Comparative effectiveness analysis of first-line immunotherapy versus chemotherapy in metastatic urothelial carcinoma of the bladder.

ABSTRACT

BACKGROUND: Clinical trials have not shown a significant overall survival (OS) difference between chemotherapy and immunotherapy as first-line agents in metastatic urothelial carcinoma (UC). However, the generalizability of these findings in a real-world setting has not yet been evaluated in comparative effectiveness studies. OBJECTIVE: To assess the effectiveness of first-line immunotherapy compared with chemotherapy regimens on OS in patients with metastatic UC of the bladder. DESIGN, SETTING, AND PARTICIPANTS: This retrospective propensity-matched study identified metastatic bladder UC patients in the National Cancer Database from 2014 to 2017 who received either first-line immunotherapy-monotherapy or multi-agent chemotherapy, and who were not treated on a clinical trial protocol. OUTCOME MEASURES AND ANALYSIS: The primary outcome was OS from the date of diagnosis to date of death or censoring at last follow-up. Patients were stratified into first-line immunotherapy and chemotherapy treatment groups. After 11 nearest-neighbor caliper-matching of propensity scores, the survival analysis was conducted using Cox regression modeling and Kaplan-Meier estimates. RESULTS AND LIMITATIONS: A total of 2,796 patients were included in the final study population, and 960 in the matched cohort (480 per treatment group). Utilization of immunotherapy increased over the time period studied as chemotherapy decreased (Immunotherapy 3%-37%; Chemotherapy 97%-63%; P < 0.001). In the overall cohort, patients who received first-line immunotherapy were older and more comorbid than those who received first-line chemotherapy (Age 73 v. 67, respectively, P < 0.001; Charlson-Deyo score ≥2 17% v. 11.5%, respectively, P < 0.001). In the matched cohort, patients who were treated with first-line immunotherapy had similar OS to those who were treated with first-line chemotherapy (HR 0.91, 95CI 0.72-1.15). Due to the retrospective nature of the study, interpretation is limited by potential selection bias from unmeasured confounding. CONCLUSIONS AND RELEVANCE Metastatic bladder UC patients who received first-line immunotherapy had similar OS to those who received first-line chemotherapy.