PK and PBPK modeling approaches to describe the lung and systemic exposure after inhalation of AZD4604
Julia Larsson (1), Ramon Hendrickx (2), Markus Fridén (3), Helena Dahlbäck (2), Camille Riff (1)
(1) Clinical Pharmacology and Quantitative Pharmacology, Clinical Pharmacology & Safety Sciences, R&D, AstraZeneca, Gothenburg, Sweden. (2) Drug Metabolism and Pharmacokinetics, Research and Early Development, Respiratory & Immunology, BioPharmaceuticals R&D, AstraZeneca, Gothenburg, Sweden. (3) Inhalation Product Development, Pharmaceutical Technology & Development, Operations, AstraZeneca, Gothenburg, Sweden.
Objectives:
AZD4604 is an inhaled JAK1 selective inhibitor being developed for treatment of moderate to severe asthma [1]. AZD4604 is currently tested in a Phase 1 setting with the aim to investigate the safety, tolerability, and pharmacokinetic (PK) properties of the molecule [2]. The dose setting for this study was guided by predictions from a physiologically based pharmacokinetic (PBPK) model derived from non-clinical data.
This work aims to construct a PK model that describes how AZD4604 reaches the systemic circulation following oral inhalation, and to use that model to better understand the lung PK properties. In addition, the observed clinical data will be compared with the PBPK model predictions to evaluate the accuracy of these predictions.
Methods:
PK data collected after single or multiple dosing from healthy subjects was used to fit the model. These subjects were administered AZD4604 in one of the following ways: By intravenous (IV) dosing, oral dosing, or by oral inhalation. A non-linear mixed effect model was then fitted to all data simultaneously independent of route of administration, to distinguish the lung properties of AZD4604 from the elimination, distribution, and oral absorption processes. All clinical work was done in R 3.5.1., PsN 4.9.0, and NONMEM 7.4.4.
The non-clinical predictions consisted of typical path lung deposition modelling of in vitro aerosol data of the inhaled drug product as described in previous studies [3]. Post-deposition events including particle dissolution, epithelial transport, tissue distribution etc. was conducted in a Matlab implementation of the models previously published [4,5].
Results:
A three-compartment linear mammillary model with first-order absorption was used to describe the IV and oral PK profiles [6]. To capture the inhaled profiles, two additional compartments were added to simulate the fast and slow absorption from the lung, respectively. An additional error term on logarithmic scale was used to capture the noise, and inter-individual variability was neglected.
The model fitted the clinical data well and was able to distinguish the lung PK properties from the distribution, elimination, and oral absorption processes. This was demonstrated by a comparison of parameter values between models, which showed that their values were retained when going from the model fitted to data after IV or oral dosing to the model fitted to all PK data. In addition to this PK modeling of clinical data, the preclinically derived PBPK model accurately predicted the observed AUC, Cmax and terminal half-life following both inhaled and intravenous dosing.
Conclusions:
The PK model successfully captured the clinical data independent of administration route and showed that the clinical and non-clinical predictions were well aligned. The different routes that an orally inhaled dose of AZD4604 can take are well distinguished and gives essential information regarding systemic exposure relative to the inhaled dose. Next steps are to update the PK model when data from mild asthmatics becomes available and use the model to inform future clinical study designs and dosing regimens.
References:
[1] Nilsson M, Rhedin M, Hendrickx R, Berglund S, Piras A, Blomgran P, Cavallin A, Collins M, Dahl G, Dekkak B, Ericsson T. Characterization of Selective and Potent JAK1 Inhibitors Intended for the Inhaled Treatment of Asthma. Drug Design, Development and Therapy. 2022 Jan 1:2901-17. doi: 10.2147/DDDT.S354291
[2] ClinicalTrials.gov. A clinical trial in healthy volunteers and patients with mild asthma to investigate a new medicine (AZD4604) for the treatment of asthma (NCT04769869). Available from: https://clinicaltrials.gov/ct2/show/NCT04769869. Accessed Feb 24, 2023
[3] Bäckman P, Tehler U, Olsson B. Predicting exposure after oral inhalation of the selective glucocorticoid receptor modulator, AZD5423, based on dose, deposition pattern, and mechanistic modeling of pulmonary disposition. Journal of aerosol medicine and pulmonary drug delivery. 2017 Apr 1;30(2):108-17. doi: 10.1089/jamp.2016.1306
[4] Boger E, Fridén M. Physiologically based pharmacokinetic/pharmacodynamic modeling accurately predicts the better bronchodilatory effect of inhaled versus oral salbutamol dosage forms. Journal of aerosol medicine and pulmonary drug delivery. 2019 Feb 1;32(1):1-2. doi: 10.1089/jamp.2017.1436
[5] Boger E, Wigström O. A partial differential equation approach to inhalation physiologically based pharmacokinetic modeling. CPT: Pharmacometrics & Systems Pharmacology. 2018 Oct;7(10):638-46. doi: 10.1002/psp4.12344
[6] Bauer RJ. NONMEM tutorial part I: description of commands and options, with simple examples of population analysis. CPT: pharmacometrics & systems pharmacology. 2019 Aug;8(8):525-37. doi : 10.1002/psp4.12404