Desipramine Population Pharmacokinetic Model and Designing CYP2D6 Drug-Drug Interaction Studies
Ivelina Gueorguieva, Kimberley Jackson, Steven A. Wrighton, Vikram P Sinha and Jenny Y. Chien
Lilly Research Laboratories, Department of Drug Disposition, Lilly Research Laboratories, Sunninghill Road, Windlesham, Surrey, UK (I.G., K.J.) and Indianapolis, Indiana, USA (J.Y.C., S.A.W.)
Objectives: The objectives of this analysis were to develop a population pharmacokinetic (PK) model to describe the pharmacokinetics of desipramine in healthy subjects, after oral administration of a 50mg dose and to examine the influence of demographic characteristics and other covariates on the PK parameters of desipramine. Secondary objectives were to propose a mechanistic population PK model for desipramine, which allowed simulation of CYP2D6-mediated inhibition when using desipramine as a probe substrate, and to identify optimal sample size and pharmacokinetic sampling schedules for extensive (EM), intermediate (IM) and ultrarapid (UM) metabolisers of substrates of CYP2D6.
Methods: Desipramine concentration-time data were available from seven studies comprising 108 healthy subjects who received a 50 mg desipramine oral dose. Different standard and semi-mechanistic pharmacokinetic models were fitted to the dataset using nonlinear mixed-effects modeling software (NONMEM, version V). Several potential covariates were tested to quantify their effect on desipramine population pharmacokinetic parameters. D-optimal sampling times (PopDes) and appropriate study designs to account for the differences in EM, IM and UM subgroups for CYP2D6 were suggested.
Results: A two-compartment model with first order absorption best described desipramine concentration-time data following oral administration. Body mass index was found to influence the apparent central distribution volume. Further, using the proposed semi-mechanistic hepatic intrinsic clearance model with Bayesian inference, mean population desipramine hepatic intrinsic clearance was estimated to be 289 L/h with between-subject variability of 86%. D-optimal sampling times for EMs were calculated to be 0.25, 20, 58, 90 and 200 hours and for the IMs subgroup similar optimal sampling times were found except now a sample at 105h instead of 90h was needed.
Conclusions: The population PK model developed is suitable to describe the behaviour of desipramine for the dose routinely used in drug-drug interaction (DDI) studies. Desipramine concentration levels can be simulated to guide clinical trial design for healthy subjects in DDI studies, which aim to assess CYP2D6 interaction potential of novel compounds incorporating sample size and optimal sampling schedules for the three CYP2D6 metabolizer subgroups.
References:
[1] Gueorguieva et al., 2007, Computer Meth Progr Biomed 86:51-61.