This randomised phase II study evaluates the safety and efficacy profile of uracil/tegafur/leucovorin combined with irinotecan (TEGAFIRI) or with oxaliplatin (TEGAFOX). One hundred and forty-three patients with measurable, non-resectable metastatic colorectal cancer were randomised in a multicentre study to receive TEGAFIRI (UFT 250 mg m(-2) day days 1-14, LV 90 mg day days 1-14, irinotecan 240 mg m(-2) day 1; q21) or TEGAFOX (UFT 250 mg m(-2) day days 1-14, LV 90 mg day days 1-14, oxaliplatin 120 mg m(-2) day 1; q21). Among 143 randomised patients, 141 were analysed (68 received TEGAFIRI and 73 TEGAFOX). The main characteristics of the two arms were well balanced. The most common grade 3-4 treatment-related adverse events were neutropenia (13% of cases with TEGAFIRI; 1% in the TEGAFOX group). Diarrhoea was prevalent in the TEGAFIRI arm (16%) vs TEGAFOX (4%). Six complete remission (CR) and 19 partial remission (PR) were recorded in the TEGAFIRI arm (odds ratio (OR): 41.7; 95% confidence limit (CL), 29.1-55.1%), and six CR and 22 PR were recorded in the TEGAFOX group, (OR: 38.9; 95% CL, 27.6-51.1). At a median time follow-up of 17 months (intequartile (IQ) range 12-23), a median survival probability of 20 and 19 months was obtained in the TEGAFIRI and TEGAFOX groups, respectively. Median time to progression was 8 months for both groups. TEGAFIRI and TEGAFOX are both effective and tolerable first-line therapies in MCRC patients. The employment of UFT/LV given in doublet combination is interesting and the presented data appear comparable to equivalent infusion regimens described in the literature. The safety profile of the two combinations also allows an evaluation with other biological agents such as monoclonal antibodies.