European Respiratory Society
Difficult-to-Treat Severe Asthma (out of print)

This book has been superseded by a newer edition. Difficult-to-treat severe asthma constitutes approximately 5–10% of the asthmatic population, and continues to present a considerable therapeutic challenge. Use of the pragmatic definitions over the past 10 years has led to an improved delineation of these patients. This monograph gathers expertise in the paediatric and adult severe asthma fields. Giving rise to a comprehensive edition that summarises what has been achieved in understanding the different phenotypes, their biomarkers and the reasons and mechanisms that underlie difficult-to-treat severe asthma as well as the refractoriness to its current treatments. This volume also provides a physician's guide to the management of severe asthma with the optimal use of the currently available therapies, while pointing to the potential approaches we need to take in order to discover more effective treatments.

  • European Respiratory Society Monographs
  1. Page v
  2. Page vi
  3. Page vii
  4. Page 1
    Abstract
    Correspondence: W.W. Busse, University of Wisconsin Hospital, K4/910 CSC, MC 9988, 600 Highland Avenue, Madison, WI 53792, USA, Email: wwb@medicine.wisc.edu

    Difficult-to-treat asthma and severe asthma are often used synonymously, and may represent the same underlying disease characteristics. In defining this group of patients, a number of features need to be considered, including level of treatment and response to treatment. Patients in this group can be considered as those requiring large doses of inhaled corticosteroids, and either achieving control or remaining symptomatic. From a variety of sources, the frequency of severe and/or difficult-to-treat asthma is greater than anticipated. In addition, clinical trials of asthma designed to escalate treatment until control is achieved have shown that the frequency of achievement of this level of control is <60%. In moving forward, attempts to establish patients’ clinical phenotypes, and eventually genotypes, promise to more-fully characterise these individuals, and, from this information, improve the predictability of response to treatment. The scope of patients with this type of asthma is greater than appreciated and remains a significant clinical problem.

  5. Page 16
    Abstract
    Correspondence: K. Sumino, Washington University School of Medicine, 660 Euclid Avenue, Campus box 8052, St Louis, MO, 63110, USA, Email ksumino@dom.wustl.edu

    Difficult-to-treat severe asthma patients should be systematically approached in the clinic to confirm the diagnosis of asthma, exclude alternative diagnosis, identify comorbidities, assess treatment compliance and evaluate for treatment-induced side-effects. During the initial evaluation process the following procedures are recommended: 1) the recording of a subjects detailed clinical history focusing on asthma, determining triggers and comorbidities that may contribute to poor asthma control; 2) a physical examination; 3) lung function tests, including spirometry with bronchodilator responses, lung volumes and diffusion capacity; and 4) laboratory tests, including serum immunoglobulin E levels and peripheral eosinophil count. Compliance with the treatment plan and inhaler technique should also be reviewed. Additional tests, including allergy skin testing, sputum eosinophil detection, multidetector row computed tomography scanning, laryngoscopy, polysomnograph and bronchoscopy may also be used as indicators for the evaluation of comorbidities and the confirmation of an asthma diagnosis. Evaluation of difficult-to-treat asthma should be performed by clinicians with special expertise in asthma or in specialised asthma centres.

  6. Page 28
    Abstract
    Correspondence: L-P. Boulet, Institut Universitaire de Cardiologie et de Pneumologie de Québec, Université Laval, Québec, QC, Canada. Email lpboulet@med.ulaval.ca

    Asthma control can be influenced by various psychological factors. Although severe psychological disorders do not seem to be more frequent in severe than in milder asthma, patients’ attitudes, personality and, particularly, triggers or conditions such as stress or anxiety/depression disorders can affect asthma outcomes. Furthermore, these conditions may sometimes explain poor adherence to maintenance therapy in severe asthma. Barriers to adherence must be identified and addressed in each patient. Cost-effective pragmatic interventions should be utilised to improve adherence, and these should ideally be long-term, patient-tailored and multifaceted. More research is required to identify the best ways of improving adherence and reducing psychological ailments to improve asthma.

