Pharmacokinetic interaction between ondansetron and cyclophosphamide during high-dose chemotherapy for breast cancer

Purpose: Both ondansetron and cyclophosphamide are thought to be metabolized by hepatic microsomal processes. The purpose of this study was to evaluate the potential pharmacokinetic interactions between ondansetron and high-dose alkylating agent chemotherapy. Methods: A total of 54 breast cancer patients receiving high-dose cyclophosphamide, cisplatin and carmustine were treated prospectively in four sequential cohorts. Cohorts I and II received continuous infusions of both ondansetron and prochlorperazine, and cohorts III and IV received a continuous infusion of ondansetron alone at the same doses. All patients received lorazepam every 4 h. A group of 75 matched historical controls had received a continuous infusion of prochlorperazine with lorazepam. Pharmacokinetic monitoring of each drug used in the high-dose chemotherapy regimen was conducted. Results: Median AUCs of cyclophosphamide in patients receiving ondansetron (73.6 mg/ml · min) were lower than those of the control patients (88.3 mg/ml · min, n = 75, P = 0.0004), but the median cisplatin AUC was approximately 10% higher and no difference in the disposition of carmustine was demonstrated. Patients treated with ondansetron displayed a higher frequency of headaches than the controls. The frequency of achieving complete emetic control was greater in the ondansetron + prochlorperazine groups compared to the ondansetron alone groups and was greater in both these groups than in the prochlorperazine alone group on the first day of therapy only. Conclusion: Ondansetron altered the systemic exposure to cyclophosphamide when these agents were administered concomitantly. Ondansetron did not substantially improve overall emetic control when used alone but may improve control in combination with prochlorperazine. Future randomized studies are needed to delineate the effect of ondansetron on the disposition of the active cyclophosphamide metabolites so that clinical implications can be addressed. Received: 28 October 1997 / Accepted: 9 March 1998