To identify a robust panel of microRNA (miRNA) signatures that can distinguish renal cell carcinoma (RCC) from normal kidney using miRNA expression levels. We performed a comprehensive meta-analysis of 29 published studies that compared the miRNA expression profiles of RCC tissues and adjacent normal tissues (NT) to determine candidate miRNAs as prognostic biomarkers for RCC. Using vote-counting strategy and robust rank aggregation method, we identified a statistically significant miRNA meta-signature of two upregulated (miR-21, miR-210) and three downregulated (miR-141, miR-200c and miR-429) miRNAs. X-tile plot was used to generate the optimum cut-off point for the 15 different deregulated miRNAs and Kaplan-Meier method was used to calculate CSS. In a cohort of 45 patients, the high expression of miR-21 (HR: 5.46, 95%CI: 2.02-53.39) and miR-210 (HR: 6.85, 95%CI: 2.13-43.36), the low expression of miR-141 (HR: 0.16, 95%CI: 0.004-0.18), miR-200c (HR: 0.08, 95%CI: 0.01-0.43) and miR-429 (HR: 0.18, 95%CI: 0.02-0.50) were associated with poor cancer-specific survival (CSS) following RCC resection. We also constructed a five-miRNAs-based classifier as a reliable prognostic and predictive tool for CSS in patients with RCC, especially in clear cell RCC (ccRCC) (HR: 5.46, 95% CI: 1.51-19.66). This method might facilitate patient counselling and individualise management of RCC.