Karine Rodriguez-Fernandez (1), Jose David Gomez-Mantilla (2), Suneet Shukla (2), Victor Mangas-Sanjuan (1,3), Sheila-Annie Peters (2).
(1) Department of Pharmacy and Pharmaceutical Technology and Parasitology, University of Valencia. Valencia, Spain; (2) Translational Medicine & Clinical Pharmacology, Boehringer Ingelheim Pharma GmbH & Co. KG, 55216 Ingelheim, Germany; (3) Interuniversity Research Institute for Molecular Recognition and Technological Development, Polytechnic University of Valencia-University of Valencia, Valencia, Spain.
Introduction: The simultaneous administration of oral dosage forms with food can have a significant impact on drug pharmacokinetics [1]. Physiologically based pharmacokinetic (PBPK) models incorporate factors influencing oral drug absorption and may be used to predict food effect [2,3]. Nevertheless, model parameterization is often challenged by in vitro – in vivo disconnect and/or parameter non-identifiability [4,5]. These challenges may be overcome by simplifying PBPK models, thereby reducing the number of parameters, while the models are still fit for purpose. PBPK models for predicting the food effect of BCS class II/IV drugs may be simplified by recognizing that solubility- limited absorption (SLA) is the main driver of positive food effect [6,7] if their exposures are not significantly impacted by intestinal efflux or metabolism.
Objectives: The aim of this study is to propose a novel approach for in silico prediction of food effect based on determining a sensitive range of dose-adjusted solubility and solubility-limited absorption for BCS class II/IV compounds.
Methods: A literature search was performed to find commercially available BCS II/IV drugs with a clinical food effect study and a solubility in fasted state simulated intestinal fluid (FaSSIF) available. A subset of compounds was selected to repeat the measure of FaSSIF solubility. The dose-adjusted FaSSIF solubility (FaSSIF/D) was calculated in the subset of compound as the ratio between the solubility in FaSSIF conditions and the approved dose strength. Then, each group of compounds were ranked from the lowest to the highest FaSSIF/D values and sub-divided into 3 areas based on an upper and a lower limit of FaSSIF/D values: drugs with FaSSIF/D > upper limit exhibited no food effect, drugs with FaSSIF/D < lower limit showed food effect, while those between the two limits lied in the sensitive range (SR) and comprised drugs with and without food effect. A second SR was determined by adding the FaSSIF/D values of the remained compounds with FaSSIF data gathered from the literature and was compared with the SR determined only with the experimentally measured FaSSIF. For drugs within SR determined with experimental FaSSIF solubility, SLA identified with simplified PBPK models in PK-Sim® was tested as a surrogate for food effect prediction, based on the determination of an optimized and an hypothetical highest solubility values. If the Cmax ratio of exposure simulated by PBPK model using the hypothetical highest solubility to exposure derived from optimized solubility was 1 or close to 1, exposure was not solubility-limited corresponding with a no food effect. In contrast, a significantly higher ratio characterized a solubility-limited exposure and a positive food effect.
Results: A total of 44 compounds with an available FaSSIF solubility value were collected from literature (70% food effect positive). The experimental FaSSIF solubility was determined in a subset of 19 compounds (63 % food effect positive). The SR was between 2.40E-05 and 1.25E-03 1/ml. The inclusion of the additional compounds with literature-reported FaSSIF data provided comparable lower SR limit to those from in-house data and a expand the upper limit to 1.25E-02 1/ml. PBPK models were available for 7 drugs within SR of FaSSIF experimentally determined. Efavirenz and carbamazepine had Cmax ratios greater than 1, which supports the positive food effect observed. In contrast, the oral exposures of digoxin, mefenamic acid, and tizanidine were not sensitive to an increase in the solubility beyond the best-fit value, which was consistent with the lack of food effect for these drugs. It was not possible to identify SLA for clopidogrel and felodipine, both with intestinal metabolism, because the observed Cmax was less than simulated with FaSSIF solubility.
Conclusions: A methodology has been proposed for in silico prediction of food effect based on determining a SR of FaSSIF/D and SLA for BCS class II/IV compounds. Based on the FaSSIF/D outside the SR, drugs with food effect can be differentiated from those without food effect. However, within the SR sensitive range, SLA identified by PBPK can serve as a good surrogate for binary prediction of food effect. This proposal allows for reliable prediction of food effect to enable decisions on the need for pilot and timing of pivotal food effect studies.
References:
[1] P.G. Welling, Influence of food and diet on gastrointestinal drug absorption: a review, Journal of pharmacokinetics and biopharmaceutics 5(4) (1977) 291-334. doi:10.1007/bf01061694.
[2] L. Cheng, H. Wong, Food Effects on Oral Drug Absorption: Application of Physiologically-Based Pharmacokinetic Modeling as a Predictive Tool, Pharmaceutics 12(7) (2020). doi:10.3390/pharmaceutics12070672.
[3] Kesisoglou F. Can PBPK Modeling Streamline Food Effect Assessments? J Clin Pharmacol. 2020 Oct;60 Suppl 1:S98-S104. doi: 10.1002/jcph.1678. PMID: 33205433.
[4] S.A. Peters, H. Dolgos, Requirements to Establishing Confidence in Physiologically Based Pharmacokinetic (PBPK) Models and Overcoming Some of the Challenges to Meeting Them, Clinical pharmacokinetics 58(11) (2019) 1355-1371. doi:10.1007/s40262-019-00790-0.
[5] S.A. Peters, Chapter 17. Absorption-related Applications of PBPK Modeling, Physiologically based pharmacokinetic (PBPK) modeling and simulations: principles, methods, and applications in the pharmaceutical industry, John Wiley & Sons, Inc. 2022.
[6] Alimpertis N, Simitopoulos A, Tsekouras AA, Macheras P. IVIVC Revised. Pharm Res. 2024 Feb;41(2):235-246. doi: 10.1007/s11095-024-03653-x. Epub 2024 Jan 8. PMID: 38191705.
[7] Simitopoulos A, Tsekouras A, Macheras P. Coupling Drug Dissolution with BCS. Pharm Res. 2024 Mar;41(3):481-491. doi: 10.1007/s11095-024-03661-x. Epub 2024 Jan 30. PMID: 38291164.
Reference: PAGE 32 (2024) Abstr 11043 [www.page-meeting.org/?abstract=11043]
Poster: Drug/Disease Modelling - Absorption & PBPK