Asthma endotypes and theratypes

Abstract

The model of asthma as a single entity has now been replaced by a much more complex biological network of distinct and interrelating inflammatory and tissue driven pathways. Individual disease manifestations (phenotypes), pathogenetic pathways (endotypes) and response to therapy (theratypes) are discussed here in the context of current stratified management of asthma in the clinic based on biomarkers measured at the point-of-care. As the current classification criteria result in significant overlaps among phenotypes, endotypes and theratypes, this paper further describes the advantage of combining precision immunology, imaging and the digital biomarkers in an unbiased approach offered by machine learning. The new European Academy of Allergy and Clinical Immunology (EAACI) nomenclature for hypersensitivity reaction is detailed as a basis for the stratified asthma management with a special focus on tissue-driven mechanisms (type V asthma), metabolic/microbiome/epigenetic/neurogenic mechanisms (type VI asthma) and direct cellular activation (type VII asthma).

Introduction: the current framework for asthma management

Type 2-high and type 2-low asthma

Asthma is a complex chronic respiratory disorder characterized by marked heterogeneity in individual patient disease manifestations (phenotypes, visible properties, treatable traits), pathogenetic pathways (endotypes) and response to therapy (theratypes).1, 2, 3, 4, 5, 6 Several asthma phenotypes have now been identified, each defined by a unique interaction between genetic and environmental factors, including inflammatory, clinical and trigger-related phenotypes. Endotypes further describe the functional or pathophysiologic mechanisms underlying the patient’s disease. Based on the major immune-inflammatory pathway involved, severe asthma can be classified as type-2 (T2) high, T2-low and mixed endotypes, and these categories share several common pathogenetic pathways such as genetic and epigenetic, metabolic, neurogenic and remodeling subtypes6, 7, 8 (Fig. 1). Less is known on the endotypes of mild asthma.⁷

Fig. 1.

Current framework of asthma phenotype. Major inflammatory pathways distinguishing T2-high and T2-low asthma endotypes are shown. Left panel (without T2 biomarkers): Viral infections drive Th1 responses leading to IFN-γ production and GSDMB expression; Th17 responses produce IL-17, recruiting neutrophils and contributing to non-T2 asthma, often associated with airway remodeling influenced by ORMDL3. Right panel (with T2 biomarkers): Air pollutants stimulate airway epithelium to release IL-33, IL-25, and TSLP, activating ILC2 and promoting IL-4, IL-13, and IL-5 production; aeroallergens engage APC to drive Th2 differentiation, resulting in elevated IgE, FeNO, and eosinophilia characteristic of T2 asthma, allergic asthma and eosinophilic asthma. APC, Antigen-presenting cell; Eos, Eosinophils; FeNO, Fractional exhaled nitric oxide; GSDMB, Gasdermin B; IFN-γ, Interferon-γ; IgE, Immunoglobulin E; IL, Interleukin; ILC2, Type 2 innate lymphoid cells; ORMDL3, Orosomucoid 1-like 3; Th, T helper; TSLP, Thymic stromal lymphopoietin; T2, Type 2.

T2-high asthma is a well-established nomenclature for a common subtype of asthma and is characterized by the release of signature cytokines interleukin (IL)-4, IL-5 and IL-13 from cells of both the innate and adaptive immune systems, followed by the involvement of effector pathways mediated by specific immunoglobulin (Ig) E, mast cells and eosinophils.^6,7,9 Several of these T2 cytokines and effectors cells and mediators are targets for pharmaceutical intervention for patients with uncontrolled severe asthma.

T2-low asthma is traditionally defined as asthma without features of T2 asthma and is usually connected with poor responsiveness to corticosteroids (both inhaled and oral) and more severe clinical course, even in childhood.10, 11, 12 Very little is known about its mechanisms, making it a significant unmet need in asthma research.¹³ The definition is arbitrary and is generally based on the presence of neutrophils in sputum, or the absence (or normal levels) of eosinophils or other T2 markers in sputum (paucigranulocytic), airway biopsies or in blood. This definition may be imprecise as we gain more knowledge from applying transcriptomics and proteomics to blood and airway samples and by using multiomics and imaging integration.14, 15, 16 Non-T2 inflammation is characterized by activation of T helper (Th)1 and Th17 cells and neutrophils, the presence of type I interferons (IFN), NOD-like receptor protein 3 (NLRP3) inflammasome activation, and an interleukin (IL)-1β and IL-17 signature.^12,17, 18, 19, 20 IL-17 induces a predominantly neutrophilic airway inflammation, mucus hypersecretion, goblet cell hyperplasia, myofibroblast differentiation, and airway smooth muscle proliferation.²¹ Th17 cells and its cytokines are implicated in mechanisms of steroid resistance in asthma via induction of glucocorticoid receptor‐β (GR‐β) expression and reduced apoptosis. Furthermore, glucocorticoid treatment also enhanced IL-17 production. The NLRP3 inflammasome activates caspase-1 to process pro-IL-1β into its mature, secreted form. Caspase-1 also cleaves gasdermin-D (GSDMD) into fragments that assemble into a plasma membrane pore releasing mature IL-1β from cells and inducing pyroptotic cell death.²² A non-canonical inflammasome comprised of caspase-4 and caspase-5 also cleaves GSDMD, with resultant activation of the NLRP3 inflammasome and caspase-1. Caspase-1 activation and GSDMD cleavage also generate neutrophil extracellular traps (NETs). In severe asthma, sputum neutrophils correlate with sputum extracellular DNA levels indicative of NET formation, while increased sputum extracellular DNA is associated with increased sputum IL-1β and asthma severity.²³ Bronchoalveolar lavage fluid from patients with severe asthma and high neutrophil counts had detectable NETs and cytoplasts that were positively correlated with IL-17 levels.²⁴ Like other gasdermin family members, gasdermin-B (GSDMB) may promote inflammasome-mediated pyroptosis in non-T2 asthma.²⁴ Additionally, expression levels of GSDMB correlated with genes involved in IFN response in airway epithelial cells, and GSDMB contributes to childhood asthma, possibly related to increased IL-17A response to viral infections.^25,26 Rhinovirus infection in patients with asthma leads to an excessive retinoic acid‑inducible gene‑I (RIG-I) inflammasome activation, which disrupts effective RIG-I-dependent type I/III interferon responses, leading to early functional antiviral impairment, prolonged viral clearance and unresolved inflammation in vitro and in vivo. Pre-exposure to house dust mite augments this phenomenon by inflammasome priming and auxiliary inhibition of early type I/III interferon responses.²⁷ Epithelial-derived cytokines (IL-33, thymic stromal lymphopoietin [TSLP], IL-25) are similarly released by drivers of type-2 low pathology such as cigarette smoke, diesel exhaust particles, microbes or proteases and thus can represent therapeutic targets in T2-low asthma.¹³ IL-33-activated gene signatures are elevated in neutrophilic and mixed granulocytic asthma corresponding with interleukin-1 receptor accessory protein (IL1RAP) co-receptor expression.²⁸ Increased expression of sphingolipid biosynthesis regulator 3 (ORMDL3) may contribute to the development of paucigranulocytic asthma, by inducing the upregulation of mediators involved in airway remodeling rather than inflammation, by promoting autophagic cell death in airway epithelial cells, by reducing the levels of serum sphingolipids and by increased airway hyperreactivity (AHR).29, 30, 31, 32

The prevalence of T2-low asthma is difficult to estimate as most studies are cross-sectional and influenced by concomitant treatment with corticosteroids, and by the presence of recognized or unrecognized airway infections or by recent exposure to outdoor or indoor pollutants. In a cohort of T2-high severe asthma patients stable on mepolizumab, a subgroup of patients with long duration of disease had worse clinical parameters, increased sputum proteins with increased markers of neutrophil activity, proinflammatory cytokines and epithelial alarmins.³³ This suggests the involvement of non-T2 inflammatory pathways that are not targeted by current T2-biologics. These pathways may represent treatable traits in severe asthma, warranting further investigation. Another study explored the physiological changes at exacerbation in patients with asthma who were T2-high and T2-low and evaluated the stability of inflammatory phenotypes during stable disease and exacerbation. At enrollment, 23.6% were T2-low. These patients had more primary care attendances and were more likely to have a previous admission to intensive care and to be receiving maintenance oral corticosteroids (OCS). At exacerbation, the T2-low events were indistinguishable from T2-high exacerbations in terms of lung function and asthma control. There was no increase in T2 biomarkers from stable to exacerbation state in the T2-low exacerbations. The inflammatory phenotype within individual patients was dynamic, with T2-low asthma being an unstable inflammatory phenotype. The inflammatory phenotype at study entry did not have a significant association with exacerbation phenotype.³⁴ Furthermore, the inflammatory phenotype might vary according to the region or exposure—a concept coined as a regiotype.⁶ Using induced sputum to compare asthma inflammatory phenotypes in high income versus low- and middle-income countries (LMIC), an epidemiological survey showed that in LMIC 61% of cases were non-eosinophilic, including in settings where corticosteroid use was low.³⁵

Defining asthma endotypes based on biomarkers

Several well-recognized biomarkers are used clinically to diagnose T2-high asthma, including blood eosinophils, fractional exhaled nitric oxide (FeNO), and specific immunoglobulin E (IgE)³⁶ (Fig. 1). An observational study comparing surrogate markers for sputum eosinophilia found blood eosinophils and FeNO to have comparable diagnostic accuracy, which was superior to total serum IgE in adult asthma patients.³⁷ Non-T2 asthma is usually a diagnosis of exclusion based on the absence of T2 biomarkers or based on the sputum inflammatory phenotype: neutrophilic or paucigranulocytic.

Evidence accumulated over 2 decades since the introduction of the first monoclonal antibody against IL-5 showed that the pathogenetic role of eosinophils is fundamentally different between asthma phenotypes.³⁸ In the IgE-driven form of T2 asthma, blood eosinophil counts are variable, mainly dependent on allergen exposure, and play only a minor effector role in allergen-induced asthma symptoms. By contrast, in the eosinophilic form of T2 asthma, eosinophils are persistently elevated and are crucial drivers of the disease. These considerations suggest that blood eosinophilia should not be considered a treatable trait in people with asthma, but only a biomarker complementing an accurate endotype diagnosis. Furthermore, there is a modest correlation between eosinophil numbers in the airway and those in the blood, but this gets weaker as asthma gets more severe and as patients are exposed to higher doses of corticosteroids.^39,40 There are still a number of unresolved issues related to the application of blood eosinophil counts in clinical practice , including: (1) the number of measurements needed to reliably predict risk and/or clinical response, and (2) whether blood eosinophils are helpful for both initiating therapy and monitoring its course.

