Immunoregulation of asthma by type 2 cytokine therapies: Treatments for all ages?

Asthma is classically considered to be a disease of type 2 immune dysfunction, since many patients exhibit the consequences of excess secretion of cytokines such as IL‐4, IL‐5, and IL‐13 concomitant with inflammation typified by eosinophils. Mouse and human disease models have determined that many of the canonical pathophysiologic features of asthma may be caused by these disordered type 2 immune pathways. As such considerable efforts have been made to develop specific drugs targeting key cytokines. There are currently available multiple biologic agents that successfully reduce the functions of IL‐4, IL‐5, and IL‐13 in patients, and many improve the course of severe asthma. However, none are curative and do not always minimize the key features of disease, such as airway hyperresponsiveness. Here, we review the current therapeutic landscape targeting type 2 immune cytokines and discuss evidence of efficacy and limitations of their use in adults and children with asthma.


Introduction
Asthma is typically characterized by eosinophilic inflammation and airway hyperresponsiveness coupled with a variety of structural airway changes that are collectively termed airway remodeling. It has traditionally been thought of as a disease of type 2 immunity due to increased levels of interleukin (IL)-13, IL-4, and IL-5. Disordered T-cell immunity and hypersecretion of these canonical type 2 cytokines are central to the immune pathophysiology associated with multiple allergic disorders, including asthma.
Therefore, these cytokine pathways have been the focus of interest as potential targets for novel asthma therapies. Data from mouse models indicate that these cytokines are responsible for many of the classical features of asthma, thus efforts from pharmaceutical companies have been directed toward blocking their function in vivo, via either neutralizing antibodies or receptor Correspondence: Prof. Clare M. Lloyd and Sejal Saglani e-mail: c.lloyd@imperial.ac.uk; s.saglani@imperial.ac.uk blocking molecules. A range of clinical trials with these agents serve as in vivo human "experiments" and have revealed important findings indicating the function of type 2 cytokines in humans that do not always reflect the data previously collected from mouse models. Here, we review the role of type 2 cytokines in driving cellular pathophysiology in asthma patients and discuss how clinical trials have added to our body of knowledge regarding the function of type 2 cytokines in human biology. Importantly, we also review the efficacy of targeting type 2 cytokines in children with asthma, how responses may vary from those in adults, and the importance of designing clinical trials in age-appropriate individuals to directly compare the efficacy and mechanism of action of type 2 biologics. 11 in mice. Although they share some common features, each also has a distinct functional profile (Fig 1). IL-4 is produced by activated T cells, mast cells, basophils, innate lymphoid cells (ILCs), and eosinophils. It is a critical cytokine for driving type 2 immunity as it facilitates differentiation of naïve Th cells into Th2 cells. This is a property unique to IL-4, since other type 2 cytokines do not drive T-cell differentiation. IL-4 is also important for dendritic cell differentiation, required to maintain immune tolerance via stimulation of regulatory T cells (Tregs). Elevated IL-4 expression can enhance the type 2 response via repression and impairment of Treg tolerogenic functions [1,2]. IL-4 also promotes class switching of immunoglobulin to IgE in B cells; increases expression of IgE receptors on the cell surface and specifically on mast cells and basophils enhances expression of the high-affinity IgE receptor [3,4]. IL-4 increases expression of VCAM on pulmonary vascular endothelial cells, thereby facilitating the transmigration of eosinophils across the vascular endothelium and increases expression of eosinophilic chemokines such as the eotaxin family, as well as inhibiting eosinophil apoptosis [5]. IL-4 also exerts effects on nonimmune cells: enhancing mucin gene expression thus promoting mucus hypersecretion, and increasing secretion of cytokines from fibroblasts, so potentially enhancing tissue remodeling [6].
IL-13 is primarily involved in the effector phase of type 2 immune reactions, and thus influences the development of many of the classical pathophysiological features of asthma due to its effects on lung structural cells such as epithelium, fibroblasts, smooth muscle cells, and endothelium as well as immune cells [7]. IL-13 inhibits macrophage production of pro-inflammatory cytokines, TNF, IL1β, and chemokines, but increases IL-12 secretion by macrophages and dendritic cells (DCs) [8]. Activation of B cells in the presence of IL-13 facilitates their proliferation and upregulates expression of CD23, MHCII, and IgM, and also promotes class switching to IgE and IgG1 [9]. Unlike IL-4, IL-13 plays no role in T-cell differentiation since naïve T cells do not express IL-13R. IL-13 can influence the recruitment of eosinophils and basophils via generation of chemokines such as eotaxin. IL-13 is closely related to IL-4, but responses evoked are generally smaller in magnitude [10]. IL-13 increases inducible nitric oxide synthase (iNOS) expression in airway epithelial cells, the predominant contributor of exhaled nitric oxide (Fe NO ), a measurement used as a clinical biomarker to indicate type 2 inflammation in the airways [11]. IL-13 also induces a variety of effects on stromal cells in the lungs that likely promote the induction of tissue remodeling: regulation of epithelial barrier function, differentiation and proliferation of mucus secreting goblet cells, induction of smooth muscle hypertrophy and enhanced contractility, production of ECM, and myofibroblast differentiation [12].
IL-5 is a pro-inflammatory cytokine that is responsible for eosinophil maturation, differentiation, activation, and migration [13]. A range of animal models have demonstrated a close association between IL-5 and eosinophil function, particularly following exposure to allergens or during helminth infections. The effects of IL-5 on eosinophilopoiesis extend from the BM to the site of local inflammation, which in the case of allergen driven inflammation is the bronchial mucosa. Raised eosinophil numbers in the blood, sputum, and airway tissue and lavage are indicative of asthma in adult patients and correlate with elevated levels of IL-5 [14][15][16]. IL-5 promotes the differentiation and maturation of CD34 + eosinophil progenitors both in BM but also in the bronchial mucosa [14,17]. In addition, IL-5 synergizes with the eotaxin family of chemokines to promote the migration of eosinophils to the asthmatic airway [18,19]. Other effects that IL-5 exerts on eosinophils specifically include apoptosis, adhesion to endothelial cells as well as to the ECM, which may reflect a potential role in development of airway remodeling.
The predominant cellular sources of IL-5 include Th2 cells and ILC2, but natural killer T cells, mast cells, and eosinophils themselves are also capable of producing IL-5. IL-5 release from Th2 cells is triggered by inhaled allergen reacting with DCs, while secretion from ILC2 is dependent on GATA3 activation induced by epithelial derived alarmins such as IL-25, IL-33, and thymic stromal lymphopoietin (TSLP) [20].

