Limited Short-term Steroid Responsiveness Is Associated With Thickening of Bronchial Basement Membrane in Severe Asthma

Transcription

1 CHEST Original Research Limited Short-term Steroid Responsiveness Is Associated With Thickening of Bronchial Basement Membrane in Severe Asthma Arnaud Bourdin, MD, PhD ; Stephane Kleis, MD ; Mohamad Chakra, MD ; Isabelle Vachier, PhD ; Fabrice Paganin, MD, PhD ; Philippe Godard, MD ; and Pascal Chanez, MD, PhD ASTHMA Background: The clinical manifestations of bronchial remodeling in asthma and the potential impact of this process on lung function remain unclear. We aimed to determine whether the presence of pathologic features of airway remodeling in patients with asthma was associated with steroid responsiveness in the short term. Methods: Sixty-three consecutive patients with severe asthma with chronic airflow impairment (post-bronchodilator, 80% predicted values) were recruited, clinically characterized, and had an initial bronchoscopy where endobronchial biopsy and BAL were performed. BAL cellular content was reported and reticular basement membrane (RBM) thickness was measured by validated repeated measures. Patients were then treated with 1 mg/kg/d of methyl prednisone, directly administered IV, for 10 days. A threshold of 15% improvement was used to discriminate responsive (group 1) and refractory patients (group 2). Results: Thirty-eight patients had a steroid responsiveness. 15% (group 1) and a thinner RBM at the biopsy level ( m m vs m m; P 5.001) compared with nonsteroid responsive group 2 patients as defined. The best predictors for being unresponsive were no long-term treatment with oral steroids and increased RBM thickness. The associated receiver operating characteristic curve indicated that RBM thickness could predict steroid responsiveness below 15% with an area under the curve of ( P ) at a threshold of 7 m m. Conclusions: Features of airway remodeling are associated with limited short-term steroid responsiveness in severe asthma. CHEST 2012; 141(6): Abbreviations: BD 5 bronchodilator; D lco 5 diffusing capacity of the lung for carbon monoxide; EBB 5 endobronchial biopsy; HRCT 5 high-resolution CT; OCS 5 oral corticosteroid; RBM 5 reticular basement membrane; SARP 5 Severe Asthma Research Program Asthma is a chronic inflammatory disease of the airways. It is commonly assumed that chronic inflammation will lead to an exaggerated repair process with bronchial deposition of scar tissue. This remodeling Manuscript received February 14, 2011; revision accepted November 1, Affiliations: Department of Respiratory Diseases, Centre Hospitalier et Universitaire (CHU) Montpellier, and Institut National de la Santé et de la Recherche Médicale (INSERM) U1046, Université Montpellier 1, Université Montpellier 2 (Drs Bourdin, Kleis, Chakra, Vachier, and Godard), Montpellier; Department of Respiratory Diseases, Groupe Hospitalier Sud Réunion (Dr Paganin), Saint Pierre de La Réunion, Réunion; and Département des Maladies Respiratoires (Dr Chanez), Assistance Publique Hôpitaux de Marseille (AP-HM), Laboratoire d immunologie, INSERM, Centre National de la Recherche Scientifique (CNRS) U600, Unité Mixte de Recherche (UMR) 6212, Université de la Méditerranée, Marseille, France. Deceased. of the airways, mainly characterized by subepithelial fibrosis, smooth muscle enlargement, and gland hyperplasia, is thought to play a role in the treatment resistance associated with chronic airflow obstruc tion that is often found in patients with severe asthma. 1,2 Nonetheless, conflicting hypotheses have been proposed since features of airway remodeling have also been Funding/Support : This study was supported by Direction de la Recherche Clinique, CHU Montpellier, France. Correspondence to: Arnaud Bourdin, MD, PhD, Department of Respiratory Diseases, Hôpital Arnaud de Villeneuve, CHU Montpellier, 371 Av. Doyen G. Giraud, Montpellier Cedex 5, France ; a-bourdin@chu-montpellier.fr 2012 American College of Chest Physicians. Reproduction of this article is prohibited without written permission from the American College of Chest Physicians. See online for more details. DOI: /chest Original Research

