Javier Zarzoso-Foj1,2, Marina Cuquerella-Gilabert1,2,3, Matilde Merino-Sanjuán1,2, Javier Reig-López1,2, VÃctor Mangas-Sanjuán1,2, Alfredo GarcÃa-Arieta4
1Universitat de València, 2Interuniversity Research Institute for Molecular Recognition and Technological Development, 3Simulation Department, Empresarios Agrupados Internacional S.A., 4Consultant for WHO Prequalification of Medicines Programme
Introduction: Physiologically-based pharmacokinetic (PBPK) modelling has recently shown powerful prediction capabilities for drug exposure in humans and other species. However, differences in formulations are not often captured with these models [1]. To address this limitation, physiologically-based biopharmaceutics models (PBBM) integrating in vitro dissolution data into PBPK frameworks have been developed to mechanistically characterize the drug release and dissolution through an in vitro-in vivo (IVIV) link [2,3]. Ibuprofen, a Biopharmaceutics Classification System (BCS) class II drug struggles in establishing biowaivers compared to high solubility BCS drugs. PBBM/PBPK frameworks can aid in identifying biopredictive in vitro testing methods, anticipating in vivo bioequivalence (BE) study outcomes and delimitating safe space for these drugs through simulations. Objectives: Therefore, the aims of this study were (i) optimizing the variability of an existing PBPK model for ibuprofen using data from BE studies, (ii) developing a PBBM model for ibuprofen integrating in vitro dissolution data at different conditions, and (iii) identifying the safe space of dissolution parameters through virtual BE (VBE) simulations under different study designs. Methods: Simcyp® Simulator (v21R1) was used to optimize the variability of the PBPK model [4] to match that of 15 BE studies across different pharmaceutical forms and doses of ibuprofen. Variability was adjusted by a stepwise workflow of increasing complexity, from intravenous administration to oral tablets. Mean test and reference formulations in vitro dissolution data and an external in vivo bioequivalence evaluation were gathered from literature [5]. Dissolution experimental in vitro conditions included phosphate 5 and 50 mM and maleate 7 mM buffers, with and without HCl pH 1.2 and 2.0 acidic pretreatments. Simcyp In Vitro Data Analysis (SIVA®, v5) toolkit was used to estimate the diffusion layer model (DLM) parameter that mainly drove the dissolution process of the reference and test formulations of ibuprofen under all in vitro conditions. After integrating biopharmaceutic data obtained in SIVA® into the PBPK model with optimized variabilities, population representative simulations were performed. Scenario showing prediction errors (PE) on Cmax within the 0.9-1.11 range and ±30% on tmax for both formulations was selected for safe space definition and VBE simulations under different study designs. Results: Cmax variability was adequately predicted after changing Vss and MRT in stomach and small intestine CV (%) to 10 and 150 %, respectively. Particle surface pH was identified as the dissolution key parameter for ibuprofen, and maleate buffer 7 mM with HCl pH 2.0 pretreatment was established as the most biopredictive media after meeting the acceptance criteria (1.01 and 1.09 PE on Cmax and -30 and -25% difference on tmax for reference and test product) when estimating surface pH values of 6.02 and 5.57 for reference and test formulations, respectively. Moreover, this dissolution media was able to predict the failure in BE of the in vivo study (90% CI Cmax: 80.48 – 93.21% and 79.69 – 91.18% for R- and S-ibuprofen, respectively), with 90% CI Cmax of 77.22 – 83.65% and 78.90 – 85.25% for R- and S-ibuprofen, respectively. A safe space for test product surface pH values of 5.64 – 6.40 was defined in order to achieve a 90% CI for Cmax ratio within the 80 – 125% range when the reference product surface pH is 6.02. R-ibuprofen was identified as the most discriminative enantiomer, and differences between Cmax ratios of both enantiomers increased when surface pH between formulations were higher. After generating dissolution profiles for the safe space, dissolution rate should be rapid (between 48-81% and 73-99% after 0.5 and 0.75 h, respectively) for formulations to be BE. VBE studies with 12, 24 and 36 individuals showed BE outcomes that are sensitive to the number of trial replicates and runs, as well as to the particle surface pH of the formulations. Small changes in surface pH led to fewer individuals needed in order to accomplish 80% probability of BE success. Conclusions: Ibuprofen particle surface pH has been identified as the in vitro parameter governing dissolution in maleate buffer 7 mM with HCl pH 2.0 pretreatment, allowing to establish an in vitro safe space useful for calculating sample sizes and to evaluate the BE success rate through PBBM/PBPK model-informed VBE simulations.
1. Loisios-Konstantinidis, I. & Dressman, J. Physiologically Based Pharmacokinetic/Pharmacodynamic Modeling to Support Waivers of In Vivo Clinical Studies: Current Status, Challenges, and Opportunities. Mol Pharm 18, 1–17 (2021). 2. Mackie, C. et al. Physiologically Based Biopharmaceutics Modeling (PBBM): Best Practices for Drug Product Quality, Regulatory and Industry Perspectives: 2023 Workshop Summary Report. Mol Pharm 21, 2065–2080 (2024). 3. Anand, O., Pepin, X. J. H., Kolhatkar, V. & Seo, P. The Use of Physiologically Based Pharmacokinetic Analyses—in Biopharmaceutics Applications -Regulatory and Industry Perspectives. Pharmaceutical Research 2022 39:8 39, 1681–1700 (2022). 4. Reig-López, J. et al. Bioequivalence risk assessment of oral formulations containing racemic ibuprofen through a chiral physiologically based pharmacokinetic model of ibuprofen enantiomers. Eur J Pharm Biopharm 199, (2024). 5. Cámara-Martinez, I. et al. Level A IVIVC for immediate release tablets confirms in vivo predictive dissolution testing for ibuprofen. Int J Pharm 614, (2022).
Reference: PAGE 33 (2025) Abstr 11764 [www.page-meeting.org/?abstract=11764]
Poster: Drug/Disease Modelling - Absorption & PBPK