Skip to content

Phenotypes, endotypes and treatable traits

TL;DR — A phenotype is an observable pattern; an endotype implies a causal biological pathway; a treatable trait is a measurable problem with an intervention. These categories overlap but are not interchangeable. Cluster analysis in the Severe Asthma Research Program separated early-onset allergic disease from later-onset, more obstructed phenotypes, yet clusters are population- and method-dependent (Moore 2010, PMID 19892860). Type-2 inflammation is the best validated endotypic axis because IgE-, IL-5/5R-, IL-4Rα- and TSLP-directed drugs reduce attacks in biomarker-enriched populations (Gans 2020, PMID 31678040). A randomized multidimensional severe-asthma trial showed that identifying and treating pulmonary, extrapulmonary and behavioral traits can improve health status, but the approach adds assessment burden (McDonald 2020, PMID 31806719). Classification should therefore serve a decision—confirm diagnosis, reduce an exposure, treat a comorbidity, choose therapy or measure response—not become a substitute for longitudinal reassessment.

Three different classification jobs

Concept Definition Example Failure mode
Phenotype Observable characteristic or cluster Early-onset allergic asthma; obesity-associated symptoms Describes co-occurrence without proving mechanism
Endotype Disease subtype defined by a causal pathway IL-5/eosinophil-dependent exacerbation biology Biomarker is mistaken for the pathway itself
Treatable trait Measurable, clinically relevant and modifiable feature Poor inhaler technique, eosinophilic inflammation, rhinitis Long checklists without evidence that treating each item improves outcomes (Fingleton 2018, PMID 29049049)

Severity and control are also distinct. Severity is inferred from the treatment required after optimization; control describes current symptoms and future risk. A patient can have few daily symptoms but high attack risk, or frequent symptoms driven by inducible laryngeal obstruction despite little airway inflammation.

Why classic severity labels were insufficient

Intermittent/mild/moderate/severe labels bundle symptom frequency, airflow, treatment and risk. They do not reliably predict inflammatory mechanism or response to targeted therapy (Padem 2019, PMID 31690376).

The molecular era added type-2 biomarkers and biologic-response evidence, while the treatable-traits framework decomposed difficult disease into addressable components. Neither eliminates the need for conventional diagnosis, spirometry and attack history.

Data-driven clusters

SARP applied unsupervised clustering to clinical variables and identified five phenotypes spanning early-onset allergic asthma with preserved function through later-onset disease with greater obstruction and health-care use (Moore 2010, PMID 19892860).

U-BIOPRED later identified adult severe-asthma clusters and linked some to sputum omics, moving from clinical phenotype toward molecular endotype (Lefaudeux 2017, PMID 27773852).

Lesson from clustering Implication
Cluster number changes with variables, distance metric and population A cluster is not automatically a natural disease entity
Severe cohorts enrich oral-steroid exposure, fixed obstruction and comorbidity Cluster prevalence does not transport to primary care
Cross-sectional clusters can change over time Longitudinal stability must be demonstrated
Omics can subdivide a clinical cluster Similar symptoms can arise from different pathways
Treatment response is the strongest validation A cluster without differential prognosis or response may have little utility

U-BIOPRED sputum molecular phenotypes were not completely stable, underscoring that infection, corticosteroids and sampling time can move patients between categories (Kermani 2021, PMID 33008937).

Type-2-high disease

Type-2-high asthma includes allergic and non-allergic routes converging on eosinophils, IL-4/IL-13 signaling, IgE and epithelial alarmins.

Observable marker/feature Pathway information Main use Main limitation
Blood eosinophils Systemic eosinophil pool Attack risk and anti-IL-5/5R eligibility Suppressed by systemic steroids; variable over time
FeNO Airway epithelial IL-4/IL-13 signaling ICS response/adherence and dupilumab/tezepelumab response probability Atopy, smoking and current ICS alter values
Sputum eosinophils Luminal airway eosinophilia Specialist phenotype and inflammation-guided care Labor-intensive and not widely available
Allergy testing/IgE Sensitization and IgE pathway Omalizumab eligibility plus exposure interpretation Sensitization is not proof an exposure drives symptoms
Nasal polyps/AERD Eosinophilic upper-airway multimorbidity Raises probability of type-2 response A clinical enrichment feature, not a single biomarker

Thresholds are administrative and trial-derived; biology is continuous. Combining eosinophils and FeNO identifies risk that either marker alone misses (Meulmeester 2025, PMID 40215991).

