A gene-based risk score model for predicting recurrence-free survival in patients with hepatocellular carcinoma

Abstract Background Hepatocellular carcinoma (HCC) remains the most frequent liver cancer, accounting for approximately 90% of primary cancers worldwide. The recurrence-free survival (RFS) HCC patients is a critical factor in devising personal treatment plan. Thus, it necessary to accurately forecast prognosis clinical practice. Methods Using Cancer Genome Atlas (TCGA) dataset, we identified genes associated with RFS. A robust likelihood-based modeling approach was used select best prognostic model. Then, GSE76427 dataset evaluate model’s effectiveness. Results We 1331 differentially expressed Seven these were selected generate validation both TCGA cohort and GEO demonstrated that 7-gene model can predict RFS patients. Meanwhile, results multivariate Cox regression analysis showed risk score could function as an independent factor. In addition, according time-dependent ROC curve, performed better predicting training set external than classical TNM staging BCLC. Furthermore, seven found be related occurrence development cancer by exploring three other databases. Conclusion Our study seven-gene signature prediction novel convenient tool. These might potential target metabolic therapy HCC.