2 reported in children in whom asthma may or may not be fully reversible. Airway remodeling can be explored in vivo in humans using investigational procedures such as highresolution CT (HRCT) imaging 3 and reticular basement membrane (RBM) thickness assessment in endo bronchial biopsies (EBBs). 4,5 Postmortem studies have revealed that assessing RBM thickness of the proximal airways (ie, those that can be explored by EBB sampling during endoscopy) correctly measures total wall thickness, which is important because it is a main feature of airway remodeling in fatal asthma cases.2 Nonetheless, the clinical consequences of airway remodeling remain unclear. We previously demonstrated the specificity of severe asthma pathology on a large-scale EBB study. In particular, an increased RBM thickness over 7 m m was specific for severe asthma when compared with mild asthma or COPD when using repeated measures of RBM at regular intervals. 4 The inflammatory patterns found in severe asthma are heterogeneous and variable over time, and might contribute to severe asthma phenotypes. 6 A recent, large cluster study (Severe Asthma Research Program [SARP]) contradicted certain previously published hypotheses. 7 Although initially the eosinophilic pattern was considered as a discriminating criterion, patients with chronic airflow obstruction in the SARP study (cluster 5) had no differences in their inflammatory patterns. 7 Steroids remain the primary treatment of asthma despite potential side effects. The potential lack of efficacy due to airway remodeling is a source of conflicting data. Some authors have demonstrated that in severe asthma treated by high doses of inhaled steroids, a directly observed triamcinolone therapy 8 improves the initial airflow impairment, while others have suggested the requirement for alternative treatments such as anti-il-5 therapies. 9 The latest reports are currently being debated because improvements in lung function using augmented steroid therapies are relatively limited despite eosinophilic signals in blood and sputum being efficiently suppressed. 10,11 Furthermore, recent studies focusing on severe eosinophilic asthma were able to demonstrate that mepolizumab, an anti-il-5 antibody, had some effect on severe asthma exacerbation rates, although this treatment did not affect or bronchial hyperresponsiveness. 9 In nine prednisone-dependent patients with sputum eosinophilia, mepolizumab improved while simultaneously decreasing daily doses of oral corticosteroid (OCS). 12 Tissue eosinophils and structural changes in the subepithelial layers were found to be responsible for the variability of the clinical responses in mild asthma. 13 Structural changes and neutrophilic inflammation are, thus, the usual culprits of the steroid unresponsiveness attributed to severe asthma. 14 It is then assumed that shortterm steroid challenge responsiveness may reflect residual eosinophilic or neutrophilic inflammation, or the cli nical and functional association with bronchial remodeling. 1,15,16 Therefore, remodeling processes may represent a potential mechanism precluding steroid responsiveness in severe asthma. We aimed to determine whether increased RBM thickness as a marker of airway remodeling could predict decreased shortterm systemic steroid challenge responsiveness in patients with asthma with chronic airflow obstruction. Materials and Methods Patient Screening Patients with severe asthma with persistent chronic airflow impairment (best reported and measured post-bronchodilator (post-bd) values, 80%) known for at least 1 year were referred to our center by their pulmonologist on behalf of the regional Difficult Asthma Network. The study was approved by an ethics committee and the local institutional review board (under the reference PHRC DRC-CHU, Montpellier, France). All patients expressed their willingness to participate in the study, and both patients and investigators signed a written informed consent. Patients were managed in our center on a 15-day basis. A rigorous clinical examination and medical history were recorded. Mild and severe