Published ahead of print on February 28, 2008, doi:10.1164/rccm.200709-1356OC
© 2008 American Thoracic Society doi: 10.1164/rccm.200709-1356OC
The Effects of Fluticasone with or without Salmeterol on Systemic Biomarkers of Inflammation in Chronic Obstructive Pulmonary Disease1 Department of Medicine (Respiratory Division), University of British Columbia, Vancouver, Canada; 2 Department of Medicine, University of Saskatchewan, Saskatoon, Canada; 3 Department of Medicine, University of Calgary, Calgary, Canada; 4 Department of Medicine, University of Alberta, Edmonton, Canada; 5 Wetaskiwin General Hospital, Wetaskiwin, Canada; 6 Lions Gate Hospital, North Vancouver, Canada; 7 Royal Alexandra Hospital, Edmonton, Canada; 8 Grey Nuns Hospital, Edmonton, Canada; 9 Lethbridge General Hospital, Lethbridge, Canada; and 10 Matsqui-Sumas-Abbotsford General Hospital, Abbotsford, Canada Correspondence and requests for reprints should be addressed to S. F. Paul Man, M.D., Room 548, Burrard Building, St. Paul's Hospital, 1081 Burrard Street, Vancouver, BC, V6Z 1Y7 Canada. E-mail: pman{at}providencehealth.bc.ca
Rationale: Small studies have suggested that inhaled corticosteroids can suppress systemic inflammation in chronic obstructive pulmonary disease (COPD). Objectives: To determine the effect of inhaled corticosteroids with or without long-acting β2-adrenergic agonist on systemic biomarkers of inflammation. Methods: We conducted a double-blind randomized placebo-controlled trial across 11 centers (n = 289 patients with FEV1 of 47.8 ± 16.2% of predicted) to compare the effects of inhaled fluticasone alone or in combination with salmeterol against placebo on circulating biomarkers of systemic inflammation over 4 weeks. The primary endpoint was C-reactive protein (CRP) level. Secondary molecules of interest were IL-6 and surfactant protein D (SP-D). Measurements and Main Results: Neither fluticasone nor the combination of fluticasone/salmeterol had a significant effect on CRP or IL-6 levels. There was, however, a significant reduction in SP-D levels with fluticasone and fluticasone/salmeterol compared with placebo (P = 0.002). Health status also improved significantly in both the fluticasone and fluticasone/salmeterol groups compared with placebo, driven mostly by improvements in the symptom scores. Changes in the circulating SP-D levels were related to changes in health status scores. FEV1 improved significantly only in the fluticasone/salmeterol group compared with placebo. Conclusions: ICS in conjunction with long-acting β2-adrenergic agonist do not reduce CRP or IL-6 levels in serum of patients with COPD over 4 weeks. They do, however, significantly reduce serum SP-D levels. These data suggest that these drugs reduce lung-specific but not generalized biomarkers of systemic inflammation in COPD. Clinical trial registered with www.clinicaltrials.gov (NCT 00120978).
Key Words: systemic inflammation • placebo • fluticasone • salmeterol
Chronic obstructive pulmonary disease (COPD) represents an increasing burden worldwide and, by 2020, its mortality rate will rank third, only behind stroke and heart disease (1). Although this figure is alarming, it is likely to be a gross underestimate of the true health and economic burden of COPD because COPD is an important risk factor for other common causes of morbidity and mortality, including cardiovascular disorders and lung cancer (2, 3). Although the pathobiology of COPD has not been fully elucidated, there is a growing recognition that systemic inflammation may play a salient role in COPD progression and morbidity (4). Systemic inflammation has been linked particularly with some extrapulmonary manifestations of COPD, including sudden deaths, arrhythmias, strokes, myocardial infarction, cancer (5–7), muscle weakness, reduced exercise tolerance, and poor health status (8). These data raise the possibility that treatments modulating systemic inflammation could improve important health outcomes in COPD. Inhaled corticosteroids with or without long-acting β2-adrenoceptor agonists (LABAs) reduce exacerbations and improve health status in COPD (9). Some studies suggest that they may also reduce systemic inflammation in COPD (10, 11), but these studies were limited by the observational nature of the study design and small sample sizes. The primary aim of this randomized, placebo-controlled, double-blind, multicenter clinical trial involving 11 centers in western Canada was to determine whether inhaled corticosteroids alone or in combination with a LABA reduce systemic inflammation in subjects with stable COPD.
