INTRODUCTION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Keywords: drug aerosol therapy; bronchial asthma; antiasthmatic agents; adrenergic -agonists

Measures of clinical response are relatively insensitive for comparing different inhalation devices. In the case of bronchodilators, the top of a dose-response curve is easily reached with normal doses, such that pulmonary function tests may fail to detect important differences in drug delivery between two formulations (1). Bioequivalence studies with inhaled corticosteroids require complex study designs, since no immediate response occurs, the variability of the response is high, and the clinical endpoints, such as symptom scores, diary records of peak expiratory flow, or use of bronchodilator medication at home remain relatively imprecise (2). Consequently, increasing attention is now directed toward other parameters that could substitute or supplement clinical response data when comparing different inhalation devices or drug formulations (3).

In a previous study involving asthmatic patients, we demonstrated that pulmonary deposition of terbutaline was predictive for its bronchodilating effect. Indeed, the increase in FEV₁ after a nominal dose of 0.25 mg of terbutaline given by Turbuhaler (AstraZeneca, Lund, Sweden) was greater than the same nominal dose of terbutaline inhaled via pressurized metered dose inhaler, pMDI, and gave the same effect as was obtained with 0.5 mg of terbutaline inhaled via pMDI. These findings were explained by the difference in amount of drug reaching the lungs, pulmonary deposition with the Turbuhaler being at least twice that with the corresponding pMDI (4).

The present study was designed to further explore the concept that lung deposition data may act as a parameter for the clinical response to inhaled drugs in asthma patients. More specifically, we investigated: (1) whether existing differences in pulmonary deposition between a pMDI and Turbuhaler might affect the bronchoprotective effect of terbutaline inhaled via these two devices; and (2) whether the bronchoprotective properties of terbutaline against methacholine and histamine were dependent on deposition, device, or both, since the receptors involved in bronchoprotection against these two substances are located at different sites within the lower airways (5).

	METHODS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Thirteen patients (nine male and four female) with stable asthma were recruited. According to the 2001 Global Initiative for Asthma (GINA) Guidelines, nine of the patients had intermittent asthma, two had mild persistent asthma, and two others had moderate persistent asthma. All patients exhibited a provocative concentration of methacholine that decreased FEV₁ by 20% (PC₂₀ methacholine) of 8 ml or less. The study was done with a double dummy, double-blind, randomized, and crossover design. The bronchoprotective effects of the following treatments with terbutaline sulfate were assessed: 0.25 mg administered via Bricanyl (terbutaline sulfate) pMDI, 0.50 mg administered via Bricanyl pMDI, 0.25 mg administered via Bricanyl Turbuhaler, and 0.50 mg administered via Bricanyl Turbuhaler. On eight different study days, the patients were challenged with methacholine or histamine (9, 10) before and 1.5, 3, and 6 h after inhalation of one of the four drug treatments. Pulmonary deposition of each of the treatment drug doses was assessed.

Pretreatment FEV₁ was within ± 15% of each patient's initial FEV₁. Pretreatment PC₂₀ was within ± 1 doubling concentration (DC) of the initial PC₂₀ (11, 12). The washout period between the study days was at least 5 d and at most 21 d.

Before each inhalation, Turbuhaler inhalers were primed and pMDIs were shaken and primed. Patients started inhalation at a lung volume close to RV. The targeted breathholding time was 10 s. The pMDIs and Turbuhaler inhalers were connected in series with a Vitalograph Compact spirometer (Vitalograph Ltd., Buckingham, UK), using specially designed, well-fitting adapters to monitor inhalation flow during the inspiratory maneuver. For the Turbuhaler, a peak inspiratory flow (PIF) of at least 50 L/min was targeted (13). For the pMDI, patients were asked to produce an inspiratory flow of about 90 L/min (4).

Pulmonary deposition of terbutaline was measured with the charcoal-block method (14, 15) and 0.125 mg of deuterated terbutaline hydrochloride was administered intravenously just after the last challenge test. Urine was sampled over a 36-h period in four fractions.

PC₂₀ was determined by linear interpolation of the logarithm of concentration versus effect for histamine and methacholine. If a 20% decrease in FEV₁ was not found at a concentration of 32 mg/ml, PC₂₀ was estimated by linear extrapolation, using the last two values of FEV₁. If the result of this extrapolation gave a PC₂₀ > 64 mg/ml, or if no extrapolation could be done, PC₂₀ was set at 64 mg/ml. The average value, E_av, of PC₂₀ and FEV₁ was computed as area under the curve (AUC) divided by the length of the measurement interval.

