Forty-eight patients with measurable metastatic colorectal cancer were treated with leucovorin (FA), 80 mg/m2 as an intravenous (i.v.) infusion over 20 h, followed by 5-fluorouracil (5-FU), 400 mg/m2, i.v. push on days 1-3. This same dose of FA was then administered as a 1-h infusion on day 11 and weekly thereafter, followed by 5-FU, 400 mg/m2, slow intravenous (i.v.) push. Thirty patients had received no prior chemotherapy, and 18 patients had received chemotherapy with various other regimens. The drugs were well tolerated, with minimal hematologic toxicity and mild to moderate gastrointestinal and mucosal toxicity. Three complete responses (CRs) and nine partial responses (PRs) were observed. Nineteen additional patients remained stable for prolonged periods of time, and 12 patients continued to progress through treatment. The median survival for responders was 52.5 weeks. The patients who had received prior chemotherapy had lower response rates than the previously untreated patients, but responses were seen in both subsets. The CR plus PR rate in previously untreated patients was 30%, with 40% having stabilization of disease. The CR plus PR rate in patients previously treated with other chemotherapy regimens was 19%, with an additional 50% having stabilization of disease. Patients with stable disease appeared to have a clinical benefit similar to that of patients with PR. Toxicity was generally mild and acceptable and consisted mainly of nausea, vomiting, and mucositis. At these dosage levels, diarrhea was not a limiting toxicity. CRs and PRs occurred only in patients with Eastern Cooperative Oncology Group (ECOG) performance status 1 and 2. Carcinoembryonic antigen (CEA) levels generally reflected the type of clinical response and could often predict responders early on in the course of treatment, but did not quantitatively correlate with the degree of response. This regimen has efficacy at least similar to that of standard 5-FU regimens, and appears to have activity as a "salvage therapy" also. The minimal toxicity of this regimen, its convenience for outpatient use, and the relatively low expense as compared with regimens employing a higher dose of FA recommend it for further investigation.