Severe Asthma in Children: Therapeutic Considerations

Louise Selby; Sejal Saglani


Curr Opin Allergy Clin Immunol. 2019;19(2):132-140. 

In This Article

Abstract and Introduction


Purpose of review: Children with poor asthma control despite maximal maintenance therapy have problematic severe asthma (PSA). A step-wise approach including objective adherence monitoring and a detailed multidisciplinary team assessment to identify modifiable factors contributing to poor control is needed prior to considering therapy escalation. Pathophysiological phenotyping in those with true severe therapy-resistant asthma (STRA) and the current array of add-on therapies will be discussed.

Recent findings: Adherence monitoring using electronic devices has shown that only 20–30% of children with PSA have STRA and need additional therapies. Omalizumab and mepolizumab are licensed for children with STRA aged 6 years and older. Although robust safety and efficacy data, with reduced exacerbations, are available for omalizumab, biomarkers predicting response to treatment are lacking. Paediatric safety data are available for mepolizumab, but efficacy data are unknown for those aged 6–11 years and minimal for those 12 years and older. A sub-group of children with STRA have neutrophilia, but the clinical significance and contribution to disease severity remains uncertain.

Summary: Most children with PSA have steroid sensitive disease which improves with adherence to maintenance inhaled corticosteroids. Add-on therapies are only needed for the minority with STRA. Paediatric efficacy data of novel biologics and biomarkers that identify the optimal add-on for each child are lacking. If we are to progress toward individualized therapy for STRA, pragmatic clinical trials of biologics in accurately phenotyped children are needed.


Problematic Severe Asthma. Severe childhood asthma which remains uncontrolled has a long-term detrimental effect on adult lung function and is associated with development of chronic obstructive pulmonary disease.[1] Problematic severe asthma (PSA) is an overarching term describing a child with a confirmed diagnosis who has persistent symptoms despite maximal conventional treatments.[2,3,4] Before escalating treatments further, PSA should be stratified into difficult asthma (with modifiable factors, such as allergen avoidance, smoke exposure, adherence to prescribed medications and psychosocial issues) or true severe therapy resistant asthma (STRA) – poor control after modifiable factors and the basics of asthma management have been addressed.[5] A step-wise approach to diagnosis and management[6] of PSA ensures that poor control is not because of a wrong diagnosis, or a potential easily remediable factor is not wrongly treated with expensive biologics (Figure 1).

Figure 1.

Flow sheet of proposed diagnostic work-up for problematic severe asthma to phenotype disease. DOT, directly observed therapy; EMD, electronic monitoring device; FeNO, fractional-exhaled nitric oxide.

Difficult Asthma. Simple modifiable factors contributing to 'difficult' disease include incorrect prescription of inhaled corticosteroids (ICS), poor inhaler technique, inappropriate device for the age of child, poor adherence and lack of family asthma education.[7] However, multiple complex factors often contribute to asthma being difficult and can only be identified following multidisciplinary team (MDT) input.[3] These include persistent exposure to aero-allergens to which the child is sensitized (house dust mite, pet dander, and moulds) and tobacco smoke. Objective confirmation of smoke exposure by measuring urinary or salivary cotinine levels helps ensure the family seek cessation advice.[8] Minimizing aeroallergen exposure in sensitized children is essential as there is a known relationship between exposure and increased disease severity.[9] A nurse-led home visit helps identify and tackle modifiable risk factors, and in a cohort of difficult asthmatic children prevented treatment escalation in just over half of all children.[7,10] Breathing pattern disorders including vocal cord dysfunction and hyperventilation may also contribute to difficult asthma and are often present with anxiety and psychosocial exacerbators. Ideally, all children with PSA should have a physiotherapy assessment to allow detection and management of dysfunctional breathing[11] and clinical psychology assessment.[10] The complex interplay of factors contributing to difficult asthma control means that the MDT evaluation may need to be undertaken during an in-patient stay. This approach showed improvement in asthma control from 18 to 69% in children with difficult asthma.[12] Moreover, immediate and sustained improvement in objective measures including spirometry, exhaled nitric oxide and exacerbations was demonstrated in 24/26 children following a 2-week in-patient assessment.[10] Alarming features prompting in-patient assessment include excess use of short-acting bronchodilator, discrepancy between symptom reporting and objective markers of disease severity, and safeguarding concerns.[10]

Objective Assessments of Adherence to Inhaled Corticosteroids. The most common modifiable factor underlying difficult asthma is poor adherence to maintenance therapy, encompassing 45–55% of all patients (adults and children).[4,13] Good adherence is defined as the administration of at least 80% of prescribed doses of ICS.[14] The British Thoracic Society Guidelines state all patients with asthma must have an annual adherence assessment,[15] and the proposed gold standard method is using an electronic monitoring device (EMD); however, other options include prescription pick up/refill. For children with PSA, we propose objective assessment is an absolute requirement prior to consideration of therapy escalation.

