Retrospective analysis identifies molecular correlates of benefit with encorafenib-based therapy in BRAF-mutant melanoma

This publication presents a retrospective exploratory analysis of genomic and transcriptomic data from the phase III COLUMBUS randomized controlled trial. The study population consisted of 921 patients with BRAF V600E/K-mutant locally advanced, unresectable, or metastatic melanoma. The original trial compared three treatment arms: encorafenib plus binimetinib, encorafenib monotherapy, and vemurafenib monotherapy. The specific dosing regimens and treatment protocols were not reported in this analysis. The clinical setting was also not reported. This analysis aimed to identify molecular correlates of treatment benefit rather than report primary clinical efficacy outcomes.

The analysis did not report results for a pre-specified primary outcome. Instead, it explored several molecular associations. The investigators reported that survival benefits with the combination of encorafenib plus binimetinib versus vemurafenib appeared greatest in patients with higher tumor mutational burden (TMB) and in those with evidence of tumor immune infiltration. No effect sizes, absolute numbers, confidence intervals, or p-values were provided for this observation. The direction of the association was reported as improved survival in these subgroups.

Key secondary exploratory outcomes focused on molecular profiling. Gene expression analysis identified three tumor subgroups, including an 'immune' subgroup that was associated with improved survival. Again, no quantitative data on the strength of this association were provided. The analysis of circulating tumor DNA (ctDNA) found that the detection of BRAF V600 alterations in baseline ctDNA samples was associated with shorter survival across treatment arms. Conversely, the clearance of these BRAF V600 alterations at cycle 2, day 1 was associated with improved survival across all treatment arms. For all these ctDNA findings, specific hazard ratios, risk differences, confidence intervals, and p-values were not reported.

Detailed safety and tolerability findings from this molecular analysis were not reported. The publication did not provide rates of adverse events, serious adverse events, or treatment discontinuations related to the encorafenib-based regimens or vemurafenib. Tolerability profiles for the subgroups defined by TMB or immune infiltration were also not described.

These results add a molecular dimension to the understanding of the COLUMBUS trial, a landmark study that established encorafenib plus binimetinib as a standard targeted therapy option in BRAF-mutant melanoma. Prior analyses from COLUMBUS demonstrated superior progression-free and overall survival for the combination over vemurafenib. This new analysis suggests that underlying tumor biology, specifically TMB and immune context, may influence the magnitude of benefit from this combination relative to an older BRAF inhibitor. The ctDNA findings align with growing evidence across oncology that ctDNA dynamics can serve as a prognostic and early pharmacodynamic biomarker.

This analysis has several important methodological limitations. First, it is explicitly a retrospective and exploratory analysis, meaning the findings are hypothesis-generating rather than confirmatory. The lack of reported statistical measures (p-values, confidence intervals) and absolute effect sizes prevents assessment of the strength and precision of the observed associations. There is a risk of bias from multiple testing across many molecular endpoints without appropriate statistical correction. Furthermore, the clinical applicability is limited because the analysis does not provide thresholds for 'high' TMB or define 'immune infiltration' in a clinically actionable way.

The clinical implications are cautious. The analysis suggests that BRAF V600 detectability in ctDNA may have utility as a marker of prognosis and early treatment response in patients with BRAF-mutant melanoma receiving targeted therapy. The observation that benefit from encorafenib plus binimetinib may be enriched in patients with higher TMB or an immune-active tumor microenvironment is intriguing but not yet practice-changing. Clinicians should interpret these findings as identifying potential biomarkers for future study, not as criteria for selecting therapy in current practice.

Several critical questions remain unanswered. Prospective validation of TMB and immune signatures as predictive biomarkers for encorafenib-based therapy is needed. The optimal cut-points for defining 'high' TMB in this context are unknown. The clinical utility of monitoring BRAF V600 ctDNA for making real-time treatment decisions has not been established. Finally, how these molecular features interact with or predict benefit from subsequent immunotherapy, a key treatment sequence in melanoma, requires investigation.