Receptor binding
Type 2 cytokines bind to a series of receptors composed of distinct but also shared chains (Fig. 2). The IL-4R is a cell surface heterodimeric complex composed of a specific immunoglobulin high-affinity alpha chain combined with a second chain that can be either the common gamma chain or the alpha chain of the IL-13R in order to facilitate the action of either IL-4 or IL-13 within different tissues [21]. IL-4Rα coupled with the common gamma chain forms the type I IL-4 receptor and binds solely IL-4. In contrast, the type II heterodimeric complex comprises IL-4Rα coupled with IL-13Rα, the low-affinity binding receptor for IL-13 that binds IL-13 and IL-4 with high affinity. Both chains of the receptor are required for cellular activation, and only the alpha chain is needed for IL-4 binding (Fig. 2). The type I receptor is mainly expressed by hematopoietic cells, primarily lymphocytes with minimal expression by nonhematopoietic cells such as epithelial or stromal cells. Conversely, the type II receptors are mainly expressed by epithelial cells, while myeloid cells express both types [21].
A soluble form of IL-4Rα is also capable of binding to IL-4 but lacks both a transmembrane and cytoplasmic domain and so cannot signal, thus can act as a decoy. This antagonistic receptor serves as an anti-inflammatory mechanism that counteracts the effects of IL-4 and might represent an autoregulatory or homeostatic mechanism [22]. Thus, sIL-4Rα has therapeutic potential as a cytokine antagonist, and clinical trials indicate it is safe and has some efficacy in moderate asthma; however, it is not currently being investigated as a therapeutic target [22,23].
IL-13 also has two receptors, but in contrast to IL-4, these comprise two separate binding chains: IL-13Rα1 and IL-13Rα2. The main high-affinity IL-13R complex comprises IL-13Rα1 in conjunction with IL-4Rα. This IL-13 Rα1 is expressed by a range of immune cells that include B cells, monocytes/macrophages, mast cells, and basophils but not T cells, as well as endothelial cells and airway epithelial cells [24]. Although IL-13Rα2 binds IL-13 with Figure 1. Functional profile of type 2 cytokines in asthma. In response to inhaled environmental triggers, hypersecretion of the type 2 cytokines, IL-4, IL-5, and IL13 drives the pathological features of asthma: IgE production, airway hyper-responsiveness, eosinophilia, goblet cell hyperplasia, and mucus hypersecretion. IL4 production by Th2, ILC2s, mast cells, DCs, and eosinophils promote IgE class switching, eosinophil transmigration, mucus hypersecretion, and fibroblast activation. Uniquely, IL4 is key for the differentiation of naïve Th cells into Th2 cells. IL5 is produced by various immune cells, however Th2 and ILC2s are the predominant cellular source. Eosinophil maturation, differentiation, and activation are all driven by IL5. IL13 has a similar functional profile to IL4, is predominantly produced by Th2 and ILC2 cells, and is a key contributor of airway remodeling. Novel biologics developed to inhibit the function of IL4/IL13 (dupilumab, lebrikizumab) and IL5 (benralizumab and mepolizumab) as well as the anti-IgE Omalizumab are depicted. To exert their biologic effects, the type 2 cytokines can bind to a series of receptors. IL4 can signal via a type I or type II receptor. Upon IL4 ligand binding to the IL4Ra receptor, the recruitment of either the common gamma chain (γc, type 1) or the alpha chain of the IL13R (IL13Rα1, type II) forms a heterodimeric complex. Type I receptors are expressed on all hematopoietic cells and are exclusive to IL4 ligand binding. Nonhematopoietic cells such as the structural airway cells express IL4Rα1, whereas myeloid cells express both the γc and IL4Rα1 receptors. IL4 type I receptor binding activates STAT6 or IRS2 to promote a type 2 (T2) response and macrophage and fibroblast activation. IL13 ligand binding to the type II receptor IL13Rα1 forms a heterodimeric complex with the IL4Rα chain. Type II receptor binding by IL4 or IL13 leads to the activation of STAT3 or STAT6 to induce airway hyper-responsiveness (AHR) and mucus hypersecretion. IL13Rα1 is expressed on a range of immune and structural airway cells. IL13 binds with higher affinity to IL13Rα2, known as a "decoy receptor", currently thought to attenuate IL13/IL4/STAT6 signaling. IL5 ligand binding to the type I IL5Rα that is only expressed on eosinophils and basophils induces the recruitment of the common βc chain to promote STAT1,2,5 signaling and modulate eosinophil biology. much higher affinity than IL-13Rα1, it does not signal and is thus considered a decoy receptor. It potentially serves to downregulate IL-13 functions, although its function is not really known. Given that IL-4 can also signal through the IL-13R complex, it operates as a second receptor for IL-4 and is referred to as type II IL-4R [21]. In contrast, the type I IL-4R complex, which consists of the IL-4R α chain and the common γ chain, is specific for IL-4 ( Fig. 1).
Alveolar macrophages are the most predominant immune cell in the lung and capable of differentiating into distinct subtypes in response to different stimuli. Resting macrophages are activated by IL-4 and IL-13, developing into a macrophage phenotype that promotes repair and remodeling [25]. Thus, both protective and pathogenic roles have been identified for these key effector cells. Ongoing studies will ascertain the benefit of targeting alveolar macrophages subtypes for the treatment of asthma [26].
The biological effects of IL-5 are mediated following binding with the specific IL-5R. This is a type I cytokine receptor consisting of a heterodimer of the IL-5Rα subunit combined with a nonspecific common βc subunit, which is also found in the receptor complexes for IL-3 and GMCSF [27]. The IL-5Rα is exclusively expressed by eosinophils and some basophils, and as a monomer is a low affinity receptor, while dimerization with the β-chain produces a high affinity receptor. The alpha subunit binds IL-5 while the β subunit is nonligand binding but facilitates signal transduction and is expressed on virtually all leukocytes.

