|
|
in Refractory Asthma
Background The development of tumor necrosis factor
Methods We measured markers of TNF-
Results As compared with patients with mild-to-moderate asthma and controls, patients with refractory asthma had increased expression of membrane-bound TNF-
Conclusions Patients with refractory asthma have evidence of up-regulation of the TNF-
(TNF-) antagonists has made it feasible to investigate the role of this cytokine in refractory asthma. activity on peripheral-blood monocytes in 10 patients with refractory asthma, 10 patients with mild-to-moderate asthma, and 10 control subjects. We also investigated the effects of treatment with the soluble TNF- receptor etanercept (25 mg twice weekly) in the patients with refractory asthma in a placebo-controlled, double-blind, crossover pilot study. , TNF- receptor 1, and TNF-–converting enzyme by peripheral-blood monocytes. In the clinical trial, as compared with placebo, 10 weeks of treatment with etanercept was associated with a significant increase in the concentration of methacholine required to provoke a 20 percent decrease in the forced expiratory volume in one second (FEV1) (mean difference in doubling concentration changes between etanercept and placebo, 3.5; 95 percent confidence interval, 0.07 to 7.0; P=0.05), an improvement in the asthma-related quality-of-life score (by 0.85 point; 95 percent confidence interval, 0.16 to 1.54 on a 7-point scale; P=0.02), and a 0.32-liter increase in post-bronchodilator FEV1 (95 percent confidence interval, 0.08 to 0.55; P=0.01). axis. (ClinicalTrials.gov number, NCT00276029
[ClinicalTrials.gov]
.)
(TNF-) is a pleiotropic inflammatory cytokine expressed in increased amounts by mast cells6 and present in increased concentrations in bronchoalveolar fluid from the airways of patients with asthma.7 It has a number of properties that might be relevant to refractory asthma.8,9,10 Interest in the role of TNF- in refractory asthma has been increased by a study showing increased concentrations of TNF- in bronchoalveolar-lavage fluid from patients with more severe asthma11 and by an uncontrolled study showing that treatment with the recombinant soluble TNF- receptor etanercept markedly improved airway hyperresponsiveness in patients with refractory asthma.11
The biologic activity of TNF- is mediated by the 26-kD transmembrane precursor protein12 (membrane-bound TNF-) as well as the 17-kD cleavage product, free TNF-.13 This cleavage is principally mediated by TNF-–converting enzyme,14 and the free TNF- subsequently forms highly active homotrimers,15 which interact with two distinct TNF- receptors on the cell surface.16 The bioactivity of TNF- is therefore likely to be reflected by increases in membrane-bound TNF-, TNF-–converting enzyme, free TNF-, cell-surface receptors, and soluble receptors. We tested the hypothesis that the TNF- axis is up-regulated in patients with refractory asthma by measuring the expression of membrane-bound TNF-, TNF- receptors, and TNF-–converting enzyme by peripheral-blood monocytes. We also performed a small, randomized, double-blind, placebo-controlled, crossover pilot study of the effects of treatment with etanercept, an agent active against TNF-, on airway hyperresponsiveness and the asthma-related quality of life in patients with refractory asthma.
Methods
Subjects
All patients with asthma had clinical features consistent with the presence of asthma and at least one of the following objective measures of airway hyperresponsiveness and variable airflow obstruction: the concentration of methacholine required to provoke a 20 percent decrease (PC20) in the forced expiratory volume in one second (FEV1) was less than 8 mg per milliliter, the FEV1 increased by at least 15 percent after the inhalation of 200 µg of albuterol, or the variation in peak flow, expressed as a percentage of the mean, exceeded 20 percent over a period of 14 days.
Patients with refractory asthma, recruited from the difficult-asthma clinic at Glenfield Hospital, Leicester, United Kingdom, met the criteria of the American Thoracic Society for this condition,17 with the exception that the daily dose of inhaled corticosteroids required to meet the definition was modified to more than 2000 µg of beclomethasone or its equivalent to reflect European practice.4 Patients met at least one major and two minor criteria for refractory asthma; all were considered to be compliant with treatment. Our assessment was based on the measurement of serum prednisolone, cortisol, and theophylline concentrations; an assessment at each patient's home by a consultant pharmacist; and an analysis of primary care records on the issuing and filling of prescriptions. We excluded patients who were thought to be symptomatic because of uncontrolled coexisting conditions such as rhinitis and gastroesophageal reflux disease. Patients were also excluded if they had any of the following: recent contact with a patient with pulmonary tuberculosis, a personal history of tuberculosis, any radiologic features suggestive of current or previous tuberculosis, or a grade III or IV tuberculin (Heaf) test.