The same patient may satisfy criteria for several biologics. A live PubMed and ClinicalTrials.gov re-query on 2026-08-30 found indirect comparisons and limited direct comparisons but did not establish a validated head-to-head selection algorithm for most overlaps. Attack history, biomarker magnitude, steroid dependence, age, allergy, nasal polyps, atopic dermatitis, dosing preference and prior response therefore all contribute.

Allergic and early-onset asthma

Allergic asthma usually begins earlier, clusters with rhinitis/eczema and links symptoms to sensitization plus exposure. IgE-mediated mast-cell activation is causal in a treatment-responsive subset, as shown by omalizumab trials (Humbert 2005, PMID 15679715).

Atopy is neither necessary nor sufficient for asthma. A positive skin test without exposure-linked disease should not be allowed to override objective airway findings.

Eosinophilic late-onset asthma

Late-onset eosinophilic disease often has fewer classic childhood-allergy features, more nasal polyps, systemic-steroid dependence and persistent obstruction. Anti-IL-5/5R trials validate eosinophil biology for attacks and steroid reduction, but eosinophils can coexist with allergy or obesity (Pavord 2012, PMID 22901886; Bleecker 2016, PMID 27609408).

Age at onset is a clinical phenotype, not an endotype. Recall is imperfect and a late diagnosis may represent relapsed or previously unrecognized childhood disease.

Type-2-low, neutrophilic and paucigranulocytic asthma

Type-2-low is a negative operational category: currently measured eosinophils and FeNO are not elevated. It can include steroid-suppressed type-2 disease, smoking/infection-associated neutrophilia, obesity-related mechanics and paucigranulocytic airway dysfunction (Hudey 2020, PMID 33160187).

Proposed subtype Evidence Treatment implication
Neutrophilic Sputum neutrophils, IL-17/NET/inflammasome signals; U-BIOPRED multi-omics identifies heterogeneity even within neutrophilic severe disease (Kermani 2024, PMID 39073027) Check infection, smoking and steroid exposure; macrolide evidence is not a universal neutrophil-targeted therapy
Paucigranulocytic Airflow/AHR without elevated sputum granulocytes Consider smooth-muscle, neural, adherence and mimic pathways (Tliba 2019, PMID 29928921)
Low biomarkers after steroids Treatment effect can create the label Repeat measures and review steroid exposure before declaring endotype

Broad non-type-2 drug development has been less successful than type-2 biologics, suggesting that more granular endotyping is needed.

Obesity-associated asthma

Obesity can increase dyspnea through mechanics, deconditioning, reflux and sleep-disordered breathing, while also modifying airway inflammation and treatment response. It does not define one phenotype (Peters 2018, PMID 29627041).

Obesity-associated asthma can be type-2-high or type-2-low. Assuming all obesity-related symptoms are non-inflammatory risks undertreating eosinophilic disease; assuming all dyspnea is asthma risks escalating inhaled therapy for mechanics or comorbidity.

Weight-loss intervention reviews support improvements in symptoms and control for some adults, but studies vary in intensity and magnitude of weight loss; airway endpoints do not always change with symptoms (Watchorn 2019, PMID 30969796).

Aspirin/NSAID-exacerbated respiratory disease

AERD/N-ERD comprises asthma, chronic rhinosinusitis with nasal polyps and respiratory reactions to cyclooxygenase-1 inhibitors. It involves dysregulated arachidonic-acid mediator biology and is frequently eosinophilic (Kowalski 2019, PMID 30216468).

Decision Evidence implication
Diagnosis A clear repeated reaction history may suffice; supervised challenge is the reference when uncertain and safe
Avoidance Nonselective COX-1 NSAIDs can provoke severe reactions; selective alternatives require individualized assessment
Aspirin desensitization Can improve upper/lower airway outcomes in selected patients but requires maintenance dosing and safety infrastructure (Stevens 2021, PMID 33307116)
Biologics Type-2 biologics can improve asthma and nasal-polyps outcomes; head-to-head sequencing with desensitization remains limited

Population prevalence estimates depend strongly on challenge versus self-report. AERD should not be inferred from asthma plus nasal polyps without a drug-reaction history.

