Comparison of prognostic scores for patients with colorectal cancer peritoneal metastases treated with cytoreductive surgery and hyperthermic intraperitoneal chemotherapy.

ABSTRACT

BACKGROUND: There are three prognostic scores for the cytoreductive surgery (CRS) and hyperthermic intraperitoneal chemotherapy (HIPEC) treatment of colorectal cancer peritoneal metastases the newly introduced COREP (colorectal peritoneal) score, the peritoneal surface disease severity score (PSDS), and the prognostic score (PS). The aim was to determine which prognostic score had the best prognostic value. METHODS: Between 2006 and 2010, a total of 77 patients with peritoneal metastases from colorectal cancer underwent CRS/HIPEC treatment. The COREP, PSDS, and PS scores were successfully applied to 56 patients (73 %) having sufficient data. The end points were prediction of open-and-close cases (n = 9), R1 resections (n = 41), and survival of <12 months (n = 18). Area under the receiver operating characteristic curves (accuracy) was compared. Subgroup analysis was performed on patients not previously used for the development of the COREP score (n = 24). Multivariable logistic regressions of the three end points were performed as well as Cox regression for overall survival. Furthermore, COREP and peritoneal cancer index were compared. RESULTS: For open-and-close case prediction, accuracy for the whole group (n = 56) and subgroup (n = 24) was 87 and 88 %, respectively for COREP; 66 and 77 % for PSDS; and 68 and 78 % for PS. For R1 resection prediction, accuracy was 81 and 81 %, 76 and 78 %, and 75 and 77 %, respectively. For prediction of survival of <12 months, accuracy was 83 and 84, 54 and 67 %, and 55 and 56 %, respectively. The COREP score was the only independent prognostic factor in all four multivariable analyses. A COREP score of ≥6 identified patients with poor survival more accurately than a PCI of >20. CONCLUSIONS: The COREP score predicted open-and-close cases, R1 resections, and poor survival better than PSDS and PS. COREP better identifies patients with poor survival than intraoperative PCI.