  7. Page 50
    Abstract
    Correspondence: S.E. Wenzel, University of Pittsburgh Asthma Institute at UPMC/UPSOM, Dept of Medicine, NW 628 Montefiore, 3459 Fifth Ave, Pittsburgh, PA 15213, USA, Email wenzelse@upmc.edu

    Severe asthma is characterised by frequent and severe exacerbations, ongoing symptoms and often low lung function. In countries where asthma medications are easily available these factors often persist despite high doses of medication. This group of characteristics in association with severe asthma suggests that definitions require a background need for high doses of “gold standard” asthma medications to prevent the asthma from becoming poorly controlled, or which remains uncontrolled despite this therapy. In addition, it is recognised that the umbrella term “severe asthma” encompasses multiple different phenotypes. Various biased and unbiased studies have led to the general appreciation that early onset allergic asthma probably contributes to the largest number of severe asthmatics. However, other phenotypes, including a late onset, eosinophilic phenotype, as well as a symptom-prone very late onset disease common in females, are also becoming better recognised. Ongoing studies linking phenotypes to genetic/epigenetic background and ongoing gene expression will strengthen the mechanistic understanding of these phenotypes and improve the search for biological therapeutic targets.

  8. Page 59
    Abstract
    Correspondence: A. Bush, Dept of Paediatrics, Royal Brompton Hospital, Sydney Street, London, SW3 6NP, UK, Email a.bush@rbh.nthames.nhs.uk

    Severe asthma accounts for a small proportion of all asthma in childhood, but consumes a disproportionate amount of healthcare resources, and leads to substantial morbidity and even premature death. Many different symptom patterns can lead to specialist referral with the diagnostic label problematic severe asthma.

    First, there is a full review to exclude other diagnoses and important comorbid conditions. Next, standard management is optimised, ideally including a home visit. As a result, approximately half of the children, for whom basic management needs to be got right, are classified as having difficult asthma. We recommend invasive testing for severe therapy-resistant asthmatics in order to try to determine the pattern and distribution of inflammation; the extent to which the asthma is steroid-sensitive; and whether the child shows persistent airflow limitation (PAL). Therapeutic choices in these children are not evidence-based, and international collaboration is essential if this is to be rectified. Finally, we review the little that is known about mechanisms of severe asthma derived from paediatric studies.

  9. Page 82
    Abstract
    Correspondence: E.R. Bleecker, Center for Genomics and Personalized Medicine, Wake Forest University Health Sciences, Medical Center Boulevard, Winston-Salem, NC 27157, USA, Email ebleeck@wfubmc.edu

    Asthma is a multifactorial disease caused by the interaction of genes and the environment. This interaction can influence the development of asthma and the progression and severity of this disease. Early genetic studies focused on known candidate genes and pathways associated with asthma susceptibility, although positive associations were not always replicated. Recent genome-wide association studies (GWAS) identified five genomic regions consistently associated with asthma: ORMDL3-GSDMB, interleukin (IL)-33, IL-1RL1, RAD50-IL-13 and HLA-DR/DQ.

    Current studies are using comprehensive phenotyping methods to investigate the role of gene variation in disease progression. A GWAS in a cohort of subjects with difficult-to-treat allergic asthma also identified RAD50/IL-13 on chromosome 5, suggesting that this region may contribute to more severe disease. Asthma severity is also influenced by specific genes that modulate pulmonary function or a pharmacogenetic response to asthma therapy. Future genetic studies will integrate genetic information from the comprehensively phenotyped cohorts and environmental exposures to predict disease risk and to develop targeted therapies for all asthma patients.

  10. Page 97
    Abstract
    Correspondence: Q. Hamid, McGill University Health Centre, Meakins-Christie Laboratories, 3626 Rue St-Urbain, Montreal H2X 2P2, QC, Canada. Email qutayba.hamid@mcgill.ca

    It has become clear that severe asthma represents a disease entity that differs from mild and moderate asthma in pathology and disease mechanisms. Even within the conventional definitions of severe/refractory asthma, depending on the types and quantities of inflammatory cells presented (e.g. eosinophils, neutrophils and mast cells) in relevant tissues and patient characteristics, further classification of refractory asthma is possible. The ability to group refractory asthma into clusters indicates that different mechanisms are responsible for the pathogenesis of different refractory asthma subclasses. The classical T-helper cell (Th) type 1/Th2 paradigm has been the focus of asthma pathogenesis for decades, and in severe asthma regulation of the Th1/Th2 cytokines production has been shown to be different from mild and moderate asthma. Furthermore, data suggest that in addition to the classic Th1 and Th2 cytokines, Th17 related cytokines might be responsible for mediating refractory asthma inflammation.

