A novel prognostic signature based on four glycolysis-related genes predicts survival and clinical risk of hepatocellular carcinoma

Abstract

Background: Hepatocellular carcinoma (HCC) is the most common cancer with limited cure and poor survival. In our study, a bioinformatic analysis was conducted to investigate the role of glycolysis in the pathogenesis and progression of HCC. Methods: Single-sample gene set enrichment analysis (ssGESA) was used to calculate enrichment scores for each sample in TCGA-LIHC and GEO14520 according to the glycolysis gene set. Weighted gene co-expression network analysis identified a gene module closely related to glycolysis, and their function was investigated. Prognostic biomarkers were screened from these genes. Cox proportional hazard model and least absolute shrinkage and selection operator regression were used to construct the prognostic signature. Kaplan–Meier (KM) and receiver operating characteristic (ROC) curve analyses evaluated the prediction performance of the prognostic signature in TCGA-LIHC and ICGC-LIRI-JP. Combination analysis data of clinical features and prognostic signature constructed a nomogram. Area under ROC curves and decision curve analysis were used to compare the nomogram and its components. Results: The glycolysis pathway was upregulated in HCC and was unfavorable for survival. The determined gene module was mainly enriched in cell proliferation. A prognostic signature (CDCA8, RAB5IF, SAP30, and UCK2) was developed and validated. KM and ROC curves showed a considerable predictive effect. The risk score derived from the signature was an independent prognostic factor. The nomogram increased prediction efficiency by combining risk signature and TNM stage and performed better than component factors in net benefit. Conclusion: The gene signature may contribute to individual risk estimation, survival prognosis, and clinical management.