Treosulfan is an alkylating agent which is increasing applied in regimens prior to allogeneic hematopoietic stem cell transplantation (HSCT) in children. It has strong myeloablative and immunosuppressive activity and, in comparison with busulfan and total body irradiation, a relatively mild toxicity profile. The optimal dose of treosulfan in pediatric patients, remains to be established. To optimize the balance between treosulfan efficacy and toxicity, pharmacokinetic (PK) monitoring may be a valuable tool. With this purpose, we recently developed a PK and limited sampling model (LSM) which allows measurement of treosulfan exposure. We here report the first results of an ongoing prospective study on the relation between the pharmacokinetic profile of treosulfan in pediatric HSCT recipients and clinical outcome.

Materials and Methods

Children (n =26; ≤ 18 years) receiving treosulfan based conditioning prior to their first HSCT in the Leiden University Medical Center, The Netherlands were included. Patients ≥ 1 year of age received treosulfan in a dose of 42 g/m2, divided over 3 days. Two patients aged 1.5 and 7 months received 30 and 36 g/m2, respectively. Treosulfan was combined with fludarabine (30 mg/m2/day, 5 days) in all patients and in n=20 patients thiotepa (8 mg/kg) was added. Serotherapy consisted of anti-thymocyte globuline or alemtuzumab. Blood samples were collected at 1.5, 3.5, 4, 5, 7 and 9 or, when a LSM was applied, at 4-5 and 6-7 hours after the start of a three hour treosulfan infusion. The individual PK profile of each patient was determined by a population PK-model with a one-compartment model. Clearance and volume of distribution were allometrically scaled using body weight, with a fixed scaling exponent of 0.75 and 1.0, respectively.

Clinical endpoints of this study were stem cell engraftment, chimerism, GvHD, organ toxicity, relapse and treatment related mortality.


The median age, clearance, and volume of distribution were 6.5 (0.13-16.8) years, 7.8 L/h (1.2 - 44.7 L/h), and 13.8 L (3.0 - 42.9 L), respectively. Within the n=24 patients receiving 42 g/m2, the mean AUC of treosulfan was 1502 mg*h/L and the inter-patient variability was 14%. The inter-occasion variability was 5.4%, based on the day 1 and day 3 results of 12 patients. Two patients received a lower dose of 30 g/m2 and 36 g/m2, resulting in a day 1 AUC of 1578 mg*h/L and 1349 mg*h/L, respectively. A linear relationship was observed between age and treosulfan exposure (R2 = 24%) and body surface area and treosulfan exposure (R2 = 29%), resulting in a lower exposure in older patients (Figure 1.)

In this cohort overall survival is 21/26 (81%). Four patients died prior to engraftment (≤18 days after HSCT), due to infection, toxicity (n = 2) and active malignant disease. Treosulfan exposure in these four patients was similar to the remaining patients of the cohort. Primary engraftment was seen in 95% of the 22 evaluable patients. Of the 21 survivors, 76% are disease free, four patients had a recurrence of initial disease, and one patient had a primary rejection followed by rapid autologous reconstitution. In 46% of the patients mucositis grade 2 or 3 was observed, 23% of the patients had skin toxicity grade 1-2 and 31% had a moderate and transient hepatic toxicity within 28 days after treosulfan administration. Acute GvHD ≥ grade II occurred in 3/26 patients. Severe hepatic toxicity occurred in three patients that received treosulfan as reduced toxicity conditioning because of their pre-existing organ toxicity. No association between treosulfan exposure and primary engraftment, survival and disease status at last follow-up was observed. Furthermore, the higher treosulfan exposure in the younger patients was not correlated with higher toxicity levels.


Our results demonstrate that PK monitoring provides novel insight in the level and inter-patient variability of treosulfan exposure in pediatric patients. To further improve early and long term clinical outcome in pediatric HSCT recipients individualized, age and possibly disease-specific dosing regimens may in the end be required. To achieve this goal, treosulfan PK monitoring as reported here will be a pivotal instrument to obtain further insight into the correlation between treosulfan dose and clinical outcome parameters. With this aim, we are currently performing a prospective multicenter pediatric study in disease-specific cohorts.


Off Label Use: Treosulfan in conditioning prior to allo-HSCT in pediatric patients.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract