Background: Asthma is a biologically heterogeneous disease characterized by diverse inflammatory mechanisms, yet the clinical translation of this heterogeneity into clearly defined, biomarker-driven endotypes remains limited in real-world populations. Although Type-2 (T2) inflammation predominates in many patients, the overlap between allergic and eosinophilic pathways and the identification of distinct endotypes with specific clinical correlates are not fully characterized. This study aimed to define biomarker-based asthma endotypes using unsupervised clustering and to explore their associated clinical and demographic features. Methods: In this cross-sectional study, asthmatic patients evaluated at a tertiary hospital were included. Demographics, body mass index (BMI), comorbidities, age of onset, disease duration, spirometry parameters, bronchodilator reversibility, allergic status, and inflammatory biomarkers; total IgE, blood eosinophil count (BEC), and fractional exhaled nitric oxide (FeNO), were retrieved from medical records. Standardized biomarker values were analyzed using hierarchical clustering to define asthma endotypes. Clinical and demographic differences between clusters were subsequently compared. Results: Among 162 patients, females predominated (69.1%), and overweight/obesity was common (median BMI 29.15 kg/m 2 ). T2 inflammation was highly prevalent: 60% had IgE ≥ 100 IU/mL, 70% had BEC ≥ 300 cells/μL, and 43.8% had FeNO ≥ 25 ppb, with allergic rhinitis present in 71.6%. Airflow limitation was frequent, as 71% had FEV 1 < 80% predicted and 86.4% lacked bronchodilator reversibility. Clustering identified four endotypes: mixed allergic-eosinophilic T2-high (35.5%), eosinophilic T2-high (24.5%), allergic T2-high (20%), and T2-low (20%). T2-high clusters showed marked female predominance and greater lung function impairment, particularly in the eosinophilic group (81.5% with FEV 1 < 80%). Allergic T2-high patients had the earliest onset, whereas T2-low patients demonstrated minimal biomarker elevation and fewer allergic comorbidities. BEC correlated with IgE (r=0.29) and FeNO (r=0.43), and multivariable analysis linked biomarkers to specific clinical features. Conclusion: T2-high inflammation predominates in this cohort, comprising 80% of patients and segregating into clinically distinct allergic, eosinophilic, and mixed endotypes. Cluster-based stratification offers meaningful phenotyping that may support personalized biologic selection and precision management strategies. Keywords: asthma, phenotypes, endotypes, T2 high inflammation, T2 low inflammation
Feteih et al. (Fri,) studied this question.