Predictive performance of two PK-PD models of D2 receptor occupancy of the antipsychotics risperidone and paliperidone in rats
Magdalena Kozielska (1), Martin Johnson (1), Venkatesh Pilla Reddy (1), An Vermeulen (2), Rik de Greef (3), Cheryl Li (4), Sarah Grimwood (4), Jing Liu (4), Geny M. M. Groothuis (1), Meindert Danhof (5), Johannes H. Proost (1)
(1) Dept. of Pharmacokinetics, Toxicology and Targeting, University Centre for Pharmacy, University of Groningen, The Netherlands; (2) Advanced PK/PD Modeling and Simulation, Johnson & Johnson Pharmaceutical Research and Development, Beerse, Belgium; (3) Pharmacokinetics, Pharmacodynamics & Pharmacometrics (P3), Merck Research Labs, Oss, The Netherlands; (4) Pfizer Global Research and Development, New London, CT 06320, USA; (5) Leiden/Amsterdam Center for Drug Research, Dept. of Pharmacology, Leiden, The Netherlands
Objectives: The level of dopamine D2 receptor occupancy is predictive of efficacy and safety in schizophrenia. Population PK-PD modelling has been used to link observed plasma and brain concentrations to receptor occupancy. The objective of this study was to compare the predictive performance of two structurally different PK-PD models for rats. In one model receptor binding was assumed to influence brain distribution of the drug and in the second model receptor occupancy was derived from brain concentration, but did not affect it.
Methods: Based on the plasma, brain and D2 receptor occupancy data for risperidone in rats, mechanism-based PK-PD models were developed previously. The model in which binding to D2 and 5-HT2A receptors was taken into account and this binding influenced brain kinetics of the drug resulted in the best fit to the data. However, if only data for higher doses were used, also the model where receptor binding did not affect brain kinetics fitted well to the data. Here, we used simulations to compare how well the two models, can predict brain concentration and receptor occupancy for different doses of risperidone.
Results: Predicted brain concentration differed between the two models, especially for lower doses. Only the model in which receptor binding influenced brain kinetics correctly predicted the brain to plasma ratio observed in the data, which was higher at lower concentrations and decreased to a relatively constant level for higher plasma concentration (when receptor binding is maximal). However, both models predicted receptor occupancy similarly well for all the doses.
Conclusions: A mechanistic model in which brain kinetics of the drug are affected by its binding to receptors is necessary to accurately predict brain to plasma ratios. However, simpler models might be sufficient to accurately predict receptor occupancy. Inclusion of binding to receptors in the drug brain kinetics may be especially important for drugs with active efflux where concentrations in brain are lower and therefore drug bound to receptors may constitute relatively large fraction of total drug in the brain.