Targeting Airway Remodeling in Severe Asthma: Is There a Window of Opportunity for Biologic Therapy Predicting Effects Using Causal Artificial Intelligence?

Airway remodeling is increasingly recognized as a major determinant of asthma progression, fixed airflow limitation, and long-term morbidity, particularly in severe disease. Although biologic therapies have transformed outcomes by reducing exacerbations and systemic corticosteroid exposure, their potential to modify structural airway trajectories-and whether a time-sensitive "window of opportunity" exists-remains uncertain. Here, we provide a progressive landscape integrating mechanistic remodeling pathways with measurable structural readouts (biopsy-derived indices and quantitative imaging) and emerging digital biomarkers derived from connected respiratory technologies. We propose an operational framework linking mechanism → biomarker → remodeling readout → timing decision, and we outline a 1-, 3-, and 5-year research roadmap in which advanced artificial intelligence (AI) methods (multimodal learning, causal inference, federated learning, and digital-twin architectures) evolve in parallel with wearable and smart-inhaler ecosystems. This landscape aims to standardize endpoints, sharpen trial design, and accelerate a shift from symptom control toward credible disease modification in severe asthma.