Treatment of primary colon cancer with peritoneal carcinomatosis: comparison of concomitant vs. delayed management.

PURPOSE: The initial dissemination of colon cancer occurs through three routes: the lymphatics, the portal blood, and the peritoneal surfaces. Although lymphatic and hematogenous metastases indicate an aggressive disease process, it is possible that dissemination to peritoneal surfaces may be only superficial contamination of the parietal and visceral peritoneum that may be treatable for cure. Unfortunately, surgery may have an adverse impact on peritoneal surface dissemination. Surgical interventions may convert a superficial process into an invasive condition with a greatly reduced prognosis. This study was conducted to test this hypothesis by the use of data prospectively recorded from patients treated for peritoneal carcinomatosis concomitant with resection of the primary colon cancer or treated for carcinomatosis after disease recurrence prompted referral. METHODS: Our first group of patients had definitive treatment of carcinomatosis simultaneous with resection of the primary colon cancer. They had cytoreductive surgery including peritonectomy procedures followed by heated intraoperative intraperitoneal chemotherapy with mitomycin C plus early postoperative intraperitoneal 5-fluorouracil. The comparison group was treated with a colon resection at an outside hospital and then later referred to us with progressive disease for treatment. The major difference between the groups is the timing of the definitive treatment of carcinomatosis. These patients were also studied by use of the completeness of the cytoreduction score and the peritoneal cancer index as prognostic indicators. Survival was the end point for all the analysis. RESULTS: Of 104 patients with peritoneal carcinomatosis from colon or rectal adenocarcinoma, five patients (4.8 percent) had definitive treatment of the peritoneal surface spread of the cancer concomitant with resection of the primary lesion. Median survival for these patients has not been reached and their five-year survival rate is 100 percent. The remainder of the patients (n = 99) were referred for local and regional recurrence after their primary cancer had been removed and there was progression of carcinomatosis. Forty-four patients (42.3 percent) had a complete cytoreduction resulting in a 24-month median survival and a 30 percent five-year survival (P < 0.0001). The other 55 patients (52.9 percent) had an incomplete cytoreduction resulting in a 12-month median survival and a 0 percent five-year survival (P < 0.0001). Patients with a peritoneal cancer index of 10 or less had a 48-month median survival and a 50 percent five-year survival rate. Patients with a peritoneal cancer index between 11 and 20 had a 24-month median survival and a 20 percent five-year survival rate (P < 0.0001). Patients with a peritoneal cancer index of more than 20 had a 12-month median survival and a 0 percent five-year survival (P < 0.0001). CONCLUSIONS: In patients with peritoneal seeding occurring at the time of resection of the primary malignancy, peritonectomy procedures and perioperative intraperitoneal chemotherapy should be performed concomitantly. By use of a quantitative scoring system, the mass of cancer present in the abdomen and pelvis at the time of treatment of carcinomatosis correlated directly with survival. Aggressive treatment of patients with peritoneal carcinomatosis requires consideration in the management of colorectal cancer.

Duke Scholars

Author David Masao Ota Surgical Oncology