Intracellular signaling pathways initiated by IL-4, IL-5, and IL-13
Engagement of either IL-4 or IL-13 with their receptor promotes activation of the signal transducer and activator of transcription (STAT6). Type I IL-4R can also activate alternative transcription factors from the insulin response substrate (IRS)-2 family, thereby linking the IL-4R with other downstream signaling pathways that include PI3K/mTORC2, AKT, AHC/MAPK, and Shp-1 [28]. Other pathways implicated in receptor signaling include STAT3 activation via IL-13Rα1 and IRS2 upregulation by Socs1/ubiquitin [29].
Cytokine binding triggers trans-phosphorylation and activation of the IL-13R cytoplasmic tyrosine kinases from the Janus family protein kinases (JAKs), which are recruited to the complex. IL-13 can recruit different combinations of JAKs depending on the tissue location of the receptor. JAK1, JAK2, and JAK3 are associated with IL-4Rα, γc, and IL-13Rα1 chains, respectively. These JAKs are then able to phosphorylate STAT6 via its SH2 domain, the principal transcription factor activated by both IL-4 and IL-13. While STAT6 induces distinct gene expression profiles in different cell types (reviewed in [28]), it activates the majority of the genes induced by IL-13 and is thus responsible for the majority of the pathophysiologic features typical of asthma [29].
Triggering of the IL-4R/STAT6 axis via IL-4 and IL-13 promotes allergic inflammation, and mice with targeted deletion of either IL-4Rα or IL-13Rα1 have provided insight into the functions of the individual receptors [30,31]. IL-4R type I is critical for the generation of type 2 responses, development of alternatively activated macrophages and fibroblast activation, while the type II receptor is essential for development of allergen induced airway hyperreactivity and mucus hypersecretion [29].
Binding of IL-5 initiates the generation of a functional IL-5Rα/βc receptor complex, which in turn initiates a network of transcription factors that consist of JAK1/2-STAT1/3/5 modules, p38, and ERK MAPK, as well as NFκB [32]. The ensuing activation of specific target genes leads to eosinophil maturation, enhanced survival/reduced apoptosis, as well as activation.