exacerbations were defined as symptom worsening that required (or not) a steroid pulse. Extensive pulmonary function tests, including spirometry, plethysmography, diffusing capacity of the lung for carbon monoxide (D lco ), and shortacting b 2 -agonist reversibility, were assessed, while short-acting b 2 -agonists were carefully withdrawn for at least 12 h and longacting b 2 -agonists for 48 h. HRCT scan of the lung and flexible bronchoscopy were performed before any intervention. HRCT scan analysis was semiquantitatively assessed in a double-blind manner according to a previously published method 17 by two radiologists. Treatments were optimized according to the best standard of care by integrating the history of the disease associating inhaled corticosteroid doses around 2,000 m g/d of equivalent beclometasone and high doses of long-acting b 2 -agonists (48 m g of formoterol or equivalent) and a third controller (leukotriene receptor antagonist, OCS). An extensive comorbidity check-up was done, including cardiac echography, bone mineral density absorptiometry, blood cortisol at 8:00 am, sleep polygraphy, and diabetes mellitus screening. Comorbidities were considered as present when cardiac echography detected a left ventricular ejection fraction below 50%, or when one or more of the following were found: noncorrected sleep apnea syndrome, noncorrected adrenal insufficiency, or de novo diabetes mellitus requiring extensive management. The Asthma Control Questionnaire (ACQ) and the Asthma Quality of Life Questionnaire (AQLQ) according to Juniper were filled out by the patients. 18,19 Inclusion criteria were rigorous in order to securely avoid COPD and other differential diagnoses. Asthma was diagnosed using a combination of clinical criteria: smoking history, 10 pack-years, highly variable lung function (at least. 12% in the follow-up year reported by the referring physician, and/or a positive metacholine challenge), normal D lco values ( 70% predicted values), a clear history of asthma, no emphysematous changes on HRCT scan, CHEST / 141 / 6 / JUNE,

3 and the absence of important comorbidities according to the previously given definition. Blood cell counts and total IgE levels were assessed. Skin prick tests for the common allergens of our region were recorded. Compliance issues were raised by the physician, and autoreported noncompliance or evidence of noncompliance (inability to manipulate the device adequately, unrecognized while supposed in use devices) were recorded. A 10-day directly observed IV challenge with 1 mg/kg/d methyl prednisone was then proposed to the patients. Spirometry was repeated on the last day of the challenge. All patients were invited to attend the outpatient clinic again and to undergo a further challenge 12 months later, but only 30% of them attended. Bronchoscopy, EBB Processing Under local anesthesia, flexible endoscopy was performed before steroid challenges were started. Trachea and bronchus examination took place before a BAL was performed in the middle lobe and EBBs in the left lower lobe. Biopsies were immediately fixed in 4% neutral-buffered formalin, then embedded in paraffin. Sections of 5 m m were cut, then placed on slides before entering the analysis process. 20 Biopsies were analyzed in a routine setting by the referent academic pathologist and qualitatively categorized for smooth muscle enlargement (only as present or absent), thickening of the RBM, and mucus gland hyperplasia. Subsequently, biopsies were assessed for RBM thickness after the slides were blinded and then stored in a random order. A CCD camera (Sony DXC950P; Sony) linked to a light microscope (Olympus BH-2; Olympus) was used to record images. Only well-orientated epithelial areas suitable for RBM thickness assessment were considered for repeated measures at regular intervals of 50 m m, according to Sullivan. 21 RBM thickness was expressed as the mean of at least 20 measures (ie, analysis of at least 1-mm length of RBM). BAL Process Aliquots obtained during bronchoscopy were routinely processed. Centrifugation and a manual cell count were performed. Eosinophils and neutrophils were expressed as percentages of the total WBC count. Statistical Analysis Steroid challenge responsiveness was judged according to a 15% improvement threshold of post-bd values. 