The effect of beclomethasone–formoterol versus placebo on chronic cough in patients with non-CF bronchiectasis: the FORZA randomised controlled trial

To the Editor:

Bronchiectasis is a chronic inflammatory condition of the airways with a high symptom burden, which includes chronic cough [1]. The current management strategy includes identifying aetiology, comorbidity and treatable traits. The role of inhaled corticosteroids (ICS) with or without long-acting beta agonists (LABA) in treatment is uncertain. ICS use is only advised for patients with coexisting asthma or COPD [1, 2]. However, up to 53.1% of bronchiectasis patients use ICS or ICS/LABA, with a third having no asthma or COPD diagnosis, according to a recent EMBARC analysis (European Multicentre Bronchiectasis Audit and Research Collaboration) [3]. A study by Martínez-García et al. [4] showed that ICS/LABA can alleviate cough in bronchiectasis patients, compared to ICS alone. In our multicentre, randomised, double-blind trial comparing beclomethasone–formoterol versus placebo, we hypothesised that ICS/LABA could reduce cough complaints in bronchiectasis patients without asthma or COPD.

The FORZA study was conducted in three Dutch hospitals: Erasmus MC, Haga Teaching Hospital, and the Franciscus Gasthuis and Vlietland Hospital. We recruited adult patients with bronchiectasis confirmed by computed tomography according to the Fleischner Society criteria and with compatible symptoms according to British Thoracic Society definition [1, 5]. Patients were on a stable treatment regimen without recent exacerbation (≥6 weeks) and had daily cough for ≥8 weeks. The main exclusion criterion was a diagnosis of asthma confirmed by spirometry reversibility and/or bronchial provocation testing, or COPD, according to Global Initiative for Asthma and Global Initiative for Chronic Obstructive Lung Disease guidelines [6, 7], along with a history of ≥10 pack-years or current smoking.

Participants were randomised to use beclomethasone–formoterol (Fostair) 200/12 μg or placebo twice daily via identical metered-dose aerosol inhalers. The primary endpoint was the change in cough-related quality of life after 3 months, measured by the Leicester Cough Questionnaire (LCQ), using ≥1.3 points as the minimum clinically important difference [8]. The study aimed to enrol 72 participants, accounting for dropouts. The minimum sample size was 66 participants to provide 80% power to detect a mean 2.7-point LCQ difference between groups with a 5% level of significance, which was based on a previous study by Chamberlain Mitchell et al. [9].

Furthermore, we assessed the mean change in forced expiratory volume in 1 s (FEV₁) % predicted, modified Medical Research Council (mMRC) Dyspnoea Scale score, Quality of Life-Bronchiectasis (QoL-B) respiratory domain score [10], pulmonary exacerbations (defined as worsening of ≥1 symptom [1] and prescription of systemic antibiotics) and adverse events.

We used the one-way ANCOVA to assess mean differences from baseline to 3 months between the beclomethasone–formoterol and placebo groups, which results in coefficients for treatment and the measurement at baseline. The treatment coefficient (β) is interpreted as the mean difference between the two groups at 3 months, corrected for values at baseline. Additionally, the observed mean differences between the two groups at 3 months are reported. The ethics review committee approved the study (ethics approval number NL61630.078.18; clinicaltrials.gov registration NCT03846570).

In total, 729 patients with diagnostic codes for bronchiectasis were reviewed for eligibility. The majority was ineligible because of a co-diagnosis of asthma (28.1%), COPD or excessive smoking history (11.2%) or current ICS use without definitive asthma or COPD (14.4%). The majority of patients already on ICS-containing treatment were unwilling to discontinue ICS to participate in the study. Other reasons for exclusion were those who did not meet the diagnostic definition due to lack of coughing or no radiologic evidence of bronchiectasis (10.0%), unwillingness or inability (e.g. language barrier) (19.2%), or comorbidity or logistical reasons which precluded participation (11.0%), especially during the SARS-CoV-2 pandemic. In total, 44 patients were screened for participation. In 10 of these patients, a positive bronchodilator reversibility (n=4), bronchial provocation test (n=2) or increase in symptoms at discontinuation of ICS (n=4) was noted during screening. The study was terminated prematurely due to insufficient inclusion.