Biologics currently available: The challenge of individualized management
Given the observed effects of type 2 cytokines in driving key features of pathology efforts to develop novel therapies for asthma have been directed toward mitigating the effects of these cytokines [33,34].
Pharmaceutical companies have developed a range of drugs designed to reduce type 2 cytokines by either neutralizing the cytokine itself or binding to the receptor. At present, the licensed indication for all available targeted treatments is only for severe asthma [35], which cannot be controlled despite high-dose inhaled and/or oral corticosteroids. Since most cytokine directed treatments available at present target either IL-5, IL-4, or IL-13, and thus type 2 immunity, predicting which specific therapeutic will be best for which individual patient can be difficult. The choice of treatment for each patient will be determined by clinical and biological biomarker results (blood eosinophils, exhaled nitric oxide, number of exacerbations in the previous 12 months), but if a patient is eligible for more than one cytokine blocking approach, then in the absence of head-to-head trials of efficacy, the pragmatic approach is often determined by physician experience, or the order in which licensing was approved.

Blocking IL-5 in severe asthma
Mepolizumab and reslizumab are both anti-IL-5 antibodies and are indicated as add-on maintenance treatments for severe eosinophilic asthma. Mepolizumab, administered subcutaneously, is licensed in patients aged ≥6 years, while reslizumab, which can only be given intravenously, is only licensed for adults aged ≥18 years. The route of administration means mepolizumab is often the preferred choice clinically.
Phase 3 clinical trials of add-on treatment with mepolizumab or reslizumab have shown reduced exacerbation rates by approximately 50% and improved health-related quality of life in adult patients with severe eosinophilic asthma with recurrent exacerbations. These outcomes were irrespective of the presence or absence of allergen sensitization [36][37][38]. Although different cut-off values for blood eosinophil counts were used in these studies (≥150 cells/mcL at screening or ≥300 cells/mcL in the previous 12 months for mepolizumab, and ≥400 cells/mcL for reslizumab), blood eosinophilia was a better predictor of a therapeutic response to anti-interleukin-5 antibody than exhaled nitric oxide (FeNO) [33,39]. Mepolizumab efficacy was assessed at different doses, but the lack of significant differences in exacerbations at the lowest dose of 75 mg compared to 750 mg resulted in the lowest dose being approved for clinical use. However, impact on reducing blood eosinophils was dose dependent, and a reduction in airway eosinophils was only seen at the highest dose [33].
Benralizumab is a monoclonal antibody that targets the alpha subunit of IL-5R. In a bronchoscopic study, benralizumab reduced eosinophil counts in the airway mucosa and in the airway lumen (sputum) by at least 90% and completely depleted blood eosinophils [40]. Phase 3 clinical trials involving predominantly adults with exacerbation-prone, severe eosinophilic asthma (baseline blood eosinophil counts of ≥300 cells/mcL), have shown addon treatment with benralizumab (administered subcutaneously every 4 or 8 weeks) significantly reduced asthma exacerbation rate and improved prebronchodilator forced expiratory volume in 1 s (FEV 1 ) compared with placebo [41,42]. The attraction of benralizumab is the efficacy of the 8-week dosing regimen, which has been recommended for its licensed use. Further realworld and open-label extension studies have confirmed the efficacy and long-term safety of benralizumab in patients with severe eosinophilic asthma [43,44].
