Pembrolizumab (pembro) plus abiraterone acetate (abi) and prednisone (p) in patients with chemotherapy-naive metastatic castration-resistant prostate cancer (mCRPC): Results from KEYNOTE-365 cohort D.

Abstract

118 Background: Abi + p is a standard of care for mCRPC. Cohort D of the phase 1b/2 KEYNOTE-365 study (NCT02861573) was used to evaluate safety and efficacy of the PD-1 inhibitor pembro + abi and p in patients (pts) who had not received chemotherapy for mCRPC. Methods: Chemotherapy-naive pts who had not previously used next-generation hormonal agents (NHAs) for mCRPC or were intolerant to enzalutamide or for whom enzalutamide was ineffective for mCRPC, whose disease progressed ≤6 months before screening, and who had ECOG PS score 0/1 were eligible. Enrolled pts received pembro 200 mg IV Q3W + abi 1000 mg PO QD and p 5 mg PO BID. Primary end points were PSA response rate (PSA decrease ≥50% from baseline), confirmed ORR per RECIST v1.1 by blinded independent central review, and safety. Secondary end points included rPFS per PCWG3-modified RECIST v1.1, DCR (CR + PR + SD or non-CR/non-PD ≥6 mo), DOR, OS, time to symptomatic skeletal-related event, radiographic bone progression, and radiographic soft tissue progression. Results: Of 103 treated pts, 35.9% had RECIST-measurable disease and 26.2% had previously received enzalutamide. Median (range) time from enrollment to data cutoff was 17.6 (9.7-27.0) months. Confirmed PSA response rate in all 103 pts was 56.3%. Overall, 78.6% of pts had a reduction in PSA level from baseline (confirmed and unconfirmed). For 37 pts with RECIST-measurable disease, ORR was 16.2% (1 CR; 5 PRs) overall, 7.7% for those who previously received enzalutamide (n = 13) and 21.7% for those who had not previously received NHAs (n = 23). Two pts with RECIST-nonmeasurable disease had a CR. DOR was not reached (NR; range, 2.1+ to 19.4+ mo); 4 pts had a response ≥12 months. DCR was 44.7% overall, 11.1% in pts who previously received enzalutamide (n = 27), and 57.3% in pts who had not previously used NHAs (n = 75). Additional analyses are listed in the table. Treatment-related AEs (TRAEs) were experienced by 90.3% of pts; 36.9% experienced grade 3-5 TRAEs. Overall, 18.4% of pts had a grade 3/4 ALT laboratory elevation and 12.6% had a grade 3/4 AST elevation. Five pts died of AEs; 1 was treatment-related (myasthenic syndrome). Conclusions: Pembro + abi and p showed antitumor activity in chemotherapy-naive pts with mCRPC. Safety was generally consistent with individual profiles of each agent, although there was an increased incidence of grade 3/4 ALT/AST laboratory elevations than reported for the individual treatments. Clinical trial information: NCT02861573. [Table: see text]