All patients classified as having mild-to-moderate asthma met the Global Initiative for Asthma18 criteria for intermittent or mild persistent asthma. All patients with refractory asthma met the Global Initiative for Asthma criteria for severe persistent asthma.
Control subjects had no respiratory symptoms, had normal spirometric values, and had a PC20 of more than 16 mg of methacholine per milliliter. All controls and patients with asthma were currently nonsmokers, with a smoking history of less than 5 pack-years, and none reported having a lower respiratory tract infection in the three months before the study. All subjects provided written informed consent to participate in the study. The protocol was approved by the Leicestershire and Rutland regional ethics committee.
Measurements
Peripheral-blood monocytes were separated from 20 ml of peripheral blood that had been treated with 500 U of heparin with the use of a density–gradient method (Histopaque 1077, Sigma). Cells were counted, incubated with antibodies labeled with fluorochrome (phycoerythrin or fluorescein isothiocyanate [FITC]) targeting CD14 (phycoerythrin), TNF receptor 1 (FITC), TNF receptor 2 (FITC), TNF-–converting enzyme (FITC), or membrane-bound TNF- (FITC) (all from R&D) or an equal concentration of an isotype-matched control — mouse IgG1 (phycoerythrin), mouse IgG2A (FITC), or mouse IgG1 (FITC) (all from R&D) — and subjected to cytometry with the use of a laser flow cytometer (FACScan, Becton Dickinson) as described previously.19 The geometric mean fluorescence for membrane-bound TNF-, TNF receptor 1, and TNF-–converting enzyme was calculated for CD14+ cells with the use of Cellquest software (Becton Dickinson).
Single-flow exhaled nitric oxide concentrations were recorded at a rate of 50 ml per second as previously described,20 and the alveolar nitric oxide concentration was derived from measurements at rates of 10, 30, 50, 100, and 200 ml per second.4 FEV1, forced vital capacity (FVC), and forced expiratory flow between 25 percent and 75 percent of FVC (FEF25–75) were measured with the use of a rolling-seal spirometer (Vitalograph). The PC20 was computed from the methacholine dose–response curve (the change in FEV1 in relation to the methacholine concentration) by linear interpolation on a log scale. Patients initially inhaled normal saline by tidal breathing for two minutes, followed by 0.03 mg of methacholine per milliliter, and concentrations were then doubled up to a maximum of 16 mg per milliliter until the FEV1, measured at intervals of up to five minutes, decreased by at least 20 percent. Patients whose FEV1 decreased by more than 20 percent after the inhalation of normal saline were assigned a PC20 value of 0.01 mg per milliliter.21 Symptoms were measured with the use of three 100-mm visual-analogue scales representing cough, wheezing, and breathlessness; higher numbers denoted more symptoms.22 The asthma-related quality of life was measured by means of the Juniper asthma quality-of-life scale; scores can range from 1 to 7 points, with higher values representing a better quality of life.23 Sputum was induced, and samples were processed as previously described.22 Details of the measurement of interleukin-8, cysteinyl leukotrienes (leukotriene C4, D4, and E4), eosinophilic cationic protein, and histamine in sputum are provided in the Supplementary Appendix, available with the full text of this article at www.nejm.org.
Crossover Trial
Patients with refractory asthma were enrolled in a randomized, double-blind, crossover study comparing the effect of 10 weeks of treatment with etanercept at the dose used in a previously reported uncontrolled study11 and placebo. An outline of the study is presented in Figure 1. Placebo (1 ml of 0.9 percent saline) or etanercept (25 mg made into a 1-ml solution with the addition of the manufacturer's diluent) was administered subcutaneously twice weekly. Etanercept was purchased by the investigators with departmental research funds, and the manufacturer had no role in the study. The order of treatment was determined with the use of a randomization sequence prepared with use of a random-number generator. Preparation and storage of the study drug were overseen by the Glenfield Hospital pharmacy. The doses of all other asthma medications were kept constant during the study.
|
We compared the geometric mean fluorescence of membrane-bound TNF-, TNF-–converting enzyme, TNF- receptor 1, and TNF- receptor 2 with that of isotype-matched controls and expressed the results as a ratio. Comparisons between groups were made with use of one-way analysis of variance, with Tukey's post hoc test for the three individual-group differences.
