Prognostic Factors of Survival in Patients with Peritoneal Metastasis from Colorectal Cancer

Objectives: The aim of this study was to analyze the prognostic factors of survival in patients with peritoneal metastasis (PM) from colorectal cancer (CRC). The type of relationship between survival and the PM time of detection was used to determine whether it was synchronous with the primary tumor or metachronous. Patients and Methods: Retrospective observational study. It included patients treated for colorectal adenocarcinoma diagnosed between January 2005 and December 2019 who presented PM at the time of diagnosis or during follow-up. Variables, such as sex, age, differentiation grade, positive adenopathy (pN+), tumor size (pT), tumor location, mucinous component, peritoneal carcinomatosis index (PCI), and KRAS mutational status, were analyzed. Results: During the study period, 1882 patients were surgically treated for CRC in our hospital. Of these, 240 patients (12.8%) were included in the study after evidence of PM. The mean age was 67 +/- 12 years (range: 32-92 years), and 114 patients were female (47.5%). The mean follow-up was 20 +/- 13 months (median 12 months). The Kaplan-Meier survival at 36 months was higher in patients with metachronous PM (24% vs. 8%; p = 0.002), WT-KRAS tumors (31% vs. 15%; p < 0.001), N0 stage (30% vs. 19%; p < 0.001), T3 stage tumors (18% vs. 19% in T4A and 3% in T4B; p > 0.001), and tumors with classic adenocarcinoma histology (18% vs. 8%; p = 0.011). Patients with a PCI of 1-10 showed a likelihood of survival at 36 months of 56%, which was longer than that found in patients with a PCI of 11-20 (8%) or a PCI of >20 (0%) (p < 0.001). In the multiple regression analysis, the factors with an independent prognostic value were: poor grade of differentiation (HR 1.995; 95% CI: 1.294-3.077), KRAS mutation (HR 1.751; 95% CI: 1.188-2.581), PCI 11-20 (HR: 9.935; 95% CI: 5.204-18.966) and PCI > 20 (HR: 4.011; 95% CI: 2.291-7.023). Conclusions: PCI should continue as the as the most useful prognostic indicator in order to assess prognostic estimations as well as therapeutic and surgical decisions, but tumor grade and KRAS mutational status may help in the treatment decision process by providing complementary information. The time of PM detection did not achieve statistical significance in the multiple regression analysis.