Published ahead of print on November 27, 2002, doi:10.1164/rccm.200209-1116OC
© 2003 American Thoracic Society Effects of Montelukast on Surrogate Inflammatory Markers in Corticosteroid-treated Patients with AsthmaAsthma and Allergy Research Group, Ninewells Hospital and Medical School, University of Dundee, Dundee, Scotland, United Kingdom Correspondence and requests for reprints should be addressed to Dr. Brian J. Lipworth, M.B.Ch.B., Asthma and Allergy Research Group, Ninewells Hospital and Medical School, University of Dundee, Dundee DD1 9SY, Scotland, UK. E-mail: b.j.lipworth{at}dundee.ac.uk
We evaluated whether montelukast conferred additive effects in patients with asthma receiving fluticasone/salmeterol (FP/SM) combination and FP alone. Twenty-two patients with mild to moderate asthma completed a double-blind, placebo-controlled study. After a 2-week run-in using FP 250 µg/SM 50 µg 1 puff twice daily, patients entered a randomized crossover period to receive additional montelukast 10 mg daily or placebo for 3 weeks each. For the first 2 weeks, they received FP/SM 1 puff BID, and then they received FP 250 µg 1 puff BID for the 3rd week. The primary outcome was adenosine monophosphate challenge threshold and recovery time; secondary outcomes included surrogate inflammatory markers and lung function. Compared with FP/SM run-in, adding montelukast to FP/SM was better (p < 0.05) than placebo for inflammatory markers but not for lung function. For adenosine monophosphate threshold, recovery, exhaled nitric oxide, and blood eosinophils, there were 1.4 (95% confidence interval, 1.1–1.8) geometric mean fold, 10 minutes (3–17 minutes), 2.1 parts per billion (0.2–3.9 parts per billion), and 88 (34–172) x 106/L differences, respectively. The combination of FP plus montelukast was superior to FP/SM for inflammatory markers but was inferior for lung function. Thus, in patients taking FP/SM or FP, montelukast conferred complimentary effects on surrogate inflammatory markers, which were dissociated from lung function. Further studies are required to evaluate whether these effects of montelukast translate into clinical benefits.
Key Words: asthma therapy • leukotriene receptor antagonist • salmeterol • fluticasone Corticosteroid responsive inflammation is the pathophysiologic hallmark of asthma. As a consequence, inhaled corticosteroids are a first-line therapy in the control of symptoms and the prevention of long-term airway remodeling (1, 2). A further component causing symptoms in asthma is bronchial smooth muscle dysfunction, which is responsive to ß2-agonist therapy. This rationale has led to increased use of inhalers using fixed combinations of corticosteroids and long-acting ß2-agonists (LABAs) (3). Combination inhalers are particularly effective on outcomes such as symptoms and lung function, which are commonly used to monitor response (4). The cysteinyl leukotrienes (C4, D4, and E4) are important proinflammatory mediators whose actions include smooth muscle contraction and proliferation, increased vascular permeability, inflammatory cell recruitment, and mucous hypersecretion. Despite treatment with inhaled corticosteroids, suppression of inflammation is often incomplete, and their effect on cysteinyl leukotriene biosynthesis is limited (5–7). Adding on therapy with a leukotriene receptor antagonist (LTRA) would therefore theoretically appear to provide a more complete and complimentary role in the suppression of airway inflammation. Various studies demonstrate LTRAs to reduce rescue treatment requirement, improve pulmonary function, and reduce symptoms (8). This additive benefit is also evident in patients receiving moderate-to-high doses of inhaled corticosteroids (9, 10). However, their exact role in the management of asthma remains to be fully established (11). For example, whether LTRAs confer additional benefit in patients who remain symptomatic despite inhaled corticosteroids plus LABAs is uncertain. In a study evaluating 72 patients with moderate to severe asthma maintained on inhaled corticosteroids and mostly taking LABAs, the addition of montelukast 10 mg daily for 14 days conferred no significant improvement in terms of peak expiratory flow and symptom scores (12); however, a limitation of this study was failure to evaluate any surrogate antiinflammatory markers (13). The purpose of this study was to evaluate the effects of add-on therapy with montelukast in patients with mild to moderate persistent asthma treated with an optimal dose of inhaled corticosteroid alone or in combination with an LABA. The primary outcome was airway hyperresponsiveness to adenosine monophosphate (AMP) and time taken to recover spontaneously after bronchial challenge. Secondary outcomes included measurements of other surrogate inflammatory markers and lung function.
Patients and Design Patients with mild-to-moderate persistent asthma were enrolled into a randomized, double-blind, placebo-controlled, crossover trial (Figure 1) .
