K. Abduljalil(1), R. Rose(1), T.N. Johnson(1), T. Cain(1), L. Gaohua(1), D. Edwards(1), M. Jamei(1), A. Rostami-Hodjegan(1,2)
(1) SIMCYP Limited (a Certara Company), Sheffield, UK, (2) School of Pharmacy and Pharmaceutical Sciences, the University of Manchester, UK
Objectives: To simulate the concentration-response of caffeine after 150mg daily dose, in a virtual healthy pregnant population using the Simcyp Simulator.
Methods: Gestational age-dependent PBPK parameters [1], including CYP1A2 activity, were incorporated into the Simcyp V12 Release 2 Simulator and added to prior in vitro information on the metabolism and kinetics of caffeine available in Simcyp. The PD custom scripting module was used to define a tolerance model, which was not available in the pre-defined library of models within the Simulator. This was then linked to the systemic concentration of a full PBPK model. The change in mean arterial pressure was used as the PD marker of the pressor effect of caffeine. The PK-PD relationship represented by an empirical tolerance model, as adopted by Shi et al., 1993 and the simulated PBPK-PD profiles were compared to the reported observed values [2, 3].
Results: Predicted caffeine plasma levels and response for different doses and routes of administrations were in agreement with the clinical observations. The predicted area under the concentration curve, AUC0-72hr, was 2.2-fold higher in pregnant compared to non-pregnant women (78 vs 35 mg/L.h). Similarly, the predicted area under the effect curve, AURC0-72hr, was 1.84-fold higher in pregnant women compared to the baseline (311 vs 169 mmHg). In comparison with non-pregnant women, who did not show the persistence of the tolerance response based on a daily dose of 150mg caffeine, tolerance to the pressor effect is still apparent in pregnant women based on this dosing regimen.
Conclusions: The link between PD and PBPK, combined with in vitro in vivo extrapolation, offers an extension of the success of PBPK in drug development [4]. In this case study, the implemented approach allowed successful simulation of the effect of caffeine observed in clinical studies and offers prediction of the response in a pregnant population. The decreased activity of CYP1A2 during pregnancy results in the reduction of drug elimination and prolongation of the effect. The custom PD tool facilitates the linkage of more flexible PD models to the Simcyp PBPK models. This can extend the capability of clinical trial simulations similar to the one shown in this study and can be used to investigate the design and power of POPPK-PD studies performed across a variety of clinical settings such as sub-populations and drug-interactions.
References:
[1] Abduljalil K et al., Clin Pharmacokinet. 2012 Jun 1;51(6):365-96.
[2] Shi J et al., Clin Pharmacol Ther. 1993 Jan;53(1):6-14.
[3] Brazier JL et al., Dev Pharmacol Ther. 1983;6(5):315-22.
[4] Rostami-Hodjegan A. Clin Pharmacol Ther. 2013 Feb;93(2):152.
Reference: PAGE 22 (2013) Abstr 2753 [www.page-meeting.org/?abstract=2753]
Poster: Absorption and Physiology-Based PK