Radical cystectomy versus trimodal therapy for non-metastatic muscle-invasive bladder cancer: Analysis of an other-cause mortality matched cohort

Document Type

Conference Proceeding

Publication Date

3-1-2024

Publication Title

Eur Urol

Abstract

Introduction & Objectives: Although Trimodal therapy (TMT) is now accepted for well-selected patients with muscle-invasive bladder cancer (MIBC), in clinical practice it is often reserved for sicker patients for whom radical cystectomy (RC) is not a feasible option. Thus, comparative effectiveness studies (TMT vs RC) based on retrospective studes are usually hindered by selection bias. To circumvent this limitation, we designed a novel approach matching patients based on their calculated other-cause mortality (OCM) risk. Using this homogeneous cohort, we tested the impact of TMT vs RC on cancer-specific mortality (CSM). Materials & Methods: The Surveillance, Epidemiology and End Results (SEER) database was queried to identify patients diagnosed with histologically confirmed T2-4 MIBC between 2004 and 2018. A Cox regression model calculating 5-years OCM was used to create a 1:1 propensity-score matched cohort of patients treated with RC vs TMT. Cumulative incidence curves depicted CSM and OCM, while Fine-Gray regression tested the impact of treatment type on CSM in the matched cohort. Patients were further stratified according to Chemotherapy receipt and clinical stage based on transurethral resection of bladder tumor (cT2 vs T3-4) and the aforementioned methodology was repeated. Results: We identified 6,587 (76%) treated with RC and 2,057 (24%) TMT. Median follow-up was 3.0 years (IQR 1.1-6.7). In the unmatched cohort, 5-year OCM and CSM rates were 14% and 40% in RC group respectively, versus 23% and 47% in TMT (all p<0.001). Our matched cohort included 6,506 patients equally distributed for treatment type (RT vs RP), with no difference in 5-years OCM. In the matched cohort, the 5-year CSM rate was 42% in RC patients versus 48% in TMT (p=0.001). This trend was confirmed at multivariable analysis, where TMT patients had a 1.54-fold higher CSM risk than their RC counterparts. After stratifying the matched cohort for Chemotherapy receipt, the 5-year CSM rate was significantly higher for TMT versus RC who did not receive Chemotherapy (48% vs 41%, p<0.001). On the other hand, no difference in CSM was recorded when compare patients treated with TMT vs RC plus Chemotherapy (46% vs 45%, p=0.2). When stratified patients based on their clinical stage, TMT shower higher CSM rates than RC both in cT2 (44% vs 26%, p<0.001) and cT3-T4 (59% vs 53%, p < 0.02), albeit not being an independent predictor of CSM in cT3-T4 patients. Conclusions: In comparison with previous literature, our study is the first to mitigate the impact of selection bias by matching based on OCM. Our findings showed invariably CSM advantage of RC over TMT in pT2 patients. Conversely, TMT prooved to be a valid option for patients who showed advanced clinical disease, as RC did not showed any oncological survival advantage in cT3-cT4 patients or for those who required chemotherapy in combination with surgery.

Volume

85

First Page

S1963

Last Page

S1963

Find It @ Sladen

Share

COinS