Patients attended the department on six occasions. The first occasion (visit 1) served as screening to determine eligibility. Inclusion criteria were an FEV1 of more than 60% predicted, a provocative concentration of AMP causing a 20% fall in FEV1 (PC20) of 200 mg/ml or less, daily asthma symptoms requiring reliever therapy on at least 5 days a week, and the use of inhaled corticosteroids alone (up to 2,000 µg/day of beclomethasone dipropionate or equivalent) or in combination with a nonsteroidal second-line agent. Tayside committee on medical research ethics approved the study, and all patients gave written consent. Patients entered an initial 2-week run-in in which their asthma medication was stopped, and the patients then commenced on fluticasone (FP) 250 µg/salmeterol (SM) 50-µg combination 1 puff twice daily (Advair Diskus, GlaxoSmithKline, Uxbridge, UK). After run-in (at visit 2), they entered the randomized crossover period and received encapsulated montelukast 10 mg daily or identical placebo at 10 P.M. for 3 weeks each. For the first 2 weeks of each randomized treatment, patients used FP/SM, after which they switched to FP propionate 250 µg 1 puff twice daily (Flixotide Diskus; GlaxoSmithKline) for the final week. Measurements were performed after FP/SM run-in (visit 2) and after FP/SM (visits 3 and 5) and FP (visits 4 and 6) treatments in both randomized limbs (Figure 1). An albuterol Diskus was supplied for the duration of the trial for "as required" use. All visits to the department took place within a 2-hour window for each patient, 14 hours after taking their last 10 P.M. study tablet and dose of inhaler.
Measurements
AMP Challenge and Recovery
Statistical Analysis
Twenty-nine patients were enrolled, of whom 22 completed the study per protocol. Demographic data at initial screening (visit 1) for patients completing the study per protocol are given in Table 1 . Blood eosinophils were available for 21 patients, as one declined venepuncture on all visits. Reasons for dropout included becoming AMP unresponsive after the initial FP/SM run-in (n = 5) (and therefore not randomized), personal reasons during the FP/SM run-in (n = 1), and asthma exacerbation while on placebo (n = 1). These seven patients were not included in the analysis, as they did not complete per protocol.
Absolute data showed no significant differences comparing before versus after run-in values, apart from a significant (p < 0.05) fall in NO and rise in AMP PC20 threshold (Table 1). For all data, responses to randomized treatments are given as a change with reference to the values after the initial 2-week run-in using FP/SM. There were no significant differences comparing the response to randomized treatments for the primary outcome according to whether they were given first or second in sequence (Table 2) . There were also no significant differences on the primary outcome (AMP PC20 threshold and recovery) comparing nonrandomized run-in on FP/SM versus randomized FP/SM plus placebo (Table 3) .
AMP PC20 Threshold Adding montelukast but not placebo to FP/SM or FP alone conferred significant improvements in the AMP PC20 threshold compared with the run-in value on FP/SM (Table 3). As change from run-in, FP/SM plus montelukast versus FP/SM plus placebo, conferred a 1.4 (95% confidence interval [CI], 1.1–1.8) geometric mean fold improvement (p < 0.05) in AMP PC20 threshold, whereas FP plus montelukast versus FP plus placebo conferred a 1.7-fold (95% CI, 1.3–2.2) improvement (p < 0.05). Comparing FP plus montelukast to FP/SM plus placebo, there was a 1.5-fold (95% CI, 1.1–2.1) improvement (p < 0.05). FP/SM plus montelukast versus FP plus montelukast and FP/SM plus placebo versus FP plus placebo demonstrated no significant differences (Figure 2A) .
After AMP challenge, the mean falls in FEV1 were 25%, 24%, 27%, and 25% for FP/SM plus montelukast, FP plus montelukast, FP/SM plus placebo, and FP plus placebo, respectively, with no significant differences between values.
AMP Challenge Recovery Time Figure 3 illustrates the individual scatter in recovery times as change from FP/SM run-in.
Exhaled NO As change from run-in, FP/SM plus montelukast versus FP/SM plus placebo conferred a significant reduction in exhaled NO amounting to a 2.1 (95% CI, 0.2–3.9) parts per billion difference. FP plus montelukast compared with FP/SM plus placebo also caused a reduction amounting to 2.5 parts per billion (95% CI, 0.1–4.9) (Figure 4A) .
Eosinophils As change from run-in, FP/SM plus montelukast and FP plus montelukast conferred a significant reduction in blood eosinophils versus FP/SM plus placebo, amounting to a 88 x 106/L (95% CI, 34–172) and 100 x 106/L (95% CI, 8–199) change, respectively (Figure 4B).