Despite the overall efficacy in large trials, all current therapies that target IL-5 or its receptor rely on the measurement of a single elevated blood eosinophil count to determine eligibility. However, variation of blood eosinophils over time has been documented in patients that were in the placebo arms of the phase 3 clinical trials [45]. Moreover, there is little relationship in children, between airway and blood eosinophils, and of the various biomarkers used to define eosinophilic asthma, since no single biomarker truly reflects airway eosinophilia, which is the target of biological therapies [46,47]. It is therefore hard to predict efficacy in the individual patient. A limitation of the therapies that target IL-5 is their relatively low impact on lung function. Other type 2 cytokines have therefore been targeted to try and exert maximal effect on airway hyperresponsiveness and lung function.

Blocking IL-13 in severe asthma
Murine studies of anti-IL-13 therapies very consistently showed an improvement in airway hyperresponsiveness, this was therefore an attractive target for type 2 asthma. Lebrikizumab, a neutralizing antibody against IL-13, which blocked its interaction with IL-4Rα, was first tested with a primary endpoint of relative change in prebronchodilator FEV 1 . Prespecified subgroups of patients were defined to try to identify the phenotype that might respond best, and this was done according to serum IgE, blood eosinophils, and serum periostin. The primary outcome was positive in the first phase 2 trial, showing greater efficacy in those with higher serum periostin levels [34]. This approach of identifying type 2 high patients using serum periostin or blood eosinophils to determine efficacy was subsequently adopted in larger phase 3 trials. The primary endpoint was rate of exacerbations, but the final outcome was not positive even in those with enhanced periostin or blood eosinophil levels [48]. The intervention was thus abandoned as a therapy for severe asthma by the company.
Tralokinumab is another IL-13 neutralizing antibody that has been tested in severe adult asthma. An early phase 2b trial of tralokinumab given every 2 weeks improved pre-bronchodilator FEV1, but it did not reduce asthma exacerbation rates in patients with uncontrolled asthma [49]. Tralokinumab was subsequently evaluated in two large multicenter phase 3 clinical trial in patients with moderate-to-severe asthma [50]. In the first, there was no significant decrease in annual exacerbation rate compared to placebo, but there was a significant exacerbation rate reduction in patients with high FeNO. However, in the second large clinical trial which prespecified treatment effect in patients with high FeNO (>37 ppb), there was no longer an improvement in exacerbation rate compared to placebo [50]. It has also failed to show an impact of oral corticosteroid reduction in patients with very severe asthma [43].
The "failed" efficacy of anti-IL-13 antibody therapy in severe asthma has been a huge lesson about the potential discrepancy between molecules that show promise from mechanistic preclinical studies and pathobiology to efficacy in clinical trials [51]. There are several explanations for lack of clinical efficacy, including the heterogeneity of the biomarkers used for severe asthma, need for increased awareness about their longitudinal variation and care before choosing a biomarker measurement at a single time-point to determine eligibility for a particular treatment. In addition, failure to block eosinophils in parallel with IL-13 may not be sufficient to reduce exacerbations, but importantly, preclinical studies show maximal impact of blocking IL-13 on reducing sputum production and goblet cell hyperplasia, so efficacy may be greatest in the clinical phenotype that includes increased mucus (sputum) production [52]. Finally, findings from pre-clinical studies that block molecules before established disease (preventive regimens), rather than after disease manifestation (therapeutic regimen), should not be extrapolated to equate to likely clinical efficacy and such models must be challenged.