15 Group 1 referred to patients who respond to steroids and group 2 as steroid refractory patients. Group 1 and 2 comparisons were performed according to data distributions using parametric or nonparametric tests. Parameters that reached P values,.10 entered a multivariate analysis whose aim was to identify independent risk factors associated with steroid responsiveness. Receiver operating characteristic curves were then performed with RBM thickness values as a discriminative tool for being steroid resistant (being in group 2). Correlations were analyzed using linear regression. Statistical analyses were carried out using SPSS for Windows, version (SPSS Inc). Results Steroid Challenge Of the 63 consecutive patients with severe asthma studied, 38 had an improvement. 15% for post-bd values at the end of the short-term steroid challenge. The 25 unresponsive patients were referred to as group 2. Clinical Assessment Disease duration (24.3 years [95% CI, 20-2]) in group 1 and 25.8 years [18-36] in group 2), baseline pulmonary function tests, and ACQ scores (3.29 [ ] vs 3.18 [ ]) did not differ. According to our definition, observance was optimal in 24 of 38 patients vs 12 of 25 patients ( P 5.30). Comparisons between groups ( Table 1 ) indicated that patients in group 1 had lower BMI and more frequent steroid short courses due to exacerbation; they Table 1 Patient Clinical Characteristics Characteristics Reversibility 15% (n 5 38) Reversibility, 15% (n 5 25) P Value Age, y 46.5 (41-63) 55 (40-68).49 Sex, F (M) 20 (18) 15 (10).56 BMI, kg/m ( ) ( ).04 Asthma duration, y 24.3 (20-29) 25.8 (18-36).57 Hospitalizations/y 1 (0-2) 0 (0-1).33 Smoking history, pack-y 0 (0-10) 0 (0-10).95 Steroids pulses/y a 2.5 (1.5-3) 2 (1-2).03 ACQ 3.29 ( ) 3.18 ( ).37 AQLQ 3.57 ( ) 3.80 ( ).74 Osteoporosis, Y/N 17/19 18/6 b.037 Rhinosinusitis, Y/N 36/2 21/4.15 Comorbidities, Y/N 2/38 2/25.99 GER, Y/N 13/25 11/14.43 ASA, Y/N 2/36 2/23.66 Atopy, Y/N 23/15 12/13.32 Premenstrual, Y/N 5/20 2/13.52 Long-term maintenance oral steroids, Y/N 18/20 4/21.01 Data are presented as means and 95% CI or counts for qualitative data. ACQ 5 Asthma Control Questionnaire; AQLQ 5 Asthma Quality of Life Questionnaire; ASA 5 aspirin intolerance; F 5 female; GER 5 gastroesophagal reflux; M 5 male; N 5 no; Y 5 yes. a Exacerbations were recorded as steroid pulses. b Data were not available for one patient. P values are unpaired t tests or Fischer exact x 2 tests Original Research

4 were more frequently under a long-term maintenance daily dose of oral steroids. Unexpectedly, osteoporosis was more frequent in group 2. There were no differences in occurrence and specificity of triggers or precipitating factors. Basal values were higher in group 2 ( Table 2 ), while lung volumes were similar. BD and steroid responsiveness were higher in group 1 and were linearly correlated among the entire population ( r , P ) ( Figs 1, 2 ). Total IgE levels were higher in group 1, without reaching levels. 500 kiu/ml. A new challenge performed 1 year later did not reveal major modifications, but data were available in only 20 patients (30% of the effective) who agreed to participate in this later visit: 10 of 13 who were in group 1 were still in the same group, the remaining three patients who belonged to this group had basal values. 80%. Among the seven who were in group 2, six still did not have any reversibility. 