EMDs provide data on the date and time of inhaler actuation. An observational prospective cohort study assessed spirometry with bronchodilator reversibility, fractional exhaled nitric oxide (FeNO), asthma control test (ACT) scores and quality of life scores before and after a median of 92 days with an EMD (median age 12.4 years). Suboptimal adherence (<80%) was demonstrated in 58%. Children with good adherence were split into those with improved control (need encouragement to maintain adherence), or those with persistent poor control (STRA). Among children with poor adherence, there was a sub-group whose control improved (likely over-treated), and a second sub-group with persistent poor control, the latter requiring an adherence intervention such as directly observed therapy (DOT). The prevalence of STRA was 18% of the entire cohort with PSA.[4] The utility of an EMD as an intervention to improve adherence is questionable. A multicentre, open-label, parallel-group trial in which all participants received EMDs for 3 months, but an intervention group had a review of their adherence data and personalized intervention strategies (such as alarm reminders) were compared. Average adherence was higher in the intervention group compared to controls (70 vs. 49%) over a 12-month period, whereas adherence rates declined over time in the control group. The intervention group had reduced exacerbations requiring oral steroids or hospitalization.[16] EMDs alone may not change adherence behaviours. The complexity and variation in adherence interventions means no single approach will work for all children and just as biological treatments need to be personalized, interventions to alter adherence behaviours also need to be individualized if they are to succeed.[17]

Although currently not used routinely in children with PSA, an alternative measure of adherence is the exhaled nitric oxide suppression test (normalization of FeNO following a period of DOT with ICS). This distinguishes patients with difficult asthma with steroid-sensitive disease from those with true steroid refractory asthma. To date, this was only undertaken following a period of DOT in hospital, but a web-based interface with integrated remote monitoring technology has now been used to deliver FeNO suppression testing at home.[18] Eight per cent of adults with difficult asthma completed the test remotely, and 64% of those had a positive FeNO suppression test with good adherence, and thus did not have STRA. Objective assessments of adherence show only 20–30% of all children with PSA have true STRA needing therapy escalation. However, there is one group of high-risk children who remain a significant challenge. Approximately 18% of children with PSA have persistent poor adherence and persistent poor control, despite all efforts to improve adherence. These children are at high risk of asthma death[19] and have been called 'refractory difficult asthma'.[6] This is a group for whom biologics administered in hospital may be the only safe option, even though, in truth, their disease is not necessarily treatment refractory.

Pathological Phenotype Should Guide add-on Therapy in Severe Therapy-resistant Asthma. As a group, children with STRA have multiple aeroallergen sensitization, eosinophilic airway inflammation and airway remodelling,[20,21] but there is a wide spectrum of disease heterogeneity. Poor control despite high-dose ICS, or regular oral steroids, means additional treatments to improve control and minimize the adverse effects of steroids need to be considered. However, investigations allowing an assessment of the airway pathological phenotype should be undertaken first (Figure 1). A detailed assessment of allergen sensitization, bronchoscopy to determine upper airway abnormalities such as tracheomalacia, lower airway inflammation and infection are important prior to adding new treatments. A trial of high-dose systemic steroids allows assessment of the specific pattern of clinical response. Whether improvement occurs in lung function, airway eosinophilia, or symptoms may enable a targeted choice of add-on treatment.[22] Children with STRA who had an improvement in FeNO after intramuscular triamcinolone were most likely to respond to omalizumab.[23] Until very recently, choice of licensed add-on therapies for paediatric STRA was limited to the monoclonal antibody to IgE, omalizumab. However, mepolizumab, an antiinterleukin (IL)-5 antibody has now been licensed and additional biologics, such as the anti-IL-5 receptor antibody, benralizumab, have been approved in adults[24–26] and may follow for children, making the need for detailed clinical and pathological phenotyping in the individual child prior to deciding add-on therapy essential.[27]

