Establishment of a molecular risk model for the prognosis of cervical cancer based on microRNA expression

BACKGROUND: Globally, the incidence of cervical cancer (CC) is highest among all tumors of the female reproductive system. Numerous studies have shown that the expression level of microRNA (miRNA) is highly correlated with cancer. This study aimed to establish a molecular prognostic model of CC based on miRNAs in order to provide more individualized treatment to CC patients. METHODS: Human tissues were selected from the Cancer Hospital (Chinese Academy of Medical Sciences and Peking Union Medical College) for miRNA gene sequencing. CC transcriptome expression and clinical data were downloaded from The Cancer Genome Atlas (TCGA). We distinguished between common differentially expressed miRNAs of CC miRNA-seq and TCGA-CC. To obtain a molecular prognostic model, R package was used to perform univariate Cox proportional hazard regression and least absolute shrinkage and selection operator (LASSO) Cox regression for common differentially-expressed miRNAs. Next, the model performance was evaluated by survival analysis, receiver operating characteristic (ROC) curve analysis, as well as univariate and multivariate analyses in the TCGA-CC dataset. Quantitative Real-time polymerase chain reaction (qPCR) detection was to verify the expression changes of miRNA. Transwell was used to verify the role of molecules in CC cell migration and invasion. RESULTS: Thirty-nine miRNAs were distinguished in TCGA-CC and CC miRNA-seq, LASSO regression analysis to obtain the risk model (risk score =−0.310× expression of hsa-miR-142-3p +0.439× expression of hsa-miR-100-3p). The survival analysis, ROC curve analysis, patient risk assessment, and univariate and multivariate analyses showed that the risk score model had good predictive ability in assessing patient survival (P<0.01), risk of death, and independent prognosis (P<0.01). qPCR detection of clinical samples and cells showed that the expression of hsa-miR-142-3p and hsa-miR-100-3p was consistent with the results of the database analysis. The Transwell results indicated that miR-142-3p is an inhibiting factor and miR-100-3p serves as a promoting factor in CC cell migration and invasion. CONCLUSIONS: Twelve miRNA-seq and TCGA-CC tissues were used to build a prognostic model for CC. We have obtained a two-miRNA risk score model. Our results provide a new strategy for the individualized diagnosis and treatment of CC.