Severe Asthma: Harnessing Biomarkers to Power Treatment Effectiveness

Asthma has been described as an “epidemic of dysregulated immunity.”¹ In particular, severe asthma that remains uncontrolled despite treatment with systematic corticosteroids poses a high risk for morbidity and mortality. The incidence of asthma-related mortality in the United States is 10.3 million, ranging from 3 million for children under the age of 18 years to 13.7 million for adults.² Dr Nicola Davies explores how phenotypes, endotypes, and biomarkers can be used to predict and assess treatment effectiveness for severe asthma.

Phenotypes and Endotypes

Severe asthma is a complex, heterogeneous set of pathologies in which multiple phenotypes and endotypes can be identified by various diagnostic tests. Phenotypes, which are the observable characteristics of a disease, include early onset obesity-exacerbated asthma; exacerbation-prone asthma; neutrophilic asthma; obesity-induced, non-neutrophilic asthma; paucigranulocytic asthma; and asthma with smoking. Endotypes, which are the specific biological mechanisms that cause the observed properties of any given disease phenotype, include early onset allergic asthma, aspirin-exacerbated respiratory disease, allergic bronchopulmonary mycosis, and severe late-onset hypereosinophilic asthma.³

Given that severe asthma is so complex, identifying phenotypes and endotypes with precision is critical for the selection of targeted therapeutic interventions. The current classification of severe asthma is largely based on the common characteristics identified as part of the Severe Asthma Research Program, as well as by guidelines set out by the American Thoracic Society Task Force, the European Respiratory Society, and the Global Initiative for Asthma. Yet, how reliable are these phenotypes and endotypes for classifying severe asthma?

Dr Tara Carr, associate professor in the Asthma and Airway Disease Research Center at The University of Arizona, says, “These phenotypes seem to cluster in individuals in whom other characteristics, such as lung function, risk of exacerbations, health care utilization, and sputum cellular characteristics are also similar.” The phenotypic clusters described by Dr Carr were initially reported within the National Heart, Lung, and Blood Institute’s Severe Asthma Research Program cohort.⁴ “Amazingly, these clusters have been almost exactly recapitulated in other asthma cohorts around the world,” she adds. The clusters include early onset allergic (mild, moderate, and severe), late-onset eosinophilic at the severe end of the spectrum, and a severe group with very low lung function that is incompletely reversible with steroid or bronchodilator treatments. Dr Carr also emphasizes that, “Other phenotypes that may overlap with these, or may be separately distinguished, include obesity-related, neutrophilic, and paucigranulocytic asthma.”

Biomarkers

Although the clinically defined phenotypic clusters of severe asthma appear to be reproducible and relatively reliable, there remains a need to further develop the use of diagnostic biomarkers to identify phenotypes, endotypes, and potential therapeutic pathways with greater precision. In the context of severe asthma, biomarkers can be used as indicators of disease processes and to characterize responses to therapeutic interventions.

Diagnostic Biomarkers

According to Dr Carr, “Aspirin-exacerbated respiratory disease is usually identified based on the clinical history of symptoms with nonsteroidal anti-inflammatory exposure, but biomarkers of arachidonic acid metabolism, such as urinary leukotriene, may help to identify these patients.” She adds that, “Allergic bronchopulmonary aspergillosis can be identified using a panel of serologic and imaging criteria.”

It is possible to identify other phenotypes through a combination of clinical history and measurements of serum immunoglobulin E (IgE) and blood eosinophils. Neutrophilic and paucigranulocytic subtypes can be identified by airway and sputum cell counts. Novel biomarkers for eosinophil activity, such as urinary bromotyrosine and eosinophil peroxidase, may help to further identify endotypes in which eosinophils play an active role in disease. Dr Carr shares that many more biomarkers are being developed, such as club cell secretory protein-16, which she says can “predict lung function decline and insulin, which may be able to identify highly inflammatory asthma.”

There are several biomarkers currently in use that can assist in the identification of the phenotypes and endotypes of severe asthma (Table 1).

