A novel microRNAs expression signature for hepatocellular carcinoma diagnosis and prognosis

Abstract

This study aims to identify prognostic microRNAs (miRNAs) biomarkers for diagnosis and survival of hepatocellular carcinoma (HCC) based on large patients cohort analysis. HCC patient cohort data were downloaded from The Cancer Genome Atlas, including paired HCC and adjacent non-cancer tissues. Receiver operating characteristic curve method was used to classify cancer and non-cancer tissues according to microRNAs expression levels. The aberrant microRNAs expression level were ranked and risked for building a prognostic miRNAs signature model. Kaplan–Meier survival was used to analyze the differences among various risk factors in accordance with miRNAs ranking scores. The study showed 33-miRNA signature, 11 were down-regulated and 22 were up-regulated through comparison between cancer samples and non-cancer samples. The maximum correct classification rate is up to 98.7%. Five microRNAs, hsa-mir-3677, hsa-mir-421, hsa-mir-326, hsa-mir-424 and hsa-mir-511-2, significantly correlated with patient survival. The survival rate and time negatively associated with lowering miRNAs index. In the low risk group, over 70% patients showed 5 years survival, while none patients survived longer than 5 years in the high risk group. MiR-424, miR-326 and miR-511 could be applied for HCC diagnostic biomarkers. These five miRNAs were significantly associated with lysosome pathway and D-Glutamine and D-glutamate metabolism pathway via Kyoto Encyclopedia of Genes and Genomes pathway analysis and Gene Ontology annotation. Conclusively, the five miRNAs expression signature could be used as HCC prognostic and diagnostic biomarkers.