Prognostic value of KRAS mutations in stage III colon cancer: post hoc analysis of the PETACC8 phase III trial dataset

Background: The prognostic value of KRAS mutations in colon adenocarcinoma is controversial. We examined this question as an ancillary study of the PETACC8 phase III trial. Patients and methods: We analyzed the pronostic impact of KRAS exon 2 mutations in stage III colon cancer patients (n = 1657) receiving adjuvant FOLFOX +/- cetuximab therapy included in the PETACC8 trial. Patients with BRAF-mutated cancers were excluded and, as no difference was found for time to recurrence (TTR) and disease-free survival (DFS) between treatment arms, both were pooled for analysis. Associations with TTR and DFS were analyzed using a Cox proportional hazards model. Results: KRAS mutations were found in 638 of 1657 tumors and linked to shorter TTR (P < 0.001). However, when specific mutations were compared with wild-type, codon 12 mutations [hazard ratio (HR) 1.67, 95% confidence interval (CI) 1.35-2.04; P < 0.001] but not codon 13 (HR 1.23, 95% CI 0.85-1.79; P = 0.26) were significantly associated with shorter TTR, independently of other covariates. The interaction test showed that, regarding tumor location (distal versus proximal), KRAS genotype affects differently on recurrence (P = 0.02) and DFS (P = 0.042). Subgroup analysis showed that KRAS only affected TTR and DFS in distal tumors (n = 1043; 692 wild type; 351 mutated), with an increased risk of relapse (HR 1.96, 95% CI 1.51-2.56; P < 0.0001) for KRAS codon 12 mutations and a borderline significance for codon 13 mutations (HR 1.59, 95% CI 1.00-2.56; P = 0.051). Conclusion: KRAS exon 2 mutations are independent predictors of shorter TTR in patients with resected stage III distal colon cancers receiving adjuvant therapy. Future clinical trials in the adjuvant setting should consider both the tumor location and KRAS mutations as important stratification factors.