Table 1. Biomarkers used to identify the phenotypes and endotypes of severe asthma (adapted from Carr et al, 2018⁵)

Phenotype or endotype *Endotypes in bold	Biomarkers
Early onset allergic asthma	Aeroallergen-specific IgE Elevated fractional exhaled nitric oxide Eosinophilia
Early onset obesity-exacerbated asthma	Aeroallergen-specific IgE Eosinophilia
Aspirin-exacerbated respiratory disease	Eosinophilia Leukotrienes
Allergic bronchopulmonary mycosis	Markedly elevated total IgE Mold-specific IgE Eosinophilia
Severe late-onset hypereosinophilic asthma	Eosinophilia
Exacerbation-prone asthma	Eosinophilia
Neutrophilic asthma	Elevated neutrophils in blood and sputum
Obesity-induced, non-neutrophilic asthma	Lack of Th 2 biomarkers Elevated IL-6 Elevated leptin
Paucigranulocytic asthma	Lack of airway inflammation
Asthma with smoking	Less eosinophilic, more neutrophilic

FeNO, fractional exhaled nitric oxide; IgE, immunoglobulin E.

Predictive Biomarkers:

In addition to profiling phenotypes and endotypes, biomarkers have potential as predictive tools for determining appropriate therapeutic pathways. The currently available biologic therapies for severe asthma were approved by the Food and Drug Administration based on total and specific IgE (omalizumab) or blood eosinophils (mepolizumab, reslizumab, benralizumab, and dupilumab). All therapies were shown in their respective trials to preferentially benefit individuals with higher blood eosinophil levels. Eosinophils and IgE are, therefore, used by some healthcare providers to select an appropriate biologic for treatment (Table 2). However, Dr Carr notes, “For many patients, both IgE and eosinophils will be present in blood at slightly or markedly elevated levels, and for these patients the pathway to treatment is less clear.”

Table 2. Biomarkers for severe asthma and associated therapies (adapted from Carr and Kraft, 2018³)

Biomarker	Type	Associated cytokines	Associated biologic agent
IgE	Predictive Dynamic	IL-4, IL-13	omalizumab
Eosinophils	Diagnostic Predictive Dynamic	IL-5	mepolizumab, reslizumab, benralizumab
IL-4, IL-13	dupilumab
Neutrophils	Diagnostic
Exhaled nitric oxide	Predictive Dynamic
Eosinophil peroxidase	Diagnostic Predictive Dynamic
Bromotyrosine	Diagnostic Predictive Dynamic
Stem cell factor/KIT	Predictive
Urinary leukotriene E4	Diagnostic Predictive Dynamic
Periostin	Predictive	IL-4, IL-13	lebrikizumab, tralokinumab, omalizumab
Single nucleotide polymorphisms	Predictive Stable

IL, interleukin; KIT, KIT proto-oncogene receptor tyrosine kinase.

*Note: Diagnostic biomarkers can identify the type of disease that is present. Predictive biomarkers can be used to determine differential responses to therapeutic interventions. Dynamic biomarkers can be used to track disease progression. Stable biomarkers can be used to predict responses to, or harm caused by, different therapeutic agents.

Some healthcare providers are looking to the Severe Asthma Research Program clusters to decipher phenotype more precisely in patients. This may result in the selection of anti-IgE therapy for allergic asthma or anti-eosinophil therapy for later onset hypereosinophilic asthma. Other biomarkers, such as the fraction of exhaled nitric oxide, may also be predictive for patients who would benefit from these medications. However, as Dr Carr highlights, “There is still work to be done toward improving these treatment decision pathways and toward identifying biomarkers that will help with choice of therapy when biologics targeted at different pathways or non-eosinophilic inflammation become available.”

Where Next?

Phenotypic clusters for severe asthma are now well established. The use of biomarkers to diagnose severe asthma and to predict differential therapeutic interventions is, however, still evolving. There remains a need to better define specific endotypes of disease and to develop further diagnostic and predictive biomarkers for these endotypes. This will improve the specificity of therapeutic interventions and, ultimately, patient outcomes.

References

1. Umetsu DT, McIntire JJ, Akbari O, Macaubus C, DeKruyff RH. Asthma: an epidemic of dysregulated immunity. Nat Immunol. 2002;3(8):715-720.

2. Centers for Disease Control and Prevention. 2015 Archived National Asthma Data. https://www.cdc.gov/asthma/archivedata/2015/2015_data.html. Updated May 2018. Accessed May 7, 2019.

3. Carr TF, Kraft M. Use of biomarkers to identify phenotypes and endotypes of severe asthma. Ann Allergy, Asthma Immunol. 2018;121(4):414-420.

4. National Institutes of Health/National Heart, Lung & Blood Institute. Severe Asthma Research Program. http://www.severeasthma.org/. Accessed May 7, 2019.

5. Carr TF, Zeki AA, Kraft M. Eosinophilic and noneosinophilic asthma. Am J Respir Crit Care Med. 2018;197:22-37.