15% and one stopped the challenge after 3 days and withdrew his consent. BAL Eosinophilia and EBB Examination Inflammatory cells including blood and BAL eosinophilia were not different between groups ( Table 3 ). EBB examination revealed that smooth muscle enlargement and thickening of the RBM were more frequently reported by the pathologist in the unresponsive group (group 2). RBM thickness was significantly higher in group 2 ( Table 3 ) when correctly measured. The qualitative assessment of the RBM (performed by a pathologist) was related to objective measurements ( m m vs m m, P 5.04). Multivariate Analysis Multiple regression analysis indicated that baseline values (before BDs) were the main predictor of steroid responsiveness ( Table 4 ). After adjustment for baseline, pre-bd, being in group 2 was associated with the absence of long-term maintenance treatment with OCSs and an increased thickness of the basement membrane. RBM Thickness and Steroid Responsiveness We quantified the ability of RBM thickness to discriminate between groups with an area under the curve computed from receiver operating characteristic curves. This area under the curve differentiated group 1 and group 2 with a value of (95% CI, ), P ( Fig 3 ). The detection cutoff was 7 mm. Discussion Asthma and severe asthma are heterogeneous conditions. Subgrouping and phenotyping these patients may allow for better management and prognosis. In this study, we demonstrated that clinically defined patients with severe asthma with chronic airflow impairment unresponsive to a short course of parenteral steroids had features of airway remodeling illustrated by an increased RBM thickness at the biopsy level. Structural bronchial airway abnormalities are well described in severe asthma. 1 They have been shown to be associated with progressive loss of airway function. 4,22 Although well documented in animals, 23 i n vivo evidence for clinical manifestations of airway remodeling are still incompletely established. In this study, we observed a particular phenotype for patients with severe asthma. Group 2 is a pattern of chronic airflow obstruction, low steroid reversibility, and high evidence of airway remodeling as measured by RBM thickness. Patients in this group were receiving less oral steroids but had more osteoporosis. This phenotype is very close to the cluster 5 Table 2 Pulmonary Function Tests Pulmonary Function Test Reversibility 15% (n 5 38) Reversibility, 15% (n 5 25) P Value Baseline, % 50.5 ( ) 56.5 (50-63).047 Baseline, L ( ) ( ).041 Postbronchodilator, % 56.5 ( ) 60 (53-67).01 Post-CS, % 74.5 ( ) 57 ( ) NA /FVC, % 59.5 (55-63) 59 (53-65).81 TLC, % 104 (99-110) 105 (94-116).95 RV, % 169 ( ) 163 ( ).35 RV/TLC, % 55 ( ) 54 (45-63).51 D lco /Va, % 97 (87-113) 95 (84-108).51 Bronchodilator responsiveness, % 12 (6-19) 8 (3-14).01 Steroid responsiveness, % 32 (21-44) 6 (1-11) NA Data are percentages of predicted values and presented as means and 95% CI. P values are unpaired t tests. Post-CS refers to after 10 days of steroids, and measured after bronchodilators. Responsiveness is measured as a percentage of improvement from basal values. D lco 5 diffusing capacity of the lung for carbon monoxide; NA 5 not applicable; post-cs 5 postcorticosteroid; RV 5 residual volume; TLC 5 total lung capacity; V a 5 alveolar volume. CHEST / 141 / 6 / JUNE,

5 Figure 1. Patterns of values given as percentages of predicted values among groups after inhaling BDs and after a 10-day steroid challenge (measured after BDs). BD 5 bronchodilator. reported in the SARP study. 7 Absence of emphysematous changes at HRCT scanning, normal D lco, a clear history of asthma, and reversibility in the past as well as absence of evidence for other chronic bronchial disease (bronchiolitis, etc) strongly support the diagnosis of severe asthma for this subgroup of patients. A retrospective examination of these patients clinical details suggested a heavy history of steroid exposure progressively stopped because of lack of evidence for efficiency and/or side effects. These data Figure 2. Correlation between the percentage of improvement after BDs at the beginning of the study and the percentage of improvement after a 10-day steroid challenge (measured after BDs). B2 5 shortacting b 2 -agonist; OCS 5 oral corticosteroid. See Figure 1 legend for expansion of other abbreviations Original Research

