Pre-Clinical PK/PD Modeling for a Dipeptidyl Peptidase-IV Inhibitor
Dittberner, S.(1), V. Duval (2), A. Staab (2), H. Fuchs (2), M. Tadayyon (2), U. Graefe-Mody (2), H.G. Schaefer (2), U. Jaehde (1)
(1) Dept. Clinical Pharmacy, Institute of Pharmacy, Rheinische Friedrich-Wilhelms-Universitaet Bonn, Bonn, Germany; (2) Boehringer Ingelheim Pharma GmbH & Co. KG, Biberach a.d.R., Germany
Background: Inhibition of Dipeptidyl Peptidase-IV (DPP-IV) is a novel strategy to treat type 2 diabetic patients. Compound X is a potent and selective non-covalent DPP-IV inhibitor, the pharmacokinetics of which showed a less than proportional increase with dose in the area under the curve of plasma concentration-time profiles (AUC). Moreover, a high in vitro plasma protein binding of the DPP-IV inhibitor was observed when concentrations were low, whereas only a relatively small fraction was bound at higher concentrations of compound X. This behavior suggests that the compound has two different kinds of binding sites. One which is saturable and has a low capacity, but a high affinity to the compound, whereas the second one has a low affinity to the compound but a high capacity.
Objectives: The objectives were firstly, to investigate using a PK model whether saturable protein binding could explain the non-linearity of the pharmacokinetics of compound X. Secondly, a pharmacokinetic/pharmacodynamic (PK/PD) model was developed to characterize the relationship between PK and ex vivo DPP-IV activity.
Methods: PK and PD profiles based on plasma samples from 6 cynomolgus monkeys (3 males and 3 females) were obtained. Each of the monkeys received four single doses (0.1, 0.3, 1.0 mg/kg oral, 1.0 mg/kg i.v.) and plasma samples were taken at several time points (oral, 9 time points: pre-dose and sequentially up to 72h; i.v, 10 time points: pre-dose and sequentially up to 48 h) after each administration. Each dose was intermitted by a wash-out period.
The population pharmacokinetic/pharmacodynamic analysis was performed based on these profiles using NONMEM.
Results: The PK data were best described by a two compartmental model (central, peripheral) and a saturable binding of compound X to protein. The estimated maximal binding capacity (Bmax) and the estimated equilibrium constant of dissociation (Kd) were in the range expected for the binding of the compound to DPP-IV, suggesting that the high affinity/low capacity binding observed might be the binding to DPP-IV. Thus, the terminal phase seems to be not only dependent on the clearance of the compound but additionally on the dissociation from DPP-IV and therefore cannot be regarded as a conventional elimination phase.
Furthermore, receptor occupancy theory could be successfully used to relate PK to ex vivo DPP-IV activity, supporting the binding hypothesis above.
Conclusions: A population PK/PD model has been successfully developed describing the non-linear pharmacokinetics of a DPP-IV inhibitor and its effect on enzyme activity.
This model will serve as a basis to support clinical PK/PD modeling.