Halofantrine (HF) is used in the treatment of uncomplicated multidrug-resistant Plasmodium falciparum malaria. Severe cardiotoxicity has been reported to be correlated with high plasma concentrations of HF but not with that of its metabolite N-debutylhalofantrine. The aim of this study was to investigate the effects of other antimalarial drugs and of ketoconazole, a typical cytochrome P-450 3A4 inhibitor, on HF metabolism by human liver microsomes. Antimalarial drug inhibitory effects were ranked as follows: primaquine > proguanil > mefloquine > quinine > quinidine > artemether > amodiaquine. Artemisine, doxycycline, sulfadoxine, and pyrimethamine showed little or no inhibition of HF metabolism. Mefloquine, quinine, quinidine, and ketoconazole used at maximal plasma concentrations inhibited N-debutylhalofantrine formation noncompetitively with Ki values of 70 microM, 49 microM, 62 microM, and 0.05 microM resulting in 7%, 49%, 26%, and 99% inhibition, respectively, in HF metabolism. In conclusion, we showed that quinine and quinidine coadministered with HF might inhibit its metabolism resulting in the potentiation of HF-induced cardiotoxicity in patients. This requires a close monitoring of ECG. For the same reasons, the concomitant administration of HF and ketoconazole must be avoided. By contrast, none of the other antimalarials studied inhibited HF metabolism and, by extrapolation, cytochrome P-450 3A4 activity.