Pharmacokinetic studies of high-dose treosulfan were carried out in seven paediatric patients (age range: 2-15 years) undergoing treosulfan-based conditioning regimen prior to allogeneic haematopoietic SCT. Treosulfan was administered intravenously in a daily dose of 10, 12 or 14 g/m(2) within 2 h. Five out of seven patients received 12 g/m(2). The plasma concentrations of treosulfan and its quantity eliminated with urine were determined using a validated HPLC method with refractometric detection. Pharmacokinetic parameters were evaluated following first dose using a two-compartment disposition model. These studies demonstrated a dose-dependent increase of area under the concentration (AUC) and maximum concentrationplasma (C(max)), but there was variability of these parameters. Rapid clearance of tresoulfan was observed, especially in 10 and 12 g/m(2) doses. Terminal half-life (t(0.5)) of treosulfan was in the range of 1.71-2.15 h, but the mean percent of parent drug eliminated with urine was 30%, range 16.3-45.4% of the total dose eliminated during the first 12 h after administration. The results of this study confirmed the linear pharmacokinetics of treosulfan, as used in children. However, variability of pharmacokinetic results observed in children studied demonstrates the need for pharmacokinetic evaluation in each paediatric patient undergoing the treosulfan-based preparative regimen, including those using different doses. This approach could enable further reduction of the risk of early and late organ toxicity related to high-dose treosulfan in paediatric patients.