Clinical factors are unable to accurately predict the absence of benefit of surgery in patients operated for resection of colorectal liver metastasis

Background: A substantial proportion of patients operated of resection for colorectal liver metastases (CRLM) with curative-intent will rapidly recur after surgery, emphasizing the need to improve the current selection process for surgery. The aim of the study is to analyze clinicopathologic prognostic factors that could better identify patients that wouldn’t benefit of surgery. Methods: A prospective database including patients operated of hepatectomy for CRLM between 2005 and 2017 was analyzed. Within this population, we selected and compared 2 groups: early relapsers (ER), defined as patients with unresectable recurrence ≤1 year postoperatively considered as having not benefited of surgery and long-term survivors (LTS), defined as patients without recurrence ≥5 years after first hepatectomy. In the entire population and in the 2 subgroups, we analyzed potential predictive factors, using uni- and multivariate analysis. Results: In total population (N = 357), 5 and 10-year disease-free survival (DFS) and overall survival (OS) are 26 and 21.5% and 44 and 25% respectively. In univariate analysis, Fong’s Clinical Risk Score (CRS) >2, mutated-KRAS, major hepatectomy and positive resection margins are significant poor prognostic factors for DFS and OS. In multivariate analysis only mutated-KRAS remains a significant poor prognostic factor for DFS (HR = 1,5 Ci:1,06-2,12 p = 0,02) and OS (HR = 1,8 Ci:1,19-2,70 p = 0,005). Comparing the 2 subgroups ER-group (77 patients) and LTS-group (64 patients), representing respectively 21 and 18% of entire population the univariate analysis showed significantly more synchronous CRLM, multiple metastases, mutated-KRAS and CRS>2 in the ER group. Of note, 25% of LTS had CRS>2. In multivariate analysis, only multiple metastases remain significantly increased in ER (p = 0.016). Conclusions: Clinical factors are unable to discriminate preoperatively the patients who will benefit of surgery for CRLM from those in whom surgery will be futile. This strongly underlines the need to identify other markers of tumor biology for better individualization of the therapeutic decision. Legal entity responsible for the study: Vincent Donckier. Funding: Has not received any funding. Disclosure: The author has declared no conflicts of interest.