FeNO is increased in patients with T2 asthma and is used as an objective biomarker of airway inflammation. Cut-off values have been established by the American Thoracic Society, the Global Initiative for Asthma, and the National Institute for Health and Care Excellence, but vary between guidance. FeNO levels can be predictive of blood and sputum eosinophil levels, but should not be used in isolation.⁴¹ High FeNO can help predict response to inhaled corticosteroids (ICS) and suppression of its levels with ICS can be used to monitor adherence. FeNO levels are also a predictor of asthma risk with elevated levels associated with increased exacerbation rates and accelerated decline in lung function; FeNO also has an emerging role in predicting response to dupilumab in severe asthma.42, 43, 44 Persistent high FeNO can identify steroid-resistant inflammation in asthma.⁴⁵ FeNO testing is noninvasive and easy to use, and has been shown to be cost-effective as a complement to clinical assessment in improving asthma management.^46,47 Unfortunately, its immediate high costs preclude its wide implementation at the point-of-care. Several confounders should be accounted as well: obesity can be associated with low FeNO levels even in the presence of sputum eosinophilia; and FeNO can also be increased by the presence of allergic rhinitis.^48,49

Allergic asthma is the most common asthma phenotype, impacting 7.3% of the general population.⁵⁰ It is usually diagnosed by the presence of specific IgE to aeroallergens in conjunction with a comprehensive history showing the allergen exposure as the major driver of asthma symptoms and control. If the history is unclear, an allergen provocation test (nasal or bronchial) done in a specialized setting is recommended to ensure a correct diagnosis.51, 52, 53 The correct diagnosis of allergic asthma is essential for recommending allergen avoidance measures and allergen immunotherapy (AIT), as add-on to the regular controller asthma treatment.⁵⁴ The addition of AIT was proven to be able to decrease the ICS dose while maintaining asthma control.^53,54 It is also cost-effective.⁵⁵ Measuring cumulative levels of IgE specific for respiratory allergens could be a useful screening method for detecting an allergic phenotype of severe asthma and may serve as a biomarker to enhance the success of IgE-targeted therapy.⁵⁶

Increased sputum neutrophils are usually used to diagnose T2-low asthma. However, it is unresolved whether this is indeed the case.^13,57 The association between asthma severity and neutrophil number has led to the use of the term neutrophilic asthma, which implies the existence of a specific asthma endotype in which neutrophils are the major cell type mediating pathobiology, severity, and symptoms. Neutrophils can be temporarily recruited into the airways in a number of circumstances—e.g., following exposure to pollutants and pathogens or in response to a diet rich in fat and carbohydrate. Upon activation in the airways, neutrophils can release their granules, DNA, and proteins, and thus could contribute to airway damage. However, this increase in neutrophil number driven by factors that promote neutrophilia might not necessarily be stable over time.

Based on the major biologicals clinical trials, blood eosinophils, FeNO and specific and total IgE are usually used to define T2 asthma endotypes: allergic asthma is defined by total serum IgE ≥30 kU/L and one or more positive perennial aeroallergen-specific IgE (≥0.35 kU/L) at baseline,^58,59 eosinophilic asthma is defined by the blood eosinophil counts ≥300 cells/µL (low cut-off ≥150 cells/µL),⁶⁰ T2-high asthma is defined by FeNO ≥35 parts per billion (ppb) (low cut-off ≥25 ppb)⁶⁰ (Fig. 1). Post hoc analyses from two adult populations showed that these classification criteria result in significant overlaps among these endotypes.⁶⁰ Thus, in a general adult asthma population without specific subtype selection, 78.0% exhibited allergic asthma; among them, 39.5% presented with eosinophilic asthma and 29.5% had T2 asthma. There were many overlapping subjects: 75.8% of those with eosinophilic asthma had allergic asthma, 41.3% of those with eosinophilic asthma had T2 asthma, 81.1% of individuals with T2 asthma had allergic asthma, and 59.2% of those with T2 asthma had eosinophilic asthma.⁶⁰

Consequently, more specific diagnostic criteria are needed (Fig. 1). One example is defining allergic asthma based on the impact of allergen exposure on asthma symptoms and control as described above using a comprehensive clinical history and if needed allergen provocation tests.^51,52,53 Another option is using combinations of biomarkers.^38,41 A third is using biomarkers measured in the target organ (nose and lung). Last, but not least, omics signatures built in an unbiased way using machine learning would provide the most precise definition of asthma endotypes.6, 7, 8^,14,16

FeNO and eosinophilia, although related, represent two distinct biomarkers resulting from two different T2 cytokine pathways: IL-4 and IL-13, which are involved in IgE synthesis and inducible nitric oxide synthase (iNOS) induction (leading to nitric oxide [NO] production), and IL-5, which drives eosinophil activation, development, and recruitment. Consequently, they provide complementary information and should be used jointly. A risk scale for predicting asthma attacks based on blood eosinophil counts and FeNO has been proposed. This scale was derived by extracting and pooling biomarker-stratified, trial-level attack rates from the control arms of several clinical trials.⁶¹

Nasal secretions and sputum reflect the local inflammatory activity and provide valuable information about the immunological reaction in the lung. Current techniques for nasal secretions sampling are mainly based on three principles: collection of spontaneous secretions, nasal washings, and absorption. Collection of spontaneous secretions is appropriate in subjects with nasal hypersecretion. Nasal washings are associated with an unpredictable, high dilution and concentrations of markers often fall below detection limits of immunological assays.⁶² Absorption seems to provide the best compromise between sufficient sample amounts and detectability of inflammatory mediators and IgE.⁶² Nasal swabs for eosinophil granule proteins or activation products hold promise.⁶³ Using formalin-fixed paraffin-embedded sputum plugs for easy assessment of sputum eosinophils numbers will soon become available at the point of care.^64,65 Streamlining the sputum processing to make dispersed fluid an easy biospecimen suitable for rapid assessment tests to identify key markers of inflammation such as eosinophilic peroxidase (EPX), myeloperoxidase, elastase, tryptase, and IgG autoantibodies (antinuclear antibodies, anti EPX, anti-MARCO [macrophage receptor with collagenous structure]) will also be soon within reach in asthma centers.66, 67, 68, 69, 70

Opportunities for asthma management in the era of precision medicine

Precision immunology, deep phenotyping and novel imaging techniques

Molecular phenotyping based on a transcriptomic analysis of bronchial epithelial and sputum cells has revolutionized the understanding of asthma pathogenesis. More than a decade ago, a T2 high inflammation cluster characterized by eosinophilia and recurrent exacerbations was initially described.⁷¹ This discovery was followed by the characterization of several T2-low clusters linked with IL-6 trans-signaling, IFN pathways, inflammasome activation and mitochondrial oxidative phosphorylation pathways.^72,73

Currently, the multi-omics unbiased approach for asthma endotyping led to major novel discoveries related to alterations of the arginine metabolome, mechanisms of ferroptotic death in airway epithelial cells in asthma, abnormal lipid metabolism or dose-dependent metabolomic changes induced by ICS in high doses.74, 75, 76, 77 The severe asthmatic and high FeNO asthmatic patients were shown to have unique endotypes that suggest changes in NO-associated taurine transport and bile acid metabolism.⁷⁸ Shotgun lipidomics of induced sputum supernatant in the U-BIOPRED patients revealed a spectrum of 9 molecular phenotypes, highlighting significant differences between the sputum lipidomes of asthma patients and healthy controls, and also within the asthma patient population.⁷⁹ Sputum lipid phenotypes with higher levels of nonendogenous, cell-derived lipids were associated with significantly worse asthma severity, worse lung function, and elevated granulocyte counts. These findings suggest a novel mechanism of increased lipid loading in the epithelial lining fluid of asthma patients resulting from the secretion of extracellular vesicles by granulocytic inflammatory cells, which could reduce the ability of pulmonary surfactants to lower surface tension in asthmatic small airways, as well as compromising its role as an immune regulator. In the same cohort, unbiased label-free quantitative mass spectrometry of sputum supernatants described 10 asthma proteotypes, with 3 highly eosinophilic, 3 highly neutrophilic, and 2 highly atopic with relatively low granulocytic inflammation.⁸⁰ Integrating biobank-scale genetics and plasma proteomics identified 70 proteins with putative causal roles in asthma risk, including known drug targets and proteins without prior genetic evidence in asthma (e.g., GTP cyclohydrolase 1 feedback regulatory protein [GCHFR], tudor and KH domain containing [TDRKH], and C-type lectin domain family 7 member A [CLEC7A]).⁸¹ The genetic architecture of causally associated proteins provided evidence for a TLR-1–IL-27 asthma axis. An integrated species-level metagenomic data with inflammatory mediators characterized the prevalence of dominant potentially pathogenic organisms in relation to the host immune responses.⁸² Neutrophilic asthma was associated with Haemophilus influenzae (H. influenzae), Moraxella catarrhalis (M. catarrhalis), Streptococcus pneumoniae (S. pneumoniae) and Tropheryma whipplei (T. whipplei) with elevated type-1 cytokines and proteases; eosinophilic asthma was associated with higher M. catarrhalis, but lower H. influenzae, and S. pneumoniae abundance. H. influenzae load was correlated with eosinophil cationic protein, elastase and IL-10. Rothia mucilaginosa was positively associated with IL-6 and negatively with fibroblastic growth factor (FGF). Bayesian network analysis also revealed close and distinct relationships of H. influenzae and M. catarrhalis with type-1 airway inflammation. The study also showed that the microbiomes and cytokine milieu were distinct between upper and lower airways.⁸²

“Breathomics” in asthma is a rapidly growing area of significant scientific interest. The repeatedly observed associations between breath volatile organic compounds (VOCs) and sputum or blood inflammatory cells suggest that breathomics are on the brink of introduction as a valuable clinically tool. However, there are major concerns about unresolved methodological issues and a general paucity of high-quality data. Numerous breath VOC studies yielded many valuable insights, which increase our awareness of interfering environmental, lifestyle and metabolic factors and of the need of a more standardized methodological approach. However, a recent study reported the lack of any significant correlations between VOCs and inflammatory markers in a well-characterized cohort of adult patients with asthma with a broad spectrum of clinical phenotypes.⁸³