Lung Function and Diary Cards
After run-in, symptoms were minimal; however, patients had fewer symptoms (p < 0.05) with FP plus montelukast versus FP plus placebo and also with FP/SM plus placebo versus FP plus placebo. However, no difference was observed with FP/SM plus montelukast versus FP/SM plus placebo. In terms of reliever use, no differences were observed between treatments (data not shown).
Our study has demonstrated that in patients with mild-to-moderate persistent asthma, the addition of montelukast conferred additional effects on surrogate inflammatory markers in patients taking FP/SM combination. This was evident on AMP PC20 threshold and recovery, exhaled NO, and eosinophils but not on lung function. In patients taking FP alone, the addition of montelukast versus placebo conferred improvements in AMP threshold and recovery but not in lung function. For inflammatory markers, adding montelukast to FP/SM was no different from adding montelukast to FP alone apart from recovery, which was quicker with the latter. Moreover, AMP threshold and recovery with FP plus montelukast was superior compared with FP/SM alone. Adding in montelukast but not placebo to FP/SM or FP alone conferred significant improvements in AMP threshold and recovery compared with the run-in value on FP/SM. For lung function, FP/SM either alone or in combination with montelukast was better than FP plus montelukast. Thus, patients taking FP alone or FP/SM combination exhibited further improvements in terms of surrogate inflammatory markers by taking montelukast. The dissociation in effects of montelukast on inflammatory markers and lung function suggests that changes in lung function are relatively distant from the inflammatory process. Moreover, airway inflammation often persists despite normal or near normal lung function (17, 18). Thus, serial monitoring of airway caliber allows no evaluation of the potential benefits of nonsteroidal antiinflammatory therapy. Our results are therefore in agreement with those of Robinson and colleagues by demonstrating that montelukast, when added to inhaled corticosteroids and LABAs, conferred no additional bronchodilator benefit (12). Our patients would have been near maximally bronchodilated on FP/SM; hence, no room for improvement in measures of caliber could reasonably be expected by adding in montelukast. Furthermore, during our trial, patients had minimal symptoms and reliever use, which probably reflected the prior use of combination inhaler therapy during the initial run-in period. It is also pertinent to consider that lung function and symptoms were secondary outcome variables and that patients were recruited on the basis of exhibiting marked AHR to AMP. The problem for everyday clinical practice is that there are no readily available reliable surrogate inflammatory markers, which may be used to optimize antiinflammatory therapy in asthma. For instance, one may have to wait several years to observe a potential benefit in terms of reducing exacerbations in an individual patient. A meta-analysis showed a significant overall benefit on exacerbations requiring systemic steroids (a 48% reduction) conferred by adding in a LTRA, although the data were limited by a small number of studies and a relatively short period of follow-up (19). We felt that the run-in on FP/SM was important to establish true baseline values in which to compare subsequently all randomized treatment responses. Indeed, no significant differences between the nonrandomized FP/SM run-in values compared with randomized FP/SM plus placebo values were observed in the primary outcomes (Table 3). This indicates that patients had been adequately stabilized during the 2-week run-in. We did find a significant reduction in exhaled NO values and improvement in AMP threshold comparing before versus after run-in. This may have reflected improved compliance with inhaled corticosteroids during the run-in. Furthermore, a higher effective inhaled corticosteroid dose was given during this period, as many of the symptomatic patients were previously enrolled on doses of 800 µg of beclomethasone equivalent or less (i.e., less than the equivalent dose of 500 µg of FP during the FP/SM run-in). Despite this, we were still able to show a significant fall in NO with the addition of montelukast to FP/SM, although in isolation this would unlikely be clinically relevant in long-term management. Our choice of primary outcome (AMP threshold and recovery) is of particular importance. Airway hyperresponsiveness to various stimuli is an integral component in asthma of all severities, whereas AMP acts indirectly via degranulation of primed mast cells causing the release of inflammatory mediators (20). Sont and colleagues highlighted the usefulness of monitoring the degree of airway hyperresponsiveness to methacholine as a tool in which to guide treatment (21). In their parallel group study involving 75 patients with asthma, it was demonstrated that titrating inhaled corticosteroid dose according to airway hyperresponsiveness plus symptoms and spirometry resulted in superior control than if the latter alone were used. The AMP PC20 is closely correlated to the degree of sputum eosinophilia and a more sensitive indicator of allergic airway inflammation than methacholine (22, 23); it may also be of greater clinical value in assessing the effects of nonsteroidal antiinflammatory treatment with montelukast. It is worth pointing out that the time taken to recover after AMP challenge is determined by the