Effect of fluticasone furoate and vilanterol on exacerbations of COPD in patients with moderate airflow obstruction
Fernando Martinez,1,2 Jørgen Vestbo,3 Julie A. Anderson,4 Robert D Brook,2 Bartolome R Celli,5 Nicholas J. Cowans6, Courtney Crim,7 Mark Dransfield8, Sally Kilbride,4 Julie Yates,7 David E Newby9, Dennis Niewoehner10*, Peter MA Calverley11* on behalf of the SUMMIT Investigators
1. Division of Pulmonary and Critical Care Medicine, Weill Cornell Medicine, New York, New York, USA
2. University of Michigan Health System, Ann Arbor, Michigan, USA
3. Division of Infection, Immunity and Respiratory Medicine, Manchester Academic Health Sciences Centre, The University of Manchester and South Manchester University Hospital NHS Foundation Trust, Manchester, UK
4. Research & Development, GlaxoSmithKline, Stockley Park, Middlesex, UK
5. Pulmonary and Critical Care Division, Brigham and Women’s Hospital, Harvard Medical School, Boston, Massachusetts, USA
6. Veramed Ltd., Twickenham, UK
7. Research & Development, GlaxoSmithKline, Research Triangle Park, North Carolina, USA
8. University of Alabama Birmingham, Birmingham, ALA.
9. Centre for Cardiovascular Science, University of Edinburgh, Edinburgh, UK
10. University of Minnesota, Minneapolis, MN.
11. University of Liverpool, Department of Medicine, Clinical Sciences Centre, University Hospital Aintree, Liverpool, UK
* Co-senior authors.
Address correspondence to:
Fernando J. Martinez, M.D., M.S.
Joan and Sanford I. Weill Department of Medicine
Weill Cornell Medical College
New York-Presbyterian Hospital/Weill Cornell Medical Center
525 East 68th St, Room M-522 Box 130
New York, NY10065; Telephone: 212.746.6420; Email:
Running head: Fluticasone furoate and vilanterol for exacerbations of COPD
Word count (main text,):
Abstract
Background: Inhaled corticosteroids have been shown to decrease exacerbations in COPD patients with moderate to severe COPD. Their effect in patients with milder airflow obstruction remains unclear.
Objective: This was an analysis of exacerbations in the Study to Understand Mortality and MorbidITy (SUMMIT) study.
Design: In a double-blind randomized controlled trial, once daily inhaled placebo, fluticasone furoate (FF, 100 μg), vilanterol (VI, 25 μg) or the combination (FF/VI) was administered. The primary outcome was all-cause mortality. Exacerbations of COPD were an additional pre-defined endpoint.
Setting: 1,368 centers in 43 countries.
Participants: 16,485 patients with moderate COPD and heightened cardiovascular risk.
Results: Compared with placebo, FF/VI reduced the rate of moderate/severe exacerbations by 29% (95% CI 22, 35; p<0.001) and the rate of hospitalized exacerbations by 27 % (95% CI 13, 39; p<0.001). These relative effects were similar regardless of whether subjects had a history of exacerbation in the year prior to the study or an FEV1 less than or ≥ 60% predicted. The number needed to treat was not influenced by baseline FEV1 but was influenced by the prior history of exacerbations. FF/VI also reduced the rate of exacerbations treated with corticosteroids alone or with corticosteroids and antibiotics but not those treated with antibiotics alone.
Conclusions: Patients with moderate chronic airflow obstruction experienced a reduction in exacerbations with FF/VI, compared with placebo, irrespective of a prior history of exacerbations or baseline FEV1.
Word count (max 250): 232
Manuscript body word count (max 3500): 2842
Key words: COPD; cardiovascular disease; fluticasone furoate; vilanterol; combination therapy, exacerbations
Introduction
Chronic obstructive pulmonary disease (COPD) is a chronic syndrome that is associated with acute worsening of symptoms(1). These events, exacerbations of COPD (eCOPD), have been associated with impaired health status as well as increased morbidity and mortality(2). Numerous pharmacotherapeutic options including long acting bronchodilators, inhaled corticosteroids (ICS), chronic antimicrobial agents, and roflumilast have been shown to decrease eCOPD frequency in select patient populations (1-3). In an effort to move towards more personalized therapy (4) several investigative groups have examined the factors associated with an increased patient level risk of suffering an eCOPD; the most frequently cited is a history of previous events with two in the previous year considered the most stable and predictive (5). The Global Initiative for Chronic Obstructive Lung Diseases (GOLD) therapeutic strategy recommends intensification of inhaled therapies, particularly the addition of ICS, in patients at risk for an exacerbation principally those with severe airflow obstruction, with a previous history two or more eCOPD or one or more requiring hospitalization (1).
