Long-term (5 year) safety of bronchial thermoplasty: Asthma Intervention Research (AIR) trial

Asthma continues to be a major health concern worldwide, with over 23 million people in the United States who suffer with this disease [1]. Approximately 5-10% of these patients are characterized as having severe persistent asthma based on continued presence of asthma symptoms despite treatment with current state-of-the-art medications [2]. Poorly controlled asthma impacts the patient's quality of life, increases healthcare utilization, and imposes both a social as well as an economic burden [3‐8].

The recent approval of the Alair^® Bronchial Thermoplasty System for delivering bronchial thermoplasty (BT) provides an additional option for managing patients with severe asthma. BT provides therapeutic benefit by reducing the amount of excess smooth muscle in the airways, with the resultant effect of reducing bronchoconstriction in response to asthma triggers. Results from the Asthma Intervention Research (AIR) Trial, the first randomized clinical trial of BT which compared BT plus standard-of-care therapy (inhaled corticosteroids (ICS) and long-acting-β₂-agonists (LABA)) to standard-of-care alone, demonstrated that the mean rate of mild exacerbations, as compared with baseline, was reduced in the BT group but was unchanged in the control group. Furthermore at 12 months, when subjects were on ICS alone, there were significantly greater improvements in the BT group than in the control group in the morning peak expiratory flow, scores on the AQLQ and ACQ, the percentage of symptom-free days and symptom scores, while fewer puffs of rescue medication were required [9]. Two additional randomized controlled trials, the Research in Severe Asthma (RISA) Trial [10], and the sham-controlled Asthma Intervention Research 2 (AIR2) Trial [11], have provided additional support for the effectiveness of BT. The safety of BT over the post-treatment period out to one year was established in all 3 randomized clinical studies. Because BT is a novel treatment that alters the amount of airway smooth muscle, and asthma is a chronic disease that is associated with structural changes in the airway wall, it is important to know whether this treatment is associated with any longer term adverse outcomes. Longer-term safety data out to 5 years are available for a small cohort of BT treated subjects from the first clinical study of patients with mild to moderate asthma [12, 13]. We now describe the safety profile of BT out to 5 years post-treatment from patients with moderate to severe asthma from the AIR Trial [9].

The recently approved Alair System for bronchial thermoplasty provides a new treatment option for patients with severe asthma that remain symptomatic despite taking inhaled corticosteroids and long-acting-β₂-agonists, the current standard-of-care medications. Three randomized clinical trials in patients with differing severity of asthma have demonstrated the safety of this device-based treatment out to one year [9‐11]. In the AIR Trial, at 12 months post-treatment, while patients were on ICS alone, these benefits included significantly greater improvements in the BT group than in the Control group in mild exacerbation rates, morning peak expiratory flow, scores on the AQLQ and ACQ, the percentage of symptom-free days and symptom scores, while fewer puffs of rescue medication were required [9]. Each of these studies has also reported on the short-term (treatment period) and long-term (out to 12 months) adverse event profile associated with the bronchial thermoplasty. The reported increase in respiratory adverse events in the short-term represented the further aggravation of the worsening of asthma symptoms that is associated with bronchoscopy in patients with asthma [14]. During the long-term post-treatment period out to 12 months, fewer subjects in the BT group reported respiratory adverse events [9‐11]. Data are now presented for the safety of this treatment over a 5 year period.

BT is a novel treatment that alters the amount of airway smooth muscle, and asthma is chronic disease that is associated with structural changes in the airway wall, it was important to know whether this treatment is associated with any longer term adverse outcomes. The participation of patients originally enrolled in the AIR Trial in this additional 4 year follow-up study provides data to support the longer-term safety of BT out to at least 5 years. The absence of serious events indicates that the integrity of the airways is not compromised over the long term as a result of BT treatment. This is supported by the observation of no deterioration of baseline FEV₁ or FVC over this period, and the ability of the airways to respond to bronchodilator administration. While the rate of decline in lung function can vary in asthma, with some subjects demonstrating equivalent declines to those without airways disease, and others suffer an accelerated decline [15‐17], it is reassuring that following BT, there was no significant loss of lung function over a 5 year period. The rate of occurrence of respiratory related adverse events remained low and stable between Year 2 and Year 5 (1.1 to 1.3 events/subject/year) and comparable to the Control group for Year 2 and Year 3 (1.2 and 1.3 events/subject/year, respectively). The frequency and method of collecting adverse events in this study was different in Year 1 compared to the other 4 years, resulting in a higher rate of respiratory adverse events in both the BT and Control groups in Year 1. During Year 1, the solicitation of adverse events was more frequent (12 office visits and 9 telephone contacts) compared to subsequent years when adverse events were solicited once a year and could be subject to recall bias. Also important is the fact that during Year 1, multiple symptoms for a given adverse event were recorded as individual adverse events as compared to the subsequent years when multiple related symptoms were recorded as a single event. Because of these methodological issues, comparisons of adverse event rates within each group for Year 1 and subsequent years are not appropriate. The lower rates of respiratory adverse events in Years 2 to 5 compared to Year 1 may reflect some underreporting as a result of recall bias at the yearly solicitation of adverse events. Every effort was made to carefully solicit adverse events from study subjects during annual evaluations and augmented with a thorough review of medical records at participating institutions. Review of medical charts is lacking for instances where a subject may have been treated for adverse event(s) at a different institution.