The prespecified primary outcome measures for the etanercept trial were the difference in the change in the PC20 from 0 to 10 weeks between the placebo and etanercept treatment phases21 and the difference in the change in the asthma quality-of-life score from 0 to 10 weeks between the treatment phases. The change in PC20 was expressed as a doubling concentration, calculated as the difference in the log PC20 after and before treatment divided by log 2. Differences between and within treatment phases were normally distributed, as assessed by the Kolmogorov–Smirnov test; they were compared with the use of paired t-tests. The effect of treatment period and order was analyzed by means of analysis of covariance. The analysis was conducted according to the intention to treat: patients who withdrew during a treatment phase for asthma-related reasons were assigned a value equal to the worst net outcome for that treatment phase; patients who withdrew for reasons unrelated to asthma were assigned the last recorded spirometric values and, for other measures, a value equal to that measured at their five-week assessment if they had completed five weeks of treatment or no change if they had not. Patients who withdrew during the washout phase were assigned baseline values from the previous treatment phase and were assumed to have had no net change during the second treatment phase. We also performed a post hoc analysis of the net change in the population that completed both phases of the study (per-protocol population). An earlier, uncontrolled study estimated a treatment effect on the PC20 of 2.5 doubling concentrations.11 Our study had a statistical power of 90 percent to detect a change of two doubling concentrations in the PC20, assuming a standard deviation of one doubling concentration within subjects.21
Prespecified secondary outcome measures were the net change in post-bronchodilator FEV1, FEF25–75, and FVC; symptom scores; exhaled nitric oxide concentrations; computed alveolar nitric oxide concentrations; differential inflammatory-cell counts in sputum; and mediator concentrations in sputum supernatant.
Multiple independent linear regression analysis was used to explore the relationship between the baseline expression of membrane-bound TNF- by peripheral-blood monocytes and the net change in primary outcomes.
Results
Baseline Characteristics
The baseline characteristics of the subjects are provided in Table 1 and Table 2. Patients with refractory asthma had a significantly lower FEV1 and FEV1 as a percentage of the predicted value, as well as FEV1:FVC ratio, after receipt of a bronchodilator than did controls or patients with mild-to-moderate asthma. There was a trend toward an increased neutrophil count in sputum from patients with refractory asthma (P=0.10). The alveolar nitric oxide concentration was significantly higher in patients with refractory asthma than in those with mild-to-moderate asthma or controls (Table 1).
|
|
Expression
The mean (±SE) ratio of fluorescence for antibody against membrane-bound TNF- to that for the isotype-matched control was 8.9±0.9 in patients with refractory asthma, 3.8±0.7 in control subjects (mean difference between groups, 5.1; 95 percent confidence interval, 2.5 to 7.7; P<0.001) (Figure 2), and 3.3±0.4 in patients with mild-to-moderate asthma (mean difference between this group and the group with refractory asthma, 5.5; 95 percent confidence interval, 2.7 to 8.4; P<0.001). The ratio of the expression of TNF- receptor 1 (P<0.001) and TNF-–converting enzyme (P<0.001), but not of TNF- receptor 2 (P=0.48), was also significantly higher in the group with refractory asthma than in the other groups (Table 1).
|
One patient withdrew for personal reasons during the washout phase after etanercept treatment owing to a change in employment, and one patient withdrew during week 4 of the second treatment phase (etanercept) owing to the death of a relative and the development of a cough productive of sputum associated with repeated isolation of Haemophilus influenzae despite a course of oral antibiotics (Table 2 and Figure 3). Data from the latter patient were treated as an asthma-related withdrawal. There were no other adverse effects. Neither the treatment period nor the order of treatment influenced values before treatment or the change in primary outcome measures (Table 2 and Figure 3).