See the online supplement for full details of methods used.
Rationale for Measurements
Study Participants
Overview of the Trial Design
Outcome Measurements
Statistical Analysis
A total of 356 subjects were screened. Of these, 67 were excluded because they failed to meet the inclusion and exclusion criteria of the study, leaving 289 subjects for analysis. Between visits 1 and 2, 29 subjects dropped out of the study and 36 dropped out between visits 2 and 3, mostly because of worsening in their symptoms or heath status. This left 224 subjects, who were randomized at visit 3 into three arms of the trial: 45 to the placebo, 87 to fluticasone and 92 to fluticasone/salmeterol combination groups. A flow diagram summarizing the distribution of the subjects is shown in Figure 1. The baseline characteristics of the study participants at each of the major visits are shown in Table 1.
Visit 1 (Enrollment) The average age of the subjects was 68.8 ± 9.0 years; 60.2% were males and 32.5% were current smokers. The average body mass index was 27.7 ± 6.0 kg/m2 and the average FEV1 was 1.36 ± 0.54 L or 47.8 ± 16.2% of predicted. A total of 58.4% of subjects were taking combination therapy, whereas 68.3% were taking an inhaled corticosteroid–containing drug at the time of enrollment. The average total SGRQ score was 42.4 ± 16.5 units. The median (interquartile range) SP-D was 104.5 (76.0–143.5) ng/ml, CRP was 3.4 (1.5–6.8) mg/L, and IL-6 was 2.2 (1.5–3.6) pg/ml.
Visit 2 (Run-in Phase; 1 mo of Fluticasone)
Visit 3 (Withdrawal Phase) Withdrawal of fluticasone for 1 month led to the following changes: the subjects' CRP changed by 0.0 (–1.3 to 1.3) mg/L (P = 0.829); IL-6 changed by –0.1 (–0.6 to 0.6) pg/ml (P = 0.728); SP-D changed by 5.6 (–4.5 to 20.1) ng/ml (P < 0.001); FEV1 changed by –1.0 (–5.4 to 2.2) % predicted (P < 0.001); and total SGRQ changed by 2.6 (–2.3 to 8.1) units (P < 0.001). The findings from the run-in and withdrawal phases are summarized in Table 2.
Visit 4 (Randomized Controlled Trial Phase)
Exacerbation and Dropouts There were significantly more exacerbations during the withdrawal than in the run-in phase. Moreover, subjects who received placebo Diskus experienced more exacerbations and were more likely to drop out compared with subjects who received fluticasone or fluticasone/salmeterol (see Table 5). All dropouts occurred because of worsening of respiratory symptoms related to the participant's COPD. There were no significant differences in exacerbations or dropouts between the fluticasone and fluticasone/salmeterol groups.
Other Adverse Events During the active treatment phase, 9% in the placebo, 6% in the fluticasone, and 10% in the combination arm complained of lower respiratory tract symptoms defined as cough and/or dyspnea (which did not result in dropping out or any additional treatment) (P = 0.620). Upper respiratory tract complaints defined as rhinorrhea, nasal congestion, and/or headache were reported in 4% of placebo-, 4% of fluticasone-, and 4% of the combination-treated groups (P = 0.908). There were no reported cases of pneumonia.
