DissolvIt : development and validation of a new in vitro dissolution apparatus for dry powder aerosols

Abstract

Today there are no recommended dissolution methods for use in the development of new inhalable drugs. In the absence of such tailored methods, standard dissolution test methods from oral- or skin administration routes are used instead. Such methods are obviously not designed to resemble the lung environment. There is also a growing demand for the next level of in vitro methods that may correlate the generated in vitro data with in vivo data for tested substances, that is to demonstrate in vitro-in vivo correlation. Being able to predict in vivo data as well as clinical outcomes, provides enormous potential in minimizing animal experiments, reduce cost, and shorten time to optimize the drug development process.

The aim of this thesis was to develop and validate a new in vitro dissolution test method to be used for dry powder aerosols. One important task in the strategy was to make the apparatus relevant to the physiology of the lung.

The DissolvIt was developed. Particles were deposited with the aerosol generator PreciseInhale on glass cover slips. At the start of an experiment, the particles on a glass cover slip are contacted with a mucus simulant. Particles are dissolved into the mucus simulant and absorbed via a hydrophilic membrane to a blood simulant pumped along the opposite side of the membrane. Samples of the blood simulant are repeatedly collected over time and are analyzed with liquid chromatography-mass spectrometry/mass spectrometry for the active pharmaceutical ingredient(s) tested. It was shown that DissolvIt discriminates between different active pharmaceutical ingredients and generates clinical-like data in the form of absorption curves to the perfusate containing Cmax and Tmax. It was also shown that the particle dose and deposition pattern affect the dissolution process, but that PreciseInhale allow sufficient control of the deposition pattern and dose, to minimize a negative influence on dissolution.

The composition of the DissolvIt mucus simulant was evaluated against more complex variants such as simulated lung fluid and Survanta, but the simulant chosen as standard for DissolvIt was based on phosphate buffer containing polyethylene oxide and L alpha phosphatidylcholine. Through physiologically based pharmaco-kinetic modeling, the standard mucus was shown to generate pharmacokinetic values closest to the in vivo profile of the same drug. The lipid content of the mucus simulant was varied, but a level of 0.4 % was chosen because the most stable data was generated and 0.4 % is close to the mean value of the lipid content in the lung lining layer. DissolvIt was also compared to the paddle over disc- and modified Transwell dissolution methods. Two parameters from the relevant lung physiology are hard to mimic in vitro; the sparse dose deposition over the lung surface and the air-blood barrier thickness. However, in a comparison with data from the ex vivo rat lung, the DissolvIt produced results, regarding both design and pharmacokinetic output such as first half-time of absorption, that most resembled data from the living lung compared with paddle over disc and modified Transwell methods.

In conclusion, the DissolvIt is a newly developed in vitro dissolution apparatus tailored for inhalable drugs, with a potential to be established as a recommended test method in the drug development process, also with respect to in vitro-in vivo correlation.