Multivariate Cluster Analyses to Characterize Asthma Heterogeneity and Benralizumab Responsiveness

Xingnan Li,Paul Newbold,Rohit Katial,Ian Hirsch,Huashi Li,Ubaldo J. Martin,Deborah A. Meyers,Eugene R. Bleecker

DOI: https://doi.org/10.1016/j.jaip.2024.04.026

IF: 11.022

2024-04-25

The Journal of Allergy and Clinical Immunology In Practice

Abstract:Background An improved understanding of how severe asthma heterogeneity affects response could inform treatment decisions. Objectives Characterize heterogeneity and benralizumab responsiveness in patients grouped by predefined Severe Asthma Research Program clusters using a multivariate approach. Methods In post-hoc analyses of the randomized, double-blind, placebo-controlled phase III SIROCCO (NCT01928771) and CALIMA (NCT01914757) studies, patients with severe asthma who received benralizumab or placebo were assigned to clusters using an established discriminant function to simultaneously analyze 11 clinical characteristics. Annualized asthma exacerbation rate, exacerbation incidence, and lung function were analyzed across clusters. Results Patients (N = 2,281) met criteria for 4 of 5 clusters: Cluster 2 (early-onset moderate asthma, n = 393), Cluster 4 (early-onset severe, n = 386), Cluster 3 (late-onset severe, n = 641), and Cluster 5 (late-onset severe, obstructed, n = 861); no patients met Cluster 1 criteria. Exacerbation rate reductions were significant in late-onset severe (−48% [95% CI: –61%, –31%], P <.0001) and late-onset severe, obstructed asthma (−50% [95% CI: –59%, –38%], P <.0001), with non-significant reductions in early-onset clusters. These differences could not be fully explained by blood eosinophil count differences. Forced expiratory volume in 1 second improvements were significant in late-onset severe (+133 mL [95% CI: 66 mL, 200 mL], P =.0001) and late-onset severe, obstructed asthma (+160 mL [95% CI: 85 mL, 235 mL], P <.0001) while maintaining acute bronchodilator responsiveness. Conclusions Benralizumab reduced exacerbations and improved lung function, primarily in late-onset asthma clusters. This multivariate approach to identify subphenotypes, potentially reflecting pathobiological mechanisms, can guide therapy beyond univariate approaches.

immunology,allergy