Fungal sensitization and ABPA

Fungal sensitization ranges from coincidental atopy through severe asthma with fungal sensitization to allergic bronchopulmonary aspergillosis (ABPA), which adds high IgE/eosinophilic responses, radiographic mucus impaction and often bronchiectasis (Agarwal 2013, PMID 23889240).

ABPA is not simply “severe fungal asthma.” Diagnostic criteria integrate asthma or another predisposing airway disease, Aspergillus sensitization, total IgE and supportive eosinophil/imaging findings; thresholds and criteria sets have evolved (Moss 2021, PMID 33529689).

Missing ABPA risks progressive bronchiectasis; overcalling it exposes patients to systemic steroids or antifungals without adequate evidence.

Exercise-induced bronchoconstriction can occur with or without chronic asthma. Objective post-exercise airflow change is more specific than exercise symptoms, which can reflect inducible laryngeal obstruction, deconditioning or cardiac limitation.

Cough-variant asthma is a phenotype of cough plus airway hyperresponsiveness/variable physiology and treatment response; eosinophilic bronchitis causes steroid-responsive cough without variable obstruction. The distinction prevents using response to empiric therapy as the sole diagnostic test.

Treatable traits framework

Traits are grouped across domains:

Pulmonary Extrapulmonary Behavioral/environmental
Eosinophilic inflammation Rhinitis/nasal polyps Inhaler technique
Variable airflow/AHR Obesity Adherence
Fixed obstruction Reflux when clinically relevant Smoking/vaping
Mucus plugging Sleep apnea Occupational/allergen exposure
Infection/bronchiectasis Anxiety/depression Low health literacy/cost barriers
Inducible laryngeal obstruction Dysfunctional breathing No written action plan

In a randomized severe-asthma trial, a multidimensional assessment identified traits and informed individualized treatment; the intervention improved quality-of-life/health-status outcomes compared with usual care (McDonald 2020, PMID 31806719).

U-BIOPRED analysis found many traits per severe-asthma patient, with trait burden associated with worse outcomes, but observational association does not prove benefit from treating every identified item (Simpson 2019, PMID 30307629).

The approach is most defensible when each trait has four elements: a reliable measure, clinical relevance, an available intervention and a predefined response assessment.

Remission and super-response

Biologic-era “clinical remission” definitions commonly combine no exacerbations, no maintenance systemic steroids, controlled symptoms and stable/improved lung function over a defined interval. Definitions vary enough that rates should not be pooled without harmonization (Menzies-Gow 2020, PMID 31866436).

Systematic review of biologic-treated severe asthma found heterogeneous remission definitions and correlates, so remission currently functions more as a target construct than a single validated endpoint (Shackleford 2025, PMID 39549709).

Remission on treatment does not necessarily mean cure: airway hyperresponsiveness, remodeling or relapse risk may persist.

Open questions

  • Which clusters are stable across geography, ancestry, treatment and time, and which merely summarize a dataset? (Moore 2010, PMID 19892860; Kermani 2021, PMID 33008937)
  • What prospective algorithm chooses among overlapping biologic eligibilities better than clinician judgment?
  • Can a type-2-low molecular classifier predict response to a specific therapy? (Hudey 2020, PMID 33160187)
  • Which treatable-trait bundle improves attacks, steroid burden and cost—not only quality of life—and in which care setting? (McDonald 2020, PMID 31806719)
  • Does biologic-associated clinical remission persist after withdrawal, and what residual biology predicts relapse? (Shackleford 2025, PMID 39549709)

