Abstract 4431: Urine microRNA profiling in bladder cancer by next-generation sequencing

Bladder cancer (BC) is one of the leading causes of cancer-related death worldwide. BC is among the most expensive cancer per patient because it requires frequent surveillance and repeated treatments over many years. The identification of new biomarkers for early BC detection, recurrence/progression is urgently needed to both improve patient outcomes and decrease health care costs. MicroRNAs (miRNAs) are aberrantly expressed in many cancers, including BC, and may be isolated from various biological specimens, including urine. To investigate miRNA signatures in surrogate tissues may be a useful alternative to reduce invasiveness of biopsies, allowing repetitive samplings during follow-up and reducing health care costs for detection, monitoring of progression and treatment. We aim to identify specific miRNA signatures in urine samples from 66 BC male patients (10 muscle invasive BC (MIBC) and 56 non-muscle invasive BC (NMIBC)) and 48 healthy controls using a Next Generation Sequencing (NGS) approach able to accurately distinguish BC patients and predict disease outcome. The measurement of miRNA levels in urine could allow to measure the levels of promising biomarkers in one of the best and closest surrogate tissue for BC, since it is in direct contact with the tissue of tumor origin. A specific miRNA signatures that could distinguish the different types of BC patients from healthy controls was found in urine. For MIBC, a 18-miRNAs signature had over 80% predictive power (PP) to recognize patients from controls (data are under validation). For NMIBC we were able to stratify cases according to grade. In particular, 23 miRNAs resulted differentially expressed among G1-G2 cases and controls (5 of them had PP>0.70), while several miRNAs resulted differentially expressed among G3 patients and controls (a 10-miRNAs signature with PP>0.97). Interestingly, we found several differentially expressed miRNAs in common among cases and some miRNAs that were differentially expressed only in specific subcategories of BC cases. NGS data were also used to search for the most constant miRNAs in the set of samples to be used as reference genes in a validation step. Twenty-three miRNAs (21 target and 2 reference miRNAs) were validated by qPCR on 177 urine samples from 113 BC case and 64 controls . Interestingly, miRNAs differentially expressed among cases and controls were able to discriminate not only BC cases from controls, but also its subcategories. This data provide evidence