The present study is a novel multicenter clinical trial designed specifically to evaluate the effects of inhaled corticosteroids and combination therapy on systemic biomarkers of inflammation. With respect to the primary endpoint, inhaled fluticasone with or without salmeterol did not significantly change CRP levels in patients with stable COPD. These medications also failed to significantly alter serum IL-6 levels. However, they significantly reduced circulating SP-D levels and improved health status and lung function over a 4-week period. Withdrawal of fluticasone, conversely, increased SP-D levels, worsened health status, and reduced lung function in patients with moderate to severe COPD. These data suggest that fluticasone-based therapy reduces circulating lung-specific but not general biomarkers of systemic inflammation. Interestingly, the reductions in SP-D levels were associated with improved health status (especially dyspnea) and lung function, supporting the notion that lung inflammation plays an important role in health outcomes of patients with COPD. Dissimilar to our previous report (11), in the present study we did not find a significant impact of fluticasone or combination product on CRP or IL-6 levels. Several explanations exist. First, the previous study evaluated a small number of patients, which may have caused a type 1 statistical error. Second, the previous cohort had milder lung function impairment. The average FEV1 was approximately 60% of predicted, whereas in the present study the average FEV1 was less than 50% of predicted. Lung function may modify the systemic action and absorption of inhaled corticosteroids (21). Third, there were fewer current smokers in the previous than in the present study. The clinical effects of corticosteroids may be modified by smoking status of patients; corticosteroids appear to be less effective in current than in former smokers (22). Fourth, a larger percentage of patients had been taking an inhaled corticosteroid long term before enrollment into this study than in the previous study and this treatment could have modified some of the inflammatory processes in the lungs. The secondary finding that inhaled fluticasone with or without salmeterol significantly reduced SP-D levels may be of relevance in stable COPD. SP-D was chosen as an endpoint for several reasons. SP-D is a large, multimeric, collagenous glycoprotein weighing approximately 43 kD that plays an important role in innate immunity and in host defense responses against inhaled microorganisms and particles (23). SP-D also has a major function in regulating surfactant homeostasis in the lungs by modulating surfactant ultrastructure and promoting reuptake of surfactant by type II pneumocytes (24). SP-D is produced mainly by type II pneumocytes in the lungs, although other cells, such as Clara cells, endothelial cells, and glandular cells in the gastrointestinal tract, can produce trace amounts of SP-D (23). Acute lung injury in general increases both lung and serum SP-D levels. Cigarette smoking, on the other hand, induces a rise in the serum levels (25) but a fall in the bronchoalveolar lavage concentrations (26). Interestingly, mice exposed to cigarette smoke demonstrate increased expression of SP-D in the lungs at both the mRNA and protein level (27). Thus, the reduced bronchoalveolar lavage expression related to smoking is likely caused by leakage of SP-D from the lungs into the systemic circulation. Within the pulmonary system, SP-D is generally beneficial in protecting the lungs from oxidant, inflammatory, and infectious stress (28). However, systemic expression of SP-D may be harmful (29). Overexpression of SP-D in the systemic circulation reduces high-density lipoprotein cholesterol levels and increases the risk of atherosclerosis (29). SP-D deficiency, on the other hand, protects mice against atherosclerosis (29). In human conditions, elevated levels of SP-D in the systemic circulation are associated with poor clinical outcomes in a variety of different settings. In acute respiratory distress syndrome of adults requiring mechanical ventilation, for example, elevated plasma SP-D levels are associated with increased risk of multiorgan failure, ventilator dependence, and even mortality (16). SP-D levels have also been noted to be elevated in other conditions such as allergic bronchopulmonary aspergillosis, community-acquired pneumonia, and interstitial lung diseases, and in these conditions, elevated SP-D levels are associated with disease severity and poor health status in these conditions (30–32). In idiopathic pulmonary fibrosis, circulating SP-D levels predict long-term survival of these patients and changes in the serum value track clinical responses to therapy (33). Elevated serum SP-D levels have also been associated with increased risk of dementia and mortality in the general elderly population (34). The mechanisms by which fluticasone and combination therapy reduce SP-D levels systemically are uncertain. In general, glucocorticoids up-regulate the expression of SP-D and other surfactant proteins in vivo and in vitro (35, 36). In theory, if anything, fluticasone and combination therapy should have caused an increase in the systemic expression of SP-D by increasing local production in the lungs. Altered production is therefore an unlikely explanation for the reduced SP-D expression related to corticosteroids. A more plausible explanation is that fluticasone and combination therapy reduced leakage of SP-D from the lung into the systemic compartments by attenuating lung inflammation and reducing epithelial and/or vascular permeability. Animal models have shown that lung