In the Study to Understand Mortality and MorbidITy (SUMMIT), we showed that inhaled treatment with the corticosteroid, fluticasone furoate, and the long-acting beta-agonist (LABA), vilanterol, did not improve survival compared with placebo in patients with moderate COPD and heightened cardiovascular risk(6). This present report describes the effects of fluticasone furoate (FF), vilanterol (VI) and their combination (FF/VI) on eCOPD, an additional pre-defined endpoint, in a COPD population with milder COPD and without a prespecified requirement of previous eCOPD where we anticipated that this therapy might be beneficial.
Methods
Details of the study design and the analysis approach were published previously(7).
Patients
All patients provided written informed consent. The study was approved by local ethics committees and was conducted in accordance with the Declaration of Helsinki and Good Clinical Practice guidelines. The study is registered on clinicaltrials.gov as NCT01313676 (GSK Study number 113782).
Study Design
This was a prospective, double blind, parallel group, placebo controlled event-driven randomized trial conducted at 1,368 centres in 43 countries. Eligible patients were 40 to 80 years old, diagnosed with COPD and had a post-bronchodilator forced expiratory volume in one second (FEV1) ≥50 and ≤70% of the predicted value(8), ratio of post-bronchodilator FEV1 to forced vital capacity (FVC) ≤0.70, and ≥2 on the modified Medical Research Council dyspnea scale. Patients had to have a history, or be at increased risk, of cardiovascular disease(6). Exclusion criteria included respiratory disorders other than COPD, lung reduction surgery, receiving long-term oxygen or oral corticosteroid therapy, severe heart failure (New York Heart Association Class IV or ejection fraction <30%) and life expectancy <3 years(7). Participants were allocated equally to one of four treatments (placebo, fluticasone furoate (FF, 100 μg; GlaxoSmithKline), vilanterol (VI, 25 µg; GlaxoSmithKline) or the combination of fluticasone furoate and vilanterol (100/25 μg; FF/VI, Relvar®/Breo®, GlaxoSmithKline)) administered once daily as a dry powder with the use of an inhaler (Ellipta®, GlaxoSmithKline). Exacerbations were assessed every 3 months and at the end of study treatment.
Outcome Measurements
ECOPD were a pre-defined outcome. COPD exacerbations were defined by use of treatment for increased respiratory symptoms in the judgment of the investigator. Treatment was based on events reported by the subject. Moderate exacerbations were defined as a symptomatic deterioration treated with antibiotic agents (AB) and/or systemic corticosteroids (SCS), whereas severe exacerbations of COPD were those leading to hospital admission. We had information on treatment with AB and SCS separately for moderate exacerbations, allowing analyses by exacerbation treatment.
Statistical Analysis
The analyses of time to first and rate of exacerbations to compare between treatment arms in the ITT population (moderate/severe, hospitalized and treated with SCS with or without ABI) were pre-specified while analyses of subgroups and analyses of exacerbations treated with other combinations (i.e. treated with SCS alone, ABI alone and or both SCS and ABI) were post-hoc. Event based number needed to treat (NNT) were calculated(9). All p-values should be interpreted as nominal as the primary outcome did not meet statistical significance (6).
To control for multiplicity, a closed testing procedure approach was planned. If significance at the 5% level was not achieved for the primary endpoint for the comparison of combination treatment with placebo, then the tests for the secondary and other efficacy endpoints would be interpreted as descriptive only. Time to first exacerbation was analyzed using a Cox proportional hazards regression model allowing for covariates of age, sex and previous exacerbations (0, 1, and 2). The frequency of eCOPD was analyzed using a negative binomial model with the number of exacerbations as the outcome and the logarithm of time during which treatment was received as an offset variable.
These same analyses were performed within each individual subgroup (smoking status, age, CV entry criteria, gender, ethnicity, race, region, % predicted FEV1, previous history of moderate/severe exacerbations and previous history of hospitalized exacerbations). Given the impact of seasons on exacerbation rates(10, 11) and therapy(12) we separated exacerbations into those which occurred in winter or summer.
Role of the funding source
Funding for SUMMIT was provided by GlaxoSmithKline. Scientific oversight of the trial was provided by a scientific steering committee composed of six academic researchers and three employees from GlaxoSmithKline, who were collectively responsible for the study design and conduct, for approval of the statistical analysis plan, and for the review and interpretation of the data. Statistical analyses were performed by a contract research organization (Veramed Ltd, Twickenham, UK; funded by GSK) on behalf of, and with oversight from, employees of the sponsor.
Results
Study Population
A total of 16,485 patients were included in the intent-to-treat efficacy (ITT) population. The patients were 65 years of age (SD 8 years); had a post-bronchodilator FEV1 % predicted at screening of 60% (SD 6%); were predominately male (75%) and had a smoking history of 41 pack-years (SD 24 pack-years). Further information is available elsewhere (6). Approximately one third (6464, 39%) of patients reported having at least one exacerbation in the year prior to study entry. Of these, 2444 (15%) patients had 2 or more exacerbations in the year prior to study entry and 2205 (13%) experienced at least one eCOPD requiring hospitalization in the previous year.
