A randomized phase 2 trial of encorafenib + binimetinib + nivolumab vs ipilimumab + nivolumab in BRAFV600-mutant melanoma brain metastases: SWOG S2000 (NCT04511013).

Abstract

LBA9507 Background: While anti-PD-1 and anti-CTLA4 immunotherapies have efficacy in the treatment of patients (pts) with asymptomatic melanoma brain metastases (MBM), their efficacy in pts with symptomatic MBM is very limited. In the Checkmate-204 trial of ipilimumab with nivolumab in MBM, 6-month progression-free survival (PFS) rate was 19% with a median PFS of only 1.2 months in symptomatic pts. In the COMBI-MB study of BRAF/MEK-inhibitors in MBM, median PFS was 5.5 months in symptomatic pts (n=17). A head-to-head approach of targeted and immunotherapy has not been tested; moreover, a combination of targeted and immunotherapy is feasible as demonstrated by prior studies. Methods: SWOG S2000 is a 1:1 randomized phase 2 trial exploring the efficacy of a front-line triplet regimen of BRAF/MEK inhibitors with anti-PD-1 monotherapy (encorafenib 450 mg qday + binimetinib 30 mg BID + nivolumab 480 mg IV q4w) versus ipilimumab 3 mg/kg + nivolumab 1 mg/kg q3w in pts with symptomatic BRAF-mutant MBM. Pts were ≥18 years old, ECOG 0-2, and prior neoadjuvant or adjuvant anti-PD-1, CTLA-4, or BRAF/MEK-inhibitors were permitted, but no systemic treatment in the metastatic setting. Steroids up to 8 mg of dexamethasone/day (or equivalent), leptomeningeal spread, and prior local therapy (radiation or surgery) for MBM were permitted, if there was at least one measurable, progressing MBM ≥ 0.5 cm. Disease assessments were performed at 6 and 12w, and then q12w from treatment start until progression. Primary objective was to compare PFS (intracranial + extracranial) per RECIST 1.1 between the arms. Results: Between September 2020 and June 2024, 30 pts with symptomatic MBM were enrolled; 1 pt was ineligible. Thirteen (45%) received prior corticosteroids and 14 pts (48%) received prior local therapy for MBM. Six-month PFS rate was 50% (95% CI 23-72%) with enco/bini/nivo, vs 29% (95% CI 9-52%) with ipi/nivo. The study met its primary endpoint with a hazard ratio (HR) of 0.51 (95% CI 0.0 – 0.92), with a statistically significant one-sided p-value of 0.07, less than 0.10 alpha pre-specified by study design. Median PFS was 6.2 months (3.0- 20.4) with enco/bini/nivo, vs 1.4 months (0.7 - 13.8) with ipi/nivo. Overall response rate (PR + CR) was 57% (31-83%) with enco/bini/nivo vs 15% (10-15%) with ipi/nivo . With enco/bini/nivo, 69% of pts had grade 3-4 toxicity, and with ipi/nivo, 75% had grade 3-5 toxicity, with one death due to cardiac arrest. Conclusions: S2000 is the first randomized trial in patients with symptomatic melanoma brain metastases. A first-line triplet regimen of enco/bini/nivo demonstrated a statistically significant improvement in PFS as compared to ipi/nivo, with a HR of 0.51. Both regimens also had toxicity rates consistent with their known profiles. In this difficult-to-treat pt population frequently requiring steroids, a triplet regimen may warrant further study. Clinical trial information: NCT04511013 .