IMPORTANCE: The prognostic value of BRAF and KRAS mutations in patients who have undergone resection for colon cancer and have been treated with combination leucovorin, fluorouracil, and oxaliplatin (FOLFOX)-based adjuvant chemotherapy is controversial, possibly owing to a lack of stratification on mismatch repair status.

OBJECTIVE: To examine the prognostic effect of BRAF and KRAS mutations in patients with stage III colon cancer treated with adjuvant FOLFOX with or without cetuximab.

DESIGN, SETTING, AND PARTICIPANTS: This study included patients with available tumor blocks of resected stage III colon adenocarcinoma who participated between December 2005 and November 2009 in the PETACC-8 phase III randomized trial. Mismatch repair, BRAF V600E, and KRAS exon 2 mutational status were determined on prospectively collected tumor blocks from 2559 patients enrolled in the PETACC-8 trial. The data were analyzed in April 2015.

INTERVENTION: Patients were randomly assigned to receive 6 months of FOLFOX4 or FOLFOX4 plus cetuximab after surgical resection for stage III colon cancer.

MAIN OUTCOMES AND MEASURES: Associations between these biomarkers and disease-free survival (DFS) and overall survival (OS) were analyzed with Cox proportional hazards models. Multivariate models were adjusted for covariates (age, sex, tumor grade, T/N stage, tumor location, Eastern Cooperative Oncology Group performance status).

RESULTS: Among the 2559 patients enrolled in the PETACC-8 trial (42.9% female; median [range] age, 60.0 [19.0-75.0] years), microsatellite instability (MSI) phenotype, KRAS, and BRAF V600E mutations were detected in, respectively, 9.9% (177 of 1791), 33.1% (588 of 1776), and 9.0% (148 of 1643) of cases. In multivariate analysis, MSI (hazard ratio [HR] for DFS: 1.10 [95% CI, 0.73-1.64], P = .67; HR for OS: 1.02 [95% CI, 0.61-1.69], P = .94) and BRAF V600E mutation (HR for DFS: 1.22 [95% CI, 0.81-1.85], P = .34; HR for OS: 1.13 [95% CI, 0.64-2.00], P = .66) were not prognostic, whereas KRAS mutation was significantly associated with shorter DFS (HR, 1.55 [95% CI, 1.23-1.95]; P < .001) and OS (HR, 1.56 [95% CI, 1.12-2.15]; P = .008). The subgroup analysis showed in patients with microsatellite-stable tumors that both KRAS (HR for DFS: 1.64 [95% CI, 1.29-2.08], P < .001; HR for OS: 1.71 [95% CI, 1.21-2.41], P = .002) and BRAF V600E mutation (HR for DFS: 1.74 [95% CI, 1.14-2.69], P = .01; HR for OS: 1.84 [95% CI, 1.01-3.36], P = .046) were independently associated with worse clinical outcomes. In patients with MSI tumors, KRAS status was not prognostic, whereas BRAF V600E mutation was associated with significantly longer DFS (HR, 0.23 [95% CI, 0.06-0.92]; P = .04) but not OS (HR, 0.19 [95% CI, 0.03-1.24]; P = .08).

CONCLUSIONS AND RELEVANCE: BRAF V600E and KRAS mutations were significantly associated with shorter DFS and OS in patients with microsatellite-stable tumors but not in patients with MSI tumors. Future trials in the adjuvant setting will have to take into account mismatch repair, BRAF, and KRAS status for stratification.

TRIAL REGISTRATION: EudraCT 2005-003463-23.