injury and associated inflammation enhance spillage of SP-D molecules into the systemic circulation, causing increases in serum SP-D levels (37). Local application of corticosteroids, on the other hand, can reduce vascular leakage (38). However, because in the present study we did not measure lung expression of SP-D or quantified capillary leakage, we cannot draw clear inferences regarding the source of serum SP-D in patients with COPD or the mechanisms by which inhaled corticosteroids reduced serum expression of SP-D. Whatever the mechanism, in the present study we found that the reduction in systemic levels of SP-D was associated with improved lung function and health status of patients with COPD, raising the possibility that SP-D may be a useful biomarker to track disease progression and clinical outcomes of patients with COPD. Because inflammation and lung function may be independent factors in the pathobiology of COPD, long-term studies powered on hard endpoints such as mortality and exacerbations are needed to validate new potential biomarkers in COPD (12). Because our study did not include any subjects without COPD, the relationship between COPD and the inflammatory biomarkers is uncertain. Previous population-based studies have reported average CRP values between 1.5 to 2.5 mg/L in adults 40 years of age and older (13, 39, 40). The CRP values were generally higher in the present study, consistent with the notion that COPD is associated with low-grade systemic inflammation. Reported IL-6 and SP-D levels are more variable than those for CRP and there are no universally accepted reference values for these measurements in the general population. Previous large epidemiologic studies have reported average serum or plasma IL-6 levels of approximately 1.5 pg/ml (41, 42) and as high as 20 pg/ml (13). The average SP-D levels in the systemic circulation in subjects without overt disease have ranged from less than 50 ng/ml to nearly 1,500 ng/ml (34, 43–45). The variability is likely due to a variety of factors, including differences in the clinical characteristics of the underlying populations and in the nature and performance of immunoassays used across the studies. SP-D levels are thus only comparable within, but not across, studies. In general, SP-D levels in the systemic circulation increase with age, current smoking status, and use of serum rather than plasma (44). There were several limitations to the study. First, the duration of the study was relatively short. Thus, the long-term impact of fluticasone and combination therapy on systemic inflammation is uncertain. A relatively short follow-up period was chosen to minimize the effects of exacerbations, infections, and inflammatory insults, which are frequent occurrences in patients with moderate to severe COPD and can variably and unpredictably perturb systemic levels of CRP and other inflammatory biomarkers (46). Pinto-Plata and colleagues demonstrated a significant relationship between the (long-term) use of inhaled corticosteroids and reduced CRP levels in a group of patients with COPD (10). Thus, we cannot discount the possibility that long-term use of inhaled corticosteroids may reduce CRP levels in COPD. Second, we did not measure inflammatory biomarkers from lung samples due to a variety of technical and logistical constraints, including lack of standardization of sample collection, invasiveness of the procedure, and poor patient tolerability, that limit their application (47). Blood measurements, on the other hand, are more robust and are standardized. Notwithstanding the limitations of lung-based measurements, Barnes and colleagues have shown that combination therapy reduces various sputum and bronchial markers of inflammation in patients with moderate to severe COPD (48). Our findings extend these findings by demonstrating a significant impact on SP-D. Third, we did not ascertain the effects of these drugs on hard clinical outcomes such as hospitalization or mortality. Fourth, we did not evaluate the potential impact of bronchodilators on the inflammatory biomarkers. A recent study suggests that bronchodilators by themselves are unlikely to impact significantly on systemic markers of inflammation (49). In summary, the present multicenter clinical trial failed to demonstrate any significant effect of inhaled fluticasone or combination therapy on systemic levels of CRP or IL-6, but they had a salutary effect on serum SP-D. The reductions in SP-D were associated with improved health status and lung function in patients with moderate to severe COPD. These data suggest that inhaled fluticasone and combination therapy reduce lung-specific but not some common biomarkers of systemic inflammation and raise the possibility of using SP-D and other lung-specific protein markers as an intermediate endpoint for future interventional studies in COPD.
The authors acknowledge and thank the site coordinators: Janet Baron (Royal University Hospital, Saskatoon); Georgina Lopez (St. Paul's Hospital, Vancouver); Anju Mainra (Lion's Gate Hospital, North Vancouver); Linda Hui, Maureen Sigurdson (Vancouver General Hospital, Vancouver); Kathy Duce (Lethbridge Regional Hospital, Lethbridge); Jill Edwards, Angie Depner (Links Clinic, Edmonton); Jennifer Barchard (Grey Nuns Community Hospital, Edmonton); Heidi Cheung (University of Alberta, Edmonton); Teena Rossitter (Wetaskiwin Lung Laboratory, Wetaskiwin); Amin Thawer, Diane Conley, Gladys Wolters (University of Calgary, Calgary). The authors also thank Mrs. Claire Gray (laboratory manager), Dr. Ted Watson (GSK), Ms. Jill Waddell (GSK), and Dr. Gerry Hagan (GSK Global) for their contributions to the project.