Overall effect on eCOPD
The rates of moderate/severe eCOPD are presented in Table 1. Treatment with FF/VI reduced the rate of eCOPD by 29% (95%confidence interval (CI) 22-35; p<0.001) compared with placebo (Table 1). Treatment with FF/VI reduced the rate of eCOPD by 19% (95% CI 12, 26; p<0.001) compared with FF, reflecting the contribution of VI to the combination, and reduced the rate of eCOPD by 21% (95% CI 14, 28; p<0.001) compared with VI, reflecting the contribution of FF to the combination. The risk of a first moderate/severe eCOPD was reduced with FF/VI compared with placebo by 21% (95% CI 14-27; p<0.001) (Figure 1). All other comparisons for time to first analyses are presented in Table e1.
Treatment with FF/VI reduced the rate of hospitalized eCOPD by 27% (95% CI 13-39; p<0.001) compared with placebo (Table 1). Treatment with FF/VI numerically reduced the rate of hospitalized eCOPD by 11% (95% CI -6, 25; p=0.204) compared with FF, assessing the contribution of VI to the combination, and the rate of eCOPD by 9% (95% CI -8, 24; p=0.282) compared with VI, assessing the contribution of FF to the combination. The risk of a first hospitalized eCOPD was reduced with FF/VI compared with placebo by 22% (95% CI 8, 33; p=0.002) (Figure 1). All comparisons for time to first analyses are presented in Table e1.
There appeared to be a higher probability of exacerbating in the winter than in the summer, although the treatment effect was very similar for the winter exacerbations as for the summer exacerbations.
Effect on eCOPD as a function of subgroups
The rate ratios for the comparison of FF/VI with placebo for moderate/severe eCOPD for subgroups of interest are presented in Fig 2A. The reductions in rate of moderate/severe eCOPD were similar for all the subgroups (including baseline %predicted FEV1 cut at 60% and whether or not patients had a prior history of eCOPD or a prior history of hospitalized eCOPD). The CIs for some subgroups are wide, reflecting the small number of subjects in that group. The event based NNTs for subgroups by previous exacerbation history are enumerated in Table e2a. There was little difference in the NNT for preventing one on-treatment moderate/severe eCOPD (10 for FEV1 60% and 12 for FEV1 > 60% predicted) (data not shown).
The rate ratios for the comparison of FF/VI with placebo for hospitalized eCOPD for subgroups of interest are presented in Fig 2B. The reductions in rate of hospitalized eCOPD were similar for most of the subgroups (including baseline %predicted FEV1 cut at 60% and whether or not patients had a prior history of eCOPD or a prior history of hospitalized eCOPD). The exception to this is for some of the race subgroups, where there were very small numbers of patients and events. The event based NNTs for the comparison of FF/VI with placebo by previous exacerbation history are enumerated in Table e2a.
The rate ratios for the comparisons of FF/VI vs VI for moderate/severe eCOPD and for hospitalized eCOPD for subgroups of interest are presented in the supplement (Fig e1a, e1b, e2a and e2b). The event based NNTs for the comparison of FF/VI vs VI by previous exacerbation history are enumerated in Table e2b.
Effect on eCOPDs treated with antibiotics and systemic corticosteroids
Treatment with FF/VI reduced the rate of eCOPD treated with both AB and SCS by 38%. Substantial effects were also seen for those eCOPD treated with SCS alone (61%), as well as those requiring SCS with or without AB (45%). Treatment with FF/VI did not affect eCOPD treated with AB alone (Table 2). For the eCOPD treated with SCS with or without antibiotics, FF and VI individually also reduced the rate of eCOPD compared with placebo, although to a lesser extent that FF/VI. Neither FF nor VI reduced the rate of eCOPD treated with AB alone compared with placebo (12% increase and 6% increase respectively) (table 2).
Discussion
SUMMIT is the largest survival study to date of an inhaled corticosteroid and long-acting beta-agonist in patients with COPD and heightened cardiovascular risk(6). This study failed to show a significant effect of treatment with inhaled fluticasone furoate and vilanterol on all-cause mortality or cardiovascular outcomes. In the current analysis we demonstrate that FF/VI decreased the rate of moderate/severe eCOPD, eCOPD requiring a hospitalization and eCOPD treated with SCS, important respiratory-related outcomes. FF/VI also reduced exacerbations treated with both SCS and AB whereas there was no effect on eCOPD treated with AB alone; for the latter FF as monotherapy actually increased the risk by 12%. Although effect estimates varied, analyses of time to first exacerbation were not substantially different from those of rates. These data in a patient population that does not meet the traditional criteria for ICS therapy suggests that the combination of an ICS and LABA may be appropriate therapy for a broader range of subjects than currently recommended by therapeutic strategy recommendations.