Combined blocking of IL-4 and IL-13 via the IL-4rα
Dupilumab is a monoclonal antibody that blocks both IL-4 and IL-13 signaling by binding the alpha subunit of the IL-4/13R. In the first phase 3 clinical trial of dupilumab involving patients with uncontrolled moderate-to-severe asthma, there was a significant reduction in severe asthma exacerbations, including those leading to emergency-department visits or hospitalization, as compared with placebo [53]. The added attraction of this antibody over those blocking IL-5 is the repeated demonstration of beneficial effect on symptom control and prebronchodilator FEV1 [54]. Reductions in exacerbations and improvements in lung function were seen among patients with blood eosinophil counts of ≥150 cells/mcL or higher, or those with FeNO ≥ 25 ppb at baseline [53,55]. However, greatest efficacy was seen in patients with blood eosinophil counts ≥300 cells/mcL, with a 47.7% reduction in exacerbations and a 0.32 L increase in FEV1.
The main adverse effect of dupilumab has been a blood hypereosinophilia, the mechanism for this is uncertain, but likely because of the IL-4-/IL-13-mediated inhibition of eosinophil migration from blood into tissues. This adverse effect is seen in 2-25% of patients; however, a significant rise of >5000 cells/mcL is seen in <2% of patients, and the finding in the majority is transient, asymptomatic, and does not require discontinuation of treatment [56].
The clinical attraction of dupilumab over IL-5 blocking agents is the beneficial effect on lung function as well as asthma exacerbations. However, currently, in the UK it can only be used after a failed trial of mepolizumab, and the disadvantage is the 2 weekly administration, compared to 4 weekly for mepolizumab and 8 weekly for benralizumab.

Blocking TSLP in severe asthma
Tezepelumab, a human monoclonal antibody specific to TSLP, is the first biologic targeting an innate epithelial cytokine that has been approved for the treatment of severe asthma. Tezepelumab is approved for patients 12 years and above who despite treatment with high-dose inhaled corticosteroids fail to achieve asthma control. As an epithelial alarmin, TSLP propagates type 2 airway inflammation in response to inhaled harmful environmental factors. More recently, a role for TSLP in mediating the interactions between airway smooth muscle cells and immune cells during inflammation has been proposed [57]. Patients with asthma have higher levels of TSLP compared to healthy controls, with TSLP levels correlating with disease severity. In Phase 2b (PATHWAY) [58] and Phase 3 (NAVIGATOR) [59], trial of tezepelumab, compared to placebo, resulted in significantly fewer exacerbations over a 52week period (71% and 44%, respectively). The NAVIGATOR trial also showed tezepelumab improved FEV 1 , asthma control, and patients reported improved quality of life compared to those on the placebo arm [59].

Challenges ahead
The challenge is now to define stable and reliable clinical phenotypes and biomarkers that will enable targeted choice of the optimal biologic for the individual patient. This may only be possible with head-to head comparison trials, but these are unlikely to be funded by the pharmaceutical industry, so there is a need for investment from independent, academic funders. This is the only way to achieve optimal cost-effectiveness and best outcomes for patients. The other critical question is whether any of these interventions will enable disease modification. There is good evidence that currently available biologics, specifically mepolizumab [60] and benralizumab, enable reduction in steroid doses, and significant steroid sparing [61]. However, studies to date have shown discontinuation of biologics results in loss of asthma control, suggesting continuous treatment is necessary to maintain clinical benefit [60]. Indeed, data from the randomized placebo-controlled study, COMET, showed that stopping long-term mepolizumab treatment worsened asthma control and increased exacerbations [62]. Unfortunately, pharma will also not want to undertake trials that may show there is no long-term need for their novel therapies, thus again necessitating independently funded academic trial designs. It is likely that disease modification may only occur by intervening earlier during disease development, making therapies that block IL-4/IL-13 very attractive targets specifically in childhood asthma, which is predominantly driven by allergen sensitization, eosinophilia, and type 2 immunity. It is important to note that the impact on features of airway remodeling and the pathological features of asthma have not been investigated following biologics. This is relevant to patients of all ages as airway remodeling, specifically thickening of the reticular basement membrane is observed in very young children and can impact lung function [63,64].

