Aims: Identification of miRNA signature to predict the prognosis of gastric cancer (GC) patients by integrating bioinformatics and experimental validation.
Methods: The miRNA expression profile and clinical data of GC were collected. The univariable and LASSO-Cox regression were used to construct the risk signature. The receiver operating characteristic (ROC) curve analysis confirmed the good performance of the prognostic model.
Results: A 3-miRNA prognostic signature was constructed, which included hsa-miR-126-3p, hsa-miR-143-5p, and hsa-miR-1275. A nomogram, including the prognostic signature to predict the overall survival, was established, and internal validation in the The Cancer Genome Atlas (TCGA) cohort was performed. We found that compared with the traditional pathological stage, the nomogram was the best at predicting the prognosis.
Conclusions: The predictive model and the nomogram will enable patients with GC to be more accurately managed in clinical practice.
Keywords: GEO; Gastric cancer; Prognosis; miRNA signature.
©2021 Qi and Zhang.