Randomised controlled trial of azithromycin in smokers with asthma

To the Editor:

Smokers with asthma have poor symptom control, accelerated decline in lung function and an attenuated response to corticosteroids compared to nonsmokers with asthma [1]. There is an unmet need for alternative or additional drugs for smokers with asthma who are unable to stop smoking [2]. Macrolide antibiotics have anti-inflammatory activity [3] and in clinical studies there is good evidence for efficacy in the treatment of diffuse pan-bronchiolitis and cystic fibrosis, as well as in preventing chronic rejection after lung transplantation [4, 5]. In asthma, chronic treatment is associated with a reduction in bronchial hyperreactivity in mild-to-moderate asthma [6] and in exacerbation rates in non-eosinophilic severe asthma [7]. To date, no studies have examined the efficacy of macrolide antibiotics exclusively in current smokers with asthma.

A randomised double-blind parallel-group trial compared azithromycin, 250 mg per day, with placebo for 12 weeks. All subjects were aged 18–70 years, were current smokers (≥5 pack-years history) with chronic asthma (>1 year duration; defined by international criteria [8]) and had to be free of exacerbation and respiratory tract infection for a minimum 6-week period prior to randomisation. A baseline visit was performed following a 4-week run-in period on inhaled corticosteroid (ICS) therapy equivalent to 400 μg beclometasone ± a long-acting β₂-agonist (LABA). Ethical approval was obtained and all subjects provided written informed consent. Study visits were performed at 4, 8 and 12 weeks. Clinic visit peak expiratory flow (PEF) after 12 weeks treatment was the primary outcome measure. A sample size of 68 was calculated to have an 80% power to detect a mean difference of 25 L·min⁻¹ in change from baseline to 12 weeks in morning PEF, the primary end-point [9], assuming a standard deviation of changes of 36 L·min⁻¹ using a two-sample t-test with a 5% two-sided significance level. Recruitment of 80 patients was planned to ensure that 68 patients completed the study. 77 subjects were randomised with 71 completing the study. Other measures of airway responsiveness PC₂₀ (provocation concentration of methacholine causing a 20% fall in forced expiratory volume in 1 s (FEV₁)), inflammation (exhaled nitric oxide fraction at 50 mL·s⁻¹ (F_eNO50)/induced sputum/blood and sputum supernatant biomarkers), symptom control (Asthma Control Questionnaire (ACQ) [10] and Leicester Cough Questionnaire (LCQ) [11]) and quality of life (Asthma Quality of Life Questionnaire (AQLQ) [10]) were assessed during the study. PEF was recorded using Piko-1 electronic peak flow meters (nSpire, Hertford, UK) and symptoms were recorded in a validated diary card [10]. Bacteriological and virological analysis of induced sputum was also undertaken. All statistical tests were two-sided and used a significance level of 5%. All data was analysed using SAS (version 9.2; SAS Institute, Cary, NC, USA). QTc was also measured at baseline and 12 weeks.

Baseline demographic data and clinical characteristics of the patients were comparable and the two groups were well balanced. Placebo versus azithromycin groups: age 42.8±9.4 years versus 46.4±8.8 years; male sex 17 (44.7%) versus 20 (51.3%); smoking history 23.6±15.8 pack-years versus 28.6±16.4 pack-years; duration of asthma 24.6±12.6 years versus 18.8±12.5 years; atopic 23 (60.1%) versus 27 (69.2%); median (interquartile range) total IgE 103 (38–291) IU·mL⁻¹ versus 165 (48–254) IU·mL⁻¹; use of ICS at screening 31 (81.6%) versus 35 (89.7%); equivalent beclomethasone dose at screening 709±564 μg versus 603±457 μg; use of LABA at randomisation 18 (47.4%) versus 15 (38.5%); pre-bronchodilator FEV₁ 81.0±16.8% predicted versus 78.3±16.4% pred; post-bronchodilator FEV₁ 89.0±15.1% pred versus 86.8±15.2% pred; FEV₁ 11.3±9.8% reversibility versus 12.3±10% reversibility; geometric mean PC₂₀ 1.06±4.10 mg·mL⁻¹ versus 1.07±3.13 mg·mL⁻¹.

At the final study visit (12 weeks) the change in mean morning clinic PEF (primary outcome), as compared with baseline, did not differ substantially between the azithromycin and placebo treatment groups (mean difference -10.3 (95% CI 47.1–26.4 L·min⁻¹), p=0.58) (table 1). There was no difference in either pre- or post-albuterol FEV₁ at 4, 8 or 12 weeks between the two groups (table 1). No differences were evident for PC₂₀ (baseline to 12 week comparison) between the azithromycin or placebo groups (table 1). None of the self-reported diary card recordings (ACQ, AQLQ or LCQ score) demonstrated any significant differences between the two groups at 4, 8 or 12 weeks. Noninvasive measures of inflammation, i.e. induced sputum, sputum supernatant cytokines, peripheral blood cytokines and F_eNO50, did not demonstrate any substantial improvements after 12-weeks of treatment with azithromycin (table 1). Bacterial colony counts did not demonstrate any treatment difference between the placebo and azithromycin groups (p=0.66, data not shown). PCR for Mycoplasma pneumonia and Chlamydophila pneumoniae were negative at both baseline and 12 weeks. QTc was unchanged following 12 weeks of treatment with azithromycin.

No suspected unexpected serious adverse events occurred during the reporting period of the study. Compliance assessed by capsule count was >90% in each group.

In the first randomised controlled study of azithromycin in smokers with asthma we found that there were no clinically important improvements in both the primary end-point and morning PEF, and in a range of secondary clinical outcomes including ACQ score, AQLQ score, spirometry and airway responsiveness, as well as measures of airway inflammation after 12 weeks of treatment. The median ACQ score of the participants recruited to the study was raised at 1.7 indicating that they had poorly controlled disease and scope for clinical improvement. We believe that the choice of a different macrolide from azithromycin or a different dose of azithromycin is unlikely to have altered our findings. This study was powered to test the hypothesis on the primary end-point, PEF. We exceeded our minimum recruitment target ensuring adequate numbers completed the study. The lack of response to treatment was such that recruiting greater numbers would be unlikely to affect either the primary end-point or the majority of secondary analyses. Taken together these findings indicate that short-term therapy with azithromycin does not improve lung function or other indices of current asthma control of smokers with mild-to-moderate asthma who are already receiving treatment with inhaled corticosteroids. This is supported by a recent study where 6 months of treatment with azithromycin in nonsmokers with severe asthma did not improve lung function [7].

Recently published data reported that azithromycin administered over 1 year reduced the rate of exacerbations in chronic obstructive pulmonary disease (COPD), although this benefit was not found in current smokers with COPD, and bronchiectasis was not an exclusion [12, 13]. Moreover, a 6-month study with azithromycin in nonsmokers with severe asthma powered to measure exacerbation frequency did not find any statistical difference between treatment and placebo groups [7], although a beneficial effect of azithromycin on reducing exacerbations was reported in patients with non-eosinophilic asthma [7]. The smokers with asthma in our study had non-eosinophilic asthma, but did not respond to azithromycin, although the duration of treatment was not long enough to assess the effects of azithromycin on exacerbations.

In conclusion, we have studied the effects of 12 weeks of azithromycin in smokers with asthma, an understudied patient group, and provide clear evidence demonstrating lack of efficacy in both clinical and laboratory outcomes. Further randomised clinical trials exploring new therapies in smokers with asthma who are unable to stop smoking are required.

• Received June 3, 2013.
• Accepted July 10, 2013.

• ©ERS 2013