Abstract
Rationale: No currently available treatment is reported to reduce the exaggerated airway wall inflammation of chronic obstructive pulmonary disease.
Objectives: We tested the hypothesis that inhaled combined long-acting β2-agonist (salmeterol) and corticosteroid (fluticasone propionate) will reduce inflammation.
Methods: Bronchial biopsies and induced sputum were taken from 140 current and former smokers (mean age, 64 yr) with moderate to severe disease, randomized in a 13-wk double-blind study to placebo (n = 73) or salmeterol/fluticasone propionate 50/500 μg (n = 67) twice daily. Biopsies were repeated at 12 wk and sputa at 8 and 13 wk. After adjustment for multiplicity, comparisons between active and placebo were made for median change from baseline in the numbers of biopsy CD8+ and CD68+ cells/mm2 and sputum neutrophils.
Measurements and Main Results: Combination therapy was associated with a reduction in biopsy CD8+ cells of −118 cells/mm2 (95% confidence interval [CI], −209 to −42; p = 0.02), a reduction of 36% over placebo (p = 0.001). CD68+ cells were unaffected by combination treatment. Sputum differential (but not total) neutrophils reduced progressively and, at Week 13, significantly with combination treatment (median treatment difference, 8.5%; 95% CI, 1.75%–15.25%; p = 0.04). The combination also significantly reduced biopsy CD45+ and CD4+ cells and cells expressing genes for tumor necrosis factor-α and IFN-γ and sputum total eosinophils (all p ⩽ 0.03). These antiinflammatory effects were accompanied by a 173-ml (95% CI, 104–242; p < 0.001) improvement in prebronchodilator FEV1.
Conclusions: The combination of salmeterol and fluticasone propionate has a broad spectrum of antiinflammatory effects in both current and former smokers with chronic obstructive pulmonary disease, which may contribute to clinical efficacy.
Chronic obstructive pulmonary disease (COPD) is a leading and rising cause of mortality worldwide (1). Treatments are available that can be used to prevent and control symptoms, reduce exacerbations, increase exercise tolerance, and improve health status (2). Long-acting β2-agonists, such as salmeterol, combine symptom control with improvement in lung function and provide clinically relevant improvements in health status. Inhaled corticosteroids are recommended for the treatment of patients with more severe disease and frequent exacerbations, and inhalation of the combination of long-acting β2-agonist and inhaled corticosteroid is more effective in improving lung function and symptoms and reducing exacerbations than either drug alone (3–5).
Progressive airflow limitation and symptoms in COPD are associated with an exaggerated inflammatory response of the lungs to noxious agents, most commonly cigarette smoke (2). Smoking causes airway inflammation even before there is detectable airflow limitation (6), and inflammation persists even after smoking cessation (7, 8). Inflammation in COPD is distinct from that in asthma (9, 10) and is characterized by a predominance of CD8+ cells at all airway levels, including the lung parenchyma (11–14). There is also an increase of CD68+ cells (monocytes/macrophages) in the bronchial subepithelium and alveoli (11, 15) and of B cells in small airways (14). Increased airflow obstruction in COPD is associated with an increase of CD8+ cells in large airways (11, 12), CD8+ cells and B lymphocytes and CD45+ leukocytes in the small airways (13, 14, 16), and CD8+ cells in the lung parenchyma (13). Neutrophils are increased in sputum and bronchoalveolar lavage fluid (17, 18), and patients with an accelerated decline in lung function have an increased sputum neutrophil differential count (19). In COPD, there is release of proinflammatory mediators that include tumor necrosis factor α (TNF-α), IFN-γ and interleukin 8 (IL-8) (20). No currently available treatment has been reported to reduce these key cells and mediators (2) in bronchial biopsies, and any reported effects in induced sputum are controversial. Therapies that attenuate the ongoing inflammation in COPD may impact on the airway and lung destruction associated with progressive disease.
The beneficial effects of the combination of a long-acting β2-agonist and inhaled corticosteroid are acknowledged clinically, yet its effects on the airway inflammation that characterizes and forms part of the definition of COPD (2) have not been reported. We therefore tested the hypothesis that, by comparison with placebo, this combination would reduce biopsy CD8+ and CD68+ cells and sputum neutrophils. We also assessed the numbers of biopsy CD45+ leukocytes, CD4+ cells, tryptase plus mast cells, and cells expressing the proinflammatory genes TNF-α and IFN-γ. Sputum eosinophils, macrophages and lymphocytes, and the concentrations of IL-8 and eosinophil cationic protein were also measured to assess inflammation in a complementary airway compartment. Some of the results of this study have been previously reported in the form of abstracts (21–28).
This was a randomized, double-blind, placebo-controlled, parallel-group multicenter study approved by local research ethics committees or institutional review boards as appropriate. All patients, who were treated as outpatients, gave written, informed consent. After a 4-wk run-in, patients with moderate to severe COPD (2) were stratified according to whether they were current or former smokers, and randomized to either salmeterol/fluticasone propionate (Seretide/Advair/Viani Diskus; GlaxoSmithKline, Greenford, UK) 50/500 μg twice daily or matching placebo for 13 wk. Patients were not treated with inhaled or oral corticosteroids for at least 4 wk and long-acting β2-agonists for at least 2 wk before the run-in. During the study, the only concurrent treatment permitted was ipratropium bromide as required.