Therapeutic Considerations for Allergic, Eosinophilic Severe Treatment-resistant Asthma: Omalizumab and Mepolizumab. Molecular-based therapies allow control of a predominant airway phenotype and may enable reduction in ICS dose,[28] use must therefore be matched to disease phenotype (Figure 1).[29] Omalizumab reduces cell-bound IgE, down-regulates IgE receptors and prevents pro-inflammatory mediator release. Childhood STRA is characterized by allergen sensitization in the majority,[2] thus making omalizumab an attractive therapeutic option. It is indicated as add-on to maintenance therapy for children aged at least 6 years with severe persistent asthma with elevated serum IgE (>30 and <1500 IU/ml) and positive specific IgE to at least one aeroallergen.[30] Many studies have assessed and confirmed real-life efficacy and safety of omalizumab in children with severe asthma (reviewed in detail elsewhere[31,32]). Overall, reduction in exacerbations and oral corticosteroid withdrawal are apparent.[33] However, only approximately two-thirds of children benefit, with no impact on lung function.[34] The Omalizumab effectiveness in patients with severe allergic asthma according to blood eosinophil count (STELLAIR) study was a retrospective review which related omalizumab effectiveness to blood eosinophils; assessing exacerbation rates, clinical evaluation and a combination of both. Mean annual exacerbation rate reduced by 70.4% and hospitalizations by 73.2% in children, but omalizumab efficacy was unrelated to blood eosinophils, suggesting this is not a good biomarker of response.[35] The presence of multiple and severe allergies in childhood STRA[36] often limits omalizumab use because of restrictive prescribing guidelines within a relatively narrow serum IgE range. Approximately 30% of children with STRA have IgE levels higher than the recommended[22] range. However, the prescribing range has been derived from adult studies and a recent report has shown omalizumab was well tolerated in children with a serum IgE higher than approved, and 10/11 had reduced need for corticosteroids, emergency department visits or hospitalizations in the year after therapy.[37] Studies that confirm the optimal serum IgE range for children are lacking. Given the majority of children have both elevated IgE and eosinophils, if a therapeutic trial of omalizumab has failed, mepolizumab needs to be considered (Figure 2).

Figure 2.

Indications for omalizumab and mepolizumab and their effects on clinical parameters and potential biomarkers of disease. Children who have both elevated IgE and eosinophilia could be considered for a trial of anti-IL-5 immunomodulation should omalizumab fail.

Antiinterleukin-5 Immunomodulators. Mepolizumab is an anti-IL-5 humanized monoclonal antibody that reduces circulating eosinophils[38] and is licensed for use as add-on therapy in severe refractory asthma in patients with a peripheral blood eosinophil count of at least 300 cells/μl in the past 12 months with at least four exacerbations requiring oral corticosteroids, or continuous use of oral corticosteroids for at least 6 months. Abundant safety and efficacy data are available in adults with severe eosinophilic asthma showing reduced exacerbation frequency (especially in those with blood eosinophils >150 cells/μl[39]), improved quality of life and asthma control.[40] Unfortunately, neither convincing efficacy nor safety data are available for children. Pooled data obtained from the Mepolizumab as Adjunctive Therapy in Patients with Severe Asthma (MENSA) and the Dose Ranging Efficacy And safety with Mepolizumab (DREAM) trials showed reduction in exacerbation rates for 12–17 year olds and no adverse events.[41] However, only approximately 30 of the total 800 participants in the trials were in this age group. Despite this scant evidence, and no efficacy data at all in children aged 6–11 years, in August 2018, the European Medicines Agency approved mepolizumab as an add-on therapy for children aged 6–17 years with STRA based on safety data being similar between children aged 6–12 years and those more than 12 years.[42] The extrapolation of efficacy and safety data to enable licensing for children is a concern. Especially when it has been reported that despite 85% of STRA children having airway eosinophilia, levels of Th2-signature cytokines IL-4, IL-5 and IL-13 were difficult to detect.[2] A sub-group had higher levels of Th2 cytokines, but efficacy of mepolizumab in all children with eosinophilic disease is highly questionable. Now that both the Federal Drugs Agency and the European Medicines Agency have approved mepolizumab in children, it is essential that we urgently undertake pragmatic trials of safety and efficacy in children.[28] At present, and until we have paediatric specific biomarkers of response, we suggest mepolizumab should be used in children who are either ineligible for, or have had a failed trial of, omalizumab (Figure 2).

Numerous biologics are now available for severe asthma with recent approval for use of reslizumab and dupilumab. These predominantly target Th2 high disease; reslizumab is an anti-IL5 antibody administered intravenously, dupilumab an antibody directed against the α subunit of the IL-4 receptor which functions by blocking the early signal transduction of the Th-2-mediated immune response[43] and downstream signalling of IL-13[44,45] and benralizumab, another antibody against the IL-5 receptor. However, trials were undertaken in patients over 12 years, including a minority of children, which means true efficacy and safety in children remains unknown, and potential biomarkers that predict response may not be applicable. Importantly, all are compared to placebo which means currently we have no way of deciding optimal add-on treatment for individual patients. Pragmatic trials assessing and comparing efficacy of biologics in children aged 6–17 years with STRA who have been accurately clinically and pathologically phenotyped are needed. If this does not happen, there is potential for us to cause harm, especially as complete depletion of circulating inflammatory cells such as eosinophils may disrupt immune homeostasis and have deleterious consequences.[46]