Bladder carcinoma is globally among the most prevalent cancers and is associated with a high mortality rate at advanced stages. Its detection relies on invasive diagnostic methods that are unpleasant for the patient. Non-invasive molecular biomarkers, such as miRNAs, could serve as alternatives for early detection and prognosis of this malignancy. We designed a computational approach that combines transcriptome profiling, survival analyses, and calculation of diagnostic power in order to isolate miRNA signatures with high diagnostic and prognostic utility. Our analysis of TCGA-BLCA data from 429 patients yielded one miRNA signature, consisting of five upregulated and three downregulated miRNAs with cumulative diagnostic power that outperforms current diagnostic methods. The same miRNAs have a strong prognostic significance since their expression is associated with the overall survival of bladder cancer patients. We evaluated the expression of this signature in 19 solid cancer types, supporting its unique diagnostic utility for bladder carcinoma. We provide computational evidence regarding the functional implications of this miRNA signature in cell cycle regulation, demonstrating its abundance in body fluids, including peripheral blood and urine. Our study characterized a novel miRNA signature with the potential to serve as a non-invasive method for bladder cancer diagnosis and prognosis.
Keywords: ROC analysis; bladder carcinoma; diagnosis and prognosis; genitourinary transcriptomics; miRNAs.