Myeloablation by intravenous busulfan and hematopoietic reconstitution with autologous marrow in a canine model.

We have previously described pharmacokinetic studies with a dimethylsulfoxide-based intravenous busulfan preparation in a canine model and in preliminary clinical trials. Using the same intravenous busulfan preparation, we carried out a dose escalation study to determine a marrow-ablative dose and to test the ability of autologous marrow to reconstitute hematopoiesis in dogs so treated. Busulfan was given intravenously at doses of 3.75 to 40 mg/kg. Marrow ablation was achieved at 20 mg/kg given either as a single dose or in four daily increments of 5 mg/kg each. There was a relative sparing of lymphocytes. A busulfan dose of 40 mg/kg resulted in severe central nervous system toxicity. Otherwise, nonhematopoietic toxicity was minimal and restricted to mild hepatic abnormalities. Four dogs were given busulfan at 20 mg/kg followed 30 hours later by infusion of autologous marrow, and all showed prompt and complete hematopoietic reconstitution. The area under the curve (AUC) determined by busulfan concentration in plasma over time was dose dependent, ranging from 12 to 100 microg x h/mL for busulfan doses of 3.75-20 mg/kg. There was a suggestion that the plasma half-life increased at the highest busulfan doses used. Intravenous administration of busulfan circumvented differences in bioavailability; nevertheless, considerable variations in the pharmacokinetic parameters were observed between individual animals. Thus, intravenous busulfan can be given safely and is effective in ablating hematopoiesis. However, factors other than absorption influence the AUC, and individualization of dosing may be required even with intravenous administration of the drug.