Javier Reig-Lopez1,2; Matilde Merino-Sanjuan1,2; Victor Mangas-Sanjuan1,2; Manuel Prado-Velasco3.
1Department of Pharmacy Technology and Parasitology, Faculty of Pharmacy, University of Valencia, Valencia, Spain. 2Interuniversity Institute of Recognition Research Molecular and Technological Development. 3Multiscale Modeling and Bioengineering Research Group, Escuela Técnica Superior de IngenierÃa, Universidad de Sevilla
Objectives: The aim of this work is to validate model predictions from PhysPK biosimulation software using a previously published semi-mechanistic model by comparing exposure metrics (AUC and Cmax values) previously generated in NONMEM with those obtained with PhysPK for each type of analyte: parent drug (PD), primary- (PM) and secondary-metabolite (SM).
Methods: A sample of 24 patients previously generated by Monte Carlo simulations in NONMEM 7.3 [1] was introduced in a parametric experiment wrote in EL language in PhysPK biosimulation software [2]. A previously semi-mechanistic model served as a basis for the analysis [3]. The study design was a single-dose study in 24 individuals receiving 100 mg of a drug. The dose is orally administered as a solid form (C8) and dissolution (E1) is considered limited by the solubility (S). The drug dissolved in lumen is either degraded or absorbed (E2). However, in this study the luminal degradation was fixed to zero. Moreover, the intestinal transit is considered as an operative absorption time (OAT1) fixed to 7h, allowing for passive diffusion, and the efflux activity was limited to 3h (OAT2). After drug absorption, the gut and liver partially metabolize the drug (E3-E6). This metabolism can be linear or non-linear, depending on the drug concentration related to KM. Then, the drug is rapidly distributed in one compartment (C4) and slowly distributed (E7) in peripheral compartment (C5). The elimination of parent drug is by intestinal (E3 and E5) and hepatic (E4 and E6) metabolism, while both metabolites are eliminated by renal excretion (E8 and E9). The simulations were performed after a single dose administration. Different scenarios were explored: different degrees of inter-individual variability, dose levels and BCS class (II and IV).
Results: Individual plasma concentration-time profiles of PD, PM and SM were represented, showing the concordance between both softwares. Relative change in exposure metrics (AUC and Cmax) were computed, considering NONMEM estimates as the reference value. Results showed that PD AUC values obtained with PhysPK biosimulation software were of the same magnitude order of those achieved with NONMEM. In general, PhysPK biosimulation software showed a relative change in PD, PM and SM AUC and Cmax values which are in agreement with the predicted value from NONMEM. The 95% confidence interval of the relative changes ranged from -12.5 to 10.7 % in the worst scenario. No statistically significant bias were detected when high versus low doses were used or high versus low variability, which demonstrates the adequacy of the non-linear processes implemented in both environments.
Conclusions: PhysPK biosimulation software is useful for estimating the pharmacokinetic parameters of a complex pharmacokinetic model, since it allows obtaining pharmacokinetic profiles similar to those obtained using NONMEM. However, these evaluation studies must be continued in order to correctly conclude the validity of the software.
References:
[1] Beal SLS, I.B.; Boeckmann A.; Bauer R.J. (2015) NONMEM Users Guides. Icon Development Solutions, Ellicott City, Maryland, USA.
[2] Gonzalez-Garcia, I., M. Prado-Velasco, C. Fernández-Teruel and S. Fudio (2017). Comparison of FO – FOCE population parameter estimation methods in PhysPK 2.0 against NONMEM 7.3. PAGE 2017. Abstracts of the Annual Meeting of the Population Approach Group in Europe Budapest, Hungary: 1-2.
[3] Mangas-Sanjuan V, Navarro-Fontestad C, García-Arieta A, Trocóniz IF, Bermejo M. Computer simulations for bioequivalence trials: Selection of analyte in BCS class II and IV drugs with first-pass metabolism, two metabolic pathways and intestinal efflux transporter. Eur J Pharm Sci. 2018 May 30;117:193-203.
Reference: PAGE 28 (2019) Abstr 9010 [www.page-meeting.org/?abstract=9010]
Poster: Methodology - New Modelling Approaches