Background: The present study tested cancer-specific (CSM) and overall mortality (OM) after partial cystectomy (PC) for variant histology bladder cancer (non–urothelial carcinoma of the urinary bladder UCUB), relative to UCUB and relative to radical cystectomy (RC). Materials and Methods: Within the Surveillance, Epidemiology, and End Results registry (2001-2016), we identified patients with stage T1-T2N0M0 non-UCUB and UCUB who had undergone PC or RC. Non-UCUB included adenocarcinoma, squamous carcinoma, neuroendocrine carcinoma, and other histologic subtypes. First, CSM and OM after PC were compared between the non-UCUB and UCUB groups. Second, CSM and OM after PC were compared with RC in the non-UCUB group. Kaplan Meier plots and multivariable Cox regression models were used before and after inverse probability of treatment weighting. Results: Overall, 248 patients (16.3%) treated with PC had had non-UCUB. Of the 248 cases, 115 (46.5%), 50 (20%), 34 (14%), and 49 (19.5%) were adenocarcinoma, squamous carcinoma, neuroendocrine carcinoma, and other histologic subtypes, respectively. The comparison between PC in the non-UCUB and PC in the UCUB group showed higher CSM (hazard ratio, 1.4; P = .03) but the same OM rates (hazard ratio, 1.1; P = .7) in the non-UCUB group. The comparison between PC and RC for the non-UCUB group showed no CSM or OM differences. Conclusions: PC for non-UCUB was associated with higher CSM compared with PC for UCUB. However, PC instead of RC for select patients with non-UCUB appears not to undermine cancer-control outcomes. Thus, the excess CSM is probably unrelated to cystectomy type but could originate from differences in the tumor biology. These results could act as hypothesis generating for the design of future trials.
- Inverse probability of treatment weighting
- Neuroendocrine carcinoma
- Other histologic subtypes
- Squamous carcinoma
ASJC Scopus subject areas