Risk-Based Lung Cancer Detection in COPD Patients

Description

The goal of the interventional study is to evaluate whether risk-based stratification using the PLCOm2012 model and a machine learning (ML) model can identify patients with chronic obstructive pulmonary disease (COPD) who are at high risk of developing lung cancer and may benefit from low-dose computed tomography (LDCT) screening. The study population includes adults aged 50 years and older with COPD and a history of smoking attending an outpatient clinic. The main question it aims to answer are: \- What is the incidence of histopathologically confirmed lung cancer following risk-based stratification? Participants will: * Undergo lung cancer risk assessment using the PLCOm2012 model and an ML-based model based on clinical and laboratory data * Be referred for LDCT if classified as high-risk * Continue standard care if classified as low-risk * Be followed through electronic health records for up to six years to assess outcomes including lung cancer incidence, adherence to LDCT, time to imaging, healthcare utilization, costs, and safety

Interventions

• Risk-based prediction models (Other)
  Patients will be stratified into high-risk and low-risk groups using both the Lung Cancer Risk Prediction Calculator for smokers (PLCOm2012) and an in-house developed machine learning model based on sex, age, smoking status, and laboratory data from routine blood sample analyses.

Eligibility criteria

Conditions

Locations

• Vejle Hospital, University Hospital of Southern Denmark — Vejle, Region Syddanmark, Denmark