Pharmacokinetic profile analyses for inhaled drugs in humans using the lung delivery and disposition model

Abstract

The kinetic clarification of lung disposition for inhaled drugs in humans via pharmacokinetic (PK) modeling aids in their development and regulation for systemic and local delivery, but remains challenging due to its multiplex nature. This study exercised our lung delivery and disposition kinetic model to derive the kinetic descriptors for the lung disposition of four drugs [calcitonin, tobramycin, ciprofloxacin and fluticasone propionate (FP)] inhaled via different inhalers from the published PK profile data.With the drug dose delivered to the lung (DTL) estimated from the corresponding γ-scintigraphy or in vivo predictive cascade impactor data, the model-based curvefitting and statistical moment analyses derived the rate constants of lung absorption (k a ) and non-absorptive disposition (k nad ). The k a values differed substantially between the drugs (0.05-1.00 h −1 ), but conformed to the lung partition-based membrane diffusion except for FP, and were inhaler/delivery/deposition-independent.The k nad values also varied widely (0.03-2.32 h −1 ), yet appeared to be explained by the presence or absence of non-absorptive disposition in the lung via mucociliary clearance, local tissue degradation, binding/sequestration and/or phagocytosis, and to be sensitive to differences in lung deposition. For FP, its k a value of 0.2 h −1 was unusually low, suggesting solubility/dissolution-limited slow lung absorption, but was comparable between two inhaler products. Thus, the difference in the PK profile was attributed to differences in the DTL and the k nad value, the latter likely originating from different aerosol sizes and regional deposition in the lung. Overall, this empirical, rather simpler model-based analysis provided a quantitative kinetic understanding of lung absorption and non-absorptive disposition for four inhaled drugs from PK profiles in humans. K E Y W O R D Scurve-fitting, inhaled drug, modeling, pharmacokinetics (PK), statistical moment analysis

Inferring pulmonary exposure based on clinical PK data: accuracy and precision of model-based deconvolution methods

“…It has to be noted that, even when selecting the right model for a drug with non-absorptive loss, the parameter estimation process resulted in unidentifiable parameters. Sakagami et al suggested that this instability can be circumvented by fixing the lung dose [ 31 ]. However, this requires detailed information about the lung dose, which is subject to great variability, both between subjects and between occasions [ 9 , 32 , 33 ].…”

A mechanistic framework for a priori pharmacokinetic predictions of orally inhaled drugs

“…However, previous modeling approaches either reduced the given complexity or lack adequate model evaluation. For example, the mucociliary clearance was described as a first-order process [15,16]. Other published population PK models did not differentiate between undissolved and dissolved drug and consider pulmonary drug absorption as a "one-way process", i.e.…”