|
As compared with placebo, etanercept significantly reduced the ratio of fluorescence of peripheral-blood monocytes for membrane-bound TNF- (geometric mean change, –6.9 vs. –0.1; mean difference between groups, 6.8; 95 percent confidence interval, 0.5 to 13.1; P=0.04). Etanercept treatment was associated with a progressive improvement in the PC20, as reflected by a doubling concentration of methacholine of 2.3 at 10 weeks, as compared with –1.2 after 10 weeks of placebo (mean difference, 3.5 doubling concentrations; 95 percent confidence interval, 0.07 to 7.0; P=0.05) (Figure 3). The Juniper asthma quality-of-life score improved by 0.84 with etanercept, as compared with a decrease of 0.02 with placebo (mean difference, 0.85; 95 percent confidence interval, 0.16 to 1.54; P=0.02) (Figure 3). The net changes in the PC20 and the asthma quality-of-life score with etanercept treatment, as compared with placebo, were both independently associated with the baseline expression of membrane-bound TNF- by peripheral-blood monocytes (adjusted R2, 0.73; P=0.004) and the treatment-associated change in expression from baseline (adjusted R2, 0.56; P=0.02). The net changes in the PC20 and the asthma quality-of-life score remained significant when the analysis was restricted to the per-protocol population (P=0.04 for both comparisons).
Secondary Outcomes
Secondary outcomes are shown in Table 3 and the Supplementary Appendix; the changes in FEV1 during the 10 weeks of etanercept and placebo administration are shown in Figure 3C. There was a greater improvement in post-bronchodilator FEV1 in the etanercept group than in the placebo group, and this difference became significant by week 5 and was 0.32 liter (95 percent confidence interval, 0.08 to 0.55; P=0.01) by week 10 (Table 3 and Figure 3). The total symptom score at 10 weeks decreased by 48 mm (out of a possible 300 mm) after treatment with etanercept, as compared with 9 mm with placebo (mean difference, 39; 95 percent confidence interval, 7 to 71; P=0.01). There were no significant differences between groups in single-flow nitric oxide concentration, calculated alveolar nitric oxide concentration, sputum total or differential cell counts, or sputum eosinophilic cationic protein, interleukin-8, or cysteinyl leukotriene concentrations (see the Supplementary Appendix). The mean histamine concentration in sputum decreased from 36.1 to 14.0 ng per milliliter with etanercept and increased from 37.0 to 41.3 ng per milliliter with placebo (mean difference in the change in histamine concentration, 26 ng per milliliter; 95 percent confidence interval, 5 to 48; P=0.02) (Table 3).
|
We have demonstrated that the TNF- axis is up-regulated in patients with refractory asthma, as evidenced by the increased expression of membrane-bound TNF-, TNF receptor 1, and TNF-–converting enzyme by peripheral-blood monocytes. Antagonism of TNF- with 10 weeks of etanercept therapy significantly reduced the expression of membrane-bound TNF- by peripheral-blood monocytes and improved the PC20, the asthma-related quality of life, FEV1, and symptom scores, as compared with placebo. The baseline expression of membrane-bound TNF- by peripheral-blood monocytes and the extent to which it was reduced by etanercept treatment were independently associated with the net improvement in both primary outcome measures.
We chose to assess TNF- activity by measuring the expression of membrane-bound TNF-, cell-surface receptor, and TNF-–converting enzyme by peripheral-blood monocytes because monocytes and macrophages are an important source of TNF- and the technique is noninvasive and suitable for repeated measurements after etanercept treatment. Membrane-bound TNF- was used to assess the response to treatment, since there is evidence that it is more closely associated with biologic activity19 and the clinical outcome of septic shock24 than are other markers. The extent to which measurements made in peripheral-blood monocytes relate to the up-regulation of TNF- in the airway is unclear. The relationship between the expression of membrane-bound TNF- and the response to etanercept suggests that the effects in peripheral-blood monocytes are relevant to those in more biologically relevant sites. More work is required to determine the main source of TNF- in the airway in patients with refractory asthma and how this relates to markers of TNF- activity on peripheral-blood monocytes.
Potential explanations for the increased TNF- activity in peripheral-blood monocytes include the coexistence of asthma with other inflammatory conditions associated with increased TNF- activity and genetic differences in the TNF- gene or genes associated with the regulation of TNF- production. The increased TNF- activity is unlikely to reflect the effects of corticosteroid treatment, since in vitro studies show that corticosteroids reduce the production of TNF- by monocytes,25 although this effect may be diminished in patients who have refractory asthma as a result of resistance to corticosteroids. An effect of other treatments cannot be excluded, although there is no strong biologic rationale to provide support for such an effect.
