Phenotypic clusters on computed tomography reflects asthma heterogeneity and severity

Area of Science:

• Pulmonary Medicine
• Radiology
• Computational Imaging

Background:

• Asthma is a complex inflammatory airway disease with diverse phenotypes.
• Quantitative computed tomography (QCT) offers precise assessment of airway and lung structure for disease characterization.
• Radiological visual analysis aids in evaluating airway involvement and remodeling in asthma.

Purpose of the Study:

• To identify distinct asthma patient clusters using QCT-derived metrics of airway and parenchymal structure.
• To correlate QCT-based asthma clusters with visual assessments performed by radiologists.
• To evaluate the association between QCT-defined asthma clusters and clinical outcomes.

Main Methods:

• Prospective study involving 61 asthma patients with varying severity.
• QCT metrics analyzed included hydraulic diameter (Dh), luminal wall thickness (WT), functional small airway disease (fSAD), and emphysema (Emph).
• Cluster analysis of QCT metrics was compared with radiologist-based visual grouping.

Main Results:

• Four distinct asthma clusters (C1-C4) were identified, showing progressive lung function decline, increased fixed obstruction, and exacerbations from C1 to C4.
• Clusters exhibited specific structural and functional characteristics: C1 (non-severe, increased WT, proximal Dh, fSAD), C2 (mixed severity, proximal bronchodilator response), C3 (severe allergic, no fSAD), and C4 (ex-smokers, high fSAD, Emph).
• QCT clusters demonstrated strong correlation with radiologist groupings and predicted clinical outcomes.

Conclusions:

• Four QCT imaging-based clusters reveal distinct structural and functional airway changes in asthma.
• These QCT clusters effectively stratify heterogeneous asthma patients.
• QCT-based clustering serves as a valuable complementary tool for predicting clinical outcomes in asthma.