6 Table 3 Inflammatory and Remodeling Patterns Biological Criteria Reversibility 15% (n 5 38) Reversibility, 15% (n 5 25) P Value Blood eosinophils, a /mm3 393 ( ) 273 ( ).14 BAL eosinophils, a % 1 (1-4) 1 (0-3).66 Blood neutrophils, a /mm3 5,004.5 (3,900-6,408) 4,914 (3,477-5,832).4 Total IgE, a kiu/ml 263 (73-384) 88.5 (13-137).06 BAL neutrophils, a % 11 (4-38) 9 (4-52).69 HRCT scan score 6 (5-7.5) 6.5 (5.5-9).54 Epithelial shedding, Y/N 11/27 9/16.59 Metaplasia, Y/N 5/33 2/23.65 Smooth muscle enlargement, b Y/N 9/18 12/6.03 Mucus gland hyperplasia, b Y/N 11/5 5/5.22 BM thickness, b m m 5.78 ( ) 7.60 ( ).001 Data are presented as means and 95% CI or as medians and interquartiles P values refer to unpaired t tests for quantitative data and Fisher exact x 2 tests for qualitative data. BM 5 basement membrane; HRCT 5 high-resolution CT. See Table 1 legend for expansion of other abbreviations. a Comparison of unpaired t tests for quantitative data done after log transformation. b Data were not available for all patients. were not available for all patients and data obtained from patient interviews were not accurate enough to be reported. While current efforts are concentrated on dividing asthma phenotypes according to T helper cell 2-driven inflammatory patterns, 24 we observed in this real-life study a potential for structural rather than inflammatory changes to limit short steroid responsiveness. Our findings contradict the prevalent hypothesis of a persistent eosinophilic-activated pathway perpetuating airway remodeling 25 as steroid-sensitive patients did not exhibit higher evidence of eosinophilic activation but did have lower signs of airway remodeling. 26 Very recently, increased RBM thickness and mucus metaplasia were demonstrated as consequences of both an eosinophilic (ie, allergen challenge) and a noneosinophilic challenge (ie, metacholine). Our findings are potentially concordant with this observation, where the most exposed to mechanical constraints are those with the greatest evidence of airway remodeling. 27 In prednisone-dependent patients with asthma with sputum eosinophilia, the ability to improve under a new medication (especially one directed toward eosinophils), is another argument for this hypothe- Table 4 Multivariate Analysis After Adjustment for Baseline Pre-BD Values Variables r 2 b SE P Value Pre-BD Post-BD BMI, kg/m Steroid pulses/y Osteoporosis Long-term maintenance oral steroids BM thickness Full model 0.765,.0001 b is the regression coefficient. SE is the SE of b. BD 5 bronchodilator. See Table 1 legend for expansion of other abbreviations. sis. 12 Even though the study was not powered to describe such a phenotype, it is noteworthy that our study investigations (particularly BAL and biopsies) were performed with all patients continuing to use their current medications, including steroids. The fact that this study reflects the real-life course of severe asthma further adds to its interest. Confirming whether these observations are reproducible over time will be important. Steroid responsiveness was found to be poorly reproducible at 1 year in a small group of patients. 13 Nonetheless, no major differences were observed in the few patients who came in 1 year after the present study, even though some improved. In the present study, steroid-sensitive patients tended to be more symptomatic and to experience more exacerbations, and it could be argued that these patients were less stable than others despite similar levels of BAL eosinophilia. Whether airway remodeling is a consequence of persistent airway inflammation or a limitation for antiinflammatory therapy responsiveness is still a matter of debate. Different inflammatory patterns that lead to enhanced airway remodeling and steroid resis tance are worthy of description. Steroid resistance may relate to either very high inflammatory responses or different inflammatory mechanisms (neutrophilic, 28 paucigranulocytic). New phenotyping efforts based on cluster analysis support the present results, even though a specific intervention, like the one in the present study, has not been previously assessed.7 Although our study has certain limitations, we do not believe that they are sufficient to discount our main findings. We acknowledge the open study design; however, the directly observed treatment was a strength of this design. Furthermore, initial investigations were performed under current treatments and, therefore, could be extrapolated to real-life conditions. Structural changes of the airways were CHEST / 141 / 6 / JUNE,

