A Comparison of Drug Transport in Pulmonary Absorption Models: Isolated Perfused rat Lungs, Respiratory Epithelial Cell Lines and Primary Cell Culture.

PURPOSE: To evaluate the ability of human airway epithelial cell layers and a simple rat isolated perfused lung (IPL) model to predict pulmonary drug absorption in rats in vivo. METHOD: The permeability of seven compounds selected to possess a range of lipophilicity was measured in two airway cell lines (Calu-3 and 16HBE14o-), in normal human bronchial epithelial (NHBE) cells and using a simple isolated perfused lungs (IPL) technique. Data from the cell layers and ex vivo lungs were compared to published absorption rates from rat lungs measured in vivo. RESULTS: A strong relationship was observed between the logarithm of the in vivo absorption half-life and the absorption half-life in the IPL (r = 0.97; excluding formoterol). Good log-linear relationships were also found between the apparent first-order absorption rate in vivo and cell layer permeability with correlation coefficients of 0.92, 0.93, 0.91 in Calu-3, 16HBE14o- and NHBE cells, respectively. CONCLUSION: The simple IPL technique provided a good prediction of drug absorption from the lungs, making it a useful method for empirical screening of drug absorption in the lungs. Permeability measurements were similar in all the respiratory epithelial cell models evaluated, with Calu-3 having the advantage for routine permeability screening purposes of being readily availability, robust and easy to culture.

In-vitro respiratory drug absorption models possess nominal functional P-glycoprotein activity.

OBJECTIVES: The P-glycoprotein (P-gp) efflux pump is known to be present within several major physiological barriers including the brain, kidney, intestine and placenta. However, the function of P-gp in the airways of the lung is unclear. The purpose of this study was to use the highly specific P-gp inhibitor GF120918A to investigate the activity of the P-gp transporter in the airways to determine whether P-gp could influence inhaled drug disposition. METHODS: P-gp activity was measured as a change in digoxin transport in the presence of GF120918A in normal human bronchial epithelial (NHBE) cells, Calu-3 cell layers and the ex-vivo rat lung. KEY FINDINGS: The efflux ratios (ERs) in NHBE and Calu-3 cells were between 0.5 and 2, in contrast to 10.7 in the Caco-2 cell control. These low levels of GF120918A-sensitive polarised digoxin transport were measured in the absorptive direction in NHBE cells (ER = 0.5) and in the secretory direction in Calu-3 cells (ER = 2), but only after 21 days in culture for both cell systems and only in Calu-3 cells at passage > 50. The airspace to perfusate transfer kinetics of digoxin in the ex-vivo rat lung were unchanged in the presence of GF120918A. CONCLUSIONS: These results demonstrated that although low levels of highly culture-dependent P-gp activity could be measured in cell-lines, these should not be interpreted to mean that P-gp is a major determinant of drug disposition in the airways of the lung.