SAN DIEGO – In patients on long-term triple therapy and up to one exacerbation in the previous year, the withdrawal of inhaled corticosteroids (ICS) led to a small decrease in lung function that was not clinically important, with no associated difference in the rates of chronic obstructive pulmonary disease (COPD) exacerbations, dyspnea or as-needed bronchodilator use.
Those are key findings from the SUNSET trial, a 26-week, randomized, double-blind, parallel-group multicenter study to assess the efficacy and safety of the switch from long-term triple therapy to indacaterol/glycopyrronium (IND/GLY, 110/50 mcg once daily) or continuation of triple therapy with tiotropium 18 mcg once daily and salmeterol/fluticasone propionate fixed-dose combination 50/500 mcg b.i.d.
Dr. Kenneth R. Chapman
Dr. Chapman, director of the asthma and airways clinic at University Health Network, Toronto, noted that relatively few patients with COPD benefit from inhaled steroids. “Given the risk of adverse events (pneumonia, osteoporosis, etc.), we’d rather not give them when they’re not needed,” he said. “Inhaled steroids seem to play only one role in COPD: They tend to reduce exacerbations in the exacerbation-prone COPD patient. That’s about 20% of the COPD population. Despite this, a great many patients end up on triple therapy [long-acting bronchodilators/long-acting muscarinic antagonist (LABA/LAMA) and ICS] needlessly.”
For the study, Dr. Chapman and his associates enrolled 1,053 patients with moderate to severe COPD who’d had no more than one exacerbation in the previous year who had used triple therapy for at least 6 months prior to study inclusion. The primary endpoint of the study was noninferiority on change from baseline in postdose trough forced expiratory volume in 1 second (FEV1) (with a noninferiority margin of –50 mL) after 26 weeks. Exacerbations, health-related quality of life as measured by the St. George’s Respiratory Questionnaire (SGRQ-C), and breathlessness as measures by the Transition Dyspnea Index also were evaluated. Of the 1,053 patients, 527 were randomized to IND/GLY and 526 to triple therapy. Their mean age was 65 years and their mean postbronchodilator FEV1 was 1.6 L.
The researchers found that ICS withdrawal led to a mean reduction in trough FEV1 of –26 mL, which exceeded the noninferiority margin. This difference between treatments on trough FEV1 was driven by the subset of patients with high blood eosinophil counts at baseline (a mean of –68 mL for patients with at least 300 cells/mcL and a mean of –13 mL for patients with fewer than 300 cells/mcL). The two treatments showed similar annualized rates of moderate/severe COPD exacerbations (rate ratio, 1.08) and all (mild/moderate/severe) exacerbations (RR, 1.07). ICS withdrawal led to a small difference in SGRQ-C (1.4 U on week 26), but no differences in Transition Dyspnea Index or use of rescue medication over 26 weeks. Safety and tolerability were balanced across the two treatment groups.
“Although we found no overall increase in exacerbations with ‘de-escalation,’ there were, of course, exacerbations that occurred during the trial,” Dr. Chapman said. “We found that they tended to occur in the minority of patients who had elevated blood eosinophil counts, especially if the counts were elevated persistently (at screening and randomization). The relevant cut-point was blood eosinophil counts above 300/UL. If exacerbations did occur in this easily identifiable subpopulation, they tended to occur early, in the first month after de-escalation. This gives physicians a simply way to identify a population they might exercise caution and a period when careful monitoring is useful.”