Results from five completed and two ongoing Phase III trials comparing 5-fluorouracil (5-FU) with 5-FU and leucovorin (LV) for the treatment of advanced, previously untreated colorectal carcinoma are reviewed. In five of these studies, 5-FU/LV was shown to provide a significantly higher response rate than 5-FU alone. Partial response rates resulting from the combination ranged from 16% to 45%, while those obtained with 5-FU alone were 5% to 18%. Complete responses with either treatment occurred in 0% to 8% of the patients treated. In two of the studies, 5-FU treatment prolonged median survival by 3 and 6 months. The planned dose intensities (DI) for single-agent 5-FU ranged from 463 to 760 mg/m2/week, and the actual delivered DI, established in two of the trials, were 531 and 533 mg/m2/week. The planned 5-FU DI for the 5-FU/LV combination ranged from 463 to 600 mg/m2/week compared with the delivered DI of 461 and 463 mg/m2/week. When used with LV, 5-FU cannot be administered at the single-agent maximum tolerated dose because of unacceptable toxicity. Nevertheless, lower doses of 5-FU, when combined with LV, had higher response rates than 5-FU alone. Significant toxicity occurred with 5-FU and with the combination. The addition of LV to 5-FU changed the type of toxicity that resulted; that toxicity varied with the schedule followed. The real, but modest improvement that resulted from the combination of 5-FU with the metabolic modulator LV offers the possibility that the administration of 5-FU/LV in the adjuvant setting will prolong survival and increase the percentage of patients cured of their disease. Furthermore, other compounds that modulate the effects of 5-FU may be capable of providing additional therapeutic benefits.