34 patients enrolled between January 2019 and April 2022. 17 participants received beclomethasone–formoterol, and 17 placebo. Mean±sd age was 53.9±18.3 years with 64.7% women. Mean±sd FEV₁ was 82.9±15.8% pred, body mass index 23.9±4.2 kg·m⁻², and former smoking history 3.8±3.3 pack-years. Five participants had a Pseudomonas aeruginosa infection. Mean±sd LCQ score was 15.7±3.1, QoL-B respiratory score was 67.2±13.3 and mMRC score 1.3±0.9. The mean number of pulmonary exacerbations in the previous year was 1.1±1.3. FACED bronchiectasis severity score was 1.2±1.2 and Bronchiectasis Severity Index score 4.3±3.5. No significant differences were observed at baseline between groups (table 1).

Due to the SARS-CoV-2 pandemic, some participants were unable or unwilling to comply with the complete study protocol. Ultimately, the primary outcome (LCQ) at 3 months was determined for 16 (placebo) versus 11 (intervention) participants, after further drop-outs due to adverse events.

At 3 months, neither group showed a significant difference in mean LCQ scores, with the placebo group showing a mean increase from baseline of 0.6 (95% CI −0.4 to +1.6) points and beclomethasone–formoterol group 0.3 (95% CI −1.3 to +1.8) points (treatment coefficient β=−0.312, p=0.708).

Furthermore, no effect of treatment was found on the QoL-B respiratory scores (placebo −0.5 (95% CI −8.2 to +7.1) versus intervention 3.2 (95% CI −9.6 to +15.9) points; treatment coefficient β=3.905, p=0.538), mMRC scores (placebo 0.0 (95% CI −0.6 to +0.6) versus intervention 0.0 (95% CI −0.3 to +0.3) points; treatment coefficient β=−0.011, p=0.978) and FEV₁ (placebo −0.8% (95% CI −3.6 to +2.0%) versus intervention 1.1% (95% CI −1.5 to +3.7%) of predicted; treatment coefficient β=2.175, p=0.264) (table 1).

Pulmonary exacerbations occurred in 4/17 (23.5%) participants in the beclomethasone–formoterol group versus 1/17 (5.8%) in the placebo group. In one participant in the beclomethasone–formoterol group with a pulmonary exacerbation, a new Stenotrophomonas maltophilia culture was obtained; in the other participants no new pathogen was detected. The number of participants with adverse events (including pulmonary exacerbations) in the beclomethasone–formoterol group was 11/17 (64.7%) versus 5/17 (29.4%) in the placebo group (chi-squared test; df=1, n=34, chi-squared=4.25, p=0.039). Oropharyngeal symptoms were the most frequent adverse events. Four participants (23.5%) in the beclomethasone–formoterol group discontinued the study medication due to adverse events, versus one (5.8%) in the placebo group.

To date, this is the only randomised, double-blind, placebo-controlled trial to investigate ICS/LABA in bronchiectasis patients with strict exclusion of patients with asthma or COPD. No treatment effect on cough, nor change in quality of life, dyspnoea, lung function or pulmonary exacerbations was observed. However, more adverse events occurred in the ICS/LABA group.

Only one other trial has investigated ICS/LABA in bronchiectasis. In a double-blind comparison of high-dose budesonide versus medium-dose budesonide–formoterol, the authors reported symptomatic benefits of ICS/LABA [4]. However, there was no comparison with placebo. A Cochrane review from 2018 showed insufficient evidence to support the use of ICS in steady-state bronchiectasis [11]. In the early presented data of an upcoming EMBARC publication, ICS use is associated with an increased exacerbation and hospitalisation risk [3]. Another prospective, but non-randomised, study of 264 patients found an association of ICS use with increased morbidity as well as all-cause mortality, while adjusting for age, sex, FEV₁ and concomitant asthma/COPD [12]. Possibly, an eosinophilic phenotype can be an indication of ICS benefit in bronchiectasis [13, 14], but prospective controlled studies are still lacking.

Bronchiectasis can be associated with nonspecific bronchial hyperreactivity [15]. Remarkably, we found that about 22.7% of the screened participants were ineligible due to new-found significant reversibility, positive bronchial provocation test or increase in symptoms after discontinuing ICS. Therefore, we recommend to assess bronchodilator responsiveness and/or bronchial hyperreactivity in all bronchiectasis patients when considering ICS and/or bronchodilators.

To conclude, the FORZA study was underpowered to give definitive results, but ICS/LABA use did not decrease cough or provide other benefits in bronchiectasis patients without asthma or COPD. Due to the occurrence of significantly more adverse events, caution should be exercised when prescribing ICS/LABA for these patients.

Shareable PDF