5-Fluorouracil enhances azidothymidine cytotoxicity: in vitro, in vivo, and biochemical studies.

Preliminary in vitro studies suggest that the combination of 5-fluorouracil (FUra) and 3'-azido-3'-deoxythymidine (AZT) is more cytotoxic than either agent alone. Therefore, a biochemical and therapeutic evaluation of this combination was initiated. Quantitation of the cytotoxicity of FUra plus AZT against the growth of HCT-8 cells in vitro revealed that 1 microM FUra (approximately 10% inhibitory concentration) increased the cytotoxicity of AZT and decreased its 50% inhibitory concentration by 60%. Similarly, incubating HCT-8 cells in 5 microM AZT (approximately 10% inhibitory concentration) decreased the 50% inhibitory concentration of FUra by over 50%. Biochemical analysis indicated that AZT did not affect FUra-induced inhibition of thymidylate synthase or [3H]-FUra incorporation into nucleic acids. In contrast, incubation in 5 microM FUra increased the incorporation of [3H]-AZT (5 microM) into the nucleic acid fraction of these cells by 52% (P less than 0.05). Therapeutic evaluation of this combination in athymic (nude) mice bearing HCT-8 xenographs revealed that, while weekly FUra (85 mg/kg) or AZT (600 mg/kg) exerts minimal antineoplastic activity (after 4 courses, treatment/control = 0.81 and 0.70, respectively), their combination, at the same doses, inhibited tumor growth by nearly 70% (P less than 0.01 versus FUra alone). FUra-related host toxicity was not increased by the addition of AZT. Higher doses of FUra alone were not more effective than FUra plus AZT. In vivo, AZT did not affect the incorporation of [3H]-FUra into the nucleic acid fraction of various murine tissues, including HCT-8 xenografts. FUra, however, increased [3H]-AZT incorporation into nucleic acids in a tissue-specific manner. In the presence of FUra, the incorporation of [3H]-AZT in spleen, liver, and gut increased 40, 21, and 4%, respectively, while in HCT-8 xenografts [3H]-AZT incorporation increased more than 2-fold. Analysis of the activities of selected enzymes involved in pyrimidine metabolism suggests that this tissue-specific effect may be related to the pyrimidine salvage capacity of these tissues. These findings are described in light of their potential impact on human colon cancer chemotherapy.