Abstract
INTRODUCTION: Data about the role of chemotherapy in sarcomatoid bladder cancer (SBC) are limited. We addressed the effect of chemotherapy in non-metastatic SBC patients treated with radical cystectomy (RC). METHODS: Using the Surveillance, Epidemiology, and End Results database (2001-2018), we identified 331 patients with non-metastatic muscle-invasive or higher SBC (T(2-4)N(0-3)M(0)). Kaplan-Meier plots and Cox regression models tested cancer-specific mortality (CSM ). Sample size and power analyses tested for power limitations. RESULTS: Of 331 SBC patients, 129 (38.9%) were exposed to chemotherapy. The rate of organ-confined stage (T(2)N(0)M(0)) was 33% in both chemotherapy-exposed and chemotherapy-naive patients. In the overall cohort, median CSM-free survival was 84 months (interquartile range [IQR] 21-NA) vs. 26 months (IQR 17-84) in chemotherapy exposed vs. chemotherapy-naive patients, respectively. In multivariable Cox regression models, chemotherapy was associated with lower CSM, without reaching statistical significance (hazard ratio [HR ] 0.72, confidence interval [CI] 0.51-1.01, p=0.054). In subgroup analyses, chemotherapy exposure in organ-confined (n=110) vs. non-organ-confined (n=221) patients resulted in a HR of 0.51 (p=0.12) vs. 0.77 (p=0.17), respectively. Power analyses, based on two-sided α=0.05, revealed values of 52%, 14%, and 43% in the entire population, organ-confined, and nonorgan-confined subgroups, respectively. CONCLUSIONS: In non-metastatic SBC treated with RC, the association between chemotherapy and lower CSM is particularly strong in organ-confined stage. A substantially larger cohort would be required to confirm the statistical significance of the recorded protective effect of chemotherapy.