摘要：Background: Genetic evaluation is essential in assessing colorectal cancer (CRC) and colorectal liver metastasis (CRLM). The aim of this study was to determine the pragmatic value of
KRAS on oncological outcomes after CRLM according to the ESMO recommendations and to query whether it is necessary to request
KRAS testing in each situation. Methods: A retrospective cohort of 126 patients who underwent surgery for hepatic resection for CRLM between 2009 and 2020 were reviewed. The patients were divided into three categories: wild-type
KRAS, mutated
KRAS and impractical
KRAS according to their oncological variables. The impractical (not tested)
KRAS group included patients with metachronous tumours and negative lymph nodes harvested. Disease-free survival (DFS), overall survival (OS) and hepatic recurrence-free survival (HRFS) were calculated by the Kaplan–Meier method, and a multivariable analysis was conducted using the Cox proportional hazards regression model. Results: Of the 108 patients identified, 35 cases had
KRAS wild-type, 50 cases had a
KRAS mutation and the remaining 23 were classified as impractical
KRAS. Significantly longer medians for OS, HRFS and DFS were found in the impractical
KRAS group. In the multivariable analyses, the
KRAS mutational gene was the only variable that was maintained through OS, HRFS and DFS. For HRFS (HR: 13.63; 95% confidence interval (CI): 1.35–100.62;
p = 0.010 for
KRAS), for DFS (HR: 10.06; 95% CI: 2.40–42.17;
p = 0.002 for
KRAS) and for OS (HR: 4.55%; 95% CI: 1.37–15.10;
p = 0.013). Conclusion: Our study considers the possibility of unnecessary
KRAS testing in patients with metachronous tumours and negative lymph nodes harvested. Combining the genetic mutational profile (i.e.,
KRAS in specific cases) with tumour characteristics helps patient selection and achieves the best prognosis after CRLM resection.