In more recent years, single-cell analysis provided the profiles of major immune and mesenchymal clusters in the human respiratory tract,^84,85 offering novel insights into the differentiation of pathogenic Th2 cells in the airways, which are enriched for genes and pathways associated with lipid and glucose metabolism.⁸⁶ Meanwhile, a framework for efficient capture of granulocytes in tissue compartments, overcoming traditional limitations of single-cell RNA sequencing, was recently published. The resulting dataset serves as a valuable resource for understanding airway granulocyte biology and inflammation, enabling detailed exploration of asthma pathogenesis.⁸⁷ Based on these methodological innovations, recent studies have examined the effects of IL-13 on human airway epithelial cells. IL-13 was shown to induce a mucus secretory program across all epithelial cell types, converting both mucus and defense secretory cells into a metaplastic state characterized by increased mucin production and secretion. In ciliated cells, IL-13 exposure led to endoplasmic reticulum (ER) stress and cell death.⁸⁸ The resulting remodeled epithelium secretes a pathologic proteome marked by mucin imbalance and depletion of innate immunity factors, ultimately impairing mucociliary clearance.⁸⁸ These insights into pathogenic type 2 immunity have paved the way for targeted biologic therapies. High-parameter flow cytometry analysis of peripheral blood mononuclear cells recently reported a new mechanism for how mepolizumab reduces airway inflammation by re-directing trafficking of inflammatory T2 lymphocytes away from airway-homing.⁸⁹ Collectively, these findings underscore the complexity of asthma pathogenesis, which involves epithelial remodeling, immune cell trafficking, and microenvironmental factors. Recent advances in next-generation sequencing and imaging technologies have enabled the application of spatial transcriptomics to study lung remodeling, shedding light on the importance of specialized niches in the context of chronic inflammation and gene–environment interactions.⁹⁰ Activation of immune cells in specialized regions of the lungs, such as the adventitial cuffs, involves specific immune regulatory mechanisms, including the production of IL‐33 and chemokines, and extracellular matrix (ECM) deposition.^91,92 Dysregulation of this niche can initiate airway remodeling via hyperactivation of mast cells (MCs) and induce vascular damage.⁹³

Sophisticated three-dimensional lung imaging using high resolution computed tomography (HRCT), ventilation imaging (single photon emission computed tomography [SPECT] and positron emission tomography [PET]), magnetic resonance imaging (MRI) and ultrahigh resolution techniques such as micro-computed tomography and synchrotron imaging is now highly developed. Optical coherence tomography (OCT) and endobronchial ultrasound enable high-resolution imaging of the large airways accessible to bronchoscopy. HRCT scanning predominantly provides measurements of airways and lung structure, as well as small airway function from the difference between inspiratory and expiratory images. Imaging-based biomarkers, including airway dimensions, blood vessel volumes, mucus scores, extent of ventilation defect, and extent of air trapping, often have increased sensitivity compared with that of traditional lung function measurements and are increasingly being used as end points in clinical trials. High mucus burden represents a distinct T2-high asthma phenotype that can contribute to chronic symptoms and acute exacerbations, potentially leading to fatal respiratory failure.⁹⁴ The mucus plug scoring system based on the spatial extent of plugging by determining the number of occluded pulmonary segments has shown a compelling relationship between mucus burden, airway obstruction, asthma control and exacerbation risk.^94,95 Differences in mucus appearance on OCT correlate with mucin content, suggesting that OCT is sensitive to differences in mucus composition as well as in volume.⁹⁶ Asthma patients with high mucus burdens respond well to biologic therapy, which clears plugs and normalizes ventilation.97, 98, 99 Ventilation heterogeneity, a classic feature of asthma, has been known to predict AHR. SPECT ventilation imaging adds spatial characterization information on the impact of peripheral airways on AHR.¹⁰⁰ Combined with quantitative computed tomography imaging, it can delineate clusters within patients with severe asthma based on differences in ventilation heterogeneity and airways resistance, thus guiding the strategies for the inhalational treatment.¹⁰¹ PET and MRI also provide information on the distribution of ventilation. The severity of airway closure measured using MRI was associated with the clinical risk of asthma exacerbations.^102,103 The response of ventilation defects to the acute bronchodilator inhalation was associated with sputum eosinophilia.¹⁰⁴ Airway closure can be widespread, distributed in a patchy, “clustered” manner. It can be the consequence of airway remodeling, parenchymal remodeling, mucus plugs or local inflammation.¹⁰⁴ Consequently, the distribution of airway closure and its response to a bronchodilator or an anti-inflammatory agent represent a reliable phenotyping and theratyping tool.

Machine-learning approaches

The application of mathematical and computational analysis, together with the modeling of biological and physiological processes, is transforming our understanding of asthma pathophysiology. A systematic biology approach to define asthma endotypes and theratypes integrates multi-omic data derived from genomic, transcriptomic, proteomic, metabolomic, lipidomic, breathomic, and metagenomic from disease-relevant tissues (lung biopsies, bronchoalveolar lavage [BAL], sputum, nasal secretions, skin, blood), together with imaging data, deep clinical phenotyping, and patient-reported outcomes.^14,16,105, 106, 107, 108, 109, 110, 111, 112, 113, 114, 115, 116 Integration of these datasets will provide a greater understanding of the molecular pathways associated with asthma in each individual patient and thus guide their personalized management (Fig. 2).

Fig. 2.

Asthma management in the era of precision medicine. Data from multiomics, single-cell/spatial transcriptomics, biomarkers, symptoms/comorbidities, and sophisticated 3D lung imaging are processed through machine learning to characterize individual molecular signatures. These signatures are combined with real-time, theratype-based personalized interventions, ultimately supporting precision medicine. BALF, Bronchoalveolar lavage fluid.

Digitally empowered asthma care

The landscape of health care is transforming rapidly as technology advancements accelerate integration of digital health technology into patient care and management. Digital health technology uses computing platforms, connectivity, software, artificial intelligence, machine learning, and sensors to manage illnesses and health risks and promote wellness with a strong emphasis on personalized health care. This includes wearable devices, mobile health, telehealth, health information technology, remote monitoring, and telemedicine.

Digital technologies can play a central role in achieving precision medicine across the patient journey, from diagnosis to remission. For diagnosis, smartphones and artificial-intelligence-enabled devices can objectively capture respiratory symptoms, while digital peak flow meters enable home assessment of airflow variability. For routine monitoring and self-management, technologies can monitor biomarkers more frequently, digital inhalers can improve medication adherence and asthma control whilst serious games can improve patient knowledge. Symptom tracking, lung function monitoring and environmental parameters can help identify exacerbation triggers. For assessing remission, digital technologies can capture most of the components of remission definitions, from exacerbations to reliever use, enabling a more objective assessment of this emerging endpoint.¹¹⁷ Digital technologies also allow for remote, objective, granular, and non-invasive data collection, offering the opportunity to move towards decentralized asthma clinical trials. This approach could facilitate recruitment of inclusive and generalizable study populations, enhance personalization and sustainability, reduce research costs, and accelerate market access for novel asthma treatments.¹¹⁸ Inhaler-based digital biomarkers can also be objectively evaluated to indicate changes in response to therapeutic interventions¹¹⁹ (Fig. 2).

The new European Academy of Allergy and Clinical Immunology (EAACI) nomenclature as a basis for stratified asthma management

The new nomenclature for hypersensitivity reactions developed by the EAACI provides a modern approach for asthma and allergic diseases, based on disease endotypes and theratypes.¹²⁰ Hypersensitivity reactions originally described by Gell and Coombs have been extended into nine different types comprising antibody-(I–III), cell-mediated (IV a–c), tissue-driven mechanisms (V–VI) and direct response to chemicals (VII). Types I–III are linked to classical and newly described clinical conditions. Type IVa–c is specified and detailed according to the current understanding of T1, T2 and T3 responses. Type V involves epithelial barrier defects and tissue remodeling, and is clinically associated with severe, persistent asthma characterized by fixed airflow obstruction, frequent exacerbations, and progressive decline in lung function. Type VI involves metabolic/microbiome/neurogenic-induced immune dysregulation and underlies heterogeneous clinical presentations linked to obesity, sex differences, and other triggers. Type VII involves direct cell activation via ion channels, receptors, or inflammatory responses to pollutants, irritants, or mechanical stress, and is clinically significant for trigger-specific exacerbations and marked airway hyperresponsiveness. It is notable that several combinations of mixed types may appear in the clinical setting.¹²⁰

Type V asthma

Type V hypersensitivity reaction includes the contribution of the tissue as initiator and further modulator of the immune-inflammatory reactions (Fig. 3). The epithelial barrier dysfunction, the airway smooth muscle (ASM) phenotype switch, the abnormal mucus production, the activation of the epithelial–mesenchymal trophic unit and the profound changes in the ECM and pulmonary blood vessels have been documented for decades as central to the pathogenesis of asthma, frequently dissociated from the immune-inflammatory pathways.5, 6, 7^,121, 122, 123, 124

Fig. 3.

Type V hypersensitivity asthma driven by epithelial barrier dysfunction and tissue remodeling. Key pathways involved in Type V asthma according to the new EAACI nomenclature are shown. Environmental insults (allergens, viruses, bacteria, pollutants, and other inhaled hazardous materials) initiate epithelial barrier dysfunction via recognition of PAMPs and DAMPs/proteases, leading to IFN responses, enhanced viral replication, and release of cytokines, chemokines and alarmins, T2 inflammation, and airway remodeling. These mediators drive angiogenesis, abnormal ECM deposition, ASM phenotypic changes, goblet cell hyperplasia with mucus plugging, and eosinophilic infiltration with IgE production, ultimately resulting in airway hyperreactivity. ASM, Airway smooth muscle; DAMPs, Damage-associated molecular patterns; EAACI, European Academy of Allergy and Clinical Immunology; ECM, Extracellular matrix; IFN, Interferon; IgE, Immunoglobulin E; PAMPs, Pathogen-associated molecular patterns; T2, Type 2.