7 Figure 3. The receiver operating characteristic curve for basement membrane thickness ability to discriminate among patients according to their steroid responsiveness (below or above 15% of values). assessed at the subepithelial levels and no quantitative data are reported for other remodeling features affecting smooth muscle, glands, or epithelium; however, the reproducibility of the method used was already validated, while there are few robust data concerning the other compart ments. 5,21 HRCT scan measurements for airway thick ness were not performed. Bronchial inflammation was assessed using BAL eosinophilia, which is a well-established methodology correlated to sputum eosinophils and exhaled nitric oxide. 29 Eosinophils can be assessed by different ways: immunostaining in EBBs using more or less specific antibodies, induced sputum, and BAL. We decided to use BAL as it was frequently reported as a good marker of eosinophilic inflammation 25 and potentially accessible to everyone. RBM thickness is unlikely to be routinely assessed, although noninvasive markers of airway remodeling are currently under validation in real-life studies. The present study argues for the development of such tools in the initial assessment of patients with severe asthma. We conclude that in severe asthma, a directly observed therapeutic test could be a criterion included in a cluster-based phenotyping analysis. The absence of response to corticosteroids is related to subepithelial thickness and is a potential signal for an important remodeling process. Acknowledgments Author contributions: Dr Bourdin takes responsibility for the integrity of the data and accuracy of the data analysis. Dr Bourdin: contributed to clinical data collection, analyzed biopsies, finalized statistical analysis, and helped write and read the manuscript. Dr Kleis: contributed to clinical data collection and helped write and read the manuscript. Dr Chakra: contributed to clinical data collection, analyzed biopsies, and helped write and read the manuscript. Dr Vachier: contributed to HRCT scan analysis, designed the study, and helped write and read the manuscript. Dr Paganin: contributed to HRCT scan analysis and helped write and read the manuscript. Dr Godard: designed the study and helped write and read the manuscript. Dr Chanez: designed the study, finalized statistical analysis, and helped write and read the manuscript. Financial/nonfinancial disclosures: The authors have reported to CHEST the following conflicts of interest : Dr Chanez has provided consultancy services for Almirall, Boehringer Ingelheim, Centocor, GlaxoSmithKline, Merck & Co Inc, AstraZeneca, Novartis, Teva Pharmaceuticals, Chiesi Farmaceutici, and Schering- Plough; has served on advisory boards for Almirall, Boehringer Ingelheim, Centocor, GlaxoSmithKline, AstraZeneca, Novartis, Teva Pharmaceuticals, Chiesi Farmaceutici, Schering-Plough, and Merck & Co Inc; has received lecture fees from Almirall, Boehringer Ingelheim, Centocor, GlaxoSmithKline, AstraZeneca, Novartis, Teva Pharmaceuticals, Chiesi Farmaceutici, Schering- Plough, Merck & Co Inc, and Boston Scientific; and has received industry-sponsored grants from Almirall, Boehringer Ingelheim, Centocor, GlaxoSmithKline, AstraZeneca, Novartis, Teva Pharmaceuticals, Chiesi Farmaceutici, and Schering-Plough. Drs Bourdin, Kleis, Chakra, Vachier, Paganin, and Godard have reported that no potential conflicts of interest exist with any companies/organizations whose products or services may be discussed in this article. Role of sponsors: The sponsor had no role in the design of the study, the collection and analysis of the data, or in the preparation of the manuscript. Other contributions: In memory of Philippe Godard, MD. References 1. Chanez P, Wenzel SE, Anderson GP, et al. Severe asthma in adults: what are the important questions? J Allergy Clin Immunol ;119 (6 ): James AL, Maxwell PS, Pearce-Pinto G, Elliot JG, Carroll NG. The relationship of reticular basement membrane thickness to airway wall remodeling in asthma. Am J