Molecular Characterization of Renal Cell Carcinoma: A Potential Three-MicroRNA Prognostic Signature

Background: Aberrantly expressed miRNAs promote renal cell carcinoma (RCC) growth and metastasis and are potentially useful biomarkers for metastatic disease. However, a consensus clinically significant miRNA signature has not been identified. To identify an miRNA signature for predicting clinical outcome in RCC patients, we used a four-pronged interconnected approach.Methods: Differentially expressed miRNAs were identified and analyzed in 113 specimens (normal kidney: 59; tumor: 54). miRNA profiling was performed in matched normal and tumor specimens from 8 patients and extended to 32 specimens. Seven aberrantly expressed miRNAs were analyzed by qPCR, and their levels were correlated with RCC subtypes and clinical outcome. miRNA signature was confirmed in The Cancer Genome Atlas RCC dataset (n = 241).Results: Discovery phase identified miR-21, miR-142-3p, miR-142-5p, miR-150, and miR-155 as significantly upregulated (2-4-fold) and miR-192 and miR-194 as downregulated (3-60-fold) in RCC; miR-155 distinguished small tumors (<4 cm) from benign oncocytomas. In univariate and multivariate analyses, miRNA combinations (miR-21+194; miR-21+142-5p+194) significantly predicted metastasis and/or disease-specific mortality; miR-21+142-5p+194 (for metastasis): P = 0.0017; OR, 0.53; 95% confidence interval (CI), 0.75-0.33; 86.7% sensitivity; 82% specificity. In the TCGA dataset, combined biomarkers associated with metastasis and overall survival (miR-21+142-5p+194: P < 0.0001; OR, 0.37; 95% CI, 0.58-0.23).Conclusions: The interconnected discovery-validation approach identified a three-miRNA signature as a potential predictor of disease outcome in RCC patients.Impact: With 10% survival at 5 years, metastatic disease presents poor prognosis for RCC patients. The three-miRNA signature discovered and validated may potentially at an early stage detect and predict metastasis, to allow early intervention for improving patient prognosis. Cancer Epidemiol Biomarkers Prev; 27(4); 464-72. ©2018 AACR.