Construction of a 13-microRNA-based signature and prognostic nomogram for predicting overall survival in patients with hepatocellular carcinoma.

BACKGROUND AND OBJECTIVE Hepatocellular carcinoma (HCC) is a common malignancy associated with a poor prognosis owing to difficulties in reliably estimating overall survival (OS). MicroRNAs (miRNAs) play critical roles in HCC initiation, progression, and metastasis and are highly correlated with patient prognosis. Thus, miRNA-based risk signatures and nomograms are urgently required for predicting OS in patients with HCC. METHODS we constructed a 13-miRNA-based signature and prognostic nomogram using 408 HCC samples and 58 normal tissues with miRNA sequencing data and clinical data from 323 patients downloaded from The Cancer Genome Atlas. 195 patients were assigned as the internal validation cohort for verification and testing. Gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes analysis (KEGG) was applied to investigate pathway enrichment for the signature. RESULTS We identified and validated a 13-miRNA risk signature highly associating with the OS of HCC patients. The signature showed good performances by calculating C-index, AUC and calibration curves. After verification and testing using an internal validation cohort, the results yielded a miRNA-based signature and a prognostic nomogram with reliable predictive accuracy. GO and KEGG demonstrated that various genes and multiple pathways were closely related to the mechanisms of HCC proliferation and metastasis. CONCLUSION We successfully identified a 13-miRNA-based signature and prognostic nomogram which are capable of predicting OS in patients with HCC.