The pharmacokinetics of doxorubicin given according to three different schedules with a similar dose-time intensity have been studied and compared in 16 women with metastatic breast cancer. Six patients were treated with doxorubicin 75 mg/m2 by i.v. bolus repeated every 3 weeks; 5 patients received doxorubicin by 4-day continuous infusion every 3 weeks (4 at 75 mg/m2 and 1 at 60 mg/m2); 5 patients received 25 mg/m2 by i.v. bolus given weekly. Timed blood samples were collected and plasma levels of doxorubicin and its metabolite doxorubicinol were measured by high-performance liquid chromatography with fluorescence detection. Peak plasma concentrations were measured, and areas under the concentration-time curves calculated. Peak plasma levels of doxorubicin were significantly lower with the 4-day infusion than with either of the bolus injections. The 4-day infusion, however, gave significantly greater total exposure to doxorubicin and doxorubicinol, as indicated by area under the concentration-time curve, than weekly or 3-weekly bolus treatment. A single bolus injection of doxorubicin 25 mg/m2 yielded a total exposure to doxorubicin approximately half that achieved with a 75 mg/m2 bolus injection. Over a 3-week period, therefore, total exposure to doxorubicin would be greater with the weekly low-dose schedule than with the 3-weekly administration. We conclude that drug scheduling has significant effects on doxorubicin pharmacokinetics.