The ABC trial is funded by GlaxoSmithKline. The funding source had no role in the study conception or design. All data analyses were conducted independently of the funding source at the James Hogg iCAPTURE Center (St. Paul's Hospital, Vancouver, BC.). The primary data are held at the James Hogg iCAPTURE Center. D.D.S. is a holder of a Canada Research Chair in COPD and a Senior Scholar Award from the Michael Smith Foundation for Health Research. M.F. is supported by the Michael Smith Distinguished Scholar Award and the BC Lung CIHR Scientist Award. This article has an online supplement, which is accessible from this issue's table of contents at www.atsjournals.org Originally Published in Press as DOI: 10.1164/rccm.200709-1356OC on March 6, 2008 Conflict of Interest Statement: D.D.S has received honoraria for speaking engagements from AstraZeneca (AZ) in 2003 ($4,000), 2004 ($3,000), and 2005 ($11,000), and from GlaxoSmithKline (GSK) in 2003 ($4,000), 2004 ($8,000), 2005 ($6,500), and 2006 ($10,000); he has also received unrestricted research funding as either the principal investigator or co-principal investigator from GSK in 2003 for $80,000 and in 2004 for $1.5 million; he has also received $3,500 from GSK for consultancy work in 2004 and $1,500 in 2006. S.F.P.M. was funded jointly by GSK and by the Canadian Institute of Health Research in 2005–2008 for a total sum of $321,750. D.D.M. received $7,500 in 2006 and $6,500 in 2007 from GSK for advisory board participation and received $3,850 in 2005, $6,250 in 2006, and $5,300 in 2007 from GSK for lecture fees. G.F has received honoraria and travel expenses for speaking at scientific meetings and continuing medical education courses from GSK for $16,500 in 2005, $7,500 in 2006, and $14,000 in 2007; he has received research funding from GSK for $135,000 between 2001 and 2005, Altana Pharma (AP) for $80,000 from 2004 to 2006 and $92,266 from GSK, AZ, and Boehringer Ingelheim, Canada, Ltd (BICL) from 2004 to 2006; he has also received honoraria and travel expenses for participating on medical advisory boards for various pharmaceutical companies (GSK, AZ, BICL, AP, Pfizer Canada, and Abbott Canada). M.F. has received fees from Merck, AZ, GSK, Novartis, and Nycomed of $22,000 in 2006, $24,000 in 2005, and $18,000 in 2005. During the same years, he received $8,000, $7,000, and $8,000, respectively, for participating in advisory boards with the same companies; all research funding received has been paid directly to UBC during the same period and totaled approximately $240,000. E.W. has participated as a speaker in seminars organized and financed by AZ, GSK, Pfizer, and Boehringer Ingelheim. E.W. has also attended an advisory board of GSK in 2006. E.W. received $47,500 from GSK as a research grant for participating in a multicenter clinical trial. E.Y. has been to three advisory board meetings with AZ in 2005 and was renumerated $1,000 per meeting; in 2006, he went to two AZ board meetings and again received $1,000 per meeting; in 2007, he was paid $1,000 for an AZ meeting. In 2006, he received $1,000 at an Altana medical advisory board meeting. E.Y. gave three lectures to general practitioners for AZ and was paid $1,000 each time; in 2006, E.Y. gave three lectures for AZ for general practitioners and was paid $750 for each of the two lectures and $1,000 for the third lecture. E.Y. also gave a lecture for Altana and was paid $1,000; in 2007 he gave three lectures for AZ and was paid $1,000 per lecture. R.R.M. has been reimbursed by GSK for several continuing medical education events (CMEs) and advisory boards totaling $7,000. W.R. has served in regional advisory board for AZ, GSK, and Nycomed, and received honoraria for participation and lecture fees. L.S.M. is a member of the GSK Western Regional Advisory Board and sponsored lectures by GSK to family physicians on COPD/asthma. E.W., as part of his responsibility as a specialist physician, has participated as a speaker for various CME lectures for a number of companies, including GSK, AZ, and Merck; the total for these talks is less than $10,000/year from all companies combined; he serves on the regional advisory board for both AZ and GSK. R.L.C. has served on the advisory board of GSK and has served as a consultant for GSK. D.W. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. W.Q.G. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. R.R. does not have a financial relationship with a commercial entity that has an interest in the subject of this manuscript. Received in original form September 12, 2007; accepted in final form February 21, 2008
Related articles in AJRCCM:
This article has been cited by other articles:
|
||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
TOP
ABSTRACT
AT A GLANCE COMMENTARY
2 test with appropriate degrees of freedom for dichotomous variables and one-way analysis of variance for continuous variables. Tukey's post hoc analysis was used to adjust for multiple (pairwise) comparisons. For nonnormally distributed variables, a Kruskal-Wallis test was used. The multiple Behrens-Fisher test (
