Abstract
Biomarkers to guide the clinical treatment of patients with renal cell carcinoma (RCC) are not yet routinely available. MicroRNAs (miRNAs) have been demonstrated to serve as biomarkers for a number of types of cancer. Based on a previous study by this group, we hypothesize that several highly differentially expressed miRNAs may serve as tissue and plasma biomarkers in patients with RCC. The expression levels of miR-210, miR-224 and miR-141 were analyzed in tissue samples from the same cohort of 78 patients with RCC, in paired pre- and post-operative plasma samples from 66 patients with clear cell RCC (ccRCC) and in 67 healthy controls by reverse transcription-quantitative polymerase chain reaction. Receiver operating characteristic (ROC) was used to evaluate the diagnostic accuracy associated with the expression of miR-210, miR-224 and miR-141. ROC curves revealed that the diagnostic accuracy (area under the curve) of tissue miR-210, miR-224, the ratio of miR-210/miR-141 (miR(210/141)), miR-224/miR-141 (miR(224/141)) and miR-210× miR-224/miR-141 (miR(210×224/141)) in ccRCC was 0.8329, 0.8511, 0.9412, 0.9898 and 0.9771, respectively. Notably, miR(224/141) demonstrated the highest accuracy among these miRNAs for discriminating ccRCC tissues from normal tissues, with a sensitivity of 97.06% and a specificity of 98.53%. The expression levels of plasma miR-210 and miR-224 were significantly increased in patients compared with healthy control patients, and were reduced postoperatively (P<0.05). The diagnostic accuracy of plasma miR-210 and miR-224 were 0.6775 (89.55% sensitivity and 48.48% specificity) and 0.6056 (88.06% sensitivity and 40.91% specificity), respectively. The present study indicated that the tissue miR-224/miR-141 ratio is a potentially powerful tool for detecting ccRCC. However, plasma miR-210 and miR-224 may not be associated with diagnosis of ccRCC.