MicroRNA expression profile and TNM staging system predict survival in patients with lung adenocarcinoma

Abstract

ObjectThe current study was performed to construct a model with microRNA (miRNA/miR) expression profile and TNM staging system for prognosis predicting in patients with lung adenocarcinoma (LUAD). MethodsDifferentially expressed miRNAs were identified from miRNA data of LUAD in The Cancer Genome Atlas (TCGA) database. Potential prognostic miRNAs and TNM classification parameters, screened out by Cox proportional hazards regression analysis, were included in the prognostic model. The prognostic model was visualized with a nomogram, and tested by calculating the C-index and drawing the calibration curve in the training set and validating set, respectively. Finally, the prognostic miRNAs were analyzed with bioinformatics tools. ResultsA total of 194 differentially expressed miRNAs were identified between LUAD tissues and matched normal tissues, including 99 up-regulated and 95 down-regulated miRNAs. miRNA index (miR.index), constructed with nine miRNAs (hsa-let-7i, hsa-mir-1976, hsa-mir-199a-1, hsa-mir-31, hsa-mir-3940, hsa-mir-450a-2, hsa-mir-4677, hsa-mir-548v and hsa-mir-6803), was an independent prognostic indicator for the survival of patients with LUAD. Bioinformatics analysis suggests that the selected miRNAs are involved in the development and progress of LUAD. ConclusionThe prognostic model constructed with nine miRNA expression profile and TNM classification parameters can predict the survival in patients with LUAD, and the predictive power of the model are warranted for further validations.

Keywords

• bioinformatics analysis
• lung adenocarcinoma
• microrna
• survival