Abstract Rationale Benralizumab, an anti-interleukin-5 receptor α monoclonal antibody, reduces eosinophilic inflammation in severe asthma, but its effects on lung structural remodeling are not fully understood. We assessed longitudinal bronchial and parenchymal changes using AI-driven quantitative CT analysis. Methods Patients with severe eosinophilic asthma underwent inspiratory and expiratory CT scans at baseline week 0 (n = 60), week 24 (n = 49), and week 48 (n = 52) following initiation of benralizumab. AI-driven quantitative analysis was performed using (1) LungQ-BA to measure bronchial widening (Bout/A and Bin/A) and wall thickening (Bwt/A and Bwa/Boa); (2) LungQ-MP to quantify mucus plug count and their volumes (mL); and (3) LungQ-VERA to assess air trapping (%) through low-attenuation region quantification. Results LungQ-MP analysis showed a significant reduction in mucus plugging following benralizumab treatment (mucus plugs p 0.001 and mucus plugs volumes p = 0.019) (Table 1; Figure 1). LungQ-BA demonstrated borderline reductions in bronchial wall thickening after intervention (Bwt/A, p = 0.036 and Bwa/Boa, p = 0.038), but no significant changes in bronchial widening (Bout/A, p = 0.051 and Bin/A, p = 0.086) (Table 1; Figure 2). LungQ-VERA analysis indicated no improvement in the percentage of air trapping (p = 0.500) (Table 1). Conclusions Benralizumab treatment was associated with a sustained reduction in mucus plugging and borderline improvement in bronchial wall thickening over 48 weeks, suggesting partial structural reversibility with anti-interleukin-5 receptor therapy. AI-driven quantitative CT analysis provides sensitive imaging biomarkers for quantifying lung structural responses in severe asthma and may complement clinical outcome measures in future studies. This abstract is funded by: CHU De Montpellier
Bourdin et al. (Fri,) studied this question.