Endobronchial biopsies were obtained 1 wk before randomization and again after 12 wk of treatment using standardized procedures, with bronchoscopy, biopsy processing (fixation and immunostaining), and quantification performed as described previously (11, 29–31). A previous study has demonstrated that bronchoscopies may be performed safely in patients with COPD (32). Three biopsies were taken from each of the lobar and subsegmental carinae. Those from the latter airway generation were used for the analyses described here unless they were of insufficient quality, in which case they were replaced by biopsies from the more proximal level. All the processing and histologic and molecular analyses of paraffin wax–embedded samples were performed centrally (Royal Brompton Hospital, London, UK). Induced sputum was collected at randomization, and after 8 and 13 wk of treatment. Sputum induction and processing were performed using previously published methods (33, 34). Total cell counts were made locally and differential counts and other sputum measurements were made at a central laboratory (Glenfield Hospital, Leicester, UK). Additional details on the methods for biopsy and sputum measurements are provided in the online supplement.
Measurements of FEV1 and FVC were made at each clinic visit, and safety was assessed by measurement of vital signs and adverse events.
The study was designed by a steering committee (N.C.B., I.D.P., M.J., P.K.J., and N.C.T.), who also approved the protocol. The sponsor (GlaxoSmithKline) funded and coordinated the study, and collected, held, and analyzed the data, which was only unblinded when data queries had been resolved, in accordance with International Conference on Harmonisation: efficacy 9 (ICH E9) guidelines. The steering committee and all authors had full access to all the data and vouch for the accuracy of the data and the data analysis. Data interpretation and writing of the manuscript were done predominantly by P.K.J., N.C.B., and I.D.P., with extensive input by all authors. The academic authors had final responsibility for the manuscript.
Biopsy CD8+ and CD68+ inflammatory cells and sputum neutrophils were chosen as coprimary endpoints. Sputum neutrophils were expressed as both total and differential counts; consequently, p values obtained were corrected for four coprimary endpoints. As data were not normally distributed, the changes from baseline for numbers of biopsy CD8+ or CD68+ cells and sputum neutrophils for each patient were ranked and analyzed using the Van Elteren extension to the Wilcoxon rank sum test (35, 36) stratified for current and former smokers. Median differences and confidence intervals were constructed using Hodges-Lehman estimates (37). Comparisons were made using all patients who had both baseline and endpoint values for biopsy and/or sputum. In addition, patients with no endpoint values who had experienced an exacerbation requiring treatment with oral corticosteroids or who were withdrawn due to perceived lack of efficacy were included in the analysis and assigned the lowest rank for change in endpoints. Additional statistical considerations are provided in the online supplement.
Secondary biopsy and sputum endpoints were analyzed similarly but with no adjustments to nominal significance levels. Post hoc analyses of percentage change were also performed, and the effects of smoking status on primary and secondary biopsy and sputum endpoints were investigated in a preliminary fashion, as the study was not powered initially for subgroup analyses. FEV1 and FVC were analyzed using analysis of covariance.
Details of patients screened, randomized, and withdrawn during the study are shown in Figure 1. Sixty-seven patients (55 male; mean age, 65 yr; range, 45–77 yr) were treated with salmeterol/fluticasone propionate and 73 (54 male; mean age, 64 yr; range 40–80 yr) with placebo. Active and placebo treatment groups were well matched for demography, smoking history, and baseline lung function (Table 1).
Placebo (n = 73) | Salmeterol/Fluticasone Propionate (n = 67) | |
|---|---|---|
| Mean (SD) age, yr | 63.9 (8.9) | 64.9 (7.6) |
| Male/female, % | 74/26 | 82/18 |
| Race: white/Asian, % | 100/0 | 99/1 |
| Current/former smokers, % | 59/41 | 63/37 |
| Mean (SD) pack-years | 44.0 (21.7) | 40.3 (24.5) |
| Median (range) pack-years | 40 (10–150) | 35 (12–177) |
| Mean (SD) % predicted FEV1* | 59 (10.8) | 58 (12.0) |
| Mean (SD) prebronchodilator FEV1, L | 1.68 (0.47) | 1.67 (0.44) |
| Mean (SD) post-bronchodilator FEV1/FVC, % | 58 (9.1) | 54 (8.9) |
| Mean (SD) reversibility to salbutamol, %† | 3.9 (3.4) | 3.9 (3.1) |
Using published criteria (33), 98.4% of sputum cytospins were evaluable. In addition, 93% of biopsies were considered evaluable. Counts were obtained from three biopsies for 89% or more of patients at baseline and 84% or more at the end of treatment. Each marker was counted throughout the study by a single observer. The error of repeat measurement for a single observer for biopsy area and cell counts (expressed as % coefficient of variation) for the primary biopsy parameters was approximately 3% and that between observers was approximately 6%. The majority of biopsies analyzed were those from the subsegmental carinae, and samples from the lobar carinae were used from nine patients only.
For inclusion in biopsy and sputum analyses, patients were required to have a baseline and an endpoint value. Not all randomized patients completed the study and some patients had missing endpoint samples (Figure 1). The numbers used for biopsy and sputum analyses are different because the patients with both evaluable baseline and endpoint samples for biopsy and sputum were not identical. In addition, two patients (one in each treatment group) with no endpoint values were assigned the lowest rank for change in endpoints and included in the analysis since they were withdrawn because of an exacerbation or perceived lack of efficacy.
Compared with placebo, salmeterol/fluticasone propionate significantly reduced the absolute number of biopsy CD8+ cells (p = 0.015; Table 2), a 36% difference (95% confidence interval [CI], 16%–56%) in favor of the combination treatment (p = 0.001; Figure 2). CD68+ cells increased in number in the placebo group and this was not altered by the combination. There was a progressive reduction in sputum neutrophil differential cell count with active treatment (Figure 3), which was significant compared with placebo at 13 wk (p = 0.037). A similar trend was seen for sputum total neutrophils, but the difference between treatments was not significant expressed as an absolute change (Table 2). However, expressed as a percentage of reduction from baseline, the difference in total neutrophils (53%) was significant (95% CI, 1%–111%; p = 0.046).
