Gülbeyaz Yildiz Türkyilmaz (1,2,3), Siv Jönsson (1), Mats O Karlsson (1)
(1) Department of Pharmaceutical Biosciences, Uppsala University, Sweden, (2) Center For Drug Research & Development And Pharmacokinetic Applications (ARGEFAR), Ege University, TURKEY, (3) Department of Pharmaceutical Technology, Faculty of Pharmacy, Ege University, Turkey
Objectives: To explore how well Pop PK models characterize the NCA parameters Cmax and AUC(0-t) in rich data setting and the influence of the absorption delay model choice.
Methods: The investigation was performed based on PopPK models for five real drug data sets with rich sampling [1-5]. Alternative models, with respect to the delay part of absorption, were evaluated for each data set (no delay, lagtime or transit compartment), using NONMEM 7.3 [6]. Simulations were conducted using nca in Perl-Speaks-NONMEM (PsN) [7, http://psn.sourceforge.net/]. The ncappc package [8] generated population and individual diagnostics were inspected and summarized across the different models and data sets. For each model, across drugs and NCA parameters, 30 population level metrics were generated; the observed population mean (Pop_Mean) compared with the 95% nonparametric prediction interval (npi), and the population mean and SD (including 95% CIs) of the NPDEs compared with the expected values (i.e. mean 0, SD 1). The individual level metrics generated across individuals, drugs and NCA parameters were the deviation of mean individual simulated NCA parameter from corresponding individual observed parameter (PPC_Outlier, outlier defined as deviation outside the 95% npi) and the NPDE of each NCA parameter (NPDE_Outlier, outlier defined as NPDE outside +/- 2 SD from mean).
Results: The transit compartment model resulted in the best fit for all drugs based on the OFV (reduction in OFV versus no delay ranged from 237 to 662). The transit model performed well for Cmax and AUC(0-t) for 5 and 4 data sets, respectively, for Pop_Mean (95% npi covered observed mean), and for 2 out of 5 drugs for both NPDE_Mean and NPDE_SD. Thus, 8 out of 30 metrics were outside the 95% PIs. For the lagtime and no delay models, the corresponding values were 10/30 and 17/30. For Cmax, across all models, the percentage identified individual subject outliers were in agreement with expectations; PPC_Outlier 3.6–6.3% vs expected 5%, and NPDE_Outlier 3.6-5.7% vs expected 4.5%. Corresponding PPC_Outlier and NPDE_Outlier values for AUC(0-t) were 2.7%-3.9%, and 2.7-3.3%, respectively.
Conclusion: Overall, the PopPK models described the mean and interindividual variability in the secondary metrics Cmax and AUC(0-t) reasonably well. The evaluation of NCA metrics was made easy by PsN and the ncappc package and choice of absorption delay model affected the quality of both Cmax and AUC(0-t) simulations.
References:
[1] Savic RM, Jonker DM, Kerbusch T, Karlsson MO. Implementation of a transit compartment model for describing drug absorption in pharmacokinetic studies. J Pharmacokinet Pharmacodyn. 2007;34(5):711-26.
[2] Bååthe S, Hamrén B, Karlsson MO, Wollbratt M, Grind M, Eriksson UG. Population pharmacokinetic of melagatran, the active form of the oral direct thrombin inhibitor ximelagatran, in atrial fibrillation patients receiving long-term anticoagulation therapy. Clin Pharmacokinet. 2006;45(8):803-19.
[3] Karlsson MO, Jonsson EN, Wiltse CG, Wade JR. Assumption testing in population pharmacokinetic models: Illustrated with an analysis of Moxonidine data from congestive heart failure patients. J Pharmacokinet Biopharm. 1998;26(2):207-46.
[4] Jönsson S, Davidse A, Wilkins J, van der Walt JS, Simonsson USH, Karlsson MO, Smith P, Mcllleron H. Population pharmacokinetic of ethambutol in south African Tuberculosis patients. Antimicrob Agents Chemother. 2011;55(9):4230-7.
[5] Karlsson MO, Sheiner LB. The importance of modelling interoccasion variability in population pharmacokinetic analyses. J Pharmacokinet Biopharm. 1993;21(6):735-50.
[6] Beal SL, Sheiner LB, Boeckmann AJ & Bauer RJ. NONMEM Users Guides. 1989-2011. Icon Development Solutions, Ellicott City, Maryland, USA
[7] Keizer RJ, Karlsson MO, Hooker A. Modeling and Simulation Workbench for NONMEM: Tutorial on Pirana, PsN, and Xpose. CPT Pharmacometrics Syst Pharmacol. 2013;2:e50.
[8] Acharya C, Hooker AC, Yildiz Türkyilmaz G, Jönsson S, Karlsson MO. A diagnostic tool for population models using non-compartmental analysis: The ncappc package for R. Computer Methods and Programs in Biomedicine (accepted for publication 2015)
Reference: PAGE 25 (2016) Abstr 5915 [www.page-meeting.org/?abstract=5915]
Poster: Methodology - Model Evaluation