Risk-Based Lung Cancer Detection in COPD Patients

Lung cancer is a leading cause of cancer-related mortality in Denmark, largely due to late-stage diagnosis. Early detection through low-dose computed tomography (LDCT) has been shown to reduce mortality in trials such as the National Lung Screening Trial and the NELSON trial, but implementation remains limited due to cost, capacity, and risk of false-positive findings. Risk-based approaches may improve screening efficiency by identifying individuals at highest risk. This prospective, single-centre interventional study evaluates the feasibility and clinical implementation of lung cancer risk stratification in patients with chronic obstructive pulmonary disease (COPD). A total of 1,000 patients will be enrolled from an outpatient clinic. At baseline, participants will undergo risk assessment using the PLCOm2012 model and a machine learning (ML)-based model based on clinical and laboratory data. Patients identified as high-risk by either model will be referred for LDCT, while low-risk patients will continue standard care. If LDCT findings are suspicious, further diagnostic evaluation will be performed according to routine clinical practice. Participants will be followed through electronic health records for up to six years, with outcome assessment at 1, 2, and 6 years. The primary outcome is the incidence of histopathologically confirmed lung cancer. Secondary outcomes include adherence to LDCT referral, time from risk assessment to LDCT, number needed to screen, healthcare utilization, healthcare costs, and safety, including adverse events related to diagnostic procedures.