Methods to determine lung distribution of inhaled drugs – could gamma scintigraphy be the gold standard?

Bioequivalence between two formulations of a systemically active drug can be accepted when circulating serum/plasma drug concentrations at each time point after dosing are similar, as the two products are assumed to have the same efficacy and safety. Regulatory authorities require this information, together with supporting in vitro data and clinical studies, for the licensing of a new or reformulated product. However when the drug effect is local and the drug is delivered to the body via this local site and simultaneously by other routes, the demonstration of bioequivalence is difficult. This is the case for an inhaled product. The therapeutic sites of action for bronchodilators and anti-inflammatory agents, used in the management of asthma and chronic obstructive pulmonary disease, are in the airway walls and inhaled therapy is the method of choice. Following inhalation, drug is either delivered to the lungs or is swallowed. The distribution is dependent on many factors (e.g. the inhaled product, inhalation technique, and condition of the patient’s airways). Thus drug enters the systemic circulation by the pulmonary and gastrointestinal route. The portion delivered to the lungs (total lung dose) is either cleared by mucociliary clearance (and then swallowed and absorbed via the gastrointestinal tract) or is absorbed through the airways. It is this pulmonary absorbed fraction which will exert a clinical effect (effective lung dose). The method of gamma scintigraphy pioneered and described by Newman [1] does not differentiate between these two fractions of the total lung dose. However, there is no study to show which lung dose (effective or total) is more important in predicting the clinical response in the airways and over the past 20 years, gamma scintigraphy has been regarded by many as the gold standard to measure lung deposition.

Gamma scintigraphy is based on reformulating an existing inhaled product to incorporate a radiolabel. In vitro studies are conducted to show that the aerodynamic particle size characteristics of the radiolabelled product (labelled and unlabelled drug) are similar to the original product. There is no requirement to show how the product was reformulated (e.g. the pellet formulations used in a Turbohaler dry powder inhaler [2]). The radiolabelled formulation is then inhaled and the amount of radioactivity is determined following imaging by a gamma camera. Owing to time constraints due to the decay of the radionuclide, imaging is restricted to two dimensions. A typical planar image is shown in Figure 2 of the review article by Newman [1]. This figure shows the deposition following inhalation from a metered dose inhaler (MDI) and when the MDI was attached to a large volume spacer. The darker areas in the lung fields for the MDI attached to the spacer demonstrates higher pulmonary deposition whereas the MDI images show more oropharyngeal deposition and more in the stomach (swallowed). This fraction of the dose remains in the spacer as shown in Figure 2 of the review article [1]. Further analysis of the data allows a template to be drawn on the planar lung images to separate amounts deposited into the central, middle (intermediate) and outer (peripheral) zones. However the clinical significance of delivering different amounts to these areas is not understood. Generally it is the total lung dose which is important rather than the distribution [3]. Thus, more sensitive three-dimensional methods (SPECT – single photon emission computed tomography, and PET – positron emission tomography) mentioned by Newman [1] may not be necessary for either a comparison between two inhaled products or identification of the total lung dose. These three-dimensional methods require much higher doses of radiation to be inhaled. PET techniques are drug rather than formulation modification specific.

Despite more than 20 years of development and publication the regulatory authorities do not regard gamma scintigraphy as the gold standard and instead favour clinical response. However most clinical studies are carried out using measurements at the flat (plateau) portion of the dose-response relationship. For instance, a doubling of the therapeutic fluticasone inhaled dose has been shown to increase the peak expiratory flow rate (PEFR) by only 4.3 l min⁻¹ from a baseline of almost 200 l min⁻¹[4]. Also for beclomethasone the FEV₁ (forced expiratory volume in the first second) increased by 0.18 l above baseline after 200 µg inhaled twice daily and by 0.21 l after 400 µg twice daily [5]. For the β₂-adrenoceptor agonists the maximum response from therapeutic inhaled doses has long been established [6]. Newman et al. (like many others) also demonstrated this in asthmatic subjects except that they also measured lung deposition using gamma scintigraphy [7]. When these subjects inhaled radiolabelled salbutamol from a MDI and a MDI attached to a large volume spacer the total lung deposition was 12.3 and 23.1% (of the dose), respectively, but there was no difference in spirometry measurements.

To overcome single steady state dose comparisons at the plateau of the dose-response relationship some have used three therapeutic doses of each product using a parallel study design and demonstrated parallelism between the dose response curves. The horizontal difference (dose axis) is then taken as the relative potency (dose ratio) between the two. A recent study using this design showed that beclomethasone formulated in a metered dose inhaler (MDI) with a hydrofluoroalkane (HFA) propellant was 2.6 times more potent than the chlorofluorocarbon propellant formulation [8]. However the 95% confidence interval (CI) was wide (1.1–11.1). The 95% CI indicate that the potency may be close to one or could be close to the difference reported by gamma scintigraphy, which has shown that the HFA MDI (formulation of beclomethasone) delivers approximately 10 times more drug to the lungs than the CFC-MDI [9]. The difference between the clinical studies and gamma scintigraphy may be due to the enhanced deposition in the peripheral zones of the lungs for the HFA product. The majority of beclomethasone from the HFA MDI formulation is emitted as ultrafine particles, which deposit in the alveolar regions whereas most of the inflammation in asthma is in the conducting airway [10]. All the particles emitted from other MDI preparations of beclomethasone (i.e. the CFC formulations) are bigger and should preferentially deposit in the large and conducting airways. This study highlights the dilemma of the regulatory authorities when evaluating the bioequivalence of inhaled products. The regulatory authorities have recently advocated the use of PD₂₀ (the methacholine dose to reduce the FEV₁ by 20%) [11]. However clinical studies have shown that this cannot differentiate between different inhaler techniques [12]. Gamma scintigraphy has also shown how total lung deposition varies for different inhalation techniques [13].