Measures such as ER Visits and hospitalizations for respiratory symptoms are generally considered to be important measures of safety, especially if an intervention results in an increase in the rate of one or more of these events. Consistent with the low rate of respiratory adverse events is the stable incidence of healthcare utilization events (hospitalizations and emergency room visits for respiratory symptoms). There was no worsening of the proportion of subjects in the BT group experiencing hospitalizations or ER visits for respiratory symptoms beyond Year 1 out to 5 years. The rates of ER visits and the proportion of subjects with ER visits in the Control group were comparable to the BT group during Years 1, 2, and 3. Longer-term safety of BT is supported further by our findings of no deterioration in measurements of static and dynamic lung volumes or diffusing capacity over the 5 years post-treatment.. In addition airway responsiveness to methacholine remained stable in the BT group out to the last measurement performed at year 3 post-treatment. Stable lung function measures in this group of patients with moderate to severe asthma support the previously reported conclusion based on lung function measures [12, 13] and high resolution CT [13] that BT-treated airways do not develop late scarring that could result in narrowing of the airways [12].

Although x-rays are not sensitive enough for the purpose of demonstrating structural abnormalities, a review of the serial x-rays obtained annually did not reveal any clinically significant findings. While the use of high resolution computed tomography (HCRT) would have been more informative, it was not performed in this study.

Bronchial hyperresponsiveness to methacholine improved in the thermoplasty group in years 2 and 3, but not in year 1. This is an interesting result in favour of a long-term efficacy of the procedure. However, the lack of follow-up of the control group on years 4 and 5 and the observational nature of the data limits the relevance of this finding.

There are several limitations to the study. Firstly, not all patients enrolled in the Asthma Intervention Research Trial participated in long-term follow-up, although none of the reasons for non-participation was due to mortality. Secondly, the follow-up period was longer in patients who received BT (at 5 years) compared to the Control group (3 years). Subjects who had been randomized to the Control group in the predecessor AIR Trial may have opted to pursue other treatment options, and for those that agreed to the longer-term follow up, it was deemed to be unethical [18] to require them to withhold other alternative treatment options (including new experimental treatments) for a period of 5 years in order to avoid confounding of data. Nevertheless a high proportion of patients who received BT (86%) participated in long-term follow-up, which provides support for the generalizability of the safety data. Thirdly, the solicitation of adverse events on a yearly basis has the potential of under reporting due to recall bias. However, the occurrence of major events such as severe exacerbations (steroid pulses for asthma symptoms), emergency room visits, and hospitalizations is less frequent and less likely to be forgotten and therefore not reported. Medical charts were reviewed to verify steroid pulses prescribed for asthma symptoms (severe exacerbations) and reported adverse events or to identify adverse events that may not have been reported by the subject, unless the subject had presented at a different medical facility for an adverse event during that year. Finally, the study was not blinded and this may add bias to the eliciting of adverse events. Despite these limitations, over the 5 year follow-up of BT treated subjects, the absence of unexpected respiratory adverse events, no increases in hospitalizations or emergency room visits for respiratory symptoms, and maintenance of stable lung function demonstrated safety of BT out to 5 years.

This report provides long-term safety data on BT in patients in whom BT was performed by trained operators, and patients were carefully observed especially in the first year. The absence of clinical complications based on adverse event reporting, healthcare utilization events, and the maintenance of stable lung function (no deterioration of FEV₁) over a 5-year period post-BT in patients with moderate to severe asthma suggest long-term safety of the procedure out to 5 years.