A new class of antiemetic agents, the 5-hydroxytryptamine (5-HT3) antagonists, have been shown to possess potent antiemetic properties in the ferret model. We conducted a phase I/II trial of the 5-HT3 antagonist BRL43694 (granisetron) in 24 chemotherapy-naïve patients who were receiving any combination of doxorubicin and/or cisplatin. The first 12 patients received 40 micrograms/kg and the second 12 received 80 micrograms/kg of granisetron intravenously before beginning chemotherapy. Nausea was assessed by a patient-completed visual analogue scale and episodes of retching recorded by the patient and an independent observer. Fifty-two percent of the 22 evaluable patients had no retching or vomiting and 32% had no nausea during the first 24 hours after chemotherapy. Pharmacokinetic measurements were performed. The disposition of granisetron was best described using a two-compartment model. The area under the plasma concentration curve (AUC) was 277 +/- 226 ng.h/mL and 359 +/- 282 ng.h/mL at 40 and 80 micrograms/kg, respectively. The total body clearance was 0.319 +/- 0.315 L/kg/hr and 0.483 +/- 0.504 L/kg/hr at the 40 and 80 micrograms/kg doses. Wide interpatient variation in model independent parameters was observed. There was no suggestion of dose-dependent efficacy at the two dose levels studied. We conclude that granisetron shows promise as a well-tolerated and effective antiemetic. Randomized trials comparing this drug with standard regimens are currently underway.