Figure 2. Median treatment differences in percentage change from baseline for biopsy endpoints. Hodges-Lehmann estimator of median treatment difference and 95% confidence interval presented. FP = fluticasone propionate; SALM = salmeterol.
[More] [Minimize]
Figure 3. Median (interquartile ranges) sputum neutrophil differential (%) through the study. White bars, baseline; light gray bars, 8 wk; dark gray bars, 13 wk. Week 13, p = 0.037 for treatment difference (between salmeterol/fluticasone propionate and placebo) in change from baseline. Week 8, p = not significant.
[More] [Minimize]Placebo | Salmeterol/Fluticasone Propionate | |||||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Baseline Median (Range*) | Week 12/13† Median (Range*) | Baseline Median (Range*) | Week 12/13† Median (Range*) | Treatment Difference (95% CI) ‡ | p Value | |||||||
| Primary Endpoints | ||||||||||||
| CD8+ cells/mm2, biopsy | 315.6 (174.9–473.6) | 285.6 (187.8–422.8) | 274.1 (173.6–538.1) | 180.9 (102.8–256.1) | −117.9 (−208.6 to −41.9) | 0.015§ | ||||||
| CD68+ cells/mm2, biopsy | 64.7 (36.8–123.6) | 107.4 (63.0–158.7) | 68.2 (35.7–142.6) | 105.2 (54.1–150.5) | −17.9 (−50.0 to 13.1) | 0.255§ | ||||||
| Neutrophils, sputum | ||||||||||||
| Differential, % | 81.00 (64.25–88.88) | 83.75 (76.00–90.50) | 80.25 (66.75–88.50) | 76.00 (55.00–93.00) | −8.50 (−15.25 to −1.75) | 0.037§ | ||||||
| Total numbers ×106 | 1.040 (0.367–4.802) | 2.069 (0.545–3.968) | 0.942 (0.324– 2.704) | 0.779 (0.372–3.235) | −0.606 (−1.554 to 0.038) | 0.130§ | ||||||
| Secondary Biopsy Endpoints | ||||||||||||
| CD45+ cells/mm2 | 706.1 (470.6–1,186.0) | 659.3 (465.4–975.9) | 782.2 (424.9–1,098.5) | 482.2 (294.9–657.5) | −285.3 (−488.9 to −100.1) | 0.006 | ||||||
| CD4+ cells/mm2 | 74.7 (36.0–255.2) | 94.2 (45.2–176.1) | 95.4 (47.6–235.1) | 42.8 (23.2–85.0) | −45.0 (−85.5 to −4.8) | 0.017 | ||||||
| Mast cells/mm2 | 155.8 (87.5–225.3) | 152.6 (102.0–208.4) | 128.6 (94.9–200.1) | 114.2 (67.0–153.3) | −28.6 (−59.4 to 2.2) | 0.083 | ||||||
| IFN-γ mRNA+ cells/mm2 | 38.7 (16.6–102.2) | 49.9 (11.1–94.7) | 37.3 (14.7–83.1) | 16.1 (8.4–32.1) | −29.7 (−55.8 to −3.6) | 0.026 | ||||||
| TNF-α mRNA+ cells/mm2 | 31.7 (14.2–64.5) | 32.6 (16.0–60.6) | 21.2 (12.7–75.1) | 14.2 (5.6–23.8) | −18.9 (−34.3 to −6.1) | 0.003 | ||||||
| Secondary Sputum Endpoints | ||||||||||||
| Total cells ×106 | 1.745 (0.619–6.334) | 2.190 (0.800–4.960) | 1.310 (0.480–4.026) | 1.220 (0.559–3.900) | −0.701 (−2.062 to 0.204) | 0.130 | ||||||
| Eosinophils | ||||||||||||
| Differential, % | 1.63 (0.63–3.25) | 1.50 (0.50–3.25) | 2.25 (0.75–5.00) | 0.75 (0.25–2.50) | −0.75 (−1.75 to 0.00) | 0.065 | ||||||
| Total numbers ×106 | 0.031 (0.007–0.075) | 0.029 (0.007–0.090) | 0.027 (0.009–0.078) | 0.017 (0.001–0.048) | −0.026 (−0.054 to −0.002) | 0.030 | ||||||
| IL-8, ng/ml | 74.3 (14.9– 226.5) | 46.2 (18.6–223.6) | 43.4 (16.3–159.3) | 26.8 (8.0–169.3) | −13.3 (−40.2 to 22.8) | 0.400 | ||||||
| ECP, ng/ml | 539.1 (255.6–1,287.0) | 669.2 (230.4–1,512.0) | 743.4 (190.8–1,602.0) | 494.1 (184.5–1,017.0) | −109.8 (−466.2 to 123.3) | 0.327 | ||||||
To investigate the individual variability in response, the proportions of patients achieving a 10, 25 or 50% change from baseline in CD8+ and CD68+ cells or a 10 or 35% decrease in neutrophils were tabulated (Table 3). The thresholds applied for differential neutrophil counts were defined by reference to data from Stanescu and colleagues (19) and those for biopsy endpoints were based on the distribution of the data because no clinically relevant limits have been defined. The results showed considerable individual variability, although the proportion of patients showing decreases in sputum neutrophils and in CD8+ cells was greater for salmeterol/fluticasone propionate than for placebo.
