ABSTRACT
BACKGROUND: First-line PD-inhibition in cisplatin-ineligible patients with locally advanced or metastatic urothelial cancer represents a novel clinical setting, with uncertainty concerning second-line outcomes. Specifying second-line treatment and outcomes will provide guidance in this new sequence. We performed a retrospective chart review to document the outcomes of these patients treated at our institution. PATIENTS AND METHODS: Our cohort consisted of 43 patients with advanced urothelial cancer receiving first-line checkpoint inhibition. Baseline factors, programmed death-ligand 1 (PD-L1) status, treatments, and outcomes during and beyond the first line were obtained. Response was scored using Response Evaluation Criteria in Solid Tumors, version 1.1 criteria. Log rank tests were used to compare outcomes in prognostic subgroups, and outcome associations with PD-L1 status were analyzed with Fisher exact tests. RESULTS: A total of 43 patients received first-line atezolizumab or pembrolizumab from June 2014 until June 2018. The median age was 76.8 years, and the population was 74% male, with 60% having visceral metastases. Reasons for cisplatin ineligibility were Eastern Cooperative Oncology Group performance status 2%, 30%; renal insufficiency, 44%, and both, 21%. First-line objective response rate (ORR) was 30%, and complete response was 14%. The median overall survival was 11.7 months. Of 29 patients progressing, 17 received second-line treatment (most commonly, gemcitabine/carboplatin [10 patients]). The second-line response rate was 33%, and the ORR was 11%. The second-line median overall survival was 6.2 months. No association was found between PD-L1 status and outcomes. CONCLUSION: Outcomes with first-line immunotherapy are consistent with historical outcomes. The ORR after first-line checkpoint inhibition falls short of historical comparators; however, the response rate compares favorably to those of chemotherapies used in previous second-line regimens. The older age and poorer performance status may have contributed to second-line outcomes.
