Integrating Quantitative CT Biomarkers to Enhance COPD Detection in the HANSE Lung Cancer Screening Program.
Authors
Affiliations (11)
Affiliations (11)
- LungenClinic Großhansdorf, Airway Research Center North (ARCN), German Center for Lung Research (DZL), Großhansdorf, Germany; Internal Medicine Department I, University Medical Center Schleswig-Holstein, Campus Kiel, Kiel, Germany. Electronic address: [email protected].
- Department of Medicine, Pulmonary and Critical Care Medicine, Clinic for Airway Infections, University Medical Centre Marburg, Philipps-University Marburg, Marburg, Germany; Institute for Lung Research, Universities of Giessen and Marburg Lung Centre, Philipps-University Marburg, Marburg, Germany.
- LungenClinic Großhansdorf, Airway Research Center North (ARCN), German Center for Lung Research (DZL), Großhansdorf, Germany.
- Institute of Medical Informatics, University of Münster, Münster, Germany.
- Department of Respiratory Medicine and Infectious Diseases, Hannover, Germany; Biomedical Research in End-Stage and Obstructive Lung Disease Hannover (BREATH), Hannover, Germany.
- Department of Respiratory Medicine, University Medical Center Schleswig-Holstein, Lübeck, Germany.
- Department of Radiology, University Medical Center Schleswig-Holstein, Lübeck, Germany.
- Biomedical Research in End-Stage and Obstructive Lung Disease Hannover (BREATH), Hannover, Germany; Department of Diagnostic and Interventional Radiology, Hannover Medical School (MHH), Hannover, Germany.
- Medical Faculty Mannheim, Heidelberg University, Heidelberg, Germany.
- LungenClinic Großhansdorf, Airway Research Center North (ARCN), German Center for Lung Research (DZL), Großhansdorf, Germany; Department of Respiratory Medicine, University Medical Center Schleswig-Holstein, Lübeck, Germany.
- Biomedical Research in End-Stage and Obstructive Lung Disease Hannover (BREATH), Hannover, Germany; Department of Diagnostic and Interventional Radiology, Hannover Medical School (MHH), Hannover, Germany; Department of Radiology, Charité - Universitätsmedizin Berlin, Corporate Member of Freie Universität Berlin, Humboldt-Universität, Berlin, Germany.
Abstract
Lung cancer screening offers an opportunity to enhance COPD detection among adults exposed to tobacco smoke, yet guidance on CT-based referral for spirometry is limited. Which low-dose CT emphysema threshold best identifies previously undiagnosed COPD in lung cancer screening subjects, and does the addition of quantitative airway biomarkers enhance detection? In adults undergoing lung cancer screening, we performed spirometry to identify previously undiagnosed COPD, defined by airflow obstruction (FEV1/FVC <0.70) in current and former smokers with ≥10 pack-years. We quantified emphysema extent, airway wall thickness (Pi10), and airway branch count on low-dose CT using AI-based software and combined these measures with clinical characteristics, mainly smoking history and dyspnea, to develop an ensemble tree-based machine-learning model (Extreme Gradient Boosting) for COPD detection. A sensitivity analysis using the lower limit of normal (LLN) definition for FEV1/FVC was additionally performed. Among 5,014 screening subjects with available spirometry, 1,115 had previously undiagnosed COPD, corresponding to a prevalence of 22.2%. In subjects without known airway disease, emphysema alone at an optimised threshold of 5.1% showed moderate performance for COPD detection (AUC 0.69 [95% CI 0.67-0.72], accuracy 66%, PPV 44%), where 41% of subjects exceeded this threshold and met criteria for confirmatory spirometry. An integrated model combining emphysema, Pi10, airway branch count, and clinical characteristics significantly improved detection (AUC 0.83 [95% CI 0.80-0.86], accuracy 78%, PPV 59%) while reducing the proportion requiring confirmatory spirometry to 34%. In a sensitivity analysis using the LLN definition of COPD, the best-performing model achieved an AUC of 0.86 (95% CI 0.83-0.89), accuracy of 84%, and PPV of 53%, while referring 22% for confirmatory spirometry. An integrated approach combining CT-derived airway biomarkers and clinical characteristics enables more efficient and targeted COPD detection within lung cancer screening programs.