Adjuvant portal liver infusion in colorectal cancer with 5-fluorouracil/heparin versus urokinase versus control. Results of a prospective randomized clinical trial (colorectal adenocarcinoma trial I).

This prospectively randomized clinical trial was carried out in four Dutch hospitals to reduce the development of metachronous liver metastases and to get a better survival in patients with colorectal malignancies after surgically radical en bloc resection of the primary tumor and the regional lymph nodes. Three hundred seventeen patients were randomized to participate in three trial arms. One group of patients was treated by surgery alone (control group); in the other patients a catheter was placed in the dilated umbilical vein and advanced until the tip was lying in the left branch of the portal vein. Fifty percent of these patients got immediate postoperative portal infusion with 1 g 5-fluorouracil (5-FU) and 5000 U heparin daily for 7 days; the others received portal vein infusion with urokinase 10.000 U/hour for 24 hours only. Three hundred four patients were eligible. Overall hospital mortality was 3.6% (11 patients) and was not influenced by adjuvant treatment. After a median follow-up of 44 months 66 patients have died with relapse and 21 as a result of other causes. The chance of developing liver metastases and other distant metastases after portal infusion with 5-FU/heparin was one third of the chance in the control group (P less than 0.001). Only an insignificant reduction of the average death rate in the 5-FU/heparin group was found. In the urokinase group no significant effect in reducing metastases or in survival was noted. Before recommending cytotoxic portal infusion as an adjuvant treatment in patients with colorectal cancer, detailed analysis of other ongoing portal infusion studies has to be awaited and careful calculations have to be made regarding how many patients really can be saved by this treatment.