An evaluation of urinary microRNA reveals a high sensitivity for bladder cancer

BACKGROUND: Urinary biomarkers are needed to improve the care and reduce the cost of managing bladder cancer. Current biomarkers struggle to identify both high and low-grade cancers due to differing molecular pathways. Changes in microRNA (miR) expression are seen in urothelial carcinogenesis in a phenotype-specific manner. We hypothesised that urinary miRs reflecting low-and high-grade pathways could detect bladder cancers and overcome differences in genetic events seen within the disease. METHODS: We investigated urinary samples (n = 121) from patients with bladder cancer (n = 68) and age-matched controls (n = 53). Fifteen miRs were quantified using real-time PCR. RESULTS: We found that miR is stable within urinary cells despite adverse handling and detected differential expression of 10 miRs from patients with cancer and controls (miRs - 15a/15b/24-1/27b/100/135b/203/212/328/1224, ANOVA P < 0.05). Individually, miR-1224-3p had the best individual performance with specificity, positive and negative predictive values and concordance of 83%, 83%, 75% and 77%, respectively. The combination of miRs-135b/15b/1224-3p detected bladder cancer with a high sensitivity (94.1%), sufficient specificity (51%) and was correct in 86% of patients (concordance). CONCLUSION: The use of this panel in patients with haematuria would have found 94% of urothelial cell carcinoma, while reducing cystoscopy rates by 26%. However, two invasive cancers (3%) would have been missed. British Journal of Cancer (2012) 107, 123-128. doi: 10.1038/bjc.2012.221 www.bjcancer.com Published online 29 May 2012 (C) 2012 Cancer Research UK