Background: The prognostic impact of RAS/RAF mutations in localised colon cancer needs clarification. Based on analysis of tumour specific DNA this study aimed at elucidating the prognostic impact of mutational status in serum using an extended panel of mutations. Methods: Stage I-III curatively resected colon adenocarcinoma patients (n = 294) were retrospectively included. Mutations and mismatch repair (MMR) status in tumor were determined at time of operation. The status (categorical) of mutated ctDNA in preoperative serum samples was obtained for patients with tissue mutations. Analyses were performed with droplet digital PCR technology. Hazard ratio (HR) for the association between mutation status and survival was estimated in multivariate analysis taking known prognostic factors into account. Primary endpoints were overall survival (OS) and disease free survival (DFS) and median follow-up was 3.9 and 3.0 years, respectively. Results: Mutational status in tumor alone (n = 189) had no prognostic impact (p = 0.22). Patients with RAS mutated DNA in both tumor and serum (n = 36) had a significantly worse prognosis, OS (HR= 2.30, 95% CI 1.27-4.15, p = 0.0057) and DFS (HR = 2.18, 95% CI 1.26-3.77, p = 0.0053). BRAF mutated DNA in serum and proficient MMR (pMMR) protein in tumor (n = 14) also reflected significantly worse survival, OS (HR= 3.45, 95%CI 1.52-7.85, p = 0.0032) and DFS (HR= 3.61, 95%CI 1.70-7.67, p = 0.0008). Mutational load had significant prognostic impact as well. Conclusions: Mutational status in tumor did not demonstrate any prognostic impact. RAS mutations in serum, and BRAF mutation in serum combined with pMMR in tumor were both strong independent prognostic factors.
CORE (COnnecting REpositories)
CORE (COnnecting REpositories)