Search

Purpose To develop a model to predict recurrence for patients with intermediate-risk (IR) non-muscle-invasive bladder cancer (NMIBC) treated with intravesical chemotherapy which can be challenging because of the heterogeneous characteristics of these patients. Methods Data from three Dutch trials were combined. Patients treated with intravesical chemotherapy with characteristics according to the IR definition of the EAU guideline 2013 were included. Uni- and multivariable Cox regression with selection methods were used to identify predictors of recurrence at 1, 2, and 5 years. An easy-readable table for recurrence probabilities was developed. An external validation was done using data from Spanish patients. Results A total of 724 patients were available for analyses, of which 305 were primary patients. Recurrences occurred in 413 patients (57 %). History of recurrences, history of intravesical treatment, grade 2, multiple tumors, and adjuvant treatment with epirubicin were relevant predictors for recurrence-free survival with hazard ratios of 1.48, 1.38, 1.22, 1.56, and 1.27, respectively. A table for recurrence probabilities was developed using these five predictors. Based on the probability of recurrence, three risk groups were identified. Patients in each of the separate risk groups should be scheduled for less or more aggressive treatment. The model showed sufficient discrimination and good predictive accuracy. External validation showed good validity. Conclusion In our model, we identified five relevant predictors for recurrence-free survival in IR-NMIBC patients treated with intravesical chemotherapy. These recurrence predictors allow the urologists to stratify patients in risk groups for recurrence that could help in deciding for an individualized treatment approach.

Objective To compare the risks according to the American Urological Association (AUA), EAU, European Organization for Research and Treatment of Cancer (EORTC) and Club Urol gico Espa ol de Tratamiento Oncologico (CUETO) classifications with real outcomes in a cohort of patients in the Netherlands, and to confirm that patients who were undertreated according to these risk models have worse outcomes than adequately treated patients. Patients and Methods Patients treated with complete transurethral resection of bladder tumour and intravesical chemotherapy were included. Not all patients would have received intravesical chemotherapy had they been treated to current standards, and thus comparison of the observed outcomes in our Dutch cohort vs expected outcomes based on the EORTC risk tables and CUETO scoring model was possible. The cohort was reclassified according to the definitions of five index patients (IPs), as defined by the AUA guidelines, and three risk groups, defined according to the EAU guidelines, to compare the outcomes of undertreated patients with those of adequately treated patients. Results A total of 1001 patients were available for comparison with the AUA definitions and 728 patients were available for comparison with the EORTC and CUETO models. There was a large overlap between the observed outcomes and expected recurrence and progression probabilities when comparison was made using the EORTC risk tables. The observed recurrence outcomes were in general higher than the expected probabilities according to the CUETO risk classification, especially in the long term. No differences in progression were found when comparing these two models to the Dutch cohort. Patients who were undertreated according to the guidelines showed, in general, a higher risk of developing recurrence and progression. Limitations are i. a. its retrospective nature and the differences in grading system. Conclusion Comparisons between the observed outcomes in our Dutch cohort and the expected outcomes based on EAU and CUETO risk models and the EORTC and AUA guidelines showed that lack of adherence to existing guidelines translates into worse outcomes.