Prognostic value of a microRNA signature as a novel biomarker in patients with lower-grade gliomas.

Abstract

MicroRNAs (miRNAs) may act as prognostic biomarkers in a variety of cancers. The aim of this study was to identify and evaluate a prognostic miRNA signature in patients with lower-grade gliomas (LGGs). miRNA expression profiles and clinical data of patients with LGGs from the Chinese Glioma Genome Atlas (CGGA; the training cohort) and The Cancer Genome Atlas (TCGA; the validation cohort) were analyzed, and the least absolute shrinkage and selection operator Cox regression model was used to identify the miRNA signature, which was combined with clinical prognostic factors to develop an individualized survival prediction model. Gene ontology analysis and Kyoto Encyclopedia of Genes and Genomes pathway analysis were conducted to reveal the biological implications of the signature. We identified a four-miRNA signature that stratified patients in the training cohort into low- or high-risk groups according to overall survival time, a finding that was verified in the validation cohort. Multivariate Cox regression analysis indicated that the four-miRNA signature was an independent prognostic biomarker, and a nomogram combining this miRNA signature with clinicopathological and molecular factors showed high prognostic accuracy for individualized survival prediction in both TCGA (C-index = 0.83) and CGGA (C-index = 0.68) cohorts. Functional annotation indicated that the major biological processes of this prognostic miRNA signature were enriched in cell cycle and DNA repair-associated biological processes. Our findings indicated that our newly discovered four-miRNA signature may be an effective independent prognostic factor for the prediction of patients with LGGs.