Optimal Time Point for Evaluation of Response to Pembrolizumab Treatment in Japanese Patients With Metastatic Urothelial Carcinoma.

Background/Aim: The duration of pembrolizumab use in actual daily practice might be shorter than that in clinical trials because termination of pembrolizumab therapy is at the discretion of the physician. We retrospectively reviewed the response to pembrolizumab in Japanese patients with metastatic urothelial carcinoma (mUC) in relation to the time to response (TTR). Patients and Methods: The records of 165 patients treated with pembrolizumab for mUC were retrospectively analyzed. Response was evaluated at 2, 4, 6 and 8 months. TTR along with time to best response were analyzed. Phase II-III clinical trials were also reviewed to compare the TTR and time to best overall response. Results: The median patient age was 70 years. The objective response rate in the total cohort was 27.1% (42 out of 155 patients). Median TTR was 2.4 months and the time to best response was 3.1 months. Radiological evaluation at each time point significantly predicted overall survival (OS). Considering the evaluation of response at 2, 4, 6 and 8 months, the response at later time points tended to predict OS better. Multivariate analysis showed that the evaluation of response at 8 months (hazard ratio=1.91, 95% confidence interval=1.16-3.16 months; p<0.01) and best response during the treatment (hazard ratio=1.69, 95% confidence interval=1.17-2.44; p<0.01) independently predicted improved OS. Conclusion: Given that response when evaluated at a later point during pembrolizumab treatment more favorably reflected improved survival than when assessed earlier, physicians may be encouraged to wait until at least the termination of pembrolizumab treatment to determine the best response. [ABSTRACT FROM AUTHOR]