  11. Page 107
    Abstract
    Correspondence: M. Gaga, 7th Respiratory Medicine Dept and Asthma Centre, Athens Chest Hospital, 152 Mesogeion Avenue, Athens 11527, Greece, Email mgaga@med.uoa.gr

    Severe asthma phenotypes have been associated with a number of allergic and nonallergic factors. Specific allergens, such as cockroach and Alternaria, as well as mould sensitivity, are associated with severe exacerbations and more severe disease, although evidence from severe asthma cohort studies, such as the European Network For Understanding Mechanisms Of Severe Asthma (ENFUMOSA) and The Epidemiology and Natural History of Asthma: Outcomes and Treatment Regimens (TENOR) studies, show that allergens may play a less important role in severe compared to mild asthma. Occupational sensitisers and environmental factors (such as outdoor air pollutants and active or passive smoking) have been implicated in the increase in asthma morbidity since the early 1960s, and can be considered as inducers of the disease or triggers of exacerbations. Infections, mainly with respiratory viruses, but also with bacteria, such as Chlamydophila pneumoniae, contribute to severe asthma exacerbations, but their role in the development of severe persistent asthma remains unknown. Moreover, in individuals with severe asthma, a number of comorbid conditions are common, including psychopathology, hormonal dysfunction, obesity, rhinosinusitis, gastro-oesophageal reflux and sleep apnoea syndrome. All of these can lead to worsening of the asthma symptoms and a more severe phenotype of the disease. In order to improve the treatment of patients with severe asthma and to control the disease, all of these factors must be taken into consideration, and it is important to manage these conditions concurrently.

  12. Page 120
    Abstract
    Correspondence: S.E. Wenzel, University of Pittsburgh Asthma Institute at UPMC/UPSOM, Dept of Medicine, NW 628 Montefiore, 3459 Fifth Ave, Pittsburgh, PA 15213, USA, Email wenzelse@upmc.edu

    Frequent and severe exacerbations are central characteristics of severe asthma. The presence of a previous asthma exacerbation is widely recognised as the best predictor of a subsequent exacerbation. The mechanisms driving frequent and severe asthma exacerbations remain poorly understood, but probably include genetic, environmental/infectious and psychosocial elements. In addition to specific triggers, asthma exacerbations in severe asthma may occur as part of the natural course of the disease or in relation to certain structural changes, including epithelial/mucus-related changes and hyperinflation. Clinical and physiological paradigms to predict asthma exacerbations have been difficult to identify and implement, although monitoring sputum eosinophils appears to show promise. Treatment options for asthma exacerbations have not changed in decades and continue to be centred around short-acting β-agonists and corticosteroids. An improved understanding of the pathobiology of asthma exacerbations is required to improve both prevention and treatment.

  13. Page 130
    Abstract
    Correspondence: T. Mauad, Dept of Pathology, São Paulo University Medical School, Av. Dr. Arnaldo 455, São Paulo-SP, 01246-903, Brazil, Email tmauad@usp.br

    Deaths due to asthma are considered uncommon, but still account for approximately 287,000 deaths annually worldwide.

    Near-fatal asthma (NFA) is usually associated with hypercapnia, respiratory acidosis and respiratory arrest requiring intubation and mechanical ventilation, and may result in death if not properly treated.

    Epidemics in asthma mortality were observed in the 1960s and 1970s in many countries, but asthma deaths have decreased since the late 1980s with better disease management.

    Factors frequently associated with NFA and fatal asthma (FA) include a previous near-fatal exacerbation of asthma, previous intubation and mechanical ventilation, and intensive care unit or hospital admission for asthma.

    On autopsy, the lungs from FA patients are usually hyperinflated, and there is mucus hypersecretion. Histological findings include epithelial detachment, thickening of the lamina reticularis of the basement membrane, increased submucosal glands and airway smooth muscle, and altered extracellular matrix composition in the large and small airways. Bronchial inflammation is present in the large and small airways, and extends to the peribronchiolar alveoli and perivascular region of the pulmonary artery.

  14. Page 147
    Abstract
    Correspondence: E.H. Bel, Academic Medical Centre Amsterdam, Meibergdreef 9, 1105 AZ Amsterdam, The Netherlands, Email e.h.bel@amc.uva.nl

    Many patients with asthma have an irreversible component of airflow limitation that is often progressive over time despite aggressive anti-asthma treatment. In some patients this leads to severe fixed airflow limitation similar to that seen in chronic obstructive pulmonary disease. The mechanisms underlying fixed airflow limitation in asthma are incompletely understood, but there is now increasing evidence that a number of genetic and environmental risk factors are associated with the development of fixed airway obstruction. An important genetic risk factor is probably related to disintegrin and metalloproteinase domain-containing protein 33 (ADAM33) expression. Other factors include adult onset of disease, absence of atopy, sensitisation to Aspergillus fumigatus, exposure to specific occupational agents, severe airway hyperresponsiveness and frequent asthma exacerbations. In this chapter the epidemiology and natural history of fixed airflow limitation in severe asthma are reviewed, risk factors and mechanisms are discussed and potential preventive and therapeutic options are proposed.

