(S-037) Mechanism-based PBPK Modeling Reveals Nonlinear Brain Pharmacokinetics of Fluvoxamine: Roles of Active Influx and Brain Metabolism

Mengxu Zhang – Division of Systems Pharmacology and Pharmacy, Predictive Pharmacology group, Leiden Academic Centre of Drug Research, Leiden University, Leiden, The Netherlands; Vivi Rottschäfer – Mathematical Institute, Leiden University, Leiden, The Netherlands; Vivi Rottschäfer – Korteweg-de Vries Institute for Mathematics, University of Amsterdam, Amsterdam, The Netherlands; Elizabeth C.M. de Lange – Division of Systems Pharmacology and Pharmacy, Predictive Pharmacology group, Leiden Academic Centre of Drug Research, Leiden University, Leiden, The Netherlands

Objectives: Fluvoxamine, a selective serotonin reuptake inhibitor (SSRI), displays nonlinear pharmacokinetics in the brain, previously attributed to efflux transporter activity [1]. However, this mechanism remains controversial due to inconsistencies with experimental findings showing brain-to-plasma unbound concentration ratios (Kp,uu,brain) greater than 1 [2]. This study aims to provide a mechanistic explanation for observed data on fluvoxamine’s nonlinear brain distribution using a physiologically based pharmacokinetic (PBPK) modeling approach via the LeiCNS-PK framework. Specifically, the study explores the roles of active influx transport and brain metabolism, offering an alternative to the efflux-based hypothesis.

Methods: A stepwise modeling strategy was employed, using published unbound plasma and brain extracellular fluid (brainECF) concentration-time data obtained via microdialysis at multiple dose levels in rats [1, 3] and serotonin (5-HT) receptor occupancy (RO) data [4] as the validation. Initially, only passive diffusion was incorporated into the LeiCNS-PK model to assess its capacity to capture the observed brainECF data. Active influx parameters were then estimated through nonlinear mixed-effects modeling in Monolix using a sequential approach based on observed plasma and brainECF data, and integrated into the PBPK model to improve the prediction performance. Finally, brain metabolic clearance [5] was introduced using in vitro–to–in vivo extrapolation (IVIVE) [6] from reported metabolic rates, enabling better prediction for the elimination phase. Additionally, 5-HT RO was predicted using a sigmoid Emax model driven by predicted unbound brainECF concentrations.

Results: Passive diffusion alone was insufficient to explain the observed nonlinearity in fluvoxamine brain kinetics. Inclusion of active influx markedly improved the fit to experimental data, however, there was still a mismatch in the elimination phase. The incorporation of brain metabolism further enhanced the model's ability to capture both concentration-time profiles, especially for the elimination phase and dose-dependent receptor occupancy.

Conclusions: This study provides a mechanistic explanation for fluvoxamine’s nonlinear brain distribution that aligns with experimental observations (Kp,uu,brain > 1), showing the essential roles of active influx and brain-specific metabolism. The developed PBPK model reconciles inconsistencies in previous assumptions and provides a robust framework for characterizing CNS drug disposition. Future directions will include evaluating the impact of CYP2D6 polymorphisms on 5-HT RO and translating model predictions to human contexts, with the goal of optimizing fluvoxamine dosing strategies to enhance therapeutic outcomes.

Citations: [1] Geldof, M., et al., Pharmacokinetic modeling of non-linear brain distribution of fluvoxamine in the rat. Pharm Res, 2008. 25(4): p. 792-804.
[2] Ma, Y., et al., Accurate prediction of K(p,uu,brain) based on experimental measurement of K(p,brain) and computed physicochemical properties of candidate compounds in CNS drug discovery. Heliyon, 2024. 10(2): p. e24304.
[3] Geldof, M., et al., Population pharmacokinetic model of fluvoxamine in rats: utility for application in animal behavioral studies. Eur J Pharm Sci, 2007. 30(1): p. 45-55.
[4] Geldof, M., et al., Pharmacokinetic-pharmacodynamic modelling of fluvoxamine 5-HT transporter occupancy in rat frontal cortex. Br J Pharmacol, 2008. 154(6): p. 1369-78.
[5] Ye, Z., et al., CYP2D6 gene polymorphism and apatinib affect the metabolic profile of fluvoxamine. Front Pharmacol, 2022. 13: p. 985159.
[6] Choi, G.W., Y.B. Lee, and H.Y. Cho, Interpretation of Non-Clinical Data for Prediction of Human Pharmacokinetic Parameters: In Vitro-In Vivo Extrapolation and Allometric Scaling. Pharmaceutics, 2019. 11(4).

Keywords: Fluvoxamine, nonlinear brain distribution, PBPK