Placebo | Salmeterol/Fluticasone Propionate | |||
|---|---|---|---|---|
| Primary Endpoints | ||||
| Neutrophils, sputum | n = 59 | n = 50 | ||
| At least 10% reduction | 6 (10%) | 19 (38%) | ||
| At least 35% reduction | 2 (3%) | 5 (10%) | ||
| CD8+ cells, biopsy | n = 68 | n = 54 | ||
| Any reduction | 38 (56%) | 38 (70%) | ||
| At least 10% reduction | 34 (50%) | 38 (70%) | ||
| At least 25% reduction | 24 (35%) | 35 (65%) | ||
| At least 50% reduction | 9 (13%) | 24 (44%) | ||
| CD68+ cells, biopsy | n = 68 | n = 54 | ||
| Any increase | 49 (72%) | 35 (65%) | ||
| At least 10% increase | 47 (69%) | 33 (61%) | ||
| At least 25% increase | 41 (60%) | 30 (56%) | ||
| At least 50% increase | 34 (50%) | 27 (50%) | ||
| Selected Secondary Endpoints | ||||
| IFN-γ mRNA+ cells, biopsy | n = 68 | n = 53 | ||
| Any reduction | 38 (56%) | 38 (72%) | ||
| At least 10% reduction | 37 (54%) | 37 (70%) | ||
| At least 25% reduction | 32 (47%) | 34 (64%) | ||
| At least 50% reduction | 27 (40%) | 28 (53%) | ||
| TNF-α mRNA+, biopsy | n = 68 | n = 54 | ||
| Any reduction | 38 (56%) | 42 (78%) | ||
| At least 10% reduction | 36 (53%) | 41 (76%) | ||
| At least 25% reduction | 32 (47%) | 35 (65%) | ||
| At least 50% reduction | 21 (31%) | 29 (54%) | ||
Compared with placebo, there were significantly greater reductions by salmeterol/fluticasone propionate in numbers of biopsy CD45+ cells, an overall measure of the total number of leukocytes. CD4+ cells, TNF-α mRNA+ cells, and IFN-γ mRNA+ cells were reduced whether expressed as absolute changes (Table 2) or percentage changes (Figure 2). There were significant positive associations at baseline between the number of CD8+ cells and the numbers of TNF-α mRNA+ and IFN-γ mRNA+ cells at baseline (ρ = 0.404, p = 0.002, and ρ = 0.407, p = 0.002, respectively). Positive associations were also found between the reduction of CD8+ cells and the relative reductions of cells expressing genes for these two proinflammatory mediators after combination treatment (ρ = 0.386, p = 0.004, and ρ = 0.443, p = 0.001, respectively).
There was a strong trend toward a reduction in absolute numbers of mast cells by the combination, but this did not achieve statistical significance (p = 0.083). However, there was a significant difference when data were expressed as percentage change (p = 0.022). Although the ratio was highest in the group given the combination, no significant difference was seen between treatments for the ratio of CD8+ to CD4+ cells (median ratio, 3.1:1 for placebo, 3.5:1 for salmeterol/fluticasone propionate) as both CD8+ and CD4+ cells had been reduced by active treatment compared with placebo.
Reductions in sputum total cell counts were seen with salmeterol/fluticasone propionate after 8 wk of treatment and at endpoint, but there was no significant difference between active and placebo treatments (Table 2). The change in total numbers of sputum eosinophils was significant in favor of active treatment but changes in eosinophil differential counts did not reach significance. IL-8 and eosinophilic cationic protein (ECP) concentrations were not significantly changed by salmeterol/fluticasone propionate in comparison with placebo.
The directions of the response to combination treatment were generally similar in smokers and former smokers, albeit the reductive effect was generally greater in the former smokers (Figure E1 of the online supplement).
Patients treated with salmeterol/fluticasone propionate showed increases in mean prebronchodilator FEV1 at each visit, which were significantly greater (p < 0.001) than with placebo (Figure 4). The mean treatment difference at the end of the study was 173 ml (95% CI, 104–242 ml; p < 0.001). Similar results were seen for FVC (mean treatment difference at end of study 170 ml; 95% CI, 41–299 ml; p = 0.010).
Figure 4. Adjusted mean changes from baseline in prebronchodilator FEV1. Solid diamonds, placebo; solid squares, salmeterol/fluticasone propionate. Treatment differences analyzed by analysis of covariance (***p < 0.001).
[More] [Minimize]Post hoc evaluation of correlations between lung function (change in FEV1 as percentage of predicted normal and change in FVC) and primary and secondary biopsy parameters showed no convincing or significant correlations. A reduction in total biopsy inflammation (i.e., CD45+ cells expressed as absolute or percentage change) corresponded to an improvement in FEV1 percent predicted (ρ = 0.25, p = 0.065, and ρ = 0.26, p = 0.057, respectively), but these relationships appeared mainly to be driven by the outlier values. For salmeterol/fluticasone propionate, reduction in sputum percentage neutrophils showed a significant association with an increase in absolute FVC (ρ = −0.30, p < 0.05).
The bronchoscopy procedure was well accepted by most patients and only four patients (three at baseline and one at endpoint) experienced adverse events (nose bleed, cough and sore throat, dyspnea, high blood pressure) associated with this procedure. Both active and placebo treatments were well tolerated. Five patients (7%) treated with placebo and 15 (22%) treated with salmeterol/fluticasone propionate experienced events that were considered to be drug-related, with no event reported by more than four patients and all events reported were predictable effects of treatment. Oral candidiasis was the most frequently recorded adverse event on active treatment (6 vs. 1%). Fewer patients in the salmeterol/fluticasone propionate group than in the placebo group experienced a worsening of COPD symptoms that required any change in normal treatment (11 [16%] vs. 24 [33%]; p = 0.025). In six patients (9%) in the combination group and eight patients (11%) in the placebo group, antibiotic treatment was required to treat the worsening symptoms, with one patient in the combination group also hospitalized to treat the worsening symptoms. One other patient (1%) in the combination group and two patients (3%) in the placebo group received oral corticosteroids to treat worsening symptoms.