  15. Page 160
    Abstract
    Correspondence: C.E. Brightling, Institute for Lung Health, Dept. of Infection, Immunity and Inflammation, University of Leicester, Leicester, LE3 9QP, UK. E-mail: ceb17@le.ac.uk

    Severe asthma is a complex heterogeneous disease with high mortality and morbidity. Although imaging is an important tool for the assessment of severe asthma, it has limited applications in a clinical setting. Novel qualitative and quantitative imaging techniques have enabled us to study the large airway architecture in detail, allowing for the assessment of the small airway structure and the ability to perform functional or novel physiological evaluations. There is an urgent need for both proof-of-concept studies and large cross-sectional and longitudinal clinical trials in severe asthma to validate and clinically correlate imaging derived measures. This will extend our current understanding of severe asthma pathophysiology and unravel the structure–function relationship. Tools that can integrate various imaging modalities with clinical and physiological data are beginning to emerge and have an unprecedented potential to discover novel severe-asthma phenotypes, and predict mortality, morbidity and the response to existing and novel pharmacological and nonpharmacological therapies.

  16. Page 182
    Abstract
    Correspondence: I. Horvath, Dept of Pulmonology, Semmelweis University, Budapest, Diós árok 1/C, 1125 Hungary, Email hildiko@elet2.sote.hu

    Understanding the biological basis of airway physiology in health and disease requires the integration of cellular and macroscopic functions. This is achieved through different scales of interaction and communication, which occur by a variety of mechanisms between the molecular, cellular and organ level. Airway physiology in severe asthma covers a wide array of data from traditional lung function measurements to recent approaches by systems physiology. This chapter focuses on some important aspects and a few recent findings that are shaping our current understanding of asthma, in particular the severe disease phenotypes.

  17. Page 189
    Abstract
    Correspondence: C. Bachert, Upper Airway Research Laboratory, Dept of Otorhinolaryngology, Ghent University Hospital, De Pintelaan 185, 9000 Ghent, Belgium, Email claus.bachert@ugent.be

    Although reiterated frequently, the relationship between upper and lower airways is not yet appreciated from a clinical standpoint and not sufficiently understood in terms of pathomechanisms. Here, we summarise the evidence for the impact of acute episodes of rhinitis, allergic, nonallergic or viral, on asthma severity and exacerbations, and differentiate subgroups of chronic rhinosinusitis in terms of their likelihood to interfere with lower airway disease and severity in the long term. We also review recent findings on the effect of staphylococcal enterotoxins, also referred to as superantigens, on mucosal inflammation and how they could serve as a link between nasal polyp disease and late-onset partially severe asthma. Finally, we provide evidence for a possible impact of superantigens on severe refractory asthma independent of sinus disease, via an amplification of the eosinophilic inflammation and polyclonal immunoglobulin E formation. The superantigen concept needs to be taken into consideration in epidemiological approaches to the problem of severe asthma, and if confirmed could eventually lead to new therapeutic targets.

  18. Page 200
    Abstract
    Correspondence: S.T. Holgate, Mail Point 810, III Division, Sir Henry Wellcome Research Laboratories, Level F, South Block, Southampton General Hospital, Southampton, SO16 6YD, UK, Email sth@soton.ac.uk

    While asthma is an inflammatory disease of the conducting airways, as the disease becomes more severe and chronic it loses its dependency on allergen exposure but becomes more responsive to other environmental exposures such as viral infection, pollution episodes and exposure to drugs and chemicals. In parallel there is increased evidence of structural remodelling of the airways involving all elements of the airway wall and spreading down the lung to involve small airways. Functional in vitro and biomarker studies applied to asthma biopsies and cells reveal evidence of a chronic wound scenario in which the airway epithelium is more sensitive to injury by environmental exposures and exhibits an impaired repair response. The net result of this is increased secretion of multiple growth factors that drive remodelling and cytokines, chemokines and mediators that sustain inflammation. We have termed the altered communication between the epithelium in asthma and the underlying mesenchymal cells the epithelial mesenchymal–trophic unit (EMTU) since many of the pathways involved are a recapitulation of those utilised in branching morphogenesis of the developing fetal lung.

