An update of the clinical and cost-effectiveness of lung cancer screening by low-dose CT
The two main components of the project are to update the systematic review and meta-analysis of randomized controlled trials (RCTs) of lung cancer screening with imaging technology (including LDCT and chest X-ray), and the economic evaluation (cost-utility analysis) of LDCT screening programs which utilized a Discrete Event Simulation (DES) model.
The DES model underlying the cost-utility analysis required a well-founded model of the natural history of lung cancer, and the ability of LDCT to detect lung cancer at early stages, achieve a stage shift and a mortality benefit. The National Lung Screening Trial (NLST) was used in the original evaluation (Snowsill et al. 2018) to calibrate the natural history model of lung cancer, which underpinned the DES model (Project ID. NLST-301).
To validate the natural history and effectiveness model calibration based on NLST data, using Bayesian Markov Chain Monte Carlo (MCMC) techniques, and update this model with evidence which has subsequently become available since the model was developed in 2017. This natural history and effectiveness model will then underpin the updated DES model for the cost-utility analysis.
Specifically:
To validate parameters for preclinical lung cancer incidence, indolent lung cancer development, preclinical lung cancer progression, clinical presentation and lung cancer mortality (the *natural history* model) obtained from the original natural history model.
To validate the sensitivity of LDCT- and chest X-ray-based screening programs, and any impact on stage-specific survival, in light of the estimated rate of indolent lung cancer development and the observed 20% reduction in lung cancer mortality (the *effectiveness* model).
Chris Hyde, University of Exeter
Jaime Peters. University of Exeter
(Others are involved in the project, but will not be involved with validation of model calibration)