This is the first demonstration that a currently available treatment can reduce the exaggerated bronchial inflammation in COPD. The combination of inhaled salmeterol and fluticasone propionate significantly reduced the absolute numbers of biopsy (CD45+) leukocytes, CD8+ cells, and CD4+ cells together with decreases in cells expressing genes for the proinflammatory mediators IFN-γ and TNF-α. In the complementary airway compartment sampled by induced sputum, combination treatment significantly reduced sputum differential neutrophils and total eosinophils. Considering the only published longitudinal data for the relationship between sputum neutrophils and lung function (19), the difference of 8.5% in favor of combination treatment is likely to be clinically significant.
The broad spectrum of antiinflammatory effects was accompanied by significant improvements in lung function. The magnitude of the improvements seen in FEV1 was similar or greater than that seen in other studies of antiinflammatory treatment used in COPD (3, 30, 38, 39).
There are a number of potential mechanisms that may contribute to such improvement. Experimental studies have shown that CD8+ cells are able to recognize viral antigen expressed on the surface of lung epithelial cells in a class I–restricted manner and these cells may destroy host cells by apoptosis directly via release of perforins and granzymes (40) or indirectly via release of TNF-α and IFN-γ. TNF-α can stimulate epithelial cells to release chemoattractants for macrophages (41) or for neutrophils, acknowledged effector cells in COPD (42). IFN-γ is also associated with generation of emphysematous lesions experimentally (43). We speculate that the effects of combination treatment most likely to be associated with clinical benefit are centered on its reduction of the sputum neutrophil differential count and biopsy CD8+ cells and the associated mediators TNF-α and IFN-γ.
A potential criticism of our study is that comparison with the effects of corticosteroid or long-acting β2-agonist alone would have been additionally informative. Although we considered the merits of a four-arm study design, we rejected this for several reasons. We wished to conduct a proof-of-principle study to ascertain whether inhaled therapy can modify inflammation in COPD. This has never been shown previously. From a practical perspective, sufficiently powered biopsy studies are extremely demanding and, based on our power calculations using the data we had obtained previously in COPD, a four-arm study would have taken several years to complete. To be adequately powered, and to ensure sufficient patients with both baseline and end-of-treatment samples, each additional treatment group would have required the recruitment of a further 65 patients. As designed, the present study represents the largest biopsy study ever to be completed in COPD. Our rejection of a four-arm study design was also influenced by clear evidence from a randomized study in patients with COPD of similar severity (and in which analyses were performed in the same biopsy central laboratory) that inhaled fluticasone propionate alone at the same dose and for the same duration as that used here had no effect on the number of biopsy subepithelial CD8+, CD68+, or CD4+ cells (key outcome measures in the current study), although significant reductions of biopsy mast cells were seen as were differences in the epithelial CD8+:CD4+ cell ratio (44, 45). Similarly, most studies have shown no effect of either inhaled or oral corticosteroids on sputum neutrophils (20, 34, 46).
We acknowledge that we cannot exclude an effect of salmeterol alone. Although no biopsy studies of long-acting β2-agonist monotherapy have been conducted in patients with COPD, nonbronchodilator effects of salmeterol are well established in vitro (47), and in asthma, salmeterol and, more recently, formoterol have been shown to reduce biopsy and sputum neutrophils and associated markers (48, 49). We chose to study the salmeterol/fluticasone propionate combination rather than salmeterol alone because combination treatment has been shown clinically to have greater effects than monotherapy in patients with COPD and salmeterol has been considered to be effective clinically because of its bronchodilator rather than antiinflammatory effects. Therefore, we considered that salmeterol/fluticasone propionate was the inhaled intervention most likely to have antiinflammatory effects in vivo. It is possible, however, that the long-acting β2-agonist may have contributed to the antiinflammatory effect, and we acknowledge that further research is needed to elucidate the antiinflammatory effects of β2-agonists alone in COPD.
Synergy between corticosteroid and long-acting β2-agonist might have enhanced the β2-agonist effect (50), resulting in a general increase of intracellular cAMP, such as that achieved by the selective PDE4 inhibitor cilomilast, which has been shown to decrease CD8+ cells and macrophages in patients with COPD (30). Considering our present results and those of previous studies, it appears that combination therapy has antiinflammatory effects not seen with inhaled corticosteroids alone. Compared with monotherapy, enhanced effects for the combination are consistent not only with the clinical data (3–5) but also with in vitro data, which demonstrate the synergistic interaction between salmeterol and fluticasone propionate (51, 52). For example, rhinovirus-induced chemokine production is decreased with the combination (51) and there is reduced production of TNF-α from alveolar macrophages (52) compared with corticosteroid alone. In these studies, salmeterol alone had no effect in either model. Similar interactions between fluticasone propionate and salmeterol in vivo are a likely explanation for the effects seen in our study. Thus, salmeterol could be enhancing the effect of the inhaled corticosteroid, or alternatively, the β2-agonist effect could be enhanced by fluticasone propionate (50).
Bronchoscopy and biopsy and sputum access the proximal bronchi only and we have not been able to assess the effects of combination therapy on small airways, the site considered to contribute most to reduced lung function in COPD (14). However, the predominance of CD8+ cells is seen in both proximal and small airways and the correlation between this cell type and impaired lung function is similar in both large and small airways and lung parenchyma (11–14). This and recent data (53) suggest that similar processes of inflammation and airway wall thickening appear to be taking place in both large and small airways. Thus, biopsy samples of large airways may be a reasonable surrogate for assessing the potential effects of treatment on small airway inflammation and remodeling. Furthermore, it is likely that the formulation of the combination treatment we used would have reached the small airways as the Diskus device delivers approximately 20% of its dose as a fine-particle, respirable fraction (i.e., particles < 5 μm) (54, 55), which would be delivered throughout the lung. Although the associations between the extent, type, and site of inflammation and lung function decline remain to be determined (56), the improvements in lung function seen in our study also indicate that sites affecting resistance to airflow were accessed by treatment.
