA) database was employed in this work to identify 62 differentially-expressed Autophagy-Related Genes (ARGs) in HCC patients. Functional enrichment studies revealed that autophagy played a tumor-promoting role in the advancement of HCC. Based on RNA sequencing of 116 tumor samples and 114 paracancerous tissue samples acquired from HCC patients at the Guangxi Medical University Cancer Hospital, 866 differentially expressed prognosis-related genes were identified. Using lasso regression analysis, two ARGs (BIRC5 and BAK1) linked to prognosis were discovered after intersecting the differential genes derived from the prognosis-related groups. A risk score based on ARGs was developed using a Cox proportional hazards regression model. RNA sequencing data were used to construct this model. Finally, the TCGA and The Human Protein Atlas Databases (THPA) were used to validate the clinical data of 116 HCC patients.Results: Elevated expression level of the BAK1 and BIRC5 genes is linked to poor prognosis. The two ARGs were used to calculate the risk score as follows: (0.0253*BAK1) + (0.0051*BIRC5). Risk score served as the independent prognostic factor as per the multivariate analysis. TCGA, THPA, and data from the Guangxi Medical University Cancer Hospital were used to confirm the predictive validity of risk scores for the prognosis of HCC patients. Conclusion:This study offers molecular insights regarding the involvement of autophagy in HCC patients, along with a probable prognostic signature for determining the outcome (prognosis) of HCC patients.