Viktoria Stachanow (1,2), Johanna Melin (1,2,3), Robin Michelet (1), Richard Ross (4), Martin J Whitaker (4), Oliver Blankenstein (5), Uta Neumann (5), Wilhelm Huisinga (6), Charlotte Kloft (1)
(1) Freie Universität Berlin, Germany, (2) Graduate Research Training Program PharMetrX, Germany, (3) Clinical Pharmacology and Quantitative Pharmacology, R&D AstraZeneca, Gothenburg, Sweden, (4) The University of Sheffield, UK, (5) Charité-Universitätsmedizin, Berlin, Germany, (6) Institute of Mathematics, Universität Potsdam, Germany
Objectives: Congenital adrenal hyperplasia (CAH) is a rare form of adrenal insufficiency leading to low or no biosynthesis of cortisol and requiring a lifelong cortisol replacement therapy. Hydrocortisone (HC, synthetic cortisol) is the recommended glucocorticoid for paediatric patients [1]. Dried blood spot (DBS) sampling allows collection of small volumes (10–20 µL) of full blood multiple times [2]. DBS sampling is therefore commonly used to quantify and monitor cortisol and disease biomarkers such as 17α-hydroxyprogesterone (17-OHP) during treatment of paediatric CAH patients. A PK/PD relation between DBS concentrations of cortisol and treatment biomarkers has not been characterised yet, leading to potentially inadequate patient outcomes. HC has complex pharmacokinetics (PK) with saturable absorption, saturable binding to corticosteroid-binding globulin (CBG), and non-saturable binding to albumin and red blood cells (RBC) [3,4]. The aim of this analysis was to inform a previously developed paediatric plasma HC PK model [5] with adult cortisol plasma data and paediatric DBS cortisol data to enable a more comprehensive characterisation of the HC PK and to quantify binding kinetics of cortisol to RBC. This PK model will then serve as a basis for subsequent PK/PD analysis.
Methods: A semi-mechanistic HC PK model based on adult plasma data from a phase I study of a novel paediatric HC formulation [6] was previously reduced to a paediatric model using sparse paediatric plasma data from a phase III study in 24 patients with adrenal insufficiency [7]. A more comprehensive dataset was created by pooling the sparse paediatric plasma cortisol dataset with DBS cortisol concentrations that were simultaneously measured during the phase III trial as well as with the phase I adult cortisol concentrations. Nonlinear mixed-effects modelling was applied using NONMEM 7.4.3 to extend the model by characterising binding of cortisol to RBC in addition to the already included binding to CBG and albumin. Covariates such as haematocrit (Hct, partly imputed) or age groups were tested on the parameters describing the RBC binding kinetics.
Results: A two-compartment model with saturable absorption as well as an additional RBC compartment with cortisol bound to erythrocytes, an estimated baseline for adults and children baselines for plasma and DBS concentrations based on pre-dose observations was chosen as final model based on drop of the objective function value (OFV) and goodness-of-fit (GOF) plots. Allometric scaling with fixed exponents of 0.75 and 1 was applied to account for the effect of body weight on the clearance and distribution parameters, respectively. The binding model included nonlinear binding of cortisol to CBG, (equilibrium dissociation constant Kd= 9.71 nmol/L) as well as linear binding to albumin and to erythrocytes, described by linear non-specific binding parameters (NSalb=4.15 and NSery=4.46, respectively). Compared to the inclusion of partly imputed Hct, the inclusion of age group as a categorical covariate on the estimated volume of the RBC compartment (VRBC) resulted in a higher drop of the OFV. The resulting VRBC for the infants and toddlers (4 months-5 years) was 15.1 L compared to 1.41 L in the neonates (2-4 weeks). The age group as a covariate also explained more than 75% of the interindividual variability (before: 142.6%, after: 32.4%). Standard model evaluation and GOF plots showed that the plasma as well as the DBS concentrations were adequately described by the model.
Conclusions: An extended nonlinear mixed-effects model capturing the complex HC PK processes with additional binding kinetics to erythrocytes was successfully developed from pooled adult plasma and paediatric plasma and DBS data. As Hct values were only available from 12 of 24 paediatric patients and it is known from literature that the Hct is higher in the first 4 weeks of life and then relatively constant in infancy and early childhood [8], age group was chosen as covariate on RBC binding parameters and explained a substantial part of the variability of the estimated volume of the RBC compartment. The next step is to investigate the PK/PD relationship between cortisol and the clinically-relevant biomarker 17-OHP sampled with DBS, to identify target 17-OHP DBS concentrations that can provide clinical guidance for monitoring paediatric CAH patients.
References:
[1] Merke, D. P. et al.: Lancet. 2005, 365: 2125-2136.
[2] Wilhelm, A.J. et al: Clin Pharmacokinet. 2014, 53: 961-973.
[3] Lentjes, E. G. W. M. et al.: J Clin Endocrinol Metab. 1999, 84: 682-687.
[4] Toothaker, R. D. et al.: J Pharmacokinet Biopharm. 1982, 10: 147-156.
[5] Melin, J.: Dissertation. Freie Universität Berlin. 2017.
[6] Melin, J. et al.: Clin Pharmacokinet. 2018, 57: 515-527.
[7] Neumann, U. et al.: Clin Endocrinol. 2018, 88: 21-29.
[8] Esan, A.J.: Hematol Transfus Int J. 2016, 3: 178-190.
Reference: PAGE () Abstr 9486 [www.page-meeting.org/?abstract=9486]
Poster: Drug/Disease Modelling - Paediatrics