The beneficial effects of etanercept-induced antagonism of TNF- on markers of asthma control support the view that TNF- contributes to the pathogenesis of refractory asthma. Our pilot study involved small numbers of patients, and the results could have been compromised by missing data, the crossover design, or the imbalance in the treatment order. Thus, the clinical findings cannot be regarded as a directive for treatment. The large effect of etanercept on measures of airway function in our study is consistent with the findings of an uncontrolled study.11 We found a significant decrease in histamine concentrations in sputum supernatant but no effects on other markers of airway inflammation, suggesting that the effect of etanercept is mediated primarily by an effect on airway smooth-muscle and mast cells.6,26,27
Our findings suggest that the beneficial effects of etanercept may be confined to patients with refractory asthma, since patients with mild-to-moderate asthma had no evidence of increased markers of TNF- activity on peripheral-blood monocytes. For this reason and because treatment with an expensive agent with potential side effects could not be justified clinically, we did not study the effects of etanercept in patients with mild-to-moderate asthma. The view that systemic dysregulation of the TNF- axis is peculiar to patients with refractory asthma is supported by a study showing that TNF- production by peripheral-blood monocytes in response to lipopolysaccharide and other stimuli was increased in patients with severe asthma but not in those whose asthma was controlled by low-dose inhaled corticosteroids.28 In addition, there is preliminary evidence that two weeks of treatment with etanercept has no effect on airway responsiveness or the bronchoconstrictor and inflammatory response to endobronchial allergen challenge in subjects with mild atopic asthma.29 However, these findings are limited by the short duration of treatment and small numbers of subjects, and further study of patients with less-severe asthma is required.
Supported by grants from Asthma UK (to Drs. Berry and Shaw) and by a grant from the University Hospitals of Leicester National Health Service Trust (to Ms. Hargadon).
Dr. Berry reports having received lecture fees from AstraZeneca; Dr. Shaw, lecture fees from AstraZeneca and GlaxoSmithKline; Dr. Bradding, lecture fees from AstraZeneca and Merck Sharp & Dohme and research grants from Schering-Plough, AstraZeneca, and Novartis; Dr. Brightling, lecture fees from AstraZeneca and GlaxoSmithKline and a research grant from Cambridge Antibody Technology; Prof. Wardlaw, lecture fees from Altana Pharmaceuticals and Merck Sharp & Dohme and research grants from AstraZeneca, GlaxoSmithKline, and Wyeth; and Prof. Pavord, lecture fees and research grants from AstraZeneca and GlaxoSmithKline. No other potential conflict of interest relevant to this article was reported.
We are indebted to Will Monterio for providing laboratory support, to Susan McKenna for preparing the injections, and to the tuberculosis nurses at Glenfield Hospital for performing the tuberculosis screening.
Source Information
From the Institute for Lung Health, University Hospital of Leicester National Health Service Trust, Glenfield Hospital, Leicester, United Kingdom.
Address reprint requests to Prof. Pavord at the Institute for Lung Health, University Hospital of Leicester National Health Service Trust, Glenfield Hospital, Groby Rd., Leicester LE3 9QP, United Kingdom, or at ian.pavord{at}uhl-tr.nhs.uk.
References
) as a novel therapeutic target in symptomatic corticosteroid dependent asthma. Thorax 2005;60:1012-1018.
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||
Related Letters:
Tumor Necrosis Factor in Refractory Asthma
Oliveri C., Polosa R., Berry M. A., Pavord I. D.
Extract |
Full Text |
PDF
N Engl J Med 2006;
354:2074-2075, May 11, 2006.
Correspondence
This article has been cited by other articles:
HOME | SUBSCRIBE | SEARCH | CURRENT ISSUE | PAST ISSUES | COLLECTIONS | PRIVACY | HELP | beta.nejm.org Comments and questions? Please contact us. The New England Journal of Medicine is owned, published, and copyrighted © 2008 Massachusetts Medical Society. All rights reserved. |
Previous
2-agonist treatment was withheld for more than 6 hours and treatment with long-acting 