References

  1. Moore WC, et al. Identification of asthma phenotypes using cluster analysis in the Severe Asthma Research Program. Am J Respir Crit Care Med. 2010;181:315-323. PMID 19892860
  2. Gans MD, Gavrilova T. Understanding the immunology of asthma: pathophysiology, biomarkers, and treatments for asthma endotypes. Med Clin North Am. 2020;104:1-14. PMID 31678040
  3. McDonald VM, et al. Targeting treatable traits in severe asthma: a randomised controlled trial. Eur Respir J. 2020. PMID 31806719
  4. Fingleton J, et al. Treatable traits of chronic airways disease. Eur Respir J. 2018. PMID 29049049
  5. Padem N, Saltoun C. Classification of asthma. Allergy Asthma Proc. 2019;40:385-388. PMID 31690376
  6. Lefaudeux D, et al. U-BIOPRED clinical adult asthma clusters linked to a subset of sputum omics. J Allergy Clin Immunol. 2017;139:1797-1807. PMID 27773852
  7. Kermani NZ, et al. Instability of sputum molecular phenotypes in U-BIOPRED severe asthma. Eur Respir J. 2021. PMID 33008937
  8. Meulmeester FL, et al. Inflammatory and clinical risk factors for asthma attacks (ORACLE2). Lancet Respir Med. 2025. PMID 40215991
  9. Humbert M, et al. Benefits of omalizumab as add-on therapy in severe persistent allergic asthma: INNOVATE. Allergy. 2005;60:309-316. PMID 15679715
  10. Pavord ID, et al. Mepolizumab for severe eosinophilic asthma (DREAM). Lancet. 2012;380:651-659. PMID 22901886
  11. Bleecker ER, et al. Benralizumab for severe asthma: SIROCCO. Lancet. 2016;388:2115-2127. PMID 27609408
  12. Hudey SN, et al. Mechanisms of non-type 2 asthma. Curr Opin Immunol. 2020;66:123-128. PMID 33160187
  13. Kermani NZ, et al. Endotypes of severe neutrophilic and eosinophilic asthma from multi-omics integration of U-BIOPRED sputum samples. Thorax. 2024. PMID 39073027
  14. Tliba O, Panettieri RA. Paucigranulocytic asthma: uncoupling airway obstruction from inflammation. J Allergy Clin Immunol. 2019;143:1287-1294. PMID 29928921
  15. Peters U, et al. Obesity and asthma. J Allergy Clin Immunol. 2018;141:1169-1179. PMID 29627041
  16. Watchorn DC, et al. Weight loss in individuals with obesity and asthma. Ann Am Thorac Soc. 2019. PMID 30969796
  17. Kowalski ML, et al. Diagnosis and management of NSAID-exacerbated respiratory disease: EAACI position paper. Allergy. 2019;74:28-39. PMID 30216468
  18. Stevens WW, et al. Aspirin desensitization followed by oral aspirin therapy in AERD: AAAAI work group report. J Allergy Clin Immunol. 2021. PMID 33307116
  19. Agarwal R, et al. Allergic bronchopulmonary aspergillosis: review and proposed diagnostic/classification criteria. Clin Exp Allergy. 2013;43:850-873. PMID 23889240
  20. Moss RB. Diagnosing allergic bronchopulmonary aspergillosis/mycosis: return to lost horizons. Mycopathologia. 2021. PMID 33529689
  21. Simpson AJ, et al. Treatable traits in the European U-BIOPRED adult asthma cohorts. Allergy. 2019;74:406-411. PMID 30307629
  22. Menzies-Gow A, et al. An expert consensus framework for asthma remission as a treatment goal. J Allergy Clin Immunol. 2020. PMID 31866436
  23. Shackleford S, et al. Clinical remission attainment, definitions, and correlates among patients with severe asthma treated with biologics: systematic review and meta-analysis. Eur Respir Rev. 2025. PMID 39549709
  24. Kuruvilla ME, et al. Understanding asthma phenotypes, endotypes, and mechanisms of disease. Clin Rev Allergy Immunol. 2019;56:219-233. PMID 30206782
  25. Nelson RK, et al. Eosinophilic asthma. J Allergy Clin Immunol Pract. 2020;8:465-473. PMID 31786254
  26. Kennedy JL, Stoner AN, Borish L. Aspirin-exacerbated respiratory disease: prevalence, diagnosis, treatment, and considerations for the future. Am J Rhinol Allergy. 2016;30:407-413. PMID 28124651
  27. Zhang C, Jiang Z, Shao C. Clinical characteristics of allergic bronchopulmonary aspergillosis. Clin Respir J. 2020;14:440-446. PMID 31901186
  28. Hiles SA, et al. Treatable traits that predict health status and treatment response in airway disease. J Allergy Clin Immunol Pract. 2021;9:1255-1264.e2. PMID 33039647