Tobias Kanacher (1) Ashish Sharma (2) José-David Gómez-Mantilla (2)
SGS Exprimo NV, Mechelen, Belgium, (2) Boehringer Ingelheim Pharma GmbH & Co. KG, Biberach, Germany, § corresponding author: jose_david.gomez_mantilla@boehringer-ingelheim.com
Objectives:
To develop a whole body PBPK model for salmeterol, capable of describing the pharmacokinetics (PK) of salmeterol following inhalation via the DISKUS® inhaler. Because this model is intended to be used as victim drug model in drug-drug interaction (DDI) simulations, close attention was paid to the precise estimation of the following aspects:
- Fraction absorbed from lung to the systemic circulation and subsequent metabolism in the liver
Fraction swallowed, which then can be subject to either pre-systemic metabolism via CYP3A4 [1] in the intestine or the liver.
Methods:
A PBPK analysis using the Open Systems Pharmacology (OSP) Suite 7.2.1. was conducted in a stepwise middle-out approach. No PK data in humans after intravenous (i.v.) dosing and only very limited data following oral (p.o.) administration to humans and different animal species were available [2]. Thus in the first step, literature mean concentration-time data from healthy volunteers inhaling a single dose of 50 µg salmeterol with simultaneous gavage of charcoal [3] to supress oral absorption were used to define distribution and clearance.
As the exact fraction deposited in the lungs and subsequently absorbed from there is unknown, different scenarios with fractions deposited ranging 10%[4] – 25%[5] were investigated. Further the effect of diverse ways of describing lung absorption were evaluated: directly from the lung cells with and without first order release or from an additional alveolar lung fluid (ALF)[6] compartment.
In the second step, data from an in-house clinical study [7] with inhaled salmeterol by healthy volunteers in the same dose and conditions but without charcoal were used to identify intestinal permeability and fraction metabolised in gut.
Model evaluation was performed with clinical data from healthy volunteers inhaling 50 µg or 100 µg salmeterol twice daily for seven days [8], [9].
Finally, the model was used to predict AUC and Cmax of salmeterol following the simultaneous oral administration of a strong CYP3A4 inhibitor. This was done by either assuming a fixed 90% reduction of the CYP3A4 clearance of salmeterol or co-administration of itraconazole, considering its dynamic concentration-time course.
Results:
A scenario-driven middle-out approach led to a robust PBPK model structure being able to describe mean PK parameters of salmeterol in healthy volunteers following inhalation.
The models with lung absorption via lung cells with first order release or via absorption from ALF resulted in equivalent estimations for distribution and clearance and matched the physicochemical properties of salmeterol better than the model with lung absorption via lung cells without first order release.
The successful model included the following assumptions:
- fraction of dose released from the device is assumed to be 90%, of this the fraction
- deposition in the lung can range from 10% – 25% and is absorbed by a fast first order kinetics
- Fraction swallowed ranges accordingly from 80% – 65%
- fraction mucociliary cleared can be neglected
- no differentiation between different lung absoprtion sides
- salmeterol mainly metabolised by CYP3A4 in gut/liver (not metabolism in lung)
- renal excretion of salmeterol is negligible (<5% dose)
Further investigation using the scenario with 20% fraction deposited to lung in population simulations with 2000 virtual individuals suggested, that variable efficiency of inhalation largely contribute to the inter-individual variability of salmeterol PK.
A DDI simulation with 90% reduced CYP3A4 clearance to mimic the effect of a strong CYP3A4 inhibitor resulted in predictions of AUC and Cmax ratios within two-fold range with clinical observations where salmeterol was given with and without ketoconazole as perpetrator. DDI simulations where itraconazole was dynamically coupled with the salmeterol model further confirmed the model.
Conclusions:
This is the first whole body PBPK model describing human PK of salmeterol after inhalation that can be used for predictions of DDIs with orally given strong CYP3A4 inhibitors.
Using a middle-out approach based on a limited human dataset from literature sources and in-house data successfully lead to a robust model structure.
The model was confirmed with salmeterol PK after multiple inhalations of different doses in humans.
This example demonstrates how PBPK models can bridge limitations in data by inclusion of prior knowledge and investigation of alternative routes of applications.
References:
[1]. G. R. Manchee et al., “The aliphatic oxidation of salmeterol to alpha-hydroxysalmeterol in human liver microsomes is catalyzed by CYP3A,” Drug Metab. Dispos. Biol. Fate Chem., vol. 24, no. 5, pp. 555–559, May 1996.
[2]. G. R. Manchee et al., “Disposition of salmeterol xinafoate in laboratory animals and humans,” Drug Metab. Dispos. Biol. Fate Chem., vol. 21, no. 6, pp. 1022–1028, Dec. 1993.
[3]. K. Soulele, P. Macheras, L. Silvestro, S. Rizea Savu, and V. Karalis, “Population pharmacokinetics of fluticasone propionate/salmeterol using two different dry powder inhalers,” Eur. J. Pharm. Sci., vol. 80, pp. 33–42, Dec. 2015.
[4]. J. A. Bennett, T. W. Harrison, and A. E. Tattersfield, “The contribution of the swallowed fraction of an inhaled dose of salmeterol to it systemic effects,” Eur. Respir. J., vol. 13, no. 2, pp. 445–448, Feb. 1999.
[5]. ARA, “Multiple-Path Particle Dosimetry Model (MPPD v 2.11) | www.ara.com.” [Online]. Available: https://www.ara.com/products/multiple-path-particle-dosimetry-model-mppd-v-211. [Accessed: 07-Nov-2017].
[6]. Massimiliano Germani, Christoph Niederalt, Tobias Kanacher, Thomas Stohr, Laurent Detalle, Thomas Wendl, and Laura Sargentini, “A Physiology based pharmacokinetic (PBPK) model to explore ALX0171 PK in infants following inhalation.” [Online]. Available: http://www.ablynx.com/uploads/data/files/poster_pbpk_acop2014.pdf. [Accessed: 20-Dec-2017].
[7]. Boehringer Ingelheim, “NCT02254174: A Randomised, Open-label Four-way Crossover Study to Evaluate Relative Bioavailability of Tiotropium and Salmeterol After Inhalation of a Fixed Combined Single Dose,” 2007.
[8]. R. Mehta, K. Riddell, A. Gupta, M. D. Louey, and R. H. Chan, “Comparison of the Pharmacokinetics of Salmeterol and Fluticasone Propionate 50/100 µg Delivered in Combination as a Dry Powder Via a Capsule-Based Inhaler and a Multi-Dose Inhaler,” Clin. Drug Investig., vol. 35, no. 5, pp. 319–326, May 2015.
[9]. R. Mehta et al., “Pharmacokinetics of fluticasone propionate and salmeterol delivered as a combination dry powder via a capsule-based inhaler and a multi-dose inhaler,” Pulm. Pharmacol. Ther., vol. 29, no. 1, pp. 66–73, Oct. 2014.
Reference: PAGE 27 (2018) Abstr 8597 [www.page-meeting.org/?abstract=8597]
Poster: Drug/Disease Modelling - Absorption & PBPK