  19. Page 208
    Abstract
    Correspondence: P.G. Gibson, Dept of Respiratory and Sleep Medicine, John Hunter Hospital, Locked Bag 1, Hunter Mail Centre, Newcastle NSW 2310, Australia, Email Peter.gibson@hnehealth.nsw.gov.au

    Severe asthma is characterised by heterogeneity at the level of both clinical expression and disease mechanisms.

    Airway inflammation is a key component of severe asthma and is heterogeneous in its pattern. This heterogeneity indicates different underlying disease mechanisms and treatment responses in severe asthma.

    There are a range of sampling procedures and biomarkers that are available for the noninvasive assessment of airway inflammation in severe asthma. Induced sputum has been extensively studied; however, other markers have undergone only limited evaluation of their clinical utility. Induced sputum markers for the identification of eosinophilic and non-eosinophilic inflammatory patterns are useful in predicting exacerbation risk and identifying potential treatment responses.

    Future work aimed at improving sampling methodologies and identifying biomarkers that are suitable for fast-throughput or automated assay platforms would add significantly to the noninvasive assessment of airway inflammation in severe asthma.

  20. Page 218
    Abstract
    Correspondence: S-E. Dahlén, Unit for Experimental Asthma and Allergy Research, National Institute of Environmental Medicine, PO Box 287, Karolinska Institutet, SE-117 77 Stockholm, Sweden, Email Sven-Erik.Dahlen@ki.se

    Arachidonic acid and related polyunsaturated fatty acids are the cellular starting material for enzymatic and non-enzymatic reactions that produce a large number of biologically active lipid mediators. Prostaglandins biosynthesised along the cyclooxygenase pathway and leukotrienes generated in 5-lipoxygenase initiated reactions are the two most well-known groups, and both classes of compounds have many known functions as mediators in asthma and inflammation.

    However, with the application of new bioanalytical approaches it is becoming increasingly evident that a number of additional compounds are formed in the lung and inflammatory cells, and sometimes molecules originating from substrates other than arachidonic acid dominate. For example, catalysis by other lipoxygenases or enzymes in the cytochrome P450 family results in lipoxins, 15-HETE, 20-HETE, resolvins and protectins. Yet another group, the isoprostanes, are formed non-enzymatically in response to oxidative stress. Moreover, some compounds, such as lipoxins, appear to be predominantly endogenous anti-inflammatory protective molecules.

    This chapter provides an overview of the different pathways, their products and observations with relevance to diagnosis and therapy of severe asthma. Some considerations that are required for successful in vivo measurements of this family of compounds will also be discussed.

  21. Page 236
    Abstract
    Correspondence: K.F. Chung, Experimental Studies Unit, National Heart and Lung Institute, Imperial College, London, SW3 6LY, UK, Email f.chung@imperial.ac.uk

    Most asthma remains controlled by inhaled corticosteroids (CS) but a significant proportion of patients do not respond to this treatment even when other controllers, such as long-acting β-agonists, are added. Patients with severe asthma continue to have uncontrolled asthma despite being established on high-dose inhaled steroids and often need additional oral CS therapy. In the latter category, there is a need to maximise the benefit/side-effect ratio by using better measures of asthma control and noninvasive biomarkers. CS-sparing agents are generally inefficacious. CS insensitivity is a feature of patients with severe asthma, and probably has similarities to previously described CS-resistant asthma. It is unclear whether those inherently less sensitive to CS develop severe asthma or whether loss of CS insensitivity is acquired. Several conditions have been associated with CS insensitivity in asthma, including obesity, cigarette smoking, vitamin D deficiency and possibly respiratory infections. Molecular mechanisms underlying CS insensitivity include: activation of mitogen-activated protein kinase pathways, particularly p38; overexpression of glucocorticoid receptor β; increased oxidative stress leading to reduction in histone deacetylase activity; or activation of inflammatory transcription factors by interferon-γ. Future treatments for severe asthma may concentrate on targets that will lead to a restoration of CS sensitivity.

