Predicting recurrence and progression in NMIBC: A multi-event and temporal analysis approach.

OBJECTIVES

Non-muscle invasive bladder cancer (NMIBC) accounts for 70-80% of all bladder cancer cases. Despite adequate Bacillus Calmette-Guérin (BCG) therapy, approximately 30% of patients fail to respond, with up to 40% recurring within a year and 20% progressing to muscle-invasive disease (MIBC). Current studies often focus on single recurrence events, failing to account for the recurrent nature of NMIBC. This study aims to identify key factors influencing recurrence and progression, utilizing advanced statistical models to capture the complexity of disease dynamics.

METHODS

This monocentric retrospective study included 385 NMIBC patients treated with BCG between 2012 and 2020. Patients were stratified by EAU intermediate-risk (IR) and high-risk (HR) categories. Multi-event analyses, including Cox proportional hazards and Prentice-Williams-Peterson Total Time (PWP-TT) models, identified predictors of recurrence and progression. Multinomial logistic regression analyzed temporal recurrence dynamics across three time intervals: within 6months of BCG initiation, 6months after the last instillation, and beyond this period.

RESULTS

Among 385 patients, 122 experienced recurrence, with 34.4% presenting as low-grade pTa, 23.8% as high-grade pTa, and 40.9% as advanced stages (pT1 or higher). Progression to MIBC occurred in 10.3%, predominantly in HR patients (HR: 13.1% vs. IR: 4.5%; P=0.017). Cox analysis identified multifocality (HR=2.07; P=0.001) and prior bladder cancer history (HR=1.72; P=0.008) as key predictors for any-grade recurrence. Multinomial regression showed high-risk patients had increased odds of advanced recurrence (pT2 or higher; OR=2.71, P=0.001). Late high-grade recurrences demonstrated similar progression and mortality risks to early refractory tumors.

CONCLUSIONS

This study highlights the heterogeneous recurrence dynamics in NMIBC. While IR patients predominantly experience low-grade recurrences, HR patients remain at persistent risk of high-grade and advanced-stage progression. These findings advocate for tailored surveillance strategies and further exploration of personalized, bladder-preserving therapies for high-risk patients.

LEVEL OF EVIDENCE: Level 3: Non-randomized controlled cohort/follow-up study (retrospective cohort).