BRAF is a critical oncogene driving tumorigenesis in multiple cancer types, including melanoma, colorectal cancer, and non-small cell lung cancer. Advances in comprehensive genome profiling reveals a growing number of alterations whose clinical significance remains undefined due to the lack of functional testing. The clinical interpretation of these variants of unknown significance in BRAF pose significant challenges for personalized treatment, particularly non-V600 variants such as CRAF-dependent class-III BRAF variants. By focusing on the functional impact of these lesser-known BRAF alterations, this study aims to elucidate their role in MAPK pathway activation and assess their potential as actionable targets in cancer therapy.

We collected BRAF variants detected in cancer patients analyzed at our center between 2021 and 2025. The cohort consists of 3520 and 972 samples for DNA and RNA, respectively. For functional testing, we selected a previously uncharacterized FNBP1::BRAF fusion and several uncharacterized non-V600 BRAF variants (E510K, G466A, E549Q, E695Q), which were introduced into plasmids via Golden Gate cloning or site-directed mutagenesis, respectively. HEK-293T cells were transfected with wild-type or mutant BRAF plasmids alone or co-transfected with CRAF. After 48 hours, proteins were isolated, and p-ERK levels analyzed by western blot. Cells transfected with FNBP1::BRAF were treated with MEK and RAF inhibitors for two hours to assess p-ERK inhibition.

BRAF variants were processed to remove artefacts, low quality variants and intronic/synonymous variants, yielding a total of 247 variants. Of those, 20% are of unknown significance, with 27% located in the kinase domain, 5% in the cysteine-rich domain, and 16% in the Ras-binding domain. Functionally, cells expressing FNBP1::BRAF, identified in a glioma patient, exhibited elevated ERK activation, with p-ERK levels similar to BRAF p.V600E-transfected cells. Treatment with RAF inhibitor Dabrafenib or MEK inhibitor Trametinib and their combination successfully reduced p-ERK levels. BRAF p.E501K and BRAF p.E549Q showed kinase impairment when mono-transfected, but CRAF presence triggered ERK hyperactivation, suggesting a class-III BRAF variant. In contrast, BRAF p.G466A led to elevated p-ERK signals, regardless of CRAF presence. BRAF p.E695Q did not cause pathway overactivation, regardless of CRAF-presence.

This study identifies FNBP1::BRAF as a previously uncharacterized potential driver of glioma. Trametinib and Dabrafenib effectively reversed its constitutive ERK phosphorylation. Additionally, BRAF p.E501K and BRAF p.E549Q are likely pathogenic class-III BRAF variant, as it exhibits CRAF dependent ERK hyperactivation. In contrast, BRAF p.G466A is a likely pathogenic BRAF variant that induces ERK hyperactivation independent of CRAF.