Epithelial damage is a pathological feature observed in all phenotypes of asthma, occurring early, before the onset of inflammation and the clinical diagnosis of asthma.^125,126 Allergens, viruses, bacteria, pollutants and other inhaled environmental insults are in first contact with the airway epithelial barrier, which forms a continuous lining of the respiratory system from the nose to the trachea, bronchi, bronchioles and finally the alveoli.127, 128, 129, 130, 131, 132, 133, 134 In addition to the physical barrier function and mucociliary clearance of foreign particles, the airway epithelium acts as chemical barrier against environmental insults by secreting, for example antimicrobial peptides, anti‐proteases and antioxidants, and is part of the innate immune system. Airway epithelial cells express pattern recognition receptors (PRRs) like toll‐like receptors (TLRs), retinoic acid‐inducible gene (RIG)‐I‐like receptors (RLRs), nucleotide‐binding oligomerization domain (NOD)‐like receptors (NLRs), C‐type lectin receptors, protease activated receptor (PAR)‐2 and purinergic receptors.¹²¹ These recognize pathogen‐associated molecular patterns (PAMPs) from inhaled microbes, parasites and allergens as well as alarmins/damage‐associated molecular patterns (DAMPs) released from dying or damaged cells. Upon recognition of PAMPs or DAMPs, PRRs activate the inflammasome, leading to caspase‐1 activity and subsequent cleavage of IL‐1β and IL‐18 into active forms, and provide downstream signaling that promotes the release of pro‐inflammatory cytokines/chemokines, including IL‐6, IL‐8, C-C motif chemokine ligand (CCL) 20, CCL17, thymic stromal lymphopoietin (TSLP), IL‐25, IL‐33 and granulocyte macrophage colony stimulating factor (GM‐CSF) that attract and/or activate cells from the innate and adaptive immune system.^121,122,127, 128, 129, 130, 131, 132, 133, 134 The epithelial derived cytokines TSLP, IL‐25, and IL‐33 activate T2 innate lymphoid cells (ILC2), which secrete T2 cytokines.¹³⁵ Dendritic cells (DCs) educated by the T2 cytokines induce the differentiation of Th2 cells, which further secrete T2 cytokines and induce specific IgE production, eosinophilic infiltration into the airways and goblet cell hyperplasia with excessive mucus production.¹³⁶ T2 cytokines further promote the epithelial barrier dysfunction.^137,138 Smoking‐induced Th17‐mediated inflammation can further reduce epithelial barrier function through Th17 cytokine IL‐17.¹³⁹ The impaired epithelial barrier function is accompanied by compromised IFN responses in asthma, resulting in increased viral replication.¹⁴⁰ Upon damage, the epithelial barrier is disrupted and releases growth factors such as epidermal growth factor (EGF) and transforming growth factor (TGF)‐β, which activate fibroblasts and myofibroblasts.¹⁴¹ Allergens, irritants and microbes (including the normal flora) enter the tissue through the disrupted barrier and induce a chronic immune-inflammatory response.^122,136 EGF and TGF‐β promote excessive deposition of ECM components, resulting in subepithelial fibrosis, airway wall thickening and increased ASM mass. In addition, release of vascular endothelial growth factor (VEGF) by airway epithelial cells increases the size of airway wall vessels and promotes angiogenesis. Last but not least, there might be a genetic predisposition to irritant-induced epithelial barrier dysfunction: lower expression of the susceptibility gene hedgehog interacting protein (HHIP) in airway epithelial cells from patients may contribute to abnormal epithelial repair.¹⁴²

Goblet cell hyperplasia and mucous plugging are a classical feature of asthma, and the functions of this mucus during a lung immune response remain elusive, despite widespread evidence of its involvement in cases of fatal asthma.^94,143 Alterations in the quality, rather than merely the quantity, of mucus are a more relevant parameter of asthma. Pathogenic mucus from asthma has a significantly higher elastic modulus, making it much stickier and harder to expectorate by coughing.¹⁴⁴ A high elastic modulus indicates extensive cross-linking of the mucus, which results in an incredibly tenacious substance that cannot be moved by the muco-ciliary escalator, is hard to expectorate, and remains lodged in the airways. The two primary mucins (MUC) found in the human lung are MUC5B and MUC5AC. MUC5B is secreted at high levels in submucosal glands and secretory cells in the distal airways, whereas MUC5AC is secreted by goblet cells. Healthy lungs contain mostly MUC5B, but a common feature of asthmatic mucus is dysregulation of this ratio, with MUC5AC being dramatically upregulated, particularly in patients with an eosinophilic asthmatic phenotype.¹⁴⁵ Serum galectin-10 is one potentially valuable marker that could point to the presence of a crystal-rich plug.¹⁴⁶ More is needed to develop clinically applicable scoring systems for mucus load in asthma on the basis of dynamic HRCT scans and other advanced imaging modalities that can pinpoint ventilation heterogeneity, perhaps supplemented with invasive bronchoscopic procedures allowing us to access mucus plugs.

ASM cells participate in the AHR as well as the inflammatory and remodeling processes observed in asthmatic subjects. The increased ASM mass may be collectively due to airway infiltration of myofibroblasts, neighboring airway smooth muscle cells in the bundle, or circulating hemopoietic progenitor cells. However, the relative contribution of each cell type is not well understood.¹⁴⁷ ASM has a vast repertoire of inflammatory receptors that, upon activation, contribute to prominent features in asthma, notably immune cell recruitment and activation, hypercontractility, proliferation, migration, and extracellular matrix protein deposition.¹⁴⁸ These phenotypic changes in ASM can be mediated by epithelial derived cytokines (TSLP), T2 (e.g., IL-4, IL-13,) and type 1 (T1) (tumor necrosis factor [TNF]-α, IFN-γ) and type 3 (T3) (IL-17A) cytokines, and by TGF-β, a key driver in paucigranulocytic asthma, highlighting roles for ASM modulating both T2 and non-T2 asthma endotypes.149, 150, 151, 152, 153, 154, 155, 156 CD4 T cell infiltration of ASMs bundles was correlated with asthma severity.¹⁵⁷ The increased expression of ECM proteins further influences the ASM phenotypic switch.¹⁵⁸ Both IL‐13 and IL‐17A enhance ASM adhesion to the ECM by activating β1 integrin.¹⁵⁹

Abnormal deposition of ECM proteins is a key factor in the development of tissue remodeling that results in symptoms and impaired lung function in these diseases. Tissue remodeling in the lungs is complex and differs between compartments. Some pathways are common but tissue remodeling around the airways and in the parenchyma has different morphologies. Hence it is critical to evaluate both common fibrotic pathways and those that are specific to different lung compartments. Immune cells use the ECM scaffold to migrate through the lungs. Using an IL‐33 mediated lung inflammation model, it demonstrated that ILC2 are actively recruited to specific locations of the airway wall, such as the adventitial cuff, at the intersection between the large blood vessels and the large airways.⁹² ILC2s were most prominent in areas with high levels of ECM. Collagen‐I fibers may induce a more polarized ILC2 morphology, resulting in increased traction, decreased velocity, and prolonged dwell time at specific lung sites, ultimately enhancing inflammation. Chemokines bind to ECM components, including glycosaminoglycans. Changes in the amount and geometry of the ECM may greatly influence the availability of these mediators to immune cells, and therefore could alter the regulation of immune cell functions locally.¹⁶⁰ Last but not least, the dynamics of ECM change with age and development: an increased reticular basement membrane formed by collagens and laminins is one of the earliest events present in childhood asthma and correlates with the severity of the disease.¹⁶¹

Angiogenesis and associated vascular remodeling are one of the pathological hallmarks of asthma. The mechanisms underlying angiogenesis in asthmatic airways and its clinical relevance represent a relatively nascent field in asthma when compared to other airway remodeling features. Angiogenesis involves the destruction of the vascular basement membrane and remodeling of the ECM, which paves way for the migration and proliferation of endothelial cells as well as the synthesis of new matrix components. Matrix metalloproteinases (MMPs) play an important role in this disruption and neovascularization process, together with various angiogenic factors (hypoxia inducible factor, VEGF, FGF-2 and angiopoietins) that are secreted by the infiltrating eosinophils, basophils, and mast cells as well as by the resident epithelial, endothelial, and ASM cells.^162,163 Several of these angiogenic factors have been tested as prognostic biomarkers: VEGF mRNA-expressing cells in the airway mucosa correlate with AHR; while angiomotin and angiostatin, when analyzed together, can distinguish between stable and exacerbated states in asthma patients.^164,165

A study exploring the airway pathology in T2-high versus T2-low severe asthma showed that the remodeling features of ASM mass and MUC5AC expression were increased in both asthma groups compared with healthy controls and were similar across T2-high and T2-low subgroups. Submucosal glands were increased in T2-intermediate and T2-low asthma. In spite of similar tissue cellular inflammation, sputum IL-4, IL-5 and CCL26 were higher in T2-high than in T2-low asthma.¹⁶⁶ Taken together, while T2 cytokine profiles differed between subgroups, airway remodeling was similarly evident in both, suggesting that remodeling might operate through a distinct pathogenetic pathway independent of T2 inflammation. This may contribute to residual disease beyond eosinophilic exacerbations.

Type VI asthma

Type VI hypersensitivity reactions include mechanisms driven by metabolic, hormonal, epigenetic, neuro-immune interactions and the interactions with the microbiome (Fig. 4). Thus, this hypersensitivity type provides an integrative framework for understanding the disease.

Fig. 4.

Type VI hypersensitivity asthma driven by metabolic, hormonal, microbiome, trained immunity, and neuro–immune interactions. Multifaceted drivers of Type VI hypersensitivity reactions in asthma according to the new EAACI nomenclature. The central circle represents the convergence of metabolic pathways (e.g., obesity-related inflammation, oxidative stress, lipid mediators), microbiome dysbiosis (e.g., gut, lung), hormonal influences (e.g., sex steroids, incretins), trained immunity (epigenetic and metabolic reprogramming of innate cells), and neurogenic inflammation (neuropeptides, neurotrophins, and neural–immune crosstalk). These interconnected mechanisms contribute to immune dysregulation, airway inflammation, remodeling, and hyperreactivity, highlighting their relevance in asthma pathogenesis and potential for stratified management. AA, Arachidonic acid; Ach, Acetylcholine; AEC, Airway epithelial cell; AERD, Aspirin-exacerbated respiratory disease; AHR, Airway hyperreactivity; ASM, Airway smooth muscle; CGRP, Calcitonin gene-related peptide; CysLTs, Cysteinyl leukotrienes; EAACI, European Academy of Allergy and Clinical Immunology; ECM, Extracellular matrix; Eos, Eosinophils; ER, Endoplasmic reticulum; GABA, γ-aminobutyric acid; GC, Glucocorticoid; GLP1RA, Glucagon-like peptide-1 receptor agonist; HDM, House dust mite; HDAC9, Histone deacetylase 9; IL, Interleukin; LAMA, Long-acting muscarinic antagonist; NANC, Non-adrenergic non-cholinergic; Neu, Neutrophils; NKA, Neurokinin A; NO, Nitric oxide; NOS, Nitric oxide synthase; ORMDL3, Orosomucoid 1-like 3; PGE, Prostaglandin E; PGD, Prostaglandin D; PNEC, Pulmonary neuroendocrine cell; RAS, Renin-angiotensin system; RAMP, Receptor activity-modifying protein; ROS, Reactive oxygen species; SCFA, Short chain fatty acids; SERCA: Sarco/endoplasmic reticulum Ca²⁺-ATPase; SP, Substance P; Th, T helper; TNF-α, Tumor necrosis factor-α.

Metabolic pathways driving asthma pathogenesis

Obesity is associated with more severe asthma, however, the mechanisms responsible are poorly understood. Obesity is associated with low-grade systemic inflammation, and it is possible that this inflammation extends to the airways, contributing to worse asthma outcomes. A recent meta-analysis that included 40 studies showed that sputum neutrophils were 5% higher in obese versus non-obese asthmatics; blood neutrophil count was also higher, as were bronchial submucosal eosinophil counts and sputum IL-5 levels. Conversely, FeNO was 4.5 ppb lower in obesity. Blood C reactive protein, IL-6 and leptin were also higher in obesity.¹⁶⁷ IL‐6 expression is increased in the context of metabolic syndrome and obesity in severe asthma and has been associated with diminished airway function.¹⁶⁸ Consequently, the different pattern of inflammation in obese asthmatics should be further explored in order to develop proper targeted interventions.

