Regional infusion for colorectal hepatic metastases. A randomized trial comparing the hepatic artery with the portal vein.

To evaluate which site of infusion maximizes tumor response to regional hepatic chemotherapy, 25 patients were randomized preoperatively to receive chemotherapy administration via hepatic artery or portal vein catheters. Of the 25 patients, five did not have catheters placed due to extrahepatic disease, and one patient did not receive infusion chemotherapy. Continuous floxuridine infusion was given at a dosage of 0.3 mg/kg/d, alternating with saline solution every two weeks. The mean percentage of hepatic tumor replacement (36% vs 41%), mean serum alkaline phosphatase level (218 vs 179 IU/mL), and mean plasma carcinoembryonic antigen level (689 vs 563 ng/mL) were similar in both groups. Four (50%) of eight hepatic artery patients responded to treatment compared with 0/11 portal vein patients. Ten of 11 portal vein patients showed progressive tumor growth; nine of these patients were crossed over to arterial infusion. Three (33%) of these nine patients had an objective tumor response with the hepatic arterial infusion. This prospective, randomized clinical trial demonstrates significantly improved tumor response after hepatic arterial infusion compared with portal vein infusion.