INTRODUCTION AND OBJECTIVE: MRI has been proposed as a new staging tool for bladder cancer (BC), but its use is limited by its high costs and availability. 29-MHz high-resolution micro-ultrasound (mUS) technology has been suggested as a potential alternative to mpMRI for the detection of prostate cancer. The feasibility of mUS to detect BC and to distinguish between non-muscle invasive BC (NMIBC) and muscle-invasive BC (MIBC) has been showed. However, no comparative analysis between mUS and mpMRI was performed. The aim of this study is to compare the diagnostic accuracy of mUS vs. MRI in differentiating NMIBC and MIBC. METHODS: This is a prospective on going study of patients with a primary diagnosis of BC with either positive urine cytology (UC) or negative UC and a tumor size >25mm from a tertiary care high volume center. Patients were prospectively enrolled within an internally approved protocol (BUS-MISS protocol; ICH-004/2019). All patients were scheduled for an endoscopic transurethral resection of bladder tumor (TURBT). mUS and MRI were performed before TURBT in every patient. Two urologists blinded to the MRI performed the mUS with the ExactVu system with an EV29L 29 MHz side-fire transducer. A 3-Tesla mpMRI was performed and the VI-RADS score was reported. We finally compared the results with the pathological report. The primary outcome was to compare the accuracy of both diagnostic methods in differentiating MIBC from NMIBC with the final histologic result. RESULTS: Between March and October 2019, we prospectively enrolled 27 consecutive patients. Three patients were excluded from the study because one had contraindications for MRI, one did not have an acceptable mUS window for the evaluation, and one revealed an inflammatory mass despite the papillary and solid looking mass. Finally, 25 patients were evaluated for MRI and 24 for mUS. Diagnostic accuracy indexes for mUS reached a sensibility, specificity, PPV, and NPV respectively of 100%, 68%, 45%, and 100% Likewise, the indexes for MRI resulted with sensibility, specificity, PPV, and NPV respectively of 100%, 32%, 32%, and 100%. Accuracy reached 75% and 48% in mUS and MRI respectively. CONCLUSIONS: These preliminary results demonstrate that the measures of diagnostic accuracy indexes demonstrate an overall superiority of the mUS compared to MRI, which instead shows an important tendency in lesion upstaging compared to mUS. A superior accuracy of mUS compared to MRI has been proven. To be underlined, the subpopulation analyzed is of greatest importance, but also more of a challenge for both methods. Further studies with a larger population are ongoing and warranted to better compare and to validate these techniques in the BC diagnostic work-up. Source of Funding: none