Rick Admiraal, MD(1,2,3,4), Charlotte van Kesteren, PharmD PhD(1,2,4), Cornelia M Jol van der Zijde, BSc(3), Maarten JD van Tol, PhD(3), Imke H. Bartelink(1), Robbert GM Bredius, MD PhD(3), Jaap Jan Boelens, MD PhD(1,4), Catherijne AJ Knibbe, PharmD PhD(2)
(1) Department of Pediatric Immunology, University Medical Centre Utrecht, the Netherlands (2) Division of Pharmacology, Leiden Academic Centre for Drug Research, Leiden University, Leiden, the Netherlands (3) Department of Pediatrics, Leiden University Medical Centre, Leiden, the Netherlands (4) U-DANCE, Tumor immunology, Laboratory Translational Immunology, University Medical Centre Utrecht, the Netherlands
Objectives: To prevent graft versus host disease (GvHD) and rejection in hematopoietic cell transplantation (HCT), children receive anti-thymocyte globulin (ATG), a polyclonal antibody depleting T-cells, as part of the conditioning regimen. The therapeutic window is critical as over-exposure may result in delayed reconstitution of donor T-cells and increased risk of viral infections. Our objective is to describe the population pharmacokinetics (PK) of Thymoglobulin as a first step towards an evidence based dosing regimen of Thymoglobulin for HCT in children.
Methods: PK data were collected for all pediatric HCT’s performed between 2004-2012 in two study centers in the Netherlands. Serum active Thymoglobulin concentrations were quantified by flow cytometry investigating the binding to a T-cell line. Since reference concentration were measured as fluorescent intensity per mg of ATG, active ATG is measured in arbitrary units (AU). Population modeling and covariate analysis was performed on active Thymoglobulin concentrations using NONMEM 7.2. The model was validated using bootstrap and NPDE.
Results: A total of 280 HCT’s in 267 patients were analyzed. A two-compartment model with saturable distribution towards the peripheral compartment (described with maximum rate Tmax and Michaelis Menten constant Tm, shown as population mean (RSE), are 155 AU/day (13.7%) and 7.1 AU/L (16.9%) respectively), as well as a parallel linear clearance (Cl; 2.1 L/day (5.7%)) and saturable clearance (described with Vmax and Km, 1.7 AU/day (18.8%) and 1.1 AU/L (21.4%) respectively), yielded a good description of the data in all age groups. The central volume of distribution (V1) was 7.9 L (5.4%), with a K21 of 1.3 day-1 (18.6%). The relationship between bodyweight and both Cl and V1 was best described by a power function with an exponent of 0.78 (10.8%) and 1.15 (6.9%), respectively. Cl was influenced by baseline lymphocyte count, with an increase of 1*109 lymphocytes leading to a 30% (18.9%) increase in Cl. Results of the validation steps were satisfactory.
Conclusion: In the validated population PK model for active Thymoglobulin in children, Cl and V1 proved dependent on body weight in a nonlinear manner, while baseline lymphocytes were linearly influencing Cl. Simulations show the current dosing regimen with a cumulative dose of 10 mg/kg over 4 days to be suboptimal, with higher bodyweight children having higher exposures when compared to low bodyweight children.
Reference: PAGE 23 (2014) Abstr 3271 [www.page-meeting.org/?abstract=3271]
Poster: Drug/Disease modeling - Paediatrics