Unveiling a BRAF Signature Proficient in Accurately Capturing Oncogenic Activity and Guiding Prognostic Prediction Across Multiple Cancers

Although <i>BRAF</i> is frequently mutated across multiple cancer types, its clinical utility as a prognostic biomarker has remained inconsistent in clinical practice, likely due to additional events modulating BRAF signaling pathways. This inconsistency has driven our investigation into the broader landscape of BRAF signaling and the development of a robust molecular signature to assess BRAF-driven oncogenic activity. To achieve this, we introduced BRAF25, a transcriptional signature designed to effectively capture BRAF oncogenic activity. Our findings reveal that 25.6% of TCGA colorectal cancer (CRC) tumors exhibit BRAF pathway activation, even in 19.4% of BRAF wild-type (WT) cases, suggesting alternative mechanisms driving pathway activation. The BRAF-active subtype, termed BAG-3 (BRAF Activity Group-3), demonstrated reduced responsiveness to chemotherapy and anti-BRAF therapy. Notably, BRAF25 subtyping addresses the limitations of using <i>BRAF</i> mutation alone to predict patient survival. We experimentally screened and validated DUSP6 as a sensitizing target for anti-BRAF therapy, enhancing BRAF inhibitor efficacy in CRC. Furthermore, pan-cancer analyses implicate the BRAF25 signature in poor prognosis across diverse BRAF-driven malignancies. In conclusion, stratifying patients by transcriptional BRAF oncogenic activity, instead of relying solely on <i>BRAF</i> mutation status, provides a more precise approach to guide clinical decision-making and improve therapeutic outcomes.