Biomarkers related to the tricarboxylic acid cycle, hypoxia response, amino acid metabolism (glutamine, L-arginine) and oxidative stress are all associated with asthma diagnosis or with increased asthma morbidity.168, 169, 170 Individuals with asthma and control subjects differ in airway epithelial cell rates of glycolysis and oxidative phosphorylation and in mitochondrial function and structure.^171,172 Obese individuals with asthma have lower L-arginine concentrations, and their airway epithelial cells produce less nitric oxide (NO) because of NO synthase (NOS) uncoupling. When this occurs, NOS preferentially makes anion superoxide instead of NO. Reduced NO bioavailability can impair bronchodilation while also affecting mitochondrial function. Loss of this NO inhibitory mechanism in obesity, in addition to increased glycolytic rates, is associated with increased maximal mitochondrial respiration, which in turn increases the production of reactive oxygen species (ROS). Unstimulated airway epithelial cells from obese subjects with asthma display increased degrees of oxidative and nitrative stress.¹⁷³

ROS play a central role in airway epithelium-mediated sensing, development of innate and adaptive immune responses, and airway remodeling and AHR.¹⁷⁴ Over-production of ROS resulting from infiltrating immune cells, particularly eosinophils and neutrophils, and a concomitant impairment of antioxidant responses lead to oxidative stress in asthma. Oxidative stress is augmented in severe asthma and during exacerbations, as well as by air pollution and obesity, both linked to T2 low asthma. Several common causes of oxidative stress were linked to acquired glucocorticoid resistance in asthma.¹⁷⁵ Antioxidant treatment may be beneficial to glucocorticoid resistant and T2-low asthma. Unfortunately, antioxidant compounds have proven clinically ineffective, partly due to poor stability and rapid in vivo metabolism.¹⁷⁶ The compartmentalized nature of ROS production and sensing, and the role of ROS in homeostatic responses and in the action of corticosteroids and β2-adrenergic receptor (β2AR) agonists, add another layer of complexity to antioxidant therapy development.

Lipid metabolism, especially in distinct cells such as T cells, macrophages, granulocytes, and non-immune cells, plays an essential role in the pathogenesis of asthma, as lipids are potent signaling molecules that regulate a multitude of cellular response. Tissue-resident alveolar macrophages display a unique metabolic profile, influenced by the concentrations of glucose and lipids in the microenvironment.¹⁷⁷ An unsupervised, group-agnostic integrative multi-omic factor analysis performed using host/bacterial (meta)transcriptomic and bacterial shotgun metagenomic datasets from bronchial brush samples paired with metabolomic/lipidomic data from bronchoalveolar lavage (BAL) samples acquired from children of 1 to 17 years old showed one signature characterizing preschool-aged recurrent wheeze and another capturing an inferred trajectory from health to wheeze and school-aged asthma. Recurrent wheeze was driven by T1-immune signatures, coupled with upregulation of immune-related lipids and metabolites, particularly those associated with neutrophils. Comparatively, progression toward asthma from ages 1 to 18 was dominated by changes related to airway epithelial cell gene expression, T2-immune responses, and constituents of the airway microbiome, such as increased Haemophilus influenzae.¹⁷⁸ A clear association was demonstrated between nonallergic childhood asthma, lower whole-blood sphingolipids, and asthma-risk 17q21 genotypes encoding ORMDL3.¹⁷⁹ ORMDL3 encodes an ER-resident transmembrane protein that regulates the activity of serine palmitoyltransferase (SPT), the first and rate-limiting enzyme for sphingolipid biosynthesis in cells. Sphingolipids are essential for formation and integrity of cellular membranes; they also serve as precursors for bioactive molecules that regulate key cellular processes and can be synthesized both de novo and through recycling pathways. In addition to its role in sphingolipid biosynthesis, ORMDL3 has been implicated in other cellular processes relevant to asthma, such as ER stress and the unfolded protein response (UPR), which are involved in asthma exacerbation.¹⁸⁰ Aside from SPT regulation, ORMDL3 is also reported to regulate the sarcoendoplasmic reticulum (SR) calcium transport ATPase (SERCA) pump, which transports calcium ions from the cytoplasm into the SR. This process is vital for maintaining homeostatic levels of calcium ions inside the ER, and any dysregulation of ER calcium levels can result in increased ER stress, UPR, and exacerbation of asthma pathogenesis. Changes in ORMDL3-dependent regulation of the SERCA pump could elicit ER stress and UPR in multiple cell types, including CD4⁺ T cells, contributing to increased incidence and/or severity of asthma in patients carrying ORMDL3 risk single nucleotide polymorphisms (SNPs). The CD4⁺ T cells from humans harboring 17q12–21 asthma risk SNPs display ∼3-fold overexpression of ORMDL3.¹⁸¹ In these patients, dynamic changes in the expression and activity of key SERCA pumps in response to T cell receptor activation would be hindered by higher ORMDL3 expression in CD4⁺ T cells. This would result in reduced calcium uptake into the ER through SERCA pumps and increased cytosolic calcium concentrations in resting T cells, altering the key rheostat of early T cell signaling.¹⁸² This dysregulation might also induce ER stress, activate the UPR, and alter T cell differentiation and function. ORMDL3 overexpression was reported to skew CD4⁺ T cell differentiation toward a Th2 and Th17 imbalance.^183,184 Altered sphingolipid synthesis via SPT regulation is the major driver of these changes in T cell differentiation and function, which may also indirectly influence calcium homeostasis and ER stress via sarco/endoplasmic reticulum Ca²⁺–ATPase 2B (SERCA2B). Furthermore, increased ORMDL3 expression induces autophagy, possibly through interacting with SERCA2, thereby inhibiting calcium uptake into the ER/SR, and induces cell death, impairing bronchial epithelial function in asthma.¹⁸⁵

Once regarded principally as regulators of smooth muscle tone and vascular permeability, arachidonic acid-derived lipid mediators are now well known to regulate features of innate and adaptive immunity. Patients with aspirin exacerbated respiratory disease (AERD) have overproduction of cysteinyl leukotrienes (cysLTs) and inflammatory prostaglandins (PGs) such as prostaglandin D₂ (PGD₂), as well as underproduction of the anti-inflammatory prostaglandin E₂ (PGE₂), high levels of mast cell activation, platelet activation, and severe T2 respiratory inflammation.^186,187 PGE₂ stabilizes and prevents the activation of MCs and ILC2s, and prevents the formation of cysLTs by inhibiting the translocation of 5-lipoxygenase.^188,189 Both cysLTs and inflammatory prostaglandins likely play a role in amplifying respiratory tract inflammation and eosinophilia.¹⁹⁰ Several studies link AERD to diminished production (or increased catabolism) of PGE₂ in the respiratory tract, associated with diminished expression and function of cyclooxygenase-2 (COX-2) and/or microsomal (m) prostaglandin E synthase (PGES)-1, the dominant PGES isoform upregulated during inflammation.^190,191 Since COX-1 is exquisitely sensitive to inhibition by aspirin, a chronic reduction in COX-2/mPGES-1-derived PGE₂ could account for aspirin-induced reactions in AERD, by removing a critical COX-1-dependent “braking” function. Urinary levels of leukotriene E₄ (LTE₄), a marker of systemic cysLT production, are characteristically elevated in patients with AERD and increase further during non-steroidal anti-inflammatory drugs-induced reactions.¹⁹⁰ The overproduction of these lipid mediators may be driven by epithelial–immune crosstalk: Epithelial cell-derived TSLP stimulates human MCs to generate PGD₂, which activates platelets to release high amounts of epithelial alarmin IL-33.^192,193 However, there are no specific clinical trials targeting the epithelial derived cytokines in AERD. As severe T2 inflammation is a hallmark of the disease, targeting its key cytokines IL-4 and IL-13 with dupilumab proved very successful within the 1st month after initiating treatment.¹⁹⁴ Omalizumab dramatically decreased urinary levels of both LTE₄ and PGD₂ metabolites and induced improvement in baseline symptoms.¹⁹⁵ These findings, along with the efficacy of dupilumab, point to a central role for both IgE- and IL-4/13-driven pathways in AERD. The marked response to omalizumab specifically supports the involvement of local IgE in driving eicosanoid production and mast cell activation, though the antigen specificity of this local IgE remains to be determined.¹⁹⁶

Overall, diets emphasizing the consumption of plant-based foods might protect against asthma development and improve asthma symptoms through their effects on systemic inflammation, oxidation, and microbial composition. Additionally, increased fruit and vegetable intake, reduced animal product consumption, and weight management might mediate cytokine release, free radical damage, and immune responses involved in the development and course of asthma.^197,198

Hormonal impact on asthma

There are marked sex differences in asthma prevalence and severity. Estrogen, progesterone, and testosterone directly interact with airway epithelial cells, ASM cells, the mononuclear phagocyte system, innate lymphoid cells, eosinophils, mast cells, T cells, and B cells.199, 200, 201 Animal studies have shown that estrogen increases T2-mediated airway inflammation and AHR, while testosterone decreases T2-mediated inflammation and protects against remodeling.^201,202 Females have increased IL-17A-mediated airway inflammation compared to males.²⁰³ In line with the protective effects of testosterone against airway inflammation and remodeling, nebulized dehydroepiandrosterone-3-sulfate—an androgen precursor—was shown to improve asthma control, while the use of oral contraceptives may worsen asthma symptoms.^204,205 The complex sex steroid signaling in the lung, its interaction with inflammatory mediators and other hormones (for example in obesity) and with the local steroid metabolism, and its age-specificity, deserve further exploration for developing targeted interventions based on sex in asthma patients. Furthermore, genetic and epigenetic factors further shape sex-related differences in asthma. The X chromosome harbors immune-regulatory genes, including TLR7 and TLR8, which amplify inflammatory responses in females.^206,207 The sex-dependent expression of IL-13 and ORMDL3 influences eosinophilic inflammation and airway remodeling.²⁰⁸ Epigenetic modifications, such as DNA methylation and microRNA regulation, further impact immune activation and corticosteroid responsiveness.²⁰⁸

Incretins may have anti-inflammatory effect in asthma, whereas hyperthyroidism can exacerbate asthma and hypothyroidism is associated with milder asthma symptom.²⁰⁹ Glucagon-like peptide-1 receptor agonists (GLP-1 RAs), originally developed for the treatment of type 2 diabetes, have attracted attention for their potential therapeutic benefits in asthma due to their anti-inflammatory properties and effects on ASM function.^210,211 However, concerns have been raised about the possibility of GLP-1RAs inducing or exacerbating asthma symptoms.²¹² Reassuring data from a large adult asthma cohort suggest that metformin was associated with a lower rate of asthma attacks, with further reductions when GLP-1 RA was added.²¹³ Furthermore, liraglutide attenuated in vivo platelet activation in an AERD murine model and in vitro activation in human platelets from patients with and without AERD.²¹⁴

