Triple bronchodilator therapy with fluticasone furoate/umeclidinium/ vilanterol may improve quality of life and reduce exacerbations compared to fluticasone furoate/vilanterol in patients with COPD
EBM Focus - Volume 13, Issue 19
Reference: IMPACT trial (N Engl J Med 2018 Apr 18 early online) (level 2 [mid-level] evidence)
- In the IMPACT trial, over 10,000 adults with significant chronic obstructive pulmonary disease (COPD) and high risk of exacerbation were randomized to fluticasone furoate/umeclidinium/vilanterol (triple therapy), fluticasone furoate/vilanterol (ICS/LABA), or umeclidinium/vilanterol (LAMA/LABA) via a single dry powder inhaler for 1 year.
- Patients allocated to triple therapy had significantly greater rates of having a clinically important improvement in quality of life and significantly lower mean annual rates of moderate-to-severe exacerbations compared to either dual therapy regimen.
- For some of the patients, the transition to the assigned therapy may not have reflected usual clinical practice, and a majority may have had asthma-COPD overlap syndrome, complicating interpretation of the findings.
For patients classified as Global Initiative for Chronic Obstructive Lung Disease (GOLD) Group D, the GOLD guidelines recommend inhaled long-acting beta-2 agonist (LABA)/long-acting muscarinic receptor antagonist (LAMA) combination in most patients, while LABA/inhaled corticosteroid (ICS) combination may be preferred in patients with potential asthma-COPD overlap syndrome. Escalation to triple therapy (ICS/LAMA/LABA) is recommended in patients who develop exacerbations while on the dual therapy regimens (GOLD 2018). In the recent IMPACT trial, 10,355 adults with symptomatic COPD (COPD Assessment Test score ≥ 10) were randomized to fluticasone furoate/umeclidinium /vilanterol (triple therapy), fluticasone furoate/vilanterol (ICS/LABA), or umeclidinium /vilanterol (LAMA/LABA) once daily via a single dry powder inhaler for 1 year. All patients had either an FEV1 < 50% predicted and ≥ 1 moderate or severe exacerbation in the previous year or an FEV1 50%-80% predicted and ≥ 2 moderate exacerbations or ≥ 1 severe exacerbation in the previous year. Baseline blood eosinophil counts were > 150 cells/mcL (eosinophilic COPD) in 57% of patients. At trial entry, 38% of patients were already taking ICS/LAMA/LABA triple therapy, 29% were taking ICS/LABA, and 8% were taking LAMA/LABA. Assessment of outcomes began immediately after the start of the intervention, and severe exacerbations were defined as resulting in hospitalization or death.
After 1 year, patients allocated to triple therapy had a significantly greater rate of a clinically important improvement in quality of life which occurred in 42% with triple therapy compared to 34% with ICS/LABA and 34% with LAMA/LABA (p < 0.001, NNT 13 for both comparisons) and a significantly lower mean annual rate of moderate-to-severe exacerbations which was 0.91 with triple therapy compared to 1.07 with ICS/LABA (rate ratio 0.85, 95% CI 0.8-0.9) and 1.21 with LAMA/LABA (rate ratio 0.75, 95% CI 0.7-0.81). All-cause mortality was significantly reduced with triple therapy compared to LAMA/LABA (1% vs. 2%, p = 0.01, NNT 100) but not compared to ICS/LABA (1% vs. 1%). Compared to LABA/LAMA, ICS/LABA was associated with reduced all-cause mortality (hazard ratio for death 0.61, 95% CI 0.4-0.93) (treatment comparisons for all-cause mortality were not corrected for multiple testing). The rate of pneumonia was significantly greater with triple therapy compared to LAMA/LABA (8% vs. 5%, p < 0.001, NNH 33) but not compared to ICS/LABA (8% vs. 7%).
In patients with significant COPD symptoms and history of prior exacerbations, triple therapy was associated with improvements in quality of life and a reduction in moderate-to-severe exacerbations compared to dual therapy with either LAMA/LABA or ICS/LABA. The absolute differences in exacerbation rates were small but important considering the clinical impact of the outcomes. Although the types of previous therapies received by the patients before randomization in the IMPACT trial were equally distributed across treatment groups, there are two points of concern regarding the transition to the assigned therapy. Firstly, the patients in this trial did not necessarily receive the sequence of treatments as they would have had in current clinical practice. Patients would generally have a step up in their therapy with additional medication if not doing well and maintain or gradually reduce the number of treatments if they were stable. In this study, patients who were already on triple therapy may have wound up on dual therapy, or some who were on LABA/ICS may have been changed to LABA/LAMA and those changes may have precipitated an exacerbation. This is of particular concern as the majority of patients (67%) were taking ICS at baseline indicating that many in this patient population may have had asthma-COPD overlap syndrome. Secondly, measurement of the outcomes began immediately after the intervention therapy. It would have been better if the early exacerbations following change in the treatment regimen were excluded from the analysis, so that exacerbations due to the sudden change in the treatment regimen or potential carryover beneficial effects of the previous regimen did not confound the assessment of the long-term benefit of the given treatment. Even more relevant for clinical practice would have been to restrict the trial inclusion criteria to patients currently on dual therapy to which placebo or medication could be added rather than have to have some patients stop taking medication that was presumably therapeutic. Despite these caveats, the findings of the IMPACT trial provide further evidence, in addition to the previous FULFIL, TRILOGY, and TRINITY trials which had also showed benefit with triple therapy, that ICS/LAMA/LABA triple therapy may be beneficial compared to ICS/LABA dual therapy, but efficacy compared to LAMA/LABA dual therapy is less clear.