Respir Crit Care Med ;166 (12 pt 1 ): Niimi A, Matsumoto H, Takemura M, Ueda T, Chin K, Mishima M. Relationship of airway wall thickness to airway sensitivity and airway reactivity in asthma. Am J Respir Crit Care Med ;168 (8 ): Bourdin A, Neveu D, Vachier I, Paganin F, Godard P, Chanez P. Specificity of basement membrane thickening in severe asthma. J Allergy Clin Immunol ;119 (6 ): Wilson JW, Li X. The measurement of reticular basement membrane and submucosal collagen in the asthmatic airway. Clin Exp Allergy ;27 (4 ): Haldar P, Pavord ID, Shaw DE, et al. Cluster analysis and clinical asthma phenotypes. Am J Respir Crit Care Med ;178 (3 ): Moore WC, Meyers DA, Wenzel SE, et al ; National Heart, Lung, and Blood Institute s Severe Asthma Research Program. Identification of asthma phenotypes using cluster analysis in the Severe Asthma Research Program. Am J Respir Crit Care Med ;181 (4 ): ten Brinke A, Zwinderman AH, Sterk PJ, Rabe KF, Bel EH. Refractory eosinophilic airway inflammation in severe asthma: effect of parenteral corticosteroids. Am J Respir Crit Care Med ;170 (6 ): Original Research

8 9. Haldar P, Brightling CE, Hargadon B, et al. Mepolizumab and exacerbations of refractory eosinophilic asthma. N Engl J Med ;360 (10 ): Flood-Page P, Swenson C, Faiferman I, et al ; International Mepolizumab Study Group. A study to evaluate safety and efficacy of mepolizumab in patients with moderate persistent asthma. Am J Respir Crit Care Med ;176 (11 ): Flood-Page PT, Menzies-Gow AN, Kay AB, Robinson DS. Eosinophil s role remains uncertain as anti-interleukin-5 only partially depletes numbers in asthmatic airway. Am J Respir Crit Care Med ;167 (2 ): Nair P, Pizzichini MM, Kjarsgaard M, et al. Mepolizumab for prednisone-dependent asthma with sputum eosinophilia. N Engl J Med ;360 (10 ): Demoly P, Jaffuel D, Mathieu M, et al. Glucocorticoid insensitive asthma: a one year clinical follow up pilot study. Thorax ;53 (12 ): Barnes PJ. New molecular targets for the treatment of neutrophilic diseases. J Allergy Clin Immunol ; 119 ( 5 ): Barnes PJ, Woolcock AJ. Difficult asthma. Eur Respir J ;12 (5 ): Wenzel SE. Asthma: defining of the persistent adult phenotypes. Lancet ;368 (9537 ): Paganin F, Séneterre E, Chanez P, et al. Computed tomography of the lungs in asthma: influence of disease severity and etiology. Am J Respir Crit Care Med ;153 (1 ): Juniper EF, Buist AS, Cox FM, Ferrie PJ, King DR. Validation of a standardized version of the Asthma Quality of Life Questionnaire. Chest ;115 (5 ): Juniper EF, O Byrne PM, Guyatt GH, Ferrie PJ, King DR. Development and validation of a questionnaire to measure asthma control. Eur Respir J ;14 (4 ): Jeffery P, Holgate S, Wenzel S ; Endobronchial Biopsy Workshop. 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 ;168 (6 pt 2 ):S1-S Sullivan P, Stephens D, Ansari T, Costello J, Jeffery P. Variation in the measurements of basement membrane thickness and inflammatory cell number in bronchial biopsies. Eur Respir J ;12 (4 ): James AL, Wenzel S. Clinical relevance of airway remodelling in airway diseases. Eur Respir J ;30 (1 ): Hirota JA, Hackett TL, Inman MD, et al. Modeling asthma in mice: what have we learned about the airway epithelium? Am J Respir Cell Mol Biol ;44 (4 ): Woodruff PG, Modrek B, Choy DF, et al. T-helper type 2-driven inflammation defines major subphenotypes of asthma. Am J Respir Crit Care Med ;180 (5 ): Bousquet J, Chanez P, Lacoste JY, et al. Eosinophilic inflammation in asthma. N Engl J Med ;323 (15 ): Chakir J, Loubaki L, Laviolette M, et al. Monitoring sputum eosinophils in mucosal inflammation and remodelling: a pilot study. Eur Respir J ;35 (1 ): Grainge CL, Lau LC, Ward JA, et al. Effect of bronchoconstriction on airway remodeling in asthma. N Engl J Med ;364 (21 ): Simpson JL, Powell H, Boyle MJ, Scott RJ, Gibson PG. Clarithromycin targets neutrophilic airway inflammation in refractory asthma. Am J Respir Crit Care Med ;177 (2 ): Silkoff PE, Lent AM, Busacker AA, et al. Exhaled nitric oxide identifies the persistent eosinophilic phenotype in severe refractory asthma. J Allergy Clin Immunol ; 116 (6 ): CHEST / 141 / 6 / JUNE,