Defining optimal selection criteria to improve prognosis and avoid futility in surgical resection of colorectal peritoneal metastases.

36 Background: Surgical resection of peritoneal metastases of colorectal cancer (CRC PM) may benefit some patients similar to hepatic metastases. This approach remains controversial in part due to inconsistent selection criteria and reported outcomes. The impact of preoperative clinical characteristics and tumor molecular profiles on survival among surgically treated patients is incompletely understood. The aim of this study was to investigate the relationship between possible predictive variables and survival in a large cohort of patients treated on a standardized clinical pathway and to develop a clinically useful patient selection tool. Methods: This retrospective cohort study utilized the database of the Catalonian peritoneal metastases regional program, established in 2006. The program provides treatment for all PM patients within an autonomous region with a population of 7.5 million and includes a single dedicated high volume surgical unit. We included all adult patients with surgically resected CRC PM. The clinical pathway includes the administration of perioperative (neoadjuvant and adjuvant) systemic chemotherapy and has historically included a dose of heated intraoperative intraperitoneal chemotherapy (HIPEC) at the time of surgery. Demographic and clinical data was analyzed with descriptive statistics. Survival and the associated predictors were analyzed with the Cox proportional hazard model with HRs used to create a predictive nomogram. Results: A total of 538 patients (mean age 59) have been treated with surgery and a complete resection (CC0) was achieved in 94% of cases. Planned preoperative systemic chemotherapy was delivered in 95% of cases, consistent with the clinical pathway. Surgical morbidity was low (urgent reoperation 6.1%, postoperative return to the ICU 2.4%, 30-day readmission 4.3%) as was 30-day mortality (0.4%). After a median follow up of 27.5 months, the median overall survival (OS) was 43.1 months. It varied considerably by subgroup: patients with low peritoneal tumor burden had the highest OS (median 49 months) and those with signet ring cell subtype the lowest (median 19 months). Factors independently associated with survival were: N stage at diagnosis, histological subtype, quantified peritoneal tumor burden (PCI score) and presence of visceral involvement. A selection tool and predictive nomogram were designed incorporating the results of the multivariable analysis. Conclusions: A set of clinical variables can be identified that independently influences patient survival after surgical resection of CRC PM that should be used as selection criteria when considering surgery. With optimized patient selection and treatment within a specialized center, excellent survival and low morbidity can be anticipated following surgery for CRC PM.