Abstract
BACKGROUND: Hepatocellular carcinoma (HCC) is classified as one of the leading malignant neoplasms worldwide, exhibiting a rising trend in its global incidence. The imperative for the development of innovative diagnostic and prognostic biomarkers is critical for improving the therapeutic and management strategies for patients with HCC. This study was undertaken to identify endoplasmic reticulum stress-associated E3 ubiquitin ligases and deubiquitinating enzymes (ERS-E3s/DUBs) and to construct a prognostic risk model for HCC. METHODS: The transcriptomic and clinical data were obtained from The Cancer Genome Atlas (TCGA) database. By analyzing the transcriptomic data treated with endoplasmic reticulum stress inducers, we identified differentially expressed E3 ubiquitin ligases (E3s) and deubiquitinating enzymes (DUBs). A prognostic risk model based on ERS-E3s/DUBs was developed using least absolute shrinkage and selection operator (LASSO) and Cox regression analyses. Utilizing the Akaike Information Criterion for computation, we ascertained the most appropriate threshold for stratifying patients into distinct categories of high-risk and low-risk cohorts. Subsequently, we predicted and analyzed the survival prognosis and Gene Ontology analyses of patients in high- and low-risk groups. RESULTS: In this study, we systematically recognized 23 ERS-E3s/DUBs and developed a liver cancer prognostic risk model founded on nine ERS-E3s/DUBs. Furthermore, we formulated a new nomogram that combines risk characteristics and clinical pathological features, which provides good predictive performance for the clinical prognosis of HCC patients. CONCLUSIONS: We identified ERS-E3s/DUBs and constructed a prognostic risk model for HCC.