Introduction:

The standard surgical treatment for muscle invasive bladder cancer (MIBC) is radical cystectomy (RC) with bilateral pelvic lymph node dissection (RC). In the absence of randomized data, we have previously demonstrated comparable oncological outcomes of RC to trimodality therapy (TMT: complete transurethral of bladder tumor [TURBT] and concurrent chemoradiation) using a propensity score-matched (PSM) multi-institutional cohort. We conducted a population level retrospective cohort study to examine the comparative effectiveness of RC and TMT with respect to oncologic outcomes in patients with MIBC.

Methods:

We utilized TriNetX capturing data from large health care organizations – acute care centers, outpatient centers and academic hospitals. Patients >18 years old with a histologic diagnosis of muscle invasive urothelial carcinoma who underwent either TMT (transurethral resection of bladder tumor followed by concurrent chemoradiation therapy - defined as radiation received within one month of systemic chemotherapy) or RC were identified using billing codes (e.g. ICD-10, CPT). The primary outcome was overall survival (OS) and the secondary outcome was metastasis-free survival (MFS). Cohorts were 1:1 propensity score matched using patient (demographic, comorbidities and risk factors), treatment (e.g. receipt of chemotherapy) and tumor (e.g. T-stage, and hydronephrosis) at a standardized mean difference(SMD) of <0.1 by nearest neighbor greedy algorithm. The association between treatment modality and outcomes were examined using cox-proportional hazard modelling to report hazard ratios (HR) and 95% confidence intervals (CIs). To assess for heterogeneity of effect, we conducted subgroup analysis stratified by age (≥80years, <80years), sex (male, female), treatment center (academic or non-academic), presence of carcinoma-in-situ, and presence of hydronephrosis.

Results:

We identified 16797 and 1161 patients who underwent RC and TMT, respectively over 2005 to 2025. After PSM, 1113 patients were identified in each cohort, and well balanced (all SMD<0.1). Follow-up of median (interquartile range) of 2.16 (3.99) years and 1.92 (3.04) years was available. Receipt of RC was associated with a decreased risk of all-cause mortality [HR 0.72, 95%CI: 0.62-0.84]. Conversely, we found no difference in the risk of metastasis [HR 0.86, 95%CI: 0.71-1.04] in patients receiving RC compared to those undergoing TMT. 

Conclusion:

In this multicenter propensity score matched retrospective cohort study with short follow up, we found that while patients undergoing TMT had greater risk of all-cause mortality, which may be explained by residual confounding from unmeasured comorbidities, metastasis-free survival was comparable to those undergoing RC. These data help inform patient counselling during treatment selection for MIBC. 

Funding: N/A