Hendrik Koffijberg

Ph.D.

University of Twente

Associate Professor and Section Chairperson

NLST (Learn more about this study)

NLST-852

Nov 15, 2021

Health economic evaluation for lung cancer screening alone or combined with screening for emphysema and coronary calcium scoring

There is increasing evidence for the benefit of low-dose CT (LDCT) screening to detect early lung cancer. However, despite these studies focusing on lung cancer (LC) alone, simultaneous screening for additional diseases could increase efficiency and screening yield, particularly for diseases with an indolent start and shared risk factors. Chest LDCT, used in LC screening, can simultaneously detect early stages of chronic obstructive pulmonary disease (COPD) through emphysema or bronchial wall evaluation and cardiovascular disease (CVD) risk from coronary calcium scoring. Both COPD and CVD pose a large burden on Western societies and share risk factors with lung cancer, such as age and smoking history. Early diagnosis or the identification of biomarkers indicating high risk of these three diseases, followed by early action can delay or stop disease progression, or provide a cure, depending on the specific disease diagnosed. Expanding LC screening to simultaneously screen for COPD or CVD may, therefore, further improve health outcomes of a screening program at marginal additional costs.

It is currently unknown whether screening for COPD or CVD is effective and whether combination screening can be cost-effective. Since long term screening impact cannot be measured, simulation modelling is necessary to estimate if combination screening could offer an attractive alternative to screening for LC only. In this project, the NLST data is requested to inform an individual-level simulation model to determine the long-term effects of such a combination screening program. The NLST data is necessary to fit probability distributions which could determine the probabilities of events occurring, such as adherence to screening, confirmed diagnoses (for one or multiple diseases per person), treatment followed and complications of treatment per individual. These probabilities could also differ based on individual patient characteristics, which can be reflected using regression models, for which demographic data is needed. It is necessary to assess how many individuals (in subgroups determined by participant data) are diagnosed with lung cancer, COPD, coronary artery calcium, or a combination of those, as well as how many and which individuals were not diagnosed with these conditions. This will be used to populate the model with incidence and prevalence of these diseases in subgroups, the true/false positive/negative test outcomes, and the disease severity in individuals diagnosed with the disease(s). Additionally, data regarding the progression of lung cancer for patients diagnosed with COPD will be incorporated in the estimation of the long term effects of screening.
Clinical trials such as NELSON and NELCINB3 have not opened data requests or have not yet gathered all data. Therefore, the NLST currently has the largest available dataset which can be used to inform this modelling study with individual-level data concerning a possible lung cancer screening program. The NLST data is crucial for the successful completion of the project funded by ZonMW in the Netherlands.