Dysregulation of renin-angiotensin system contributes to the pro-inflammatory, pro-oxidative, and pro-fibrotic processes that occur in pulmonary diseases like asthma, chronic obstructive pulmonary disease, idiopathic pulmonary fibrosis, and acute lung injury. Inhibition of the pro-inflammatory angiotensin-converting enzyme (ACE)–angiotensin (Ang) II axis and activation of the protective ACE2–Ang-(1-7)–Mas receptor axis have each demonstrated varying degrees of efficacy in experimental respiratory disease models and human trials.²¹⁵

Impact of the microbiome on asthma

The effect of microbiota dysbiosis on asthma heterogeneity and natural evolution seems intuitive, notwithstanding that until recently the lung and gut microbiome composition have been addressed in studies on phenotypes and endotypes of asthma.^216,217

Proteobacteria appear repeatedly to be the most dominant phylum overrepresented in the airways of patients with asthma compared with non-asthmatic volunteers across several human studies.^218,219 The Proteobacteria phylum is represented by potentially pathogenic bacteria, including those that belong to the genera Haemophilus, Moraxella, and Neisseria.²²⁰ Patients with neutrophilic asthma, usually receiving high ICS doses, demonstrate a less diverse bacterial load with relative enrichment in Haemophilus and Moraxella species, and a reduction in the relative abundance of Streptococcus, Gemella, and Porphyromonas taxa compared with patients with eosinophilic asthma.^221,222 In contrast to reproducible findings in neutrophilic asthma, the status of the microbiota in eosinophilic and T2 high asthma is less clear and more heterogeneous, potentially reflecting again the differences in underlying endotypes or mixed responses caused by ICS treatment. Subjects with atopic asthma demonstrate enrichments in bacteria from the genera Fusobacterium and Porphyromonas and the Sphingomonodaceae family, and decreased relative abundance of members of Mogibacteriaceae family and Lactobacillales order.²²³ Increased numbers of sputum eosinophils have been connected with the presence of T. whipplei.²²² Streptococcus abundance was also reported to be increased in patients with severe asthma compared with healthy control,²²⁴ while Actinobacteria abundance was associated with molecular indicators of steroid responsiveness (FK506 binding protein, FKBP5).²¹⁸

The concept of the gut–lung connection was born out of the observation that different lung diseases can be influenced by intestinal microenvironment changes and vice versa. The microbiota is an important factor responsible for interactions between these two sites in asthma.²²⁵ Many studies show that early life is the most important period during which microbiota dysbiosis in the gut may lead to the development of many respiratory diseases, as the gut microbiota has a significant influence on immune cell maturation and resistance to pathogens.²²⁶ In adults, both obese and non-obese asthma patients with severe disease have reduced fecal levels of Akkermansia muciniphila. This depletion may have a causal role in disease pathogenesis, as supported by murine models showing that a loss of this bacterium exacerbates acute and chronic airway inflammation.²²⁷ The effects of the gut microbiota on asthma are at least partially mediated by bacterial metabolites, such as histamine or short chain fatty acids (SCFAs), which may influence immune responses in distal parts of the body. For instance, the abundance of histamine-secreting bacteria is significantly higher in fecal samples of asthma patients compared with non-asthmatic volunteers.²²⁸ In parallel, SCFAs have been shown to activate olfactory receptors in the lung,²²⁹ particularly OR51E2 expressed on ASM cells, where they slow cytoskeletal remodeling and reduce ASM proliferation.²²⁹ Soluble fiber supplementation, essential for the generation of SCFA, has been found to decrease sputum eosinophilia and sputum histone deacetylase 9 gene (HDAC9) expression in asthma patients.²³⁰ These observations establish a non-immune gut–lung mechanism that ties gut microbial communities to asthma endotypes through both metabolic and epigenetic regulation.

Trained immunity in asthma pathogenesis

Trained immunity refers to the enhanced response of innate immune cells to subsequent encounters with antigens, facilitated by epigenetic and metabolic modifications.²³¹ While having beneficial effects against pathogens, it may worsen inflammatory diseases by promoting disease progression through mechanisms such as inflammasome activation and epigenetic reprogramming.²³² Early-life viral infections have been shown to promote sustained innate immune memory (trained immunity) in macrophages, which drives the differentiation of naive Th cells toward Th2 and Th17 cells.²³³ Combined with allergen exposure, they induce metabolic reprogramming in lung macrophages leading to allergic asthma in childhood.²³³ Even a short exposure to allergens leads to a fundamental reprogramming of lipid mediator metabolism, with macrophages representing particularly plastic responder cells.²³⁴ Pollution-induced trained immunity accounts for an enhanced inflammatory response (TNF, IL-6, and IL-8) and defines a special pediatric asthma endotype that can be depicted by enhanced H3K27ac marks in circulating monocytes.²³⁵ Trained immunity was also linked to the differential antiviral cytokine production between asthmatic patients and healthy subjects.²³⁶

Distinct neonatal DNA methylation modules are associated with the capacity to mount innate immune responses, underscoring a link between neonatal epigenetics and childhood asthma risk. The fetal innate immune system can be trained by microbial exposures during pregnancy, thus shedding new light on the mechanisms explaining asthma-protected farm children.237, 238, 239

Bacterial lysates can promote trained immunity against pathogens. In a randomized controlled trial, mucosal bacterial immunotherapy based on whole inactivated bacteria shows safety and clinical efficacy against recurrent wheezing attacks in children.²⁴⁰ Innate immune cells trained with specific stimuli might also acquire anti-inflammatory features and promote tolerance, which may have important implications for chronic inflammatory diseases such as asthma. Recent findings showed that allergoid–mannan conjugates, which are next generation vaccines for AIT, can reprogram monocytes into tolerogenic DCs by mechanisms depending on metabolic and epigenetic rewiring.²⁴¹

Neurogenic inflammation in asthma

In addition to the classical cholinergic and adrenergic systems, human airways are also innervated by non-adrenergic, non-cholinergic (NANC) pathways comprising an excitatory branch that mediates bronchoconstriction via tachykinins, substance P (SP) and neurokinin (NK) A, and an inhibitory branch that induces bronchodilation via vasoactive intestinal peptide (VIP) and NO. Imbalances between excitatory and inhibitory NANC signaling also contribute to AHR. The airway epithelium contains pulmonary neuroendocrine cells (PNEC) filled with neurotransmitters (serotonin and gamma-aminobutyric acid [GABA]) and neuropeptides (SP, NKA, VIP, calcitonin-gene related peptide [CGRP], and nociception/orphanin FQ(N/OFQ)), which are released upon various environmental exposures. Together with classical neurotransmitters such as acetylcholine (ACh) and neuropeptide Y (NPY) from autonomic nerve fibers, these mediators create a complex neurochemical environment within the airway wall. This environment is further modified by neurotrophins (e.g., nerve growth factor and brain-derived neurotrophic factor) secreted by structural and inflammatory cells, including epithelial cells, lymphocytes, and eosinophils.^242,243 Under pathological conditions, this finely tuned neurochemical network can be disrupted. For instance, allergens such as house dust mites have been shown to induce PNECs hyperplasia, which may contribute to exacerbated asthma response.²⁴⁴

The release of SP, NKA, and serotonin may exacerbate the inflammatory response, while VIP, N/OFQ, and GABA show anti-inflammatory activity. The GABA-ergic system was however linked to mucus overproduction.²⁴⁵ Tachykinins may cause vasodilatation, plasma exudation, and mucus secretion, whereas CGRP may contribute to hyperemia of inflammation. Their effect may be amplified further by loss of the major degrading enzyme, neutral endopeptidase, from epithelial cells. The CGRP–receptor activity‑modifying protein 1 (RAMP1) axis enhances Th2 and Th9 responses; and the SP–NK1 receptor (NK1R) axis promotes the synthesis of chemokines in eosinophils, mast cells, and neutrophils.²⁴⁶ In contrast, VIP–VIP receptor 1 (VPAC1) and N/OFQ–N/OFQ peptide receptor (NOP) axes cause bronchodilation and anti-inflammatory and anti-remodeling effects.247, 248, 249 Sputum SP concentration was significantly higher in adults with asthma and significantly corelated with the eosinophil cell count in induced sputum and with the forced expiratory volume in 1 s (FEV₁)/forced vital capacity (FVC) ratio.²⁵⁰ Asthmatic children also exhibited an increased level of SP, along with a higher proportion of M2 macrophages in their bronchoalveolar lavage fluid.²⁵¹ Mechanically, SP interaction with its NK1R activates the phosphoinositide 3-kinase (PI3K)/protein kinase B (AKT)/nuclear factor kappa-light-chain-enhancer of activated B cells (NF-κB) pathway, which further triggers NLR family pyrin domain containing 3 (NLRP3)-mediated pyroptotic cell death in the bronchial epithelial cells.²⁵² The clinical relevance of this pathway is underscored by the positive correlation observed between epithelial mucus content and epithelial SP expression.²⁵³ Of interest, asthmatic subjects with Mycoplasma pneumoniae, compared with those without, showed higher baseline epithelial neurokinin 1 (NK1) expression, which was significantly reduced after antibiotic treatment.²⁵³ This infection-induced upregulation of NK1R may amplify epithelial responses to SP, exacerbating airway inflammation.

