Approximately 20-30% of patients with metastatic renal cell carcinoma (mRCC) in first-line treatment with tyrosine kinase inhibitors (TKIs) do not respond due to primary resistance to this drug. At present, suitable robust biomarkers for prediction of a response are not available. Therefore, the aim of this study was to evaluate a panel of microRNAs (miRNAs) in nephrectomy specimens for use as predictive biomarkers for TKI resistance. Archived formalin-fixed, paraffin embedded nephrectomy samples from 60 mRCC patients treated with first-line TKIs (sunitinib, n = 51; pazopanib, n = 6; sorafenib, n = 3) were categorized into responders and non-responders. Using the standard Response Evaluation Criteria in Solid Tumors, patients with progressive disease within 3 months after the start of treatment with TKI were considered as non-responders and those patients with stable disease and complete or partial response under the TKI treatment for at least 6 months as responders. Based on a miRNA microarray expression profile in the two stratified groups of patients, seven differentially expressed miRNAs were validated using droplet digital reverse-transcription quantitative real-time polymerase chain reaction (RT-qPCR) assays in the two groups. Receiver operating characteristic curve analysis and binary logistic regression of response prediction were performed. MiR-9-5p and miR-489-3p were able to discriminate between the two groups. MiR-9-5p, as the most significant miRNA, improved the correct prediction of primary resistance against TKIs in comparison to that of conventional clinicopathological variables. The results of the decision curve analyses, Kaplan-Meier analyses and Cox regression analyses confirmed the potential of miR-9-5p in the prediction of response to TKIs and the prediction of progression-free survival after the initiation of TKI treatment.
