No single agent or combination chemotherapy protocol, wit the exception of gemcitabine, has so far proven superior to standard bolus 5-fluorouracil regimes for the treatment of advanced pancreatic cancer. The present phase II trial was designed to stud whether the effectivity of 5-fluorouracil can be improved with a weekly high-dose 5-fluorouracil schedule. 26 patients with cytologically or histologically verified, metastasized (n = 21) or locally advanced (n = 5) previously untreated adenocarcinoma of the pancreas were included in this study. Treatment consisted of weekly applications of 2,600 mg/m(2) 5-fluorouracil as 24-h infusion on days 1, 8, 15, 22, 29, 36. Treatment was repeated at day 50 and was continued until disease progression. Primary endpoints of the study were response rates and toxicity, secondary endpoints were survival and clinical benefit in terms of performance status, body weight and analgesia consumption. Toxicity of the regimen wa mild wit only four instances of grade-3 toxicity. Response rates were 8% (95% CI = 1.2-13.7) with two partial remissions. Improvement of at least one parameter of clinical benefit for greater than or equal to four most weeks was observed for 11.5% of the study patients. The most prominent effect was a transient stabilization of objective tumor measurements (48% [95% CI = 27.8-68.7]) and individual parameters of clinical benefit (50-75%). Median survival was 248 days (95% CI = 164-459) for all patients included in the study. The present study indicates that the high-dose 5-fluorouracil regimen shows weak activity in advanced pancreatic cancer which seems comparable to gemcitabine.