Response to Pembrolizumab After Dose-Reduced Cisplatin Plus Gemcitabine Chemotherapy Is Inferior to That After Carboplatin Plus Gemcitabine Chemotherapy in Cisplatin-Unfit Patients With Advanced Urothelial Carcinoma.

INTRODUCTION: Response to pembrolizumab after first-line chemotherapy is vital to prolonged survival in advanced, unresectable, and/or metastatic urothelial carcinoma (aUC). However, there are sparse clinical data on host-tumor immune modification by first-line platinum-based chemotherapy. This study investigated the association between response to first-line gemcitabine plus cisplatin (GC) or carboplatin (GCarbo) chemotherapy and response to subsequent pembrolizumab treatment. PATIENTS AND METHODS: A multicenter-derived database registered 454 patients diagnosed with aUC between 2008 and 2020. Of these, 108 patients who received first-line GC or GCarbo followed by second-line or later pembrolizumab were eligible for investigation and were classified into 3 groups: 48 receiving full-dose GC, 21 receiving dose-reduced GC, and 39 receiving GCarbo. Overall survival (OS) was calculated using the Kaplan-Meier method and compared using the log-rank test. Possible factors associated with the response to pembrolizumab were evaluated using binary logistic regression methods. RESULTS: The rate of patients undergoing surgical removal of the primary organ was higher and creatinine clearance was lower in the dose-reduced GC and GCarbo groups than in the full-dose GC groups. Pembrolizumab responders had significantly better survival benefits than nonresponders. The rate of pembrolizumab responders was much higher in first-line chemotherapy responders than in first-line chemotherapy nonresponders. In contrast to the full-dose GC and GCarbo groups, the pembrolizumab responder rate was lower, and no association was observed between response to first-line chemotherapy and response to pembrolizumab in the dose-reduced GC group. CONCLUSION: Cisplatin and carboplatin may play an important role in the antitumor immune response, which could impact the outcome of subsequent pembrolizumab treatment. Given that the rate of response to pembrolizumab after dose-reduced GC chemotherapy was relatively low, this regimen is not recommended for cis-unfit patients with aUC. Further studies are required to understand the mechanisms responsible for the cross-reactivity of platinum and immune checkpoint inhibitors.