threshold concentration given, indicating that a higher PC20 would consequently be expected to result in a longer recovery time. Despite this, our data clearly show that the addition of montelukast to FP/SM and FP alone results in the recovery being shortened rather than prolonged, despite producing a higher PC20 threshold value. Moreover, we observed changes in AMP threshold and recovery despite no significant changes in FEV1, suggesting that the effects were independent of airway geometry and reflected genuine antiinflammatory activity. It is also relevant to note that we found no differences in the maximum fall in FEV1 after AMP challenge, despite the improvements in recovery time. Our data demonstrated that montelukast plus FP/SM versus montelukast plus FP showed no difference in terms of surrogate inflammatory markers, but as expected, the former treatment produced significant improvements in lung function due to the smooth muscle effects of SM. It would therefore appear reasonable that in terms of optimizing antiinflammatory activity, inhaled corticosteroids plus a LTRA would provide greatest effects. Additive effects on methacholine hyperreactivity and exhaled NO were observed in a previous study using a LTRA, where the combination of zafirlukast plus low-dose inhaled corticosteroid was comparable to medium-dose inhaled corticosteroid alone (24). In a large multicenter study in patients with moderate to severe asthma suboptimally controlled on budesonide 800 µg/day, patients were randomized to receive budesonide 1,600 µg/day or budesonide 800 µg/day plus montelukast over 3 months (25). Both treatments showed comparable improvements in terms of lung function, a reduction in blood eosinophils and exacerbations, probably due to additive nonsteroidal antiinflammatory activity of montelukast. It does have to be borne in mind that although short-term improvements in noninvasive surrogate markers such as airway hyperresponsiveness to AMP exhaled NO and blood eosinophils are often observed with montelukast, whether they translate into commensurate improvements in exacerbations requires more definitive long-term evaluation. Triple therapy with inhaled corticosteroid/LABA combination plus montelukast could be considered beneficial in patients in whom bronchodilation and antiinflammatory activity are both important, or in other words in patients with more severe asthma in which airway caliber is compromised. In a recent study, Fowler and colleagues initially optimized asthma control with high-dose inhaled corticosteroids (26). Subsequent stepping down to use low-dose FP/SM, rather than using a medium dose of inhaled corticosteroid alone, resulted in superior lung function and quality of life, despite worse antiinflammatory activity. However, in Fowler and colleagues' study, triple therapy with FP/SM plus montelukast was not evaluated. Nonetheless, triple therapy would be prohibitively more expensive for routine use in primary or secondary care. In this study, we evaluated add on effects of montelukast in patients using 500 µg/day of FP to ensure that the top of the dose–response curve for antiinflammatory activity was achieved. For example, in the study by Meijer and colleagues (27), no significant additional antiinflammatory effects was gained when 500 µg/day was compared with 2,000 µg/day of FP dry powder on evaluation of different surrogate inflammatory markers, including AMP PC20, methacholine PC20, sputum eosinophils, and eosinophil cationic protein. Because we had achieved the plateau for inhaled corticosteroid effects, our present data suggest that there is a further nonsteroidal inflammatory component to asthma not controlled on FP/SM, which can be suppressed with montelukast. The fact that there is a leukotriene-mediated component to asthma is also reflected in previous studies using high-dose inhaled or oral corticosteroid, which have shown little effect on leukotriene synthesis after allergen challenge (28, 29). Furthermore, in a cross-sectional study despite high-dose inhaled corticosteroids, sputum levels of cysteinyl leukotrienes remained elevated compared with control subjects (7). Our results demonstrate that SM confers no discernable antiinflammatory properties in patients receiving inhaled corticosteroids, as FP plus SM was no different from FP alone on all surrogate inflammatory markers. This is similar to other studies, which also suggest that LABAs do not exert meaningful antiinflammatory activity in vivo (30–33). Moreover, we observed no potentiation of antiinflammatory effects when comparing FP/SM versus FP alone. This in turn suggests that in vitro data showing synergy between inhaled corticosteroids and LABAs on nuclear glucocorticoid receptor translocation and smooth muscle proliferation may not be clinically relevant (34, 35). In conclusion, in patients with mild-to-moderate persistent asthma taking FP/SM or FP alone, adding montelukast conferred complimentary activity on surrogate inflammatory markers, which was dissociated from effects on airway caliber. Thus, monitoring lung function alone may miss potentially beneficial antiinflammatory effects of adding in a LTRA. Further long-term studies are indicated to assess whether these effects translate into clinically meaningful reductions in exacerbations.
Supported by the University of Dundee departmental anonymous grant. Received in original form September 30, 2002; accepted in final form November 25, 2002
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
METHODS
error. 