The difference between treating children and adults
Although the biologics that target IL-5 and the IL-4rα have been approved for use in children with severe asthma, there has been limited evidence of efficacy in children compared to adults, and licensing has been approved by extrapolation of adult data to children. However, generation of evidence in children is important because the immune system develops postnatally and it is acknowledged that there may be different immune mechanisms at play during early life compared to in adulthood [65][66][67]. It is clear that the timing of allergen exposure, during this "window of opportunity," influences the nature and magnitude of the immune response to the allergen and impact development of pathology. Thus, age is an important consideration when using biologic reagents, which influence key immune pathways.

Efficacy of anti-IL-5 biologics in children
Until recently, only 37 adolescents aged 12-18 years had been included in placebo-controlled trials of mepolizumab, which overall had included >1800 subjects. Despite this very small number of adolescents being included, and no convincing evidence of efficacy in a post hoc analysis of these subjects [68], approval for use in children aged 6-16 years with severe asthma was licensed by both the FDA and EMA. The same biomarkers were approved as used in the adult studies. The need to consider the impact of maturing immunity and very different ranges of blood eosinophils in healthy children compared to adults [69], and evidence of a marked airway eosinophilia, which is relatively steroid resistant in children, were highlighted as significant reasons for urgent ageappropriate trials before licensing [70]. Three years after license was approved, the first trial of efficacy of mepolizumab in a pediatric trial has been published. However, the inclusion criteria were very different to those used in adult studies, and the population was urban, inner city, and predominantly African American or Hispanic. The MUPPITS-2 trial included children aged 6-16 years, and for inclusion, they only needed two exacerbations in the previous year with blood eosinophils of ≥150 cells/mcL. In contrast, current prescribing guidance is at least four exacerbations in the previous year with blood eosinophils at least 300 cells/mcL or three exacerbations with eosinophils of ≥400 cells/mcL. Despite the differences in biomarkers and severity of disease, in this exacerbation prone urban population, mepolizumab resulted in a 27% reduction in exacerbation rate over 52 weeks, mean exacerbation rate in mepolizumab group was 0.96 compared to 1.3 in the placebo group [71]. Although this was a significant reduction, it is hard to argue this was a clinically meaningful reduction and the authors acknowledge the effect was much lower than seen in adult studies. It is hard to know whether the effect was lower because of the urban and minority population, or because lower blood eosinophils were used as the cut-off, or whether number of exacerbations prior to entry were lower than in the adult studies, or whether the role of eosinophils is very different in children to adults. However, this highlights the need for specific clinical trials in children. In contrast to adult studies, there was no benefit of mepolizumab on any secondary clinical endpoints including symptoms scores, lung function, or exhaled nitric oxide.
An interesting finding was that mepolizumab showed the greatest effect during Autumn, when viral exacerbations are at their peak. This is consistent with the finding that omalizumab also has a specific pre-seasonal effect [72], with improved antiviral immunity [73] to rhinovirus infection during Autumn. Given the primary effect of all current type 2 targeting biologics, and the continued viral seasonal peak in admissions for children, the mechanisms of action of the biologics and interactions between anti-viral immunity and type 2 immunity need specific investigation, as pre-seasonal prescribing may be an attractive approach, especially as the long-term effects of blocking type 2 immunity remain unknown.

