PURPOSE Pembrolizumab has previously shown antitumor activity against programmed death ligand 1 (PD-L1)–positive metastatic castration-resistant prostate cancer (mCRPC). Here, we assessed the antitumor activity and safety of pembrolizumab in three parallel cohorts of a larger mCRPC population. METHODS The phase II KEYNOTE-199 study included three cohorts of patients with mCRPC treated with docetaxel and one or more targeted endocrine therapies. Cohorts 1 and 2 enrolled patients with RECIST-measurable PD-L1–positive and PD-L1–negative disease, respectively. Cohort 3 enrolled patients with bone-predominant disease, regardless of PD-L1 expression. All patients received pembrolizumab 200 mg every 3 weeks for up to 35 cycles. The primary end point was objective response rate per RECIST v1.1 assessed by central review in cohorts 1 and 2. Secondary end points included disease control rate, duration of response, overall survival (OS), and safety. RESULTS Two hundred fifty-eight patients were enrolled: 133 in cohort 1, 66 in cohort 2, and 59 in cohort 3. Objective response rate was 5% (95% CI, 2% to 11%) in cohort 1 and 3% (95% CI,, 1% to 11%) in cohort 2. Median duration of response was not reached (range, 1.9 to $ 21.8 months) and 10.6 months (range, 4.4 to 16.8 months), respectively. Disease control rate was 10% in cohort 1, 9% in cohort 2, and 22% in cohort 3. Median OS was 9.5 months in cohort 1, 7.9 months in cohort 2, and 14.1 months in cohort 3. Treatment-related adverse events occurred in 60% of patients, were of grade 3 to 5 severity in 15%, and led to discontinuation of treatment in 5%. CONCLUSION Pembrolizumab monotherapy shows antitumor activity with an acceptable safety profile in a subset of patients with RECIST-measurable and bone-predominant mCRPC previously treated with docetaxel and targeted endocrine therapy. Observed responses seem to be durable, and OS estimates are encouraging.
Bibliographical noteFunding Information:
Funded by Merck Sharp & Dohme, a subsidiary of Merck & Co (Kenilworth, NJ). The authors thank the patients and their families and caregivers for participating in this trial, along with all investigators and site personnel. The authors also thank the following employees of Merck Sharp & Dohme, a subsidiary of Merck & Co, (Kenilworth, NJ): Scot Ebbinghaus for critical review of the manuscript draft, and Andrew Albright, Razvan Cristescu, Xinwei Sher, and Jialin Xu for assistance in generating and analyzing biomarker data. Medical writing and editorial assistance were provided by Melanie Leiby, an employee of Merck Sharp & Dohme.
© 2019 by American Society of Clinical Oncology