The other methods used to identify deposition, highlighted in Newman’s review [1], are indirect methods of evaluating the total lung dose. These are based on traditional pharmacokinetic methods to demonstrate bioequivalence using either plasma drug concentration [14, 15] or urine excretion [16, 17]. Unlike gamma scintigraphy these identify the effective lung dose and use the product available for the patient. Some of the methods are also able to identify the systemic delivery from the inhaled product [17] and could be used to compare the safety of two inhaled products. These pharmacokinetic methods have been shown to differentiate between the fractions delivered to the body via the pulmonary and gastrointestinal routes and in the future the use of these may increase at the expense of gamma scintigraphy. In combination with in-vitro characterization of the emitted dose and with supporting clinical evidence such approaches are the most likely candidates for the gold standard to identify bioequivalence between inhaled products.

Modifications to the inhaled product are necessary for gamma scintigraphic methods and consequently the product tested is not the one which has the product licence. Thus accurate in vitro analysis is necessary. In vitro studies have shown changes to the aerodynamic particle characteristics of the emitted dose when a radiolabel is incorporated into the formulation [18]. The study used an Andersen Cascade Impactor, which provides extensive characterization of the respirable fraction of an emitted dose. This report is not consistent with the in vitro data of the published gamma scintigraphy papers. The majority of these publications use a five stage Multiple Stage Liquid Impinger (MSLI) which only characterizes the respirable fraction of the dose as the amounts between 6.8 and 3.1, 3.1–1.7 and < 1.7 µm. Other impingers (e.g. Andersen) characterize the respirable fraction in more detail. Also, most of the in vitro validations for these radiolabelled formulations use multiple doses whereas therapeutic doses are inhaled for the gamma scintigraphy studies. Newman et al.[19] describe the use of 40 beclomethasone doses, from a MDI, using a multistage liquid impinger (MSLI) to validate their radiolabelled formulations. El-Araud et al.[20] have shown changes to the aerodynamic particle size characteristics when 2, 5, 10, 20, 30 and 40 doses of beclomethasone from a MDI are measured by the MSLI. Also, papers frequently describe data as a percentage of the nominal dose rather than the emitted dose. The latter is essential because changes to the formulation may affect the amount of dose emitted. Tissue attention standardization [21] and validation is another aspect of gamma scintigraphy that needs to be improved because this factor can alter the results obtained.

Gamma scintigraphy identifies the total dose deposited in the lungs but over-estimates the effective lung dose. Drug delivered to the lungs will be removed by direct absorption into the systemic circulation or by mucociliary clearance. Only the effective lung dose will have a therapeutic effect as it passes through the walls of the airways. This is shown by higher gamma scintigraphy values than those of urinary excretion with charcoal block [16]. Similar results have been reported for sodium cromoglycate in that gamma scintigraphy indicates a total lung deposition of 8.8%[22] while urinary excretion suggests less than 3%[23].

Other aspects to consider are that a gamma scintigraphy study is expensive to commission but the visual images are striking and good for marketing. Recently studies have been commissioned for financial interests rather than scientific merit. Finally and most important, although radiation doses are within safety standards, radioactivity will be concentrated at bifurcations especially in the large airways together with local areas in the oropharyngeal area and in the stomach. The long-term effects of this have yet to be established.

Gamma scintigraphy provided some of the first quantifiable data for the pulmonary deposition of inhalations and highlighted the importance of inhalation techniques and methods. Initial studies revolved round the scientific/clinical merit of the method but further validation has been limited. Gamma scintigraphy has many advantages and provides striking visual images but the problems highlighted above need to be addressed and thus a gold standard cannot be awarded. In-vitro aerodynamic particle size characterization is not affected by patient variables but methods need to be standardized. Recently the Pharmacopeias have started this process of harmonization. Also in vitro/vivo correlations are necessary so that these methods are not regarded as a quality control measure. The potential of pharmacokinetic methods has been demonstrated over the past decade and scientific research to validate these is expanding. Although the proposed pharmacokinetic methods are indirect measurements they are useful for comparing inhaled products, simple to carry out and use the original product intended for patient use. These methods can also evaluate the systemic delivery of the drug. A combination of all these methods will continue to be used to demonstrate that two inhaled products are similar. The goal is to show bioequivalence but for inhaled products this has never been defined. Dr Newman has highlighted the differences between inhaled products and their use in the case studies of his review [1]. Since therapeutic doses are at the top of dose response curves and lung deposition varies between different products, methods and patient use, what is more important? If the emitted dose and the particle size characteristics of two products are similar they will have the same deposition characteristics. Patients can be titrated according to their clinical response so consistency of dose emission is important. The primary reason why the new MDI formulation of beclomethasone using the HFA propellant recommends half the dose of the existing CFC formulation was because of double systemic delivery. This was demonstrated by pharmacokinetic studies [24]. Thus over the next decade these pharmacokinetic methods will be more widely used and should replace gamma scintigraphy as the leading contender for the gold standard.