Efficacy of IL-4 receptor blockade in childhood severe asthma
In contrast to mepolizumab, dupilumab has only been licensed for children with severe asthma after data from a pediatric clinical trial was generated.
In children with severe asthma and those having recurrent asthma exacerbations, a post hoc analysis of the LIBERTY ASTHMA QUEST study suggested that dupilumab assessed in adults may be efficacious in children [74].LIBERTY ASTHMA QUEST was a phase 3, randomized, double-blind, placebocontrolled, parallel-group trial, whereby 1902 patients were recruited, aged ≥12, with persistent asthma. Patients taking continuous ICS therapy that required one-to-two additional asthma controller therapies were randomized and add-on dupilumab therapy, subcutaneously, every 2 weeks was compared to a matched placebo. The primary end points assessed were rates of severe exacerbations over a 52-week period and change in FEV 1 [75]. With these data, a placebo-controlled trial was conducted in 6-to 11-year-olds with uncontrolled moderate-to-severe asthma to receive subcutaneous dupilumab or placebo 2 weekly for 52 weeks [76].Two primary efficacy cohorts were included, a group with inflammation of 150 blood eosinophils/μL or exhaled nitric oxide of 20 ppb, or a blood eosinophil count of >300/μL. In the former, efficacy population assessed the primary endpoint, annualized asthma exacerbation rate, significantly reduced, and there were associated improvements in asthma control (symptom score) and lung function in children treated with dupilumab compared to placebo. Of note, similar significant dupilumab beneficial improvements were observed in children with an eosinophil count of >300/μL. The safety of dupilumab was similar to that of placebo in the study. In contrast to the mepolizumab effect in children, the overall effect of dupilumab in children was much larger.
The annualized rate of severe asthma exacerbations was 0.31 with dupilumab and 0.75 with placebo (relative risk reduction in the dupilumab group, 59.3%) [76]. The effect size on exacerbations, together with an improvement in lung function, and equal effects in those with high or low blood eosinophils, suggests blocking IL-4 and IL-13 via the IL-4 receptor is a more favorable mechanism for children with severe asthma.
A post hoc analysis comparing efficacy of dupilumab in subjects aged 12 years and older with "allergic asthma" to those without "allergic asthma" has been performed. Allergic asthma phenotype was defined as serum total IgE ≥ 30 IU/mL and sensitizations to 1 or more perennial aeroallergen. Of 1902 patients included in the analysis, 83.3% had eosinophils ≥150 cells/mcL and/or FENO ≥ 25 ppb; 56.9% had evidence for allergic asthma. Dupilumab significantly reduced the rate of severe asthma exacerbations in patients with (48.8%) and without (64.0%) evidence of allergic asthma and improved lung function, irrespective of whether they showed evidence of an allergic asthma phenotype.
Given that dupilumab shows efficacy in children with eosinophils >150 cells/mcL, and with equal efficacy regardless of allergen sensitization, this means its use is not as limited as is currently the case for the anti-IgE monoclonal antibody, omalizumab, suggests that dupilumab may be the biologic with potentially greatest efficacy in children. However, the key now is an urgent need for head-to-head trials that directly compare efficacy of different biologics currently licensed to really understand the optimal biomarkers and predictors of response, but also to understand the mechanism of action and enable the correct choice for the individual child. There are still a group of nonresponders for all current biologics, so it is key to identify markers of response and nonresponse if we are to progress. A summary of the efficacy of the currently available biologics for treating adult and pediatric severe asthma is provided in Table 1.

Summary
The range of treatments for severe asthma has increased over the last few years with the introduction of biological agents that suppress the activity of type 2 cytokines. While clinical trials and real-world evidence show that they effectively reduce exacerbations and, in the case of anti-IL-5/R agents, suppress eosinophilia, there are unanswered key questions. Given the data from multiple animal models of allergic disease that show suppression of type 2 cytokine activity ameliorates allergen driven lung dysfunction, it is not clear why this does not seem to be the case in patients. These differences are likely that mouse models exhibit dominant type 2 immunopathology and do not reflect the real-world complexity of the disease in humans. IL-13 in particular has been shown to promote airway hyperreactivity in mice yet suppression in patients via a variety of different biologics does not affect lung function [51]. Similarly, the effect on remodeling pathways has not yet been described, although there is some indication that benralizumab reduces airway smooth muscle mass [77]. It may be necessary to use agents that target more than one cytokine. This strategy has been employed with a novel IL-4Rα/IL-5-bispecific antibody that targeted multiple type 2 cytokines. Simultaneous blockade of IL-4, IL-5, and IL-13 resulted in ameliorated goblet cell hyperplasia as well as reduced lung function [78]. Similarly, vaccination with dual targeting IL-4 and IL-13 kinoids reduced type2 inflammation in HDM-exposed mice [79]. To date, the longer term consequences of suppressing type 2 immunity has not been determined. It is accepted that many of the cells and molecules that are involved in type 2 immune reactions are also involved in key homeostatic pathways such as energy regulation, thermoregulation, and tissue repair [80], so it would be important to understand how type 2 biologics affect these functions. This is particularly important when considering the agents as treatments for childhood asthma. Going forward, it will be important to determine how type 2 biologics influence the full range of asthma characteristics and how suppression of type 2 immunity affects immune function over longer time frames.

Conflict of interest:
The authors declare no commercial or financial conflict of interest.
Data availability statement: Data sharing is not applicable to this article as no new data were created or analyzed in this study.