The aim of this study was to investigate whether N-(phosphonacetyl)-L-aspartic acid (PALA) can enhance the activity of low-dose methotrexate (LD-MTX) modulated infusional 5-fluorouracil (5-FU) in patients with advanced colorectal cancer. 198 patients were randomised either to (i) 5-FU 60 mg/kg as a continuous infusion over 48 h, to be given weekly four times and thereafter every 2 weeks, with methotrexate 40 mg/m(2) administered just before 5-FU (control regimen) or to (ii) PALA 250 mg/m(2) as a 15 min infusion administered 24 h before 5-FU and methotrexate which was given as described in the control regimen. The response rate was 13% in the patients randomised to the control arm and 7% in the patients randomised to the experimental arm. If stabilisation of the disease was also considered as a positive response, these figures become 54 and 46%, respectively. All these differences did not reach statistical significance. The median durations of progression-free survival (PFS) in the two treatment groups were not significantly different. The median duration of survival was 13.1 months in the control arm and 11.9 months in the experimental arm (P=0.31). No benefit was obtained by adding PALA to LD-MTX+infusional FU. Taking into account data from US trials, the modulating effect of PALA, although 'promising' in phase II, could not be substantiated in randomised studies.