The ACh–M3 axis induces goblet cell metaplasia, ECM deposition, and bronchoconstriction.²⁵⁴ In contrast, the ACh–α7nAChR axis diminishes the synthesis of TNF-α, IL-1, and IL-6, attenuating lung inflammation.²⁵⁵ Patients with asthma are more sensitive to M3 receptor signaling as compared to healthy subjects, partially due to an enhanced ability to open large Ca²⁺ channels in ASM.²⁵⁶ Consequently, in addition to bronchodilation, long acting antimuscarinic agents (LAMA) also exert anti-inflammatory and anti-fibrotic effects by inhibiting muscarinic receptors present in neutrophils, macrophages, fibroblasts and ASM cells.²⁵⁴ Unfortunately, 75% of severe asthmatics on triple therapy remain uncontrolled.²⁵⁷ The relatively low success rate of LAMA in severe asthma might arise from the lack of a phenotype-guided prescription. The visible properties of an asthma patient theratype benefiting from LAMA addition were recently described.²⁵⁴

Communication between local sensory neurons, immune cells, and resident lung stromal cells appears to be a key driver of AHR, airway obstruction, and inflammation.^242,258 Lung-innervating neurons of the jugular nodose complex express the high-affinity IgE receptor Fc epsilon receptor (FcεR). Vagal nociceptor neurons expressing high-affinity immunoglobulin epsilon receptor gamma subunit (FcεR1γ) are directly depolarized upon encountering allergen–IgE complexes, leading to action potential firing, calcium influx, and release of SP.²⁵⁹ Increased in vivo circulating levels of IgE following allergen sensitization enhances the responsiveness of FcεR1 to immune complexes in both mouse jugular nodose complex neurons and human induced pluripotent stem cell-derived nociceptors.²⁵⁹ SP released from these neurons can then act on local immune cells. Specifically, SP-mediated Mas-related G protein-coupled receptors (MRGPR) B2 activation on MCs contributes to airway inflammation and goblet cell hyperplasia.²⁶⁰ These responses are modulated by β-arrestin 2, which promotes MCs recruitment to facilitate their activation through FcεRI.²⁶⁰ In addition to these neuronal pathways, PNECs represent another key sensory population in the airway epithelium. Their functional importance is underscored by studies showing that bronchial thermoplasty attenuates airway disease by interfering with PNEC secretion and synapse formation.²⁶¹

Type VII asthma

Type VII hypersensitivity covers direct cell activation leading to tissue damage. This activation can occur through mechano- or chemical-sensitive ion channels, through transmembrane G protein-coupled receptors (GPCRs) or through direct activation of intracellular pathways (Fig. 5). These pathogenic mechanisms are central to asthma following exposure to pollutants, cold temperature, microbes, allergens, various metabolites or drugs or as a consequence of extensive airway remodeling. They also define asthma theratypes such as those with primary or acquired resistance to corticosteroids.

Fig. 5.

Type VII asthma induced by direct cell activation. Key pathways involved in Type VII hypersensitivity reactions in asthma according to the new EAACI nomenclature. Direct activation of airway cells occurs through multiple non-immune mechanisms, including chemical pollutants (e.g., PFAS), environmental irritants (cold, mechanical stress, microplastics, Gram-negative bacteria), proteases (e.g., HDM Der p 1), epithelial cells and macrophages, and specific receptor agonists. AEC, Airway epithelial cell; AIM2, Absent in melanoma 2; ASM, Airway smooth muscle; cAMP, Cyclic adenosine monophosphate; EAACI, European Academy of Allergy and Clinical Immunology; HDM, House dust mite; IL, Interleukin; MФ, Macrophage; MC, Mast cell; MRGPR, Mas-related G protein-coupled receptor; OR, Olfactory receptor; PAR, Protease-activated receptor; PFAS, Per- and polyfluoroalkyl substances; TAS2R, Type 2 taste receptor; TRP, Transient receptor potential.

More than 300,000 new substances have been introduced into human lives after 1960s without adequate assessment on their health and environmental impact. Many of them have ended up as pollutants. Perfluoroalkyl substances (PFAS) are widely used in various manufacturing processes. Accumulation of these chemicals has adverse effects on human health, including inflammation in multiple organs. Double-stranded DNA receptor absent in melanoma 2 (AIM2) can recognize perfluorooctane sulfonate, a common form of PFAS, to trigger IL-1β secretion and pyroptosis.²⁶²

Transient receptor potential ankyrin-1 (TRPA1) and TRP vanilloid-1 (TRPV1) are nonselective cation channels that act as chemosensors for many exogenous irritants and endogenous ligands including proinflammatory mediators.^263,264 Besides the C-fibers, TRPA1 and TRPV1 are also expressed by nonneuronal cells including epithelial cells, ASM cells and fibroblasts.^265,266 An in vitro study claimed that although either TRPV1 or TRPA1 activation causes airway neurogenic inflammation, solely TRPA1 activation orchestrates an additional inflammatory response which is not neurogenic.²⁶⁶ TRPA1 polymorphisms were correlated with reduced asthma control.²⁶⁷ A relationship between higher TRPA1 expression and activity and lower TRPV1 expression and activity was recently described.²⁶⁸ The TRPV1 I585I/V genotype was associated with increased TRPA1 expression by primary airway epithelial cells and with amplified responses to selected air pollution particles in vitro. However, the TRPV1 I585I/V genotype was not associated with worse asthma symptom control among children exposed to tobacco smoke, whereas other TRPA1 and TRPV1 variants were.²⁶⁸ Co-exposure to polystyrene microplastics and di-(2-ethylhexyl) phthalate was shown to amplify oxidative stress and inflammatory responses in allergic asthma by activating the TRPA1–p38 mitogen-activated protein kinase (MAPK) pathway.²⁶⁹ Both TRPV1 and TRPA1 function as receptors for N-acyl homoserine lactones, quorum sensing molecules produced by Gram-negative bacteria.²⁷⁰ Another function of the TRPA1 channel is to sense cold temperature.²⁷¹ In a mouse model of cold-induced exacerbation of allergic airway disease, blocking TRPA1 with a selective antagonist resulted in reduced inflammation, airway remodeling and AHR. This study highlighted a potential new mechanism of cold-induced airway disease through TRPA1.²⁷² Of note, TRPA1 is not involved in body temperature regulation at basal levels or under cold challenge, making it a viable target for cold-induced exacerbations of lung disease without unwanted temperature effects on the body.

Cysteine and serine proteases contribute to inflammation by activating protease and Mas‑related G protein-coupled receptors (MRGPRs). Der p1, a major allergen molecule from the house dust mite, containing cysteine protease, activates protease-activated receptor 2 (PAR-2) and Mas-related G-protein coupled receptor member X1 (MRGPRX1) and induces the release of the pro-inflammatory cytokine IL-6 from cells expressing these receptors.^273,274 Mas-related G-protein coupled receptor member X2 (MRGPRX2) is expressed predominantly on MCs.²⁷⁵ MCs are increased especially in the lungs of patients who died from asthma and have been related to AHR.²⁷⁶ A recent study showed that MRGPRX2 expression is significantly upregulated in MCs from asthmatic lungs as compared to those from non-asthmatic lungs, and the main activator of MRGPRX2 is the hemokinin-1, a neuropeptide produced by human bronchial cells and macrophages resulting in ASM contraction.²⁷⁷ Thus, selective MRGPRX antagonists could serve as novel targets for the modulation of T1 asthma (MRGPRX1) and of AHR/near-fatal asthma (MRGPRX2).

Bitter taste receptors belong to type II taste receptors (TAS2R), a family of the seven transmembrane GPCRs expressed on a variety of cell types: Tuft cells, ciliated epithelial cells, mast cells, neutrophils, monocytes, eosinophils, lymphocytes and ASM cells.278, 279, 280 There are 25 distinct TAS2R subtypes in humans.²⁶⁴ The nuances in signaling differences between TAS2R subtypes, if any, are unclear due to lack of receptor subtype-specific ligands and expression of multiple TAS2R subtypes on an individual cell. Their expression is increased in severe asthma.²⁸¹ A seminal paper established the potential of targeting TAS2R to stimulate ASM relaxation and bronchodilation.²⁸² Critically, this therapeutic potential targeting TAS2R appears to be preserved even in the context of airway inflammation, since TAS2R expression and signaling, along with TAS2R-mediated ASM relaxation and bronchodilation, remain unaltered under inflammatory conditions.²⁸³ TAS2R-mediated ASM relaxation is unaffected by tachyphylaxis, as opposed to β₂-adrenergic receptor (β₂ AR)_, suggesting that in the conditions where asthmatics are refractory to beta agonist treatment, TAS2R agonists can be useful for stimulating bronchodilation. Further, ASM cells express multiple subtypes of TAS2Rs (at least 3–4 of them at a level higher than β₂AR), all of which are known to relax ASM. This presents an opportunity to explore multiple subtypes of TAS2R either singularly or in combination for drug development, as opposed to the β₂AR, which has only one subtype. TAS2R agonists have been shown to exert pleiotropic effects on different airway cell types. In ASM cells, they significantly inhibit the development of AHR and the cell cycle progression by inhibiting PI3K-mediated pro-mitogenic signaling.^284,285 In MCs, stimulation with TAS2R agonists decreases the release of pro-inflammatory mediators such as histamine and eicosanoids.²⁷⁸ In immune cells, TAS2R agonists also exert anti-inflammatory effects. For instance, chloroquine may act on monocytes, while azithromycin promotes macrophage polarization from M1 pro-inflammatory phenotype to M2 anti-inflammatory/reparative phenotype.^286,287

Several human olfactory receptors (ORs) were identified on the ASM cells.²²⁹ Olfactory receptor family 1 subfamily D member 2 (OR1D2) and olfactory receptor family 2 subfamily AG member 1 (OR2AG1) are expressed at the RNA and protein levels in ASM cells.²⁸⁸ Specific agonists for OR2AG1 and OR1D2 trigger transient calcium increases in ASM via a cyclic adenosine monophosphate (cAMP)-dependent signal transduction cascade. Furthermore, the activation of OR2AG1 inhibited the histamine-induced contraction of the ASM, whereas the stimulation of OR1D2 led to an increase in cell contractility.²⁸⁸ In addition, OR1D2 activation induced the secretion of IL-8 and GM-CSF.²⁸⁸ Overall, ORs exert diverse and cell type-specific effects on ASM function. ORs do not uniformly promote bronchodilation; instead, they modulate a range of cellular responses including cytoskeletal remodeling, contractility, and cytokine secretion—processes that collectively contribute to two cardinal features of asthma: airway remodeling and hyperplasia.²⁸⁹

A critical aspect of the biology of the lungs is its capacity to respond to mechanical stretch. The sensing of the physical movements in the lungs is regulated by mechanosensitive ion channels such as the TRP superfamily and Piezo (PZ) channels (i.e., Piezo 1 and Piezo 2), both of which are widely expressed throughout the lungs.²⁹⁰ It was recently shown that pulmonary fibrosis could be driven by elevated mechanical tension in alveolar type 2 cells inducing a TGF‐β signaling loop, leading to impaired alveolar regeneration.²⁹¹ Recent studies have revealed cell type-specific and context-dependent roles for the mechanosensitive Piezo channels in asthma pathogenesis. In lung fibroblasts, Piezo1 mediates mechanical stretch-induced ECM production via extracellular signal-regulated kinase (ERK) phosphorylation and Ca²⁺ influx, suggesting a role in airway remodeling.²⁹² Similarly, in ASM cells, mechanical stretch activates pro-inflammatory mechanisms through the interaction between the mechanosensitive piezo channels and the key Ca²⁺ regulatory protein stromal interaction molecule 1 (STIM1), implicating a role of Piezo in pro-inflammatory responses.²⁹³ In contrast to these pro-inflammatory and pro-remodeling effects, Piezo1 serves as a negative regulator of ILC2 function. Upon ILC2 activation, Piezo1 expression is induced, which restrains ILC2 activity by suppressing cytokine production and ultimately decreasing AHR.²⁹⁴ This dual functionality highlights the functional complexity of mechanical signaling in the airways. Blocking mechanical pathways using inhibitors such as gadolinium (i.e., Piezo 1 inhibitor) has been shown to reduce inflammation and mucous secretion.²⁹⁵ These findings underscore the need for cell type-specific targeting strategies when considering Piezo channels as therapeutic targets in asthma.

