Haini Wen1, Dr Muhammad Waqas Sadiq2, Professor Lena Friberg1, Associate Professor Elin Svensson1,3
1Department of Pharmacy, Uppsala University, 2Clinical Pharmacology & Quantitative Pharmacology, Clinical Pharmacology & Safety Sciences, R&D, 3Department of Pharmacy, Radboud University Medical Center
Introduction
Pulmonary drug administration offers significant advantages for treating respiratory diseases, including rapid onset of action, targeted delivery to lung tissues, and reduced systemic exposure.[1] However, developing inhaled medications present unique challenges, particularly in understanding intrapulmonary pharmacokinetics (PK). The lung’s complex structure, consisting of conducting airways and alveolar spaces, results in heterogeneous drug distribution across different regions.[2] Methods for assessing lung PK include tissue biopsies, bronchoalveolar lavage (BAL), and bronchosorption.[3] While biopsies and BAL are widely used, they have drawbacks such as lack of spatial information and risk of potential biases.[4,5] Bronchosorption, a newer technique for sampling intrapulmonary drug concentrations, improves spatial resolution by collecting samples directly from bronchial regions. However, it remains invasive due to its reliance on bronchoscopy, which requires the insertion of a bronchoscope into the airways and limits sampling frequency.[3] Consequently, many studies still rely on epithelial lining fluid (ELF)-to-plasma concentration ratios or area under the curve (AUC) comparisons to evaluate pulmonary drug distribution.6 Population pharmacokinetic (PopPK) modeling, when combined with limited sampling techniques, helps address uncertainties in drug disposition. Although one- or two-compartment models are commonly used to describe intrapulmonary pharmacokinetics, their empirical parameters are difficult to interpret and often poorly predict lung PK profiles. This study aimed to develop a population pulmonary physiologically based pharmacokinetic (PBPK) model for salbutamol to combine advantages of population and PBPK approaches. Additionally, different sampling strategies to improve the accuracy and reliability of lung PK data were explored, of value for ultimately advancing the development of inhaled therapies.
Methods
This population PBPK analysis utilized data from a previously published PK study on the lung disposition of inhaled salbutamol (NCT03524066).[3] The study included 13 healthy, non-smoking volunteers (18–50 years old) who received a single 200 µg dose of salbutamol via a spacer. Each participant underwent two lung sampling sessions using bronchoscopy at one of the following time intervals after drug administration: 1 and 6 hours, 2 and 9 hours, 3 and 12 hours, or 4 and 18 hours. Plasma samples were obtained at 0.25, 0.5, 1, 2, 4, 8, 12, and 24 hours post-dose. Drug concentrations were quantified using liquid chromatography with tandem mass spectrometry (LC-MS/MS). Plasma, BAL and bronchosorption samples underwent protein precipitation, while biopsy samples were homogenized before processing. ELF drug concentrations in BAL were corrected for dilution using the urea method. A total of 101 plasma and 100 lung PK samples (32 bronchosorption, 32 biopsy, and 36 BAL) were available for analysis. A minimal PBPK (mPBPK) model for inhaled salbutamol was developed using NONMEM with first-order conditional estimation. Systemic PK parameters were first estimated using plasma data and subsequently fixed to inform lung PK modeling. The lung was compartmentalized into tracheobronchial, bronchiolar, and alveolar regions following Weibel’s lung morphometry.[7] Pulmonary absorption was described as a passive diffusion process, with the unbound lung-to-plasma partition coefficient (Kp,u,lung) and the effective permeability (Peff) as key parameters. Deposition fractions for lung regions were estimated based on typical path models.[2] The impact of permeability on lung PK was assessed by simulating hypothetical molecules with various permeabilities. Sampling strategies for bronchosorption were evaluated via stochastic simulation-estimation (SSE). Staggered sampling strategies, dividing participants into three groups sampled at distinct time points, were compared with uniform sampling methods for precision and bias in estimating Kp,u,lung and Peff.
Results
Plasma PK was best described by a one-compartment model with simultaneous first-order absorption from the lung and gut. The mPBPK model adequately described pulmonary PK data across sampling methods. Final estimates for Kp,u,lung and Peff were 11.2 and 0.55 m/h, respectively. No significant inter-individual variability (IIV) or differences between left and right lung were detected. Simulation results indicated that drugs with lower permeability exhibited prolonged distribution phases in ELF. BAL-derived lung-to-plasma ratios showed high variability over time, with values changing approximately 1000-fold over 8 hours post-dose. The performance of different bronchosorption sampling strategies varied with drug permeability in estimating PK parameters. When subject numbers were limited (<15), uniform sampling led to high bias in some cases. For salbutamol, uniform sampling at 2 hours performed better, while lower permeability compounds favoured later sampling. Staggered strategies demonstrated better accuracy and precision across different permeability conditions, with early staggered sampling (0.5, 4, 10 hours) preferable for salbutamol and moderately lower permeability drugs, while late staggered sampling (2, 6, 24 hours) performed better for low-permeability compounds.
Conclusion
A pulmonary population PBPK model for salbutamol was developed to describe intrapulmonary PK of salbutamol. By evaluating different sampling strategies, the study provides insights into optimizing PK assessments while minimizing patient burden. The developed mPBPK framework can support the integration and interpretation of preclinical and clinical data, ultimately facilitating the advancement of inhaled drug therapies.
[1]. Patton, J. S. & Byron, P. R. Inhaling medicines: delivering drugs to the body through the lungs. Nat. Rev. Drug Discov. 6, 67–74 (2007). [2]. Boger, E. & Wigström, O. A Partial Differential Equation Approach to Inhalation Physiologically Based Pharmacokinetic Modeling. CPT Pharmacomet. Syst. Pharmacol. 7, 638–646 (2018). [3]. Sadiq, M. W. et al. Lung pharmacokinetics of inhaled and systemic drugs: A clinical evaluation. Br. J. Pharmacol. 178, 4440–4451 (2021). [4]. Mouton, J. W. et al. Tissue concentrations: do we ever learn? J. Antimicrob. Chemother. 61, 235–237 (2008). [5]. Haeger, S. et al. The bronchoalveolar lavage dilution conundrum: an updated view on a long-standing problem. Am. J. Physiol.-Lung Cell. Mol. Physiol. 327, L807–L813 (2024). [6]. Drwiega, E. N. & Rodvold, K. A. Penetration of Antibacterial Agents into Pulmonary Epithelial Lining Fluid: An Update. Clin. Pharmacokinet. 61, 17–46 (2022). [7]. Weibel, E. R. Geometry and Dimensions of Airways of Conductive and Transitory Zones. In Morphometry Hum. Lung (Weibel, E. R.) 110–135 (Springer, Berlin, Heidelberg, 1963).doi:10.1007/978-3-642-87553-3_10
Reference: PAGE 33 (2025) Abstr 11747 [www.page-meeting.org/?abstract=11747]
Poster: Drug/Disease Modelling - Other Topics