1083P KEYMAKER 02B: A randomized trial of pembrolizumab (pembro) alone or with investigational agents as first-line treatment for advanced melanoma

Substudy 02B of the phase 1/2 KEYMAKER umbrella trial (NCT04305054) is evaluating first-line pembro (anti–PD-1) alone or with investigational agents for advanced melanoma. We present results from patients (pts) treated with pembro + vibostolimab (vibo; anti-TIGIT; arm 1), pembro alone (arm 2), quavonlimab (qmab; anti–CTLA-4) coformulated with pembro (qmab/pembro; arm 3), and qmab/pembro + lenvatinib (len; multitargeted TKI; arm 4).

Pts were aged ≥18 y with previously untreated unresectable stage III or IV cutaneous melanoma who had measurable disease per RECIST v1.1 and an ECOG PS of 0 or 1, but no active brain metastases. Pts were randomly assigned to open arms: pembro 200 mg IV Q3W + vibo 200 mg IV Q3W for ≤35 cycles (arm 1), pembro 400 mg IV Q6W for ≤18 cycles (arm 2), qmab 25 mg/pembro 400 mg IV Q6W for ≤18 cycles (arm 3), or qmab 25 mg/pembro 400 mg IV Q6W for ≤18 cycles + len 20 mg PO QD (arm 4). Primary end points were safety and ORR. DOR was secondary and PFS was exploratory.

Overall, 232 pts were assigned to arm 1 (n = 90), arm 2 (n = 24), arm 3 (n = 31), or arm 4 (n = 87). Median (range) follow-up was 20.3 mo (10.7-42.4) for arm 1, 35.5 mo (18.3-43.1) for arm 2, 27.2 mo (18.2-33.6) for arm 3, and 14.9 mo (7.1-35.9) for arm 4. Efficacy data are reported in the table. Treatment-related AEs (TRAEs) occurred in 88% of pts in arm 1, 79% in arm 2, 97% in arm 3, and 97% in arm 4. Grade 3-5 TRAEs occurred in 31%, 29%, 42%, and 63% of pts, respectively. Three pts (3%) in arm 1 died due to immune-mediated AEs (myasthenic syndrome, myositis, and encephalitis).

Preliminary results from KEYMAKER-U02B showed promising antitumor activity for first-line pembro + vibo and qmab/pembro. Safety was generally manageable. Additional treatment arms will be reported when available. Adjuvant vibo/pembro is being explored in high-risk stage IIB-IV melanoma in the KEYVIBE-010 study.