External validation of the Uromonitor®‐version 2 urine test as a biomarker for optimisation of non‐muscle‐invasive bladder cancer management
J. Rubio‐Briones, Félix Guerrero‐Ramos, Asier Mercadé Sánchez, Ines Bezana Abadía, Ruth Martín Rodríguez, Antonio Alcaraz, Maite Miqueleiz Legaz, Fernando Lozano, Miguel Arrabal Martín, Adrián Luis Cardo, Victoria Gonzalo Rodríguez, J. Caño Velasco, Miriam Serrano Liesa, C. Ferreiro Pareja, R. Espílez, Ignacio de González‐Valcárcel de Torres, D. Sánchez Zalabardo, Cristina Esquinas, Lourdes Mengual, Ángel Borque‐Fernando, Luis Esteban, Joan Palou Redorta, Fred Witjes, Luis Martínez‐Piñeiro
Abstract
OBJECTIVES: To conduct an external multicentre validation to evaluate Uromonitor®-version 2 (v2; U-monitor, Porto, Portugal) as a biomarker for detecting non-muscle-invasive bladder cancer (NMIBC) recurrence in an independent patient cohort in a real-world-evidence setting and compare it to cytology. PATIENTS AND METHODS: The 'External Validation of Uromonitor as a Biomarker for Optimization of NMIBC Management by the Club Urológico Español de Tratamiento Oncológico Group' (EVALUATION-CUETO) study is a prospective, multicentre, observational study (NCT05864599) conducted in Spain. Inclusion criteria included any risk group of patients with NMIBC undergoing follow-up. The most recent (index) NMIBC tumours of any grade had to have been diagnosed within the past 3 months to 2 years. Regarding biomarkers, data from cystoscopy, cytology, and the Uromonitor-v2 test were collected. Predictive performance parameters, including accuracy, area under the curve, negative and positive predictive values (NPV, PPV), sensitivity, and specificity, were calculated. RESULTS: A total of 1178 patients with a mean age of 72 years and a predominant male representation (80%) were included. Among the positive Uromonitor-v2 cases, 64 (57.66%) exhibited a TERT promoter mutation, and 59 (53.15%) harboured a FGFR3 mutation. No KRAS mutations were found. For the total population, Uromonitor-v2 demonstrated high specificity (0.93) and NPV (0.93), but low sensitivity (0.36) and PPV (0.36), indicating a limited ability to detect all recurrences and confirm positive cases. Performance varied across tumour grades, with low-grade tumours showing a low sensitivity of 0.32 and high-grade tumours achieving a higher but insufficient sensitivity of 0.44. Specificity remained consistent at 0.93 across all groups. Uromonitor-v2 achieved higher accuracy compared to cytology alone, with the combined approach further improving overall predictive performance. CONCLUSION: The strong specificity and NPV observed across all studies suggest a role of Uromonitor-v2 as a potential tool for NMIBC follow-up, particularly in ruling out recurrence and clarifying doubtful cystoscopy results. However, the unexpectedly low sensitivity reported in this external multicentre validation suggests the need for further investigation before routine clinical implementation.