N. Goyal (1), M. Beerahee (2)
(1) Clinical Pharmacology, Modeling and Simulation, King of Prussia, PA, USA; (2) Clinical Pharmacology, Modeling and Simulation, Stevenage, UK
Objectives: To develop a K-PD model to describe the kinetics of effect of a novel FLAP inhibitor (GSK2190915) on the functional biomarker-cysteinyl leukotriene LTE4, as measured in urine samples. Compare performance of this K-PD model to a corresponding PK-PD model using data from single and repeat dose escalating study in healthy volunteers [1].
Background: K-PD modeling presents an appealing methodology to support clinical drug development especially where it may not be feasible or possible to obtain time course of systemic drug concentrations to correlate with biomarker or clinical endpoints [2-5].
Methods: Data: Serial drug samples (PK) and urine biomarker samples of LTE4 (PD) were available from healthy volunteers receiving 50,150,300,600 or 1000 mg single dose (n=40) and 10,50,150 or 450 mg once daily oral repeat dose (n=32) of GSK2190915 for 11 days.
Model: The PK-PD model was a two compartment PK model with first order absorption and an indirect PD response model with inhibition of rate of synthesis of LTE4. The K-PD model completely ignored the PK data and utilised only PD information. The modeling was performed with NONMEM (NMVI) [6].
Results: The K-PD model described the data with similar efficiency as the PK-PD model. The parameter estimates are presented in the table below. The EDK50 represents the drug’s in vivo apparent potency at steady state. This EDK50 estimate when adjusted for the systemic clearance is comparable to the IC50 estimate calculated from the PK-PD model. The K-PD model run time was 4 times faster than the PK-PD model.
| Parameter | PK-PD | K-PD |
| CL (L/hr) | 7.7 (52.1) |
|
| V2 (L) | 82.2 (65.7) |
|
| IC50 (ng/ml) | 25.6 (152.3) |
|
| Kin (pg/mg Cr/hr) | 7.83 (63.3) | 11.7 |
| Kout (1/hr) | 0.213 (28.6) | 0.28 (70.8) |
| KDE (1/hr) |
| 0.03 (107.2) |
| EDK50 (mg/hr) |
| 0.16 (101.9) |
| NMVI Run Time (mins) | 58 | 14 |
Conclusions: This work demonstrates the value of K-PD modeling in providing a good description of kinetics of drug effect even in absence of systemic drug concentrations. The K-PD model for GSK2190915 provides a valuable tool to support its clinical drug development; e.g. paediatric studies where plasma samples may not be available. Certain limitations exist with generalizing the K-PD approach across untested dosing routes or regimens. Diligent use of K-PD methodology may obviate requiring systemic concentrations in clinical studies where appropriate..
References:
[1] Bain G et al, Pharmacodynamics, pharmacokinetics and safety of GSK2190915, a novel oral anti-inflammatory 5-lipoxygenase-activating protein inhibitor British Journal of Clinical Pharmacology, 2012: p. Manuscript in Press.
[2] Verotta D and Sheiner L, Semiparametric analysis of non-steady-state pharmacodynamic data. J Pharmacokinet Biopharm, 1991. 19(6): p. 691-712.
[3] Fisher D and Wright P, Are plasma concentration values necessary for pharmacodynamic modeling of muscle relaxants? Anesthesiology, 1997. 86(3): p. 567-75.
[4] Gabrielsson J, Jusko W, and Alari L, Modeling of dose-response-time data: four examples of estimating the turnover parameters and generating kinetic functions from response profiles. Biopharm Drug Dispos, 2000. 21(2): p. 41-52.
[5] Jacqmin P, et al., Modelling response time profiles in the absence of drug concentrations: definition and performance evaluation of the K-PD model. J Pharmacokinet Pharmacodyn, 2007. 34(1): p. 57-85.
[6] Beal S, Sheiner L, and Boeckmann A, NONMEM User’s Guides. NMVI (2006) Icon Development Solutions: Ellicott City, Maryland, USA.
Reference: PAGE 21 (2012) Abstr 2461 [www.page-meeting.org/?abstract=2461]
Poster: Other Modelling Applications