Increases in sputum neutrophils, biopsy CD68+ cells (monocyte/macrophage), and other cells were observed in the placebo control group. An increase of CD68+ cells with placebo was also seen in a previous biopsy study of patients with stable COPD (30). The reasons for these increases are presently unclear. As patients were recruited when their disease was stable, deterioration in their condition was a likely outcome when no active treatment was given. This requires further study, as do the effects of smoking status on the results seen. The present study was not powered to investigate effects by smoking status, but a preliminary post hoc analysis indicated that there are antiinflammatory effects seen in both current and former smokers, with a trend for the magnitude of the effect to be greater in former smokers, although there is variability in response.
In summary, we report here results of the largest multicenter clinical trial yet in COPD involving repeat bronchial biopsy and sputum samples in patients with disease of severity Global Initiative for Chronic Obstructive Lung Disease stage II or III. The data demonstrate broad-spectrum antiinflammatory effects of a currently available treatment, salmeterol/fluticasone propionate combination. We prospectively identified key parameters, biopsy CD8+ cells and sputum neutrophils, shown previously to relate to severity and disease progression and show antiinflammatory effects of the combination of a magnitude likely to be clinically significant. These findings may support the consideration that combination treatment applied earlier than currently proposed in guidelines may be helpful. They provide a rationale for further investigation of salmeterol/fluticasone propionate in disease progression and survival. The TORCH (Toward a Revolution in COPD Health) trial (57) is currently investigating the effect of salmeterol/fluticasone propionate combination and its components on all-cause mortality. Its results will determine whether the same combination that we have shown attenuates the exaggerated inflammation of COPD will also improve the prognosis of patients with this chronic condition.
The authors thank each subject who consented to be part of this investigation and all clinical investigators who recruited and cared for the study patients. They thank Linda Peachey and Hui Gong (Royal Brompton Hospital) who processed and sectioned the biopsy samples. They are indebted to Nicola Thomas, GlaxoSmithKline, who analyzed the data; to Angela Barbaro, Data Management, GlaxoSmithKline; and to Dr. Sam Lim, Eva Gomez, and Richard Follows, GlaxoSmithKline, who helped to organize and coordinate the study; and to Charlotte Huskisson for technical assistance with the preparation of the manuscript.
| 1. | Murray C, Lopez A. Alternative projections of mortality and disability by cause 1990–2020: global burden of disease study. Lancet 1997;349:1498–1504. |
| 2. | Global Initiative for Chronic Obstructive Lung Disease. Global strategy for the diagnosis, management, and prevention of chronic obstructive lung disease. NHLBI/WHO workshop report. Bethesda, National Heart Lung and Blood Institute, April 2001; update of the Management sections, GOLD website. Available from: www.goldcopd.com (accessed March 2, 2005). |
| 3. | Calverley P, Pauwels R, Vestbo J, Jones P, Pride N, Gulsvik A, Anderson J, Maden C. Combining salmeterol and fluticasone in the treatment of chronic obstructive pulmonary disease. Lancet 2003;361:449–456. |
| 4. | Szafranski W, Cukier A, Ramirez R, Menga G, Sansores R, Nahabedian S, Peterson S, Olsson H. Efficacy and safety of budesonide/formoterol in the management of chronic obstructive pulmonary disease. Eur Respir J 2003;21:74–81. |
| 5. | Mahler DA, Wire P, Horstman D, Chang C-N, Yates J, Fischer T, Shah T. Effectiveness of fluticasone propionate and salmeterol combination delivered via the Diskus device in the treatment of chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2002;166:1084–1091. |
| 6. | Costabel U, Bross KJ, Reuter C, Ruhle KH, Matthys H. Alterations in immunoregulatory T-cell subsets in cigarette smokers: a phenotypic analysis of bronchoalveolar and blood lymphocytes. Chest 1986;90:39– 44. |
| 7. | Turato G, Di Stefano A, Maestrelli P, Mapp CE, Ruggieri MP, Roggeri A, Fabbri LM, Saetta M. Effect of smoking cessation on airway inflammation in chronic bronchitis. Am J Respir Crit Care Med 1995;152:1262–1267. |
| 8. | Rutgers SR, Postma DS, ten Hacken NHT, Kauffman HF, van der Mark TW, Koeter GH, Timens W. Ongoing airway inflammation in patients with COPD who do not currently smoke. Thorax 2000;55:12–18. |
| 9. | Jeffery PK. Comparison of the structural and inflammatory features of COPD and asthma. Chest 2000;117:251s–260s. |
| 10. | Fabbri LM, Romagnoli M, Corbetta L, Casoni G, Busljetic K, Turato G, Ligabue G, Ciaccia A, Saetta M, Papi A. Differences in airway inflammation in patients with fixed airflow obstruction due to asthma or chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2003;167:418–424. |
