Effect of KRAS and BRAF Mutations on Survival of Metastatic Colorectal Cancer After Liver Resection: A Systematic Review and Meta-Analysis

Treatment of metastatic colorectal cancer includes resection of liver metastases, however, no biomarkers drive selection of patients. We performed a meta-analysis including 1833 patients treated with complete liver resection, showing that Kirsten rat sarcoma viral oncogene homolog and b-viral oncogene homolog B1 mutations are negatively associated with survival. This supports integration of mutational status into a combined score to better identify candidates for resection of colorectal cancer liver metastases. The purpose of the study was to evaluate whether the mutational status of Kirsten rat sarcoma viral oncogene homolog (KRAS) or b-viral oncogene homolog B1 (BRAF) could be an independent prognostic factor in the subset of patients with colorectal cancer liver metastases (CRLM) who undergo complete liver resection. A systematic literature review was performed to identify articles reporting relapse-free survival (RFS) and/or overall survival (OS) of patients who underwent complete liver resection for CRLM, stratified according to KRAS and BRAF mutational status. Hazard ratios (HRs) from multivariate analyses were pooled in the meta-analysis. Eleven studies, including 1833 patients, were eligible for the meta-analysis. Nine of them reported OS stratified according to KRAS mutation. The pooled analysis revealed that KRAS mutation was negatively associated with OS (HR, 1.674; 95% confidence interval [CI], 1.341-2.089; P < .001). Nine among 11 studies reported RFS stratified according to KRAS mutation and HRs in multivariate analysis were available in 7. In a pooled analysis, KRAS mutation was negatively associated with RFS (HR, 1.529; 95% CI, 1.287-1.817; P < .001). In 3 studies HRs of the multivariate analysis regarding the OS according to BRAF mutational status were also available, showing a negative association with OS (HR, 3.055; 95% CI, 1.794-5.204; P < .001). KRAS mutations are negatively associated with OS and RFS in patients who undergo complete liver resection for CRLM. A similar negative effect on OS was observed also for BRAF mutation, although fewer studies were included. These data support integration of KRAS and BRAF mutational status into a combined predictive score for prospective assessment of outcome after resection of CRLM in clinical studies.