For statistical analysis, PC₂₀ at 1.5 h after administration of the study drug, and the average PC₂₀ during the 6-h period after study drug intake (AUC/6), were used. A multiplicative analysis of variance (ANOVA) model with the fixed factors of patient, period, and treatment, and using baseline PC₂₀ as a covariate, was used to compare the treatments. Changes in PC₂₀ were expressed as DCs. The relative dose-potency for the Turbuhaler and pMDI was estimated by fitting parallel dose-response lines to the means from the ANOVA and then estimating the shift of these lines. Confidence limits for the relative dose-potency values were constructed with Fieller's theorem. The pulmonary deposition of study drug was compared for the various treatments by using a multiplicative ANOVA model with the fixed factors of patient, period, and treatment. The pooled estimate from the two doses of study drug within each inhaler device was used when comparing the Turbuhaler and pMDI.

	RESULTS

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Patient Characteristics

The 13 participating asthmatic patients had a mean age of 28 yr (range: 19 to 48 yr). Four of the patients were ex-smokers. All participants completed the study. The mean FEV₁ at the initial visit was 3.42 L (range: 2.63 to 4.64 L) or 85% (75% to 95%) of the predicted value. The mean PC₂₀ for methacholine at the initial visit was 5.0 mg/ml (range: 1.2 to 7.9 mg/ml). None of the patients needed to inhale rescue medication during the study days.

Inhalation Technique

Median PIF was 69 L/min (range: 38 to 97 L/min) for the Turbuhaler and 99 L/min (range: 63 to 227 L/min) for the pMDI. Breathholding time was 10 s for all inhalations, except three.

Completeness of Urine Collections

The fraction of labeled terbutaline excreted by each individual patient, on each of the eight study days, was between 40% and 60% of the dose administered on 91 of the total of 104 study days. For seven patients, however, one of the estimates deviated from the rest, having a much lower value, and for two patients, this was true for more than one of the estimated values. Comparison of the rate of excretion of inhaled versus intravenously administered terbutaline for each patient indicated that only 15 of a total of 416 urine collections were incomplete.

Pulmonary Deposition

The mean pulmonary deposition after inhalation of 0.25 and 0.5 mg of terbutaline was 4.8% (95% confidence interval [CI]: 3.8% to 6.1%) and 7.4% (95% CI: 5.8% to 9.5%) of the nominal dose, respectively, for the pMDI, and 20.8% (95% CI: 16.4% to 26.6%) and 16.9% (95% CI: 13.2% to 21.7%), respectively, for the Turbuhaler. The mean Turbuhaler/pMDI deposition ratio was 3.14 (95% CI: 2.46 to 4.01).

Calculation of PC₂₀

A total of 312 bronchoprovocation tests were performed after inhalation of the study medication. Extrapolation was done in 15 tests. PC₂₀ was set at 64 mg/ml on 32 occasions.

Pretreatment PC₂₀ and Presaline FEV₁

Differences in pretreatment PC₂₀ between the histamine and methacholine components of the study were small; PC₂₀ averaged 4.12 mg/ml on the histamine and 3.71 mg/ml on the methacholine study days.

Mean pretreatment FEV₁ was 3.40 L on the methacholine challenge study days and 3.50 L on the histamine challenge study days. At the first postdrug challenge, mean FEV₁ had increased from 0.13 L to 0.32 L on the methacholine study days and from 0.32 L to 0.40 L on the histamine study days. By the second posttreatment methacholine challenge (after 3 h), and by the third posttreatment histamine challenge (after 6 h), FEV₁ had returned to its value recorded at 1.5 h before treatment.

Bronchoprotection against Histamine

The bronchoprotective activity of the different study drug treatments against histamine, expressed as DCs and 95% CI at 1.5 h and as an integrated value over 6 h, is summarized in Table 1. All treatments caused the PC₂₀ histamine to increase at 1.5 h and 3 h after inhalation (Figure 1), with a return to baseline values after 6 h. The magnitude of this increase was dependent on the inhaled dose of drug and the formulation. Protection by terbutaline delivered via Turbuhaler significantly exceeded the protective effects of terbutaline inhaled via pMDI, as calculated over the 0- to 6-h interval. The relative dose needed to obtain the same protective effect with the pMDI as with the Turbuhaler over the 0- to 6-h period was 2.1 (1.2 to 28) (i.e., significantly larger than 1).

View larger version (22K): [in this window] [in a new window]   Figure 1.   Time course of change in PC20 histamine, expressed as DCs from pretreatment PC20 (time = 0) after inhalation of 0.25 mg and 0.5 mg terbutaline sulfate either via pMDI or Turbuhaler.

