Phase 1b/2 study of pembrolizumab plus lenvatinib or pembrolizumab coformulated with vibostolimab in participants (pts) with metastatic neuroendocrine prostate cancer (NEPC): KEYNOTE-365 cohorts F and H.

5055 Background: Effective therapy for NEPC is an unmet need due to poor prognosis and no established standard of care. The phase 1b/2 KEYNOTE-365 study (NCT02861573) evaluated the efficacy and safety of pembrolizumab (pembro) + lenvatinib (lenva; cohort F) or coformulated pembro/vibostolimab (vibo; cohort H) in pts with NEPC. Methods: Pts with pathologically confirmed treatment-emergent neuroendocrine (t-NE) or de novo metastatic NEPC (small cell carcinoma, large cell neuroendocrine carcinoma, or mixed morphology per central review), progression ≤6 mo before screening, and an ECOG PS of 0 or 1 were enrolled. Prior androgen deprivation therapy for metastatic disease was required for pts with t-NE. Prior chemotherapy regimens (≤2) for metastatic castration-resistant prostate cancer and second-generation hormonal agents (≤2) were permitted. Pts in cohort F received pembro 200 mg IV Q3W + lenva 20 mg PO QD. Pts in cohort H received pembro/vibo 200 mg/200 mg IV Q3W. Pembro or pembro/vibo were given for ≤35 cycles; lenva was given until discontinuation criteria were met. Primary end points were safety, ORR per RECIST v1.1 by blinded independent central review (BICR), and confirmed prostate-specific antigen (PSA) response rate (≥50% decrease from baseline measured twice ≥3 weeks apart). Secondary end points included ORR, time to PSA progression, radiographic PFS (rPFS) per Prostate Cancer Working Group 3 (PCWG3)–modified RECIST (mRECIST) v1.1 by BICR, DOR and DCR per RECIST v1.1 by BICR, and OS. Results: As of August 25, 2025, 14 and 20 pts received ≥1 dose of study treatment in cohort F or cohort H, respectively. Median follow-up was 25.6 mo (range, 22.9-35.1) and 31.4 mo (range, 23.2-45.9), respectively. Among pts with RECIST-measurable disease, confirmed ORR was 23.1% (3/13; 95% CI, 5.0-53.8) in cohort F and 12.5% (2/16; 95% CI, 1.6-38.3) in cohort H. Additional efficacy results are shown in the table. Treatment-related adverse events (TRAEs) occurred in 85.7% of pts in cohort F (64.3% grade 3 or 4) and 55.0% of pts in cohort H (25.0% grade 3 or 4). No grade 5 TRAEs occurred in either cohort. In cohorts F and H, TRAEs resulted in treatment discontinuation in 21.4% and 10.0% of pts, respectively, and immune-mediated AEs occurred in 35.7% and 15.0%. Conclusions: Pembro + lenva or pembro/vibo had a manageable safety profile and only modest antitumor activity in pts with NEPC. Clinical trial information: NCT02861573 . Cohort Fn = 14 Cohort Hn = 20 PSA response rate (95% CI), % 14.3 (1.8-42.8) 10.0 (1.2-31.7) Median time to PSA progression (95% CI), mo NR (NR-NR) NR (2.9-NR) ORR per mRECIST v1.1 (95% CI), % 15.4 (1.9-45.4) 12.5 (1.6-38.3) DCR per RECIST v1.1 (95% CI), % 23.1 (5.0-53.8) 12.5 (1.6-38.3) Median DOR per RECIST 1.1 (95% CI), mo NR (1.1-NR) NR (8.3-NR) Median rPFS per mRECIST v1.1 (95% CI), mo 3.9 (2.1-NR) 2.0 (1.6-4.3) Median OS (95% CI), mo 4.3 (2.8-NR) 7.0 (2.5-12.9)