A novel seven-microRNA signature predicts prognosis in colorectal adenocarcinoma

Abstract

Background Colorectal carcinoma, primarily colorectal adenocarcinoma (CRA), is one of the most common malignancies, ranking third, while contributing the second cause of cancer death in the world. MicroRNAs, a type of non-coding RNA, play an important role in regulating cancer-related cell biology. To simply and accurately predict survival risk for CRA patients, we identified a novel seven-miRNA signature. The microRNA (miRNA) expression profiles along with clinical data of 512 CRA patients were downloaded from The Cancer Genome Atlas (TCGA) database, and 415 patients with complete clinical information were further divided into training set and test set randomly. To construct the prognostic signature in the training set, a series of Cox regression analyses were performed, including univariate regression, least absolute shrinkage and selectionator operator (LASSO) - Cox regression, and multivariate regression. Results Seven predictive miRNAs (miR-153-2, miR-3199-2, miR-144, miR-887, miR-561, miR-3684, and miR-505) were ultimately screened. The ROC curves for 5-year survival in the training set, test set and entire set were 0.889, 0.742, and 0.816, respectively. Kaplan-Meier analysis results of the three sets all showed p <0.05. Further analyses demonstrated that the signature was an independent prognostic risk factor for CRA patients, and its predictive ability was superior to age and TNM stage. Functional enrichment analysis revealed that p53 and ErbB pathways were related to the prognostic regulation of miRNAs in the signature in CRA patients.Conclusion Our study demonstrated the potential of this novel seven-miRNA signature to independently predict overall survival in patients with CRA. Functional enrichment analysis further revealed the possible regulatory role of miRNAs in the signature in CRA-related cell biological processes and signaling pathways.