Evaluation of a Mechanism-Based Pharmacokinetic-Pharmacodynamic Model for D2 Receptor Occupancy of Olanzapine in Rats
Martin Johnson, Magdalena Kozielska, Venkatesh Pilla Reddy, An Vermeulen, Rik de Greef, Cheryl Li, Sarah Grimwood, Jing Liu, Geny M. M. Groothuis, Meindert Danhof and Johannes H. Proost
 Department of Pharmacokinetics, Toxicology and Targeting, University Centre for Pharmacy, University of Groningen, The Netherlands;  Advanced PK/PD Modeling and Simulation, Johnson & Johnson Pharmaceutical Research and Development, Beerse, Belgium ;  Pharmacokinetics, Pharmacodynamics & Pharmacometrics (P3), Merck Research Labs, Oss, The Netherlands ;  Pfizer Global Research and Development, New London, CT 06320, USA  Leiden/Amsterdam Center for Drug Research, Dept. of Pharmacology, Leiden, The Netherlands
Objective: Two structurally different mechanism-based pharmacokinetic-pharmacodynamic (PK-PD) models were developed to predict the time course of dopamine receptor occupancy (D2RO) in rat striatum following the administration of olanzapine, an atypical antipsychotic drug. These models included the characterization of association and dissociation rate constants (Kon and Koff) as the determinant of time delay between the brain concentration and D2 receptor occupancy. Model A was developed with an assumption that receptor binding affects the free concentration of olanzapine in the striatum. Model B was developed with an alternative assumption where receptor binding does not affect the free drug concentration in the striatum. The objective of this study is to evaluate these assumptions in a systematic way by testing both models and assessing parameter(s) sensitivity using a simulation based approach.
Methods: A population approach was utilized to quantify both the pharmacokinetics and pharmacodynamics of olanzapine using the drug exposure (plasma and brain concentration) and D2 RO profile obtained at various doses (0.1-30 mg/kg) administered by different routes. A two-compartment pharmacokinetic model was used to explain the plasma pharmacokinetic (PK) profile. Two structurally different binding models were developed to characterize the D2 receptor binding at striatum and were fitted sequentially to the PK data. The effect of binding was evaluated using a dataset simulated from Model A including low dose levels (0.01-0.03 mg/kg). PK-PD parameters were estimated for this simulated dataset using Model A and Model B. The parameters were estimated using NONMEM VI, level 2.0. The binding parameters (Kon, Koff) and Bmax were subjected to a parameter sensitivity analysis, where these parameters were perturbated with different range of values and doses. The effect of these perturbations on the D2RO profile was examined.
Results: The PK-PD time profiles were well described by both models. PK-PD parameter estimates obtained from models A and B did not differ significantly for both the real dataset and simulated dataset including low doses. Bmax did not influence D2RO when perturbated to different values, whereas Kon, Koff did influence to some extent.
Conclusion: The relatively simple model (Model B) recapitulated the essential features of the Model A to predict D2RO and reduced the need for Bmax which is difficult to identify from the available data/information.