Bronchoprotection against Methacholine

All treatments caused an increase in PC₂₀ methacholine at the first posttreatment bronchoprovocation, an effect that decreased after 3 h and had completely disappeared after 6 h (Table 2; Figure 2), the increase depended on the inhaled dose and formulation of the study drug. The mean bronchoprotection over 6 h provided by terbutaline delivered via Turbuhaler exceeded that of terbutaline delivered via pMDI. The relative dose needed to obtain the same protective effect with the pMDI as with the Turbuhaler was numerically larger for methacholine (ratio: 3.2) than for histamine (ratio: 2.1). The difference for methacholine did not reach statistical significance, since no dose response could be shown between the different doses.

View larger version (22K): [in this window] [in a new window]   Figure 2.   Time course of change in PC20 methacholine, expressed as DCs from pretreatment PC20 (time = 0) after inhalation of 0.25 mg and 0.5 mg terbutaline sulfate either via pMDI or Turbuhaler.

Comparison of Bronchoprotective Effects against Histamine and Methacholine

For a given dose and formulation, terbutaline induced a protective effect that was systematically greater for histamine than for methacholine, with the difference reaching statistical significance when averaged over 6 h. Thus, the bronchoprotection against histamine provided by 0.25 mg of terbutaline inhaled via Turbuhaler was 1.34 (range: 1.14 to 1.54) DCs and only 0.85 (range 0.58 to 1.12) DCs against methacholine. A 20% decrease in FEV₁ was not reached on 34 of 156 occasions after inhalation of the highest histamine concentration (20 of which were at 1.5 h and 14 at 3 h after drug inhalation), whereas this was the case in only 14 of 156 tests with methacholine (10 of which were at 1.5 h and 4 at 3 h).

	DISCUSSION

TOP ABSTRACT INTRODUCTION METHODS RESULTS DISCUSSION REFERENCES

Few studies have examined correlations between lung deposition of inhaled medications and clinical response in the same patients. Whole-lung deposition data predicted the bronchodilating effect of inhaled ₂-agonists in two studies (4, 16), and one other study (17) predicted the bronchoprotective action of sodium cromoglycate against allergen challenge. Studies relating pulmonary deposition of ₂-agonists to their bronchoprotective effects have not been performed previously. The present study demonstrates that a threefold greater deposition of terbutaline from the Turbuhaler than from a pMDI was translated into a two- to threefold increase in clinical effect, since the relative doses needed to obtain the same protective effect of terbutaline with the pMDI as opposed to the Turbuhaler were 2.1 and 3.2. These observations support our hypothesis that pulmonary bioavailability is predictive of the overall clinical effect of inhaled drugs in studies done with a correct dosing interval. Thus, pulmonary bioavailability may, in the drug development process, be used as a surrogate for the clinical response to inhaled drugs. This could lead to a significant saving of time in the process of drug evaluation. It should be noted, however, that it is still premature to claim that the pulmonary bioavailability of a drug can substitute for clinical response from a regulatory perspective.

Our findings are not only of relevance for regulatory authorities, who have suggested the use bronchoprovocation studies for equivalence testing (18); they should also prompt clinicians to take the deposition characteristics of formulations into account when switching from one to another delivery device containing the same active drug. Indeed, both the bronchoprotective and the bronchodilating effects (4) of terbutaline are more strongly related to the amount of drug deposited in the lungs than to the nominal metered or delivered dose of the inhaler. The data in the present study, however, may not be extrapolated to other pMDI/dry powder formulation pairs in the absence of lung deposition data (19).

At least twice as much terbutaline was deposited in the lungs of our subjects with the Turbuhaler as with the corresponding pMDI. This difference in pulmonary deposition is in accord with the findings in our previous study (4). In that study, pulmonary deposition of terbutaline in patients with mild obstructive airways disease was 19.0% and 22.0% of the nominal dose with the Turbuhaler and 8.1% and 8.3% with the pMDI. These values are somewhat higher than those reported in the present study. The discrepancy is not explained by the use of a different inspiratory flow or the incompleteness of patients' urine collections, since the inspiratory flow was quite similar to that reported in our previous study, and 96% of the urine collections were complete.

It cannot be excluded that bronchoconstriction induced by methacholine or histamine might have altered the deposition pattern of terbutaline within the airways, since the drug was inhaled 10 to 15 min after the end of the first bronchoprovocation challenge. The more constricted airways seen after challenge would, if anything, have resulted in a more central deposition as compared with that in our previous study (4). Indeed, preferential pulmonary deposition in central airways and less deep penetration in more peripheral airways have been reported in patients with lower FEV₁ (20) and in association with a bronchoprovocation test with methacholine (21). More central deposition would probably increase the portion of drug that is removed from the airways by mucociliary clearance, eventually affecting the measurement of total lung deposition with the pharmacokinetic method (22).

