Adjuvant encorafenib and binimetinib in stage II melanoma: an early-closed trial with an unanswered question

Design

Randomised placebo-controlled trial, closed early; primary endpoint amended to safety

Population

110 adults with resected stage IIB/IIC BRAF V600E/K cutaneous melanoma

Primary outcome

Safety (amended); recurrence-free survival secondary

Effect

12-month RFS 86% (65–95%) vs 70% (46–85%); 33% discontinued for adverse events

EORTC 2139/COLUMBUS-AD randomised adults with resected stage IIB or IIC BRAF V600E/K-mutant cutaneous melanoma to one year of encorafenib plus binimetinib or placebo. It planned 815 patients but closed after 110, and the primary endpoint was changed from recurrence-free survival to safety. Results were published in July.

Among 54 patients who started treatment, 24% had grade 3 or worse treatment-related adverse events and 33% stopped permanently because of an adverse event. Recurrence-free survival at 12 months was 86% (95% CI 65–95%) with treatment against 70% (46–85%) with placebo; distant metastasis-free survival was 92% against 82%. Median follow-up was only 12 and 7 months.

The overlapping intervals and short follow-up mean efficacy is not established. A third of patients stopping treatment is a real cost for a group, many of whom would never relapse. For now, stage II patients should be counselled on the basis of established options and trials, not this result.

• Test for BRAF mutation in resected high-risk stage II melanoma where adjuvant options are being discussed
• Explain to patients that adjuvant targeted therapy in stage II remains unproven
• Refer eligible patients to adjuvant trials where available
• Keep structured skin and nodal surveillance regardless of adjuvant decisions

Why it matters

Stage II melanoma can recur as often as some stage III disease, and this trial could not show whether targeted adjuvant therapy helps.

Don't overread it

Recurrence-free survival became a secondary endpoint in a trial closed at 13% of planned size — this is not evidence of benefit.

The statistics, in plain English

Recurrence-free survival of 86% against 70% looks large, but the confidence intervals (65–95% and 46–85%) overlap widely, which is what happens with about 55 patients per arm. The trial was stopped long before it could answer its efficacy question, and follow-up differed between arms, which can bias the comparison.

Read the rest in the app

You have read your two free briefings this month. The app carries all 27 specialties, every morning, free — and this finding is waiting in it.

Get Daily Doctor — free

Scan to keep reading on your phone. No account needed to start.