11003 Introduction: The BRAFV600E mutation (BRAF) causes constitutive activation of the RAS-induced MAPK pathway and is found in 10% of colon cancers. B-RAF inhibitors are in early clinical development, but their development in CRC will be challenging unless subsets of patients (pts) with higher rates of BRAF can be defined. The mutation rate in rectal tumors, the concordance between primary and metastases, and the prognostic/predictive significance of BRAF are current gaps in knowledge.
METHODS
481 primary tumors and 80 matched primary-metastasis (prim-met) pairs were analysed from a pre-defined cohort of pts with CRC based on age (≥ 70 vs < 70 years), gender, tumor site (right-R, left-L and rectum), stage (A to C vs D) and ≥ 2 years follow-up. BRAF was assessed by routine sequencing of exon 15 and by a mutant-specific PCR assay. KRAS (KRAS-codon 12 and 13) and MSI (Bethesda markers) status were also examined.
RESULTS
Overall prevalence of BRAF was 11%. BRAF (see table ) was independently associated with increasing age, female gender and R-sided cancer, but not with stage. Mutations were rare in rectal cancers. BRAF was associated with inferior overall survival in stage D pts (log-rank, p = 0.0003; HR 0.38, 95% CI, 0.10-0.51). Survival analysis will be further stratified by treatment received. No difference in outcome was seen in preliminary analysis of earlier stage cancers. Mutation frequencies in the prim-met pairs were 38% (30/80) and 1.3% (1/80) for KRAS and BRAF, respectively. Overall concordance was 88% (70/80) for KRAS and 100% (80/80) for BRAF status.
CONCLUSIONS
The development of selective B-Raf inhibitors in CRC is potentially more attractive due to the ability to define patient subsets with a higher prevalence of BRAF mutations. Analysis of the primary tumor reliably predicts the status of metastatic disease in the same patient. The association between BRAFand poor outcome will need to be considered when interpreting the result of studies targeting this mutation. [Table: see text] [Table: see text].