  22. Page 253
    Abstract
    Correspondence: E.D. Bateman, University of Cape Town Lung Institute, George Street, Mowbray 7700, Cape Town, South Africa, Email eric.bateman@uct.ac.za

    Bronchodilators play a central role in the symptomatic treatment of severe asthma. β2-adrenergic agonists are the most effective and widely used bronchodilator, but recognition of the potential adverse effects of their misuse (e.g. monotherapy or at too high a dose) underlines the need for caution in their use. β2-agonists with long duration of action, particularly those used both as controllers (in association with inhaled corticosteroids) and for acute relief of intermittent symptoms, has provided another means of addressing symptoms and reducing exacerbations.

    Another promising development for patients with severe asthma is the use of long-acting anti-muscarinics as controller therapy. Theophylline is of limited value, mainly because of its weak bronchodilator properties and side-effect profile. The challenge in managing severe asthma is to reduce the need for and reliance on bronchodilators and give pre-eminence to controller therapy such as inhaled corticosteroids. In addition, in some patients, the use of other classes of controller therapy may improve the clinical impact of bronchodilator use, and reduce reliance on bronchodilators.

  23. Page 268
    Abstract
    Correspondence: P. Chanez, Clinique des bronches de l’allergie et du sommeil, APHM Hopital Nord, Université de la Méditerranée, Chemin des bourrelys, 130015 Marseille France, Email pascal.chanez@univmed.fr

    There is an urgent need for new treatments particularly with anti-inflammatory properties since not all patients with asthma respond adequately to current anti-asthma treatments, and the response of many patients with severe asthma to corticosteroids and bronchodilators is suboptimal. The optimal use of corticosteroids and bronchodilators should be determined based on evidence. Improved drugs in these categories may be of some benefit. Anti-immunoglobulin E therapy has now been established for the treatment of severe allergic asthma. More specific anti-inflammatory targets are provided by the range of mediators and second messengers that contribute to the chronic inflammatory and remodelling processes in asthma. The targeted inhibition of interleukin (IL)-4/IL-13, IL-5 and IL-17 are currently being investigated. Anti-tumour necrosis factor-α approaches have been ineffective in severe asthma patients and have been associated with infections and malignancies. Other approaches target corticosteroid insensitivity in severe asthma, such as devising small molecule inhibitors of kinases. Bronchial thermoplasty has been reported to provide some benefit to patients with severe asthma. Future specific therapies will be targeted towards specific phenotypes of severe asthma.

  24. Page 282
    Abstract
    Correspondence: A. Menzies-Gow, Royal Brompton Hospital, Sydney Street, London, SW3 6NP, UK, Email a.menzies-gow@rbht.nhs.uk

    Severe asthma remains a frustrating condition for both the patient and the clinician. Significant advances have been made over the last decade and there is now a consensus for these patients to receive a thorough systematic assessment to ensure that asthma is the correct diagnosis, that patients adhere to their prescribed therapies and that any comorbidities are treated.

    Several cohorts of patients with severe asthma have been published. They have clearly demonstrated that severe asthma is a heterogeneous condition comprising of several different phenotypes. This heterogeneity, to a certain extent, explains the failure of any one medication to be effective for the whole severe asthma population.

    When examining the treatment options, it is apparent that there is little evidence to guide the clinician. The majority of clinical trials have been carried out in milder asthmatics and the results extrapolated to the more severe population. The hope for the next decade lies with identifying all the relevant phenotypes of severe asthma to allow effective targeted therapy for the individual patient.

  25. Page 297
    Abstract
    Correspondence: K.F. Chung, National Heart & Lung Institute, Imperial College, Dovehouse St, London, SW3 6LY, UK, Email f.chung@imperial.ac.uk

    The publication of the first definitions more than 10 years ago has helped to focus attention on severe asthma and there has been progress in understanding about this condition. Severe asthma is heterogeneous in terms of its natural history, pattern of airflow obstruction, inflammatory component and therapeutic response to corticosteroid therapy, and it is necessary to define and validate its phenotypes. There have also been advances in understanding of the inflammatory process and the airway wall remodelling changes of severe asthma as compared to non-severe asthma; clearly the inflammatory process is different and the remodelling process appears more accentuated in severe asthma. Some measures of innate immune responses may be diminished and corticosteroid responsiveness is impaired. The complex heterogeneity of severe asthma may require an unbiased approach, with integration of high-dimensional datasets and systematic modelling of perturbed networks. Definition of the phenotypes of asthma is also important for emerging target-specific therapies. Looking to the future, phenotypic characterisation will become a crucial step, together with the definition of pathophysiological and inflammatory mechanisms for each phenotype. Such an integrative approach will afford a better chance of discovering the target-specific treatments that will be effective in specific phenotypes of severe asthma.