Phase I and clinical pharmacological study of mercaptopurine administered as a prolonged intravenous infusion.

The bioavailability of oral mercaptopurine (MP) is poor, and plasma levels following p.o. dosing are highly variable. In an attempt to circumvent these problems, we conducted a Phase I trial and clinical pharmacological study of MP administered as a prolonged i.v. infusion. An infusion rate of 50 mg/sq m/h, which was designed to achieve therapeutic drug levels in plasma, was used in all patients. The infusion duration was escalated in 12-h increments. Thirty-eight patients were evaluated. The dose-limiting toxicity was mucositis. Other reversible toxicities were myelosuppression and hepatotoxicity. An infusion duration of 48 h was found to be safe, unassociated with dose-limiting toxicity. Objective responses were seen in five patients. The mean plasma steady-state MP concentration achieved was 6.9 microM with little interpatient variability seen. Allopurinol coadministration had no effect on the plasma pharmacokinetics of i.v. MP. However, allopurinol did alter the urinary metabolite pattern, decreasing thiouric acid and increasing MP and thioxanthine levels. The steady-state cerebrospinal fluid:plasma ratio for MP was 0.27, suggesting that this approach may be of value in the treatment of central nervous system cancer. MP can be safely administered as a 48-h i.v. infusion at a dose rate which reliably achieves MP levels associated with optimal antileukemic activity in vitro.