Overall survival benefit of doublet versus triplet BRAF inhibitor–based therapy: Insights from the BEETS study (JACCRO CC-18).

3540 Background: Cetuximab plus encorafenib, with or without binimetinib, is established as one of the standard treatment options for BRAF V600E-mutated metastatic colorectal cancer (mCRC) in the second- or later-line setting. Although the triplet therapy (encorafenib/cetuximab/binimetinib; T) is approved in Japan and commonly used for patients with poor prognostic features based on guideline recommendations, objective biomarkers to guide optimal patient selection remain lacking. Given the BREAKWATER trial support for first-line doublet therapy (encorafenib/cetuximab; D) with chemotherapy, the clinical role of MEK inhibitor–containing regimens may become more limited; nevertheless, the development of biomarkers capable of identifying patients who are likely to benefit from MEK inhibition remains an important unmet need. We therefore performed a prospective translational analysis within the BEETS study to identify biomarkers that guide treatment selection between T and D. Methods: BEETS (JACCRO CC-18: UMIN000045530) was a multicenter, prospective biomarker study of second/third-line encorafenib-based D or T regimen. We prospectively collected paired pre- and post-treatment blood samples, and analyzed pretreatment tumor-educated platelet RNA by RNA sequencing (TEP-seq). To address baseline imbalances, propensity score matching (PSM) was performed using ECOG PS, number of metastatic organs (≥3), C-reactive protein (>1 mg/dL), and prior primary tumor resection. Interaction between gene expression (z-score) and treatment arm was evaluated using Cox proportional hazards models. Candidate biomarkers were prioritized based on |log2(HR)| ≥ 0.5 and P < 0.05. Results: Of 203 enrolled patients, 189 were evaluable for TEP-seq. After PSM, 81 patients per arm were analyzed. In the matched cohort, median overall survival (OS) was longer for D than T (17.4 months [m] vs. 10.2 m, HR 1.80; 95% CI:1.24–2.61). Interaction screening identified 44 genes for which higher expression was associated with better outcomes on D but worse outcomes on T, and six genes showing the opposite pattern. Among these candidates, RAD51C expression emerged as a representative marker; when dichotomized at the arm-specific median, higher expression favored T but indicated worse outcomes with D ( RAD51C Low vs. High; T 7.8m vs. 10.6m, HR 0.79, 95%CI 0.49–1.28; D 28.2m vs. 12.9m, HR 2.02, 95%CI 1.16–3.53). Pathway-level exploration suggested a possible association between elevated RAD51C expression and increased MAPK-related signaling activity, potentially explaining the requirement for more intensive MEK inhibition in these patients. Conclusions: While OS was superior with D regimen in this matched cohort, RAD51C expression was identified as a candidate biomarker for identifying patients who may derive preferential benefit from MEK inhibition. Clinical trial information: UMIN000045530.