17 patients with metastasizing colorectal cancer were treated in a phase II-study with systemic intravenous chemotherapy (Petrelli N, Proc ASCO 286, 1987) consisting of leucovorin 500 mg/m2 in a 2 hr infusion and 5-fluorouracil (5-FU) 600 mg/m2 bolus one hour after the commencement of the leucovorin infusion. Limiting toxicities were gastrointestinal, in form of nausea/vomiting of WHO grade 3 (n = 1) and of diarrhoea of maximal grade 4 (n = 5), as well as haematological with a maximal nadir of leucocytes of WHO grade 4 (n = 1) and of thrombocytes of WHO grade 1 (n = 1). Of the 14 evaluable patients an objective response was achieved in 21% of cases (CR: n = 1, PR: n = 2). 64% of the patients showed no change and 14% showed progression of their disease, 7 out of the 17 patients have died; the median survival was 24 months. With respect to objective remission, our result of 21% response rate is lower than the median of 32% of 24 different studies comprising 695 patients; nevertheless, the present response rate is within the range observed in the published studies of between 0% and 67%. In order to estimate the antitumoral efficacy of the new regimen 5-FU/leucovorin more objectively, a retrospective comparison of the therapy regimens used most frequently in disseminated colorectal cancer (5-FU monotherapy, 5-FU/methotrexate (MTX), 5-FU/MTX/low dose leucovorin, 5-FU/cisplatin, 5-FU/leucovorin) has been performed. The continuous administration of 5-FU monotherapy, with an objective response rate of 26%, was superior to standard 5-FU monotherapy as a bolus injection or short-term infusion.(ABSTRACT TRUNCATED AT 250 WORDS)