Severe steroid-resistant asthma in adults is defined by the failure to achieve a >15% improvement in FEV₁ after 14 days of oral steroid treatment.²⁹⁶ This condition affects 5–10% of patients with severe asthma and represents a major unmet need in current asthma management, contributing to the substantial healthcare burden resulting from prolonged and frequent hospitalizations and complications associated with OCS use.²⁹⁷ Steroid resistance can be either inherited (primary) or acquired (secondary). Corticosteroids readily diffuse into the airway tissue, bind and activate cytosolic glucocorticoid receptors (GRs). The anti-inflammatory effects are mediated by the GRα isoform via ligand-dependent transcription factors, while GRβ acts as a dominant inhibitor of GRα. Defective GRα expression and activity impairs the anti-inflammatory effects of glucocorticoids and plays important roles in the induction of steroid resistance (Fig. 6). Conversely, one of the important molecular mechanisms contributing to primary corticosteroid resistance is the overexpression of the GRβ isoform.²⁹⁸ Over 3000 single-nucleotide polymorphisms have been described for the gene encoding GR.²⁹⁹ In addition, various GR protein isoforms result from 11 transcription start sites, and differential mRNA splicing leads to further GR protein variants; each can be modified post-translationally and alter steroid responsiveness. Furthermore, some GR isoforms are expressed in a cell-type-specific manner or in a sub-cellular location. Defective nuclear translocation of GRs reduces expression and activity of histone deacetylase (HDAC) 2 in patients with asthma,³⁰⁰ and this reduction is associated with steroid insensitivity and more severe disease.³⁰⁰ Dysregulation of GR signaling pathways, including changes in GR phosphorylation, makes a substantial contribution to the inherent glucocorticoid resistance in severe asthma.³⁰¹ Higher levels of IL-2, IL-4 and IL-13 expression result in local cytokine secretion which further alters GRα translocation. Pro-inflammatory transcription factors, including NF-κB, activator protein-1, and IFN regulatory factor-1, competitively interact with GR for binding sites on DNA as well as for transcriptional coactivators. Such competitive interaction limits GR’s ability to suppress the production of pro-inflammatory cytokines.²⁹⁹ In severe asthma, there is a dysregulation of the NF-κB and MAPK pathways, whereby the ASM cells can proliferate and contribute to airway remodeling despite corticosteroid treatment.³⁰² In addition, exposure to smoke or pollutants increases oxidative stress, which further compromises GR signaling and contributes to corticosteroid resistance.³⁰³ PI3K-delta inhibitors that reestablish HDAC2 function together with p38 MAPK inhibitors decreasing GR phosphorylation, are currently in clinical development.^304,305

Fig. 6.

Mechanism of glucocorticoids resistance in severe asthma. Glucocorticoids bind to cytosolic GRα, inducing its nuclear translocation where the ligand-activated GRα complex recruits HDAC2 to GRE, resulting in gene suppression and anti-inflammatory effects. Resistance arises from multiple defects, including overexpression of the dominant-negative GRβ isoform (①), defective GRα nuclear translocation (②), reduced HDAC2 expression/activity (③), cytokine-induced alterations (IL-2, IL-4, IL-13) (⑤), and competitive interference by pro-inflammatory transcription factors (NF-κB, NF-AT1, IRF1) (⑥). GC, Glucocorticoid; GRα, Glucocorticoid receptor α; GRβ, Glucocorticoid receptor β; GRE, Glucocorticoid response element; HDAC2, Histone deacetylase 2; IL, Interleukin; IRF1, Interferon regulatory factor 1; NF-AT1, Nuclear factor of activated T cells 1; NF-κB, Nuclear factor κB.

Understanding how to tackle the complex pathogenetic pathways classified as type V, VI or VII asthma opens novel opportunities for personalized asthma management. This approach complements current biologicals targeting the T2 pathway and brings us closer to achieving biological remission in asthma (Table 1).

Table 1.

Novel asthma theratypes uncovered by the concept of type V, VI and VII asthma.

Target	Interventions by asthma type	Clinical impacts
Chronic inflammation	Fruit, fibers and vegetable intake, reduced animal product consumption, and weight management170,197,198,229,230 (Type VI) Long-acting antimuscarinic agents254,255 (Type VI) Metformin, glucagon-like peptide-1 receptor agonists212,213 (Type VI) Nebulized dehydroepiandrosterone-3-sulfate204 (Type VI) Inhibition of the renin-angiotensin system215 (Type VI) Mucosal bacterial immunotherapy240 (Type VI) Innate immune cells trained with specific stimuli (e.g., allergoid-mannan conjugates)241 (Type VI) Decreased release of mediators from MCs with TAS2R agonists278 (Type VI) Anti-inflammatory actions on monocytes with bitter taste receptor agonists (chloroquine)286 (Type VI) Phenotypic shift in macrophages to an anti-inflammatory/reparative phenotype with TAS2R agonists such as macrolides287 (Type VI) TRPA1 inhibitors266 (Type VII)	Asthma associated with obesity/metabolic syndrome
T1 asthma	Selective MRGPRX1 antagonists273,274,277 (Type VII)	Non-eosinophilic asthma
AHR/near-fatal asthma (MCs driven)	Selective MRGPRX2 antagonists275, 276, 277 (Type VII)	AHR/near-fatal asthma
Cold-induced respiratory disease	TRPA1 inhibitors271,272 (Type VII)	Cold exposure as major trigger of asthma symptoms or exacerbations
Airway obstruction	Mucosal bacterial immunotherapy240 (Type VI) Long-acting antimuscarinic agents254 (Type VI) TAS2R agonists in asthmatics refractory to beta agonist treatment282 (Type VII)	Fixed airway obstruction Asthmatics refractory to β- agonist treatment
Airway hyperreactivity	Long-acting antimuscarinic agents254 (Type VI) Piezo1 activation restrains ILC2 activity by reducing cytokine production, resulting in decreased AHR294 (Type VII) TAS2R agonists282 (Type VII)	Asthmatics refractory to β- agonist treatment
Restoration of the epithelial barrier function and integrity	Topical steroids111,131 (Type VII) AIT restoring the antiviral interferon response306 (Type VII) Cannabinoids (WIN55212-2) restoring rhinovirus-induced epithelial barrier disruption241 (Type VI)	Benefits to all asthma phenotypes as epithelial barrier function and integrity is a major protective factor
Epithelial derived cytokines	Anti TSLP and anti-IL-33 biologics7,99 (Type VII)	Non-eosinophilic asthma
Mucous secretion	Long-acting antimuscarinic agents254 (Type VI) Antibodies directed against key epitopes of the CLC crystallization interface146 (Type VII) Mucolytic approaches that involve disruption of disulfide bonds or weakening of non-covalent mucin interactions145 (Type VI) Gene or transcript directed therapies that target each mucin145 (Type V) Fucosyltransferase 2 inhibitors to inhibit mucin fucosylation145 (Type V) Blocking mechanical pathways using the Piezo 1 inhibitor gadolinium290, 291, 292 (Type VII)	High mucus burden (HRCT score) not responding to T2 -targeting biologics
Cytoskeletal remodeling and hyperplasia	Long-acting antimuscarinic agents254 (Type VI) Inhibition of the renin-angiotensin system215 (Type VI) Specific agonists of the olfactory receptors OR2AG1 and OR1D2288 (Type VII) TAS2R agonists (inhibition of the cell cycle progression by inhibiting phosphoinositide 3-kinase-mediated pro-mitogenic signaling in ASM cells)285 (Type VII)	Asthma with fixed airway obstruction or with accelerated lung function decline
Communication between local sensory neurons, immune cells, and resident lung stromal cells	Bronchial thermoplasty significantly reduced or interrupted the signaling between PNECs and other cells by interfering with their secretion and synapse formation246 (Type VI)	Asthma with fixed airway obstruction or with accelerated lung function decline
Corticosteroid resistance	Phosphoinositide 3-kinase-delta inhibitors that reestablish HDAC2 function (Type VII) p38 mitogen-activated protein kinase inhibitors decreasing GR phosphorylation304,305 (Type VII)	Asthma needing maintenance OCS

AIT, Allergen immunotherapy; AHR, Airway hyperreactivity; ASM, Airway smooth muscle; CLC, Charcot-Leyden crystals; GR, Glucocorticoid receptor; HDAC, Histone deacetylase; HRCT, High-resolution computed tomography; IL, Interleukin; ILC, Innate lymphoid cells; MCs, Mast cells; MRGPR, Mas-related G protein-coupled receptor; OCS, Oral corticosteroids; OR, Olfactory receptors; PNEC, Pulmonary neuro-endocrine cells; TAS2R, Type 2 taste receptor; TSLP, Thymic stromal lymphopoietin; TRP, Transient receptor potential.

Conclusion

The current stratified approach in asthma using a few available biomarkers measured at the point-of-care leads to a significant overlap between endotype classification. It also did not solve the significant incomplete response rate to biologicals targeting the T2 pathway. Machine-learning models combining precision immunology data with imaging biomarkers and deep phenotyping provide an unbiased novel view on asthma pathogenetic pathways and related biomarker signatures guiding personalized management of asthma patients. Digital health technology can impact asthma care by identifying and educating about environmental triggers, prompting earlier recognition of asthma symptoms, and improving medication adherence and inhaler device technique. The new EAACI nomenclature introducing the V, VI and VII hypersensitivity reactions provides the basis for defining new asthma theratypes.

However, there are still ongoing and future challenges we have to confront in asthma diagnosis and treatment, including the standardization of multi-omics for clinical use, the development of cost-effective point-of-care biomarker panels and therapies for non-T2 targets, and the integration of digital health data into the decision-making process. These underscore the need to prioritize the integration of diverse data streams into asthma theratypes-based clinical decision-support systems to achieve clinical remission of asthma,³⁰⁷ and ultimately improve patient outcomes.

CRediT authorship contribution statement

Ioana Agache: Writing – review & editing, Writing – original draft, Supervision, Project administration, Investigation, Conceptualization. Shengjie Li: Visualization, Project administration. Yang Zheng: Visualization, Project administration. Yadong Gao: Writing – review & editing, Visualization, Project administration.

Declaration of competing interest

The authors declare that they have no known competing financial interests or personal relationships that could have appeared to influence the work reported in this paper.

Edited by: Peifang Wei