| 11. | O'Shaughnessy T, Ansari TW, Barnes NC, Jeffery PK. Inflammation in bronchial biopsies of subjects with chronic bronchitis: inverse relationship of CD8+ T lymphocytes with FEV1. Am J Respir Crit Care Med 1997;155:852–857. |
| 12. | Lams BEA, Sousa AR, Rees PJ, Lee TH. Subepithelial immunopathology of large airways in smokers with and without chronic obstructive pulmonary disease. Eur Respir J 2000;15:512–516. |
| 13. | Saetta M, Di Stefano A, Turato G, Facchini FM, Corbino L, Mapp CE, Maestrelli P, Ciaccia A, Fabbri LM. CD8+ T-lymphocytes in peripheral airways of smokers with chronic obstructive pulmonary disease. Am J Respir Crit Care Med 1998;157:822–826. |
| 14. | Hogg JC, Chu F, Utokaparch S, Woods R, Elliott WM, Buzatu L, Cherniack RM, Rogers RM, Sciurba FC, Coxson HO, et al. The nature of small-airway obstruction in chronic obstructive pulmonary disease. N Engl J Med 2004;350:2645–2653. |
| 15. | Retamales I, Elliott WM, Meshi B, Coxson HO, Pare PD, Sciurba FC, Rogers RM, Hayashi S, Hogg JC. Amplification of inflammation in emphysema and its association with latent adenoviral infection. Am J Respir Crit Care Med 2001;164:469–473. |
| 16. | Turato G, Zuin R, Miniati M, Baraldo S, Rea F, Beghe B, Monti S, Formichi B, Boschetto P, Harari S, et al. Airway inflammation in severe chronic obstructive pulmonary disease: relationship with lung function and radiologic emphysema. Am J Respir Crit Care Med 2002;166:105–110. |
| 17. | Ronchi MC, Piragino C, Rosi E, Amendola M, Duranti R, Scano G. Role of sputum differential cell count in detecting airway inflammation in patients with chronic bronchial asthma or COPD. Thorax 1996;51:1001–1004. |
| 18. | Riise GC, Ahlstedt S, Larsson S, Enander I, Jones I, Larsson P, Andersson B. Bronchial inflammation in chronic bronchitis assessed by measurement of cell products in bronchial lavage. Thorax 1995;5:360–365. |
| 19. | Stanescu D, Sanna A, Veriter C, Kostianev S, Calcagni PG, Fabbri LM, Maestrelli P. Airways obstruction, chronic expectoration, and rapid decline of FEV1 in smokers are associated with increased levels of sputum neutrophils. Thorax 1996;51:267–271. |
| 20. | Keatings VS, Jatakanon A, Worsdell YM, Barnes PJ. Effects of inhaled and oral glucocorticoids on inflammatory indices in asthma and COPD. Am J Respir Crit Care Med 1997;155:542–548. |
| 21. | Barnes NC, Qiu Y, Pavord I, Parker D, Johnson M, Thomson N, Jeffery PK. Salmeterol/fluticasone propionate (SFC): anti-inflammatory effects in COPD [abstract]. Proc Am Thorac Soc 2005;2:A543. |
| 22. | Zhu J, Qiu Y, Barnes NC, Johnson M, Pavord I, Jeffery PK. The effect of salmeterol/fluticasone propionate (SFC) on pro-inflammatory gene expression in COPD [abstract]. Proc Am Thorac Soc 2005;2:A127. |
| 23. | Qiu Y, Parker D, Barnes NC, Johnson M, Pavord I, Jeffery PK. The effect of salmeterol/fluticasone propionate (SFC) on eosinophils and mast cells in COPD [abstract]. Proc Am Thorac Soc 2005;2:A132. |
| 24. | Qiu Y, Davis P, Zhu J, Peachey L, Barnes NC, Pavord I, Johnson M, Jeffery PK. Effects of salmeterol/fluticasone propionate (SFC) on airway inflammation in COPD: a placebo-controlled study of endobronchial biopsies. Eur Respir J 2005;26:203S. |
| 25. | Qiu Y, Davis P, Zhu J, Peachey L, Barnes NC, Pavord I, Johnson M, Jeffery PK. Anti-inflammatory effects of salmeterol/fluticasone propionate (SFC) on airway T-lymphocyte populations in COPD. Eur Respir J 2005;26:203S. |
| 26. | Pavord I, Parker D, Barnes NC, Johnson M, Jeffery PK. Effects of salmeterol/fluticasone propionate (SFC) on airway inflammation in induced sputum in COPD. Eur Respir J 2005;26:203S. |
| 27. | Barnes NC, Qiu Y, Zhu J, Pavord I, Johnson M, Thomson NC, Jeffery PK. Anti-inflammatory effects of salmeterol/fluticasone propionate (SFC) by smoking status in COPD. Eur Respir J 2005;26:204S. |
| 28. | Barnes NC, Qiu YS, Pavord I, Parker D, Johnson M, Thomson M, Jeffery PK. Clinical improvement with salmeterol/fluticasone propionate is associated with reduction in inflammation in patients with COPD. Eur Respir J 2005;26:717S. |
| 29. | Jeffery P, Holgate S, Wenzel S. Methods for the assessment of endobronchial biopsies in clinical research: application to studies of pathogenesis and the effects of treatment. Am J Respir Crit Care Med 2003;168:1s–17s. |
| 30. | Gamble E, Grootendorst DC, Brightling CE, Troy S, Qiu Y, Parker D, Matin D, Majumdar S, Vignola AM, Kroegel C, et al. Antiinflammatory effects of the phosphodiesterase-4 inhibitor cilomilast (Ariflo) in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2003;168:976–982. |
| 31. | Zhu J, Qiu YS, Majumdar S, Gamble E, Matin D, Turato G, Fabbri LM, Barnes N, Saetta M, Jeffery PK. Exacerbations of bronchitis. Bronchial eosinophilia and gene expression for interleukin-4, interleukin-5, and eosinophil chemoattractants. Am J Respir Crit Care Med 2001;164:109–116. |
