Background
Phase 3, randomized, double-blind, placebo-controlled EORTC trial (EORTC-2139-MG) in fully resected stage IIB/IIC cutaneous melanoma harbouring a BRAF V600E or V600K mutation — a population with high recurrence risk and, until recently, no targeted adjuvant option. The study planned to randomize 815 patients to demonstrate superiority for recurrence-free survival. Accrual was terminated prematurely; the protocol was then amended so that safety became the primary endpoint and RFS a secondary endpoint. Between June 9, 2022 and October 9, 2023, 339 patients were screened for a BRAF mutation and 110 were randomized. 87 (79%) had BRAF V600E and 39 (35%) had AJCC8 stage IIC disease.
Results
Interventions and follow up: Arm A: Encorafenib 450 mg PO once daily + binimetinib 45 mg PO twice daily for 12 months (n=55; 54 initiated treatment)
Arm B: Placebo for 12 months (n=55)
Primary endpoint: Safety (after protocol amendment); RFS was the original primary endpoint and became secondary
mFollow up: 12 mo (encorafenib/binimetinib) and 7 mo (placebo); data cutoff Nov 19, 2024 after the last patient discontinued
Results: RFS at 12 months: 86% (95% CI 65-95) vs 70% (95% CI 46-85)
DMFS at 12 months: 92% (95% CI 77-97) vs 82% (95% CI 55-93)
Hazard ratios and P values: NR — the amended trial was not powered for a formal efficacy comparison
OS: NR
Arm B: Placebo for 12 months (n=55)
Primary endpoint: Safety (after protocol amendment); RFS was the original primary endpoint and became secondary
mFollow up: 12 mo (encorafenib/binimetinib) and 7 mo (placebo); data cutoff Nov 19, 2024 after the last patient discontinued
Results: RFS at 12 months: 86% (95% CI 65-95) vs 70% (95% CI 46-85)
DMFS at 12 months: 92% (95% CI 77-97) vs 82% (95% CI 55-93)
Hazard ratios and P values: NR — the amended trial was not powered for a formal efficacy comparison
OS: NR
Adverse events
Grade ≥3 treatment-related (among 54 who started encorafenib/binimetinib): 13 (24%)
Discontinuation: 18 (33%) had an AE leading to permanent treatment discontinuation
Overall profile: consistent with the established encorafenib/binimetinib class effect (pyrexia, arthralgia, fatigue, transaminitis, ocular and cardiac events); no new safety signal identified
Discontinuation: 18 (33%) had an AE leading to permanent treatment discontinuation
Overall profile: consistent with the established encorafenib/binimetinib class effect (pyrexia, arthralgia, fatigue, transaminitis, ocular and cardiac events); no new safety signal identified
Conclusions
COLUMBUS-AD demonstrated a consistent and manageable safety profile and numerically encouraging 12-month RFS (86% vs 70%) and DMFS (92% vs 82%) for one year of adjuvant encorafenib plus binimetinib in resected stage IIB/IIC BRAF V600E/K melanoma. Because accrual stopped at 110 of a planned 815 patients and the primary endpoint was switched to safety, the efficacy signal is hypothesis-generating and cannot establish benefit. The trial leaves adjuvant BRAF/MEK inhibition in stage II melanoma an open question.
Key Limitations
Accrual terminated at 110 of 815 planned patients (13% of target), so the trial has no power for its original RFS comparison and no hazard ratio or P value is reported. Switching the primary endpoint from RFS to safety by amendment after accrual failure means the efficacy readout is descriptive and post hoc in spirit. Follow-up was short and markedly unbalanced (12 vs 7 months), which inflates apparent early separation. Confidence intervals around the RFS estimates are very wide and overlap substantially (86%, 95% CI 65-95 vs 70%, 95% CI 46-85). A 33% permanent discontinuation rate for adverse events in a curative-intent adjuvant population of otherwise well patients is a meaningful tolerability concern. No OS or long-term DMFS data.
Clinical Context
Adjuvant therapy for resected stage IIB/IIC melanoma is established with anti-PD-1: pembrolizumab (KEYNOTE-716) and nivolumab (CheckMate-76K) are both FDA-approved in this setting and are the NCCN- and ESMO-endorsed options alongside observation. Adjuvant BRAF/MEK inhibition with dabrafenib + trametinib is approved only for resected stage III BRAF V600-mutant melanoma (COMBI-AD); encorafenib + binimetinib is approved for unresectable or metastatic BRAF V600-mutant melanoma (COLUMBUS) but has no adjuvant indication. COLUMBUS-AD does not change practice: anti-PD-1 remains the adjuvant option for high-risk stage II disease regardless of BRAF status, and BRAF/MEK inhibitors should not be used adjuvantly in stage II outside a trial.