Three of the four treatments with terbutaline caused the PC₂₀ histamine to increase by at least one DC for about 3 h, which is in accord with previous reports (10, 23). However, the magnitude and reproducibility of changes in bronchoprotection after administration of ₂-agonists have not been systematically reported in the literature. For methacholine, there was a sharper drop in PC₂₀ between 1.5 h and 3 h. Although differences in the overall efficacy of treatments with the two inhalers reached statistical significance only after challenge with histamine, the effect of the 0.5 mg pMDI dose of terbutaline was consistently greater than that obtained with the 0.25 mg pMDI dose for both bronchoconstrictive agents. This indicates that the chosen pMDI doses were on the steep part of the dose-response curve, an important issue in comparative studies (24). The dose-response relationship for the bronchoprotective effect of ₂-agonists is characterized by a dose-dependent, steep part at lower doses, followed by a flat part once higher doses of ₂-agonist are inhaled (12, 19). This "ceiling effect" probably explains why the increase in terbutaline dose from 0.25 to 0.5 mg delivered via Turbuhaler did not yield a statistically significant increase in protection. However, it cannot be excluded that a significant difference could have been obtained if patients with more severe asthma and lower baseline PC₂₀ values had been included in the study. In our previous study with asthmatic patients, the difference between the bronchodilating effect of 0.25 and 0.5 mg of terbutaline inhaled via Turbuhaler also failed to reach statistical significance (4), the relationship between bronchodilating effect and inhaled dose of ₂-agonist exhibiting a similar sigmoid dose- response curve to that of the bronchoprotective effect (25).

Histamine and methacholine are both bronchoconstricting agents, but with different mechanisms of action. Methacholine is believed to exert its bronchoconstricting effects via muscarinic receptors located in central airways (5), whereas histamine also acts on the more peripheral airways through the stimulation of H₁ receptors (6) and via an indirect, vagal effect (26). Despite these differences, excellent correlations have been reported between the level of airway responsiveness to histamine and to methacholine under baseline conditions (27). Comparative data on the bronchoprotective effects of ₂-agonists against methacholine and histamine are scarce (31). The present findings do not support the existence of clinically important differences between the pMDI and Turbuhaler in terms of distribution of deposited terbutaline within the airways, since the bronchoprotective effects against histamine and methacholine of 0.5 mg of terbutaline inhaled via pMDI equaled those of 0.25 mg inhaled via Turbuhaler.

It remains unclear why the overall protective effect of terbutaline against histamine was greater than that against methacholine. It is well known that the degree of bronchoconstriction may affect airway responsiveness to some extent (12). In the present study, FEV₁ tended to remain slightly above baseline over the 0- to 6-h study period on the histamine study days, whereas a progressive decrease in FEV₁ was noticed before the third and fourth challenge on the methacholine study days. This is in accord with previous data (32). The currently observed changes in FEV₁ were, however, too small to explain differences over time in airway responsiveness to histamine or methacholine. Indeed, in one study with asthmatic patients (33), changes in FEV₁ of as great as 700 ml did not affect the PC₂₀ for either methacholine or histamine, and in other studies (34, 35), changes in airway geometry accounted only for 30% of the intrasubject variance in airway responsiveness.

Tachyphylaxis to histamine may develop in patients with asthma if successive bronchoprovocation tests are performed within 3 h of one another, an occurrence not observed with acetylcholine or methacholine (36). The absence of placebo study days in our study prevents an estimation of the exact role of tachyphylaxis in the apparently superior protective action of terbutaline against histamine. Conversely, it is unlikely that a period effect could have enhanced the PC₂₀ histamine, since the duration of histamine tachyphylaxis did not exceed 6 h (36) and the washout period between two study days always exceeded 5 d. A distribution of terbutaline in the airways that is more in accord with that of histamine receptors than of muscarinic receptors is the most likely explanation for the better bronchoprotective action of terbutaline against histamine.

On the basis of the present study, it appears that pulmonary deposition data reflect the relative bronchoprotective effects of ₂-agonists. Taken together with previous observations of this as also being the case with the bronchodilator response, our findings provide convincing evidence for the ability of pulmonary drug deposition, measured either with a pharmacokinetic or a scintigraphic method, to predict the overall clinical response to inhaled ₂-agonists. Whether this is also the case in patients with other pulmonary disorders, such as chronic obstructive pulmonary disease, or for other categories of inhaled drugs, such as glucocorticosteroids, remains to be investigated.