| 32. | Hattotuwa K, Gamble EA, O'Shaughnessy T, Jeffery PK, Barnes NC. Safety of bronchoscopy, biopsy, and BAL in research patients with COPD. Chest 2002;122:1909–1912. |
| 33. | Ethimiadis A, Spanevello A. Methods of sputum processing for cell counts, immunocytochemistry and in situ hybridization. Eur Respir J 2002;20:19s–23s. |
| 34. | Brightling CE, Monterio W, Ward R, Parker D, Morgan MD, Wardlaw AJ, Pavord ID. Sputum eosinophilia and the short-term response to prednisolone in chronic obstructive pulmonary disease: a randomized controlled trial. Lancet 2000;356:1480–1485. |
| 35. | Lehmann E. Nonparametrics: statistical methods based on ranks. San Francisco, CA: Holden-Day; 1975. pp. 132–137. |
| 36. | van Elteren P. On the combination of independent two-sample tests of Wilcoxon. Bull Int Stat Inst 1960;37:351–361. |
| 37. | Hollander M, Wolfe DA. Nonparametric statistical methods. New York: John Wiley and Sons; 1973. |
| 38. | Rabe KF, Bateman ED, O'Donnell D, Witte S, Bredenbröker D, Bethke TD. Roflumilast—an oral anti-inflammatory treatment for chronic obstructive pulmonary disease: a randomised controlled trial. Lancet 2005;366:563–571. |
| 39. | Rama FSF, Jardin JR, Atallah A, Castro AA, Mazzini R, Goldstein R, Lacasse Y, Cendon S. Efficacy of theophylline in people with stable chronic obstructive pulmonary disease: a systematic review and meta-analysis. Respir Med 2005;99:135–144. |
| 40. | Chrysofakis G, Tzanakis N, Kyriakoy D, Tsoumakidou M, Tsiligianni I, Klimathianaki M, Siafakas NM. Perforin expression and cytotoxic activity of sputum CD8+ lymphocytes in patients with COPD. Chest 2004;125:71–76. |
| 41. | Zhao MQ, Stoler MH, Liu AN, Wei B, Soguero C, Hahn YS, Enelow RI. Alveolar epithelial cell chemokine expression triggered by antigen-specific cytolytic CD8+ T cell recognition. J Clin Invest 2000;106:49–58. |
| 42. | Stockley RA. Neutrophils and the pathogenesis of COPD. Chest 2002;121:151s–155s. |
| 43. | Wang Z, Zheng T, Zhu Z, Homer RJ, Riese RJ, Chapman HA, Shapiro SD, Elias JA. Interferon gamma induction of pulmonary emphysema in the adult murine lung. J Exp Med 2000;192:1587–1600. |
| 44. | Hattotuwa KL, Gizycki M, Ansari TW, Gizycki M, Jeffery PK, Barnes NC. The effects of inhaled fluticasone on airway inflammation in chronic obstructive pulmonary disease: a double-blind placebo-controlled biopsy study. Am J Respir Crit Care Med 2002;165:1592–1596. |
| 45. | Gizycki M, Hattotuwa K, Barnes N, Jeffery PK. Effects of fluticasone propionate on inflammatory cells in COPD: an ultrastructural examination of endobronchial biopsy tissue. Thorax 2002;57:799–803. |
| 46. | Brightling CE, McKenna S, Hargadon B, Birring S, Green R, Siva R, Berry M, Parker D, Monteiro W, Pavord ID, et al. Sputum eosinophilia and the short-term response to inhaled mometasone in chronic obstructive pulmonary disease. Thorax 2005;60:193–198. |
| 47. | Johnson M, Rennard S. Alternative mechanisms for long-acting β2-adrenergic agonists in COPD. Chest 2001;120:258–270. |
| 48. | Jeffery PK, Venge P, Gizycki MJ, Egerod I, Dahl R, Faurschou P. Effects of salmeterol on mucosal inflammation in asthma: a placebo-controlled study. Eur Respir J 2002;20:1378–1385. |
| 49. | Maneechotesuwan K, Essilfie-Quaye S, Meah S, Kelly C, Kharitinov SA, Adcock IM, Barnes PJ. Formoterol attenuates neutrophilic airway inflammation in asthma. Chest 2005;128:1936–1942. |
| 50. | Adcock IM, Maneechotesuwan K, Usmani O. Molecular interactions between glucocorticoids and long-acting β2 agonists. J Allergy Clin Immunol 2002;110:S261–S268. |
| 51. | Edwards M, Johnson M, Johnson S. Synergistic effects of salmeterol and fluticasone in human rhinovirus induced pro-inflammatory cytokine production [abstract]. Am J Respir Crit Care Med 2003;167:A399. |
| 52. | Seeto LJ, Burgess JK, Johnson PR, Black JL, Roth M, Lim S. Effect of fluticasone & salmeterol on human alveolar macrophage cytokine production in patients with chronic obstructive pulmonary disease (COPD) [abstract]. Am J Respir Crit Care Med 2003;167:A318. |
| 53. | Nakano Y, Wong JC, de Jong PA, Buzatu L, Nagao T, Coxson HO, Elliott WM, Hogg JC, Pare PD. The prediction of small airway dimensions using computed tomography. Am J Respir Crit Care Med 2005;171:142–146. |
| 54. | Malley TR, Ashurst IC, Daniels GE, Prime D, Stock NJ, Haywood PA. Fine particle mass of a dry powder inhaler device containing a combination of salmeterol and fluticasone propionate. Eur Respir J 1998;12:8s. |
| 55. | Hill LS, Slater AL. A comparison of the performance of two modern multidose dry powder asthma inhalers. Respir Med 1998;92:105–110. |
| 56. | Lapperre TS, Snoeck-Stroband JB, Gosman MME, Stolk J, Sont JK, Jansen DF, Kerstjens HAM, Postma DS, Sterk PJ. Dissociation of lung function and airway inflammation in chronic obstructive pulmonary disease. Am J Respir Crit Care Med 2004;170:499–504. |
| 57. | Vestbo J and the TORCH Study Group. The TORCH (Towards a Revolution in COPD Health) survival study protocol. Eur Respir J 2004;24:206–210. |
