Although the efficacy of 5-fluorouracil (5-FU) modulated by leucovorin is well established for advanced colorectal cancer, the question of the most effective regimen and optimal dose of leucovorin remains unanswered. This prospective randomized trial compares low-dose (group 1) and high-dose (group 2) leucovorin, combined with the same dose of 5-FU to determine whether high-dose leucovorin was more beneficial than low-dose on overall survival. Inclusion criteria were: unresectable metastatic colorectal carcinoma, with or without evaluable tumor response; a performance status of less than grade 3 (World Health Organization classification); and no previous chemotherapy for metastases. Forty-two patients were randomized in group 1 (leucovorin, 20 mg/m2/day, days 1 through 5) and 41 patients in group 2 (leucovorin, 200 mg/m2/day, days 1-5). All the patients in the two groups received a 1-hour infusion of 400 mg/m2/day 5-FU every 4 weeks. The two groups were matched with no statistically significant differences in gender ratio, site of primary tumor, performance status, and tumor extent. Toxicity in the two regimens was low and not significantly different between the two groups. Overall median survival was 346 days in group 1 and 323 days in group 2 and was not significantly different between the two groups. At 1 year, the test of equivalence was significant (p < 0.01), demonstrating an absence of more than 20% benefit in 1-year survival for the high-dose regimen. The use of high-dose leucovorin combined with 5-FU in the 5-day regimen does not significantly improve overall survival for patients who have metastatic colorectal cancer.