5-Fluorouracil (5-FU) remains the standard chemotherapy for gastrointestinal cancer, particularly colorectal cancer, with response rates of 8% to 33% and median survivals of 24 to 44 weeks. The schedule of delivery for 5-FU has been addressed in a number of clinical trials over the past two decades, but the optimal schedule based upon more recent studies is not clear. The prospective comparative trial by Ansfield et al., investigated bolus delivery in four schedules and the daily X 5 loading regimen was superior to the less intensive regimens. However, two randomized trials of bolus 5-day vs. continuous 5-day infusions have indicated that the infusion schedule is superior: in colorectal cancer, the study by Seifert et al. demonstrated a response rate of 42% for infusion vs. 21% for bolus; and in head and neck cancers 5-FU administered in conjunction with cisplatin achieved a response rate of 76% for infusion vs. 20% for bolus delivery. Further support for the superiority of the infusion schedule is provided by the comparative trials of hepatic arterial infusion (HAI) and systemic venous infusion, in which response rates are equivalent regardless of the route of delivery. In order to definitively establish an improved therapeutic effect for the infusion schedule, additional prospective trials comparing standard bolus schedules to infusion schedules are necessary, and should address other issues, such as the optimal duration of infusion and relative cost-effectiveness.