Abstract:
HTML
Purpose: To estimate the stage-shift and mortality reduction due to CT lung cancer screening.
Methods and Materials: A parametric natural history model is developed to estimate the size at which lung cancer transitions from local to regional and from regional to distant stages. The model assumes exponential growth from the smallest screen detectable tumor size and volume-dependent hazards for clinical detection and stage transitions. Maximum likelihood estimates are obtained based on observable tumor sizes and stages of clinically detected lung cancer patients. The natural history model is embedded into a larger Monte-Carlo simulation model that generates the individual life histories of current and former smokers undergoing CT screening. The screening simulator assumes that the detection threshold for CT is 5mm and the mean doubling time of lung cancer is 6 months.
Results: SEER data on lung cancer cases detected after 1988 were used to derive parameter estimates for the natural history model, separately for males and females. The model reproduces the data well for larger tumors. Smaller tumors are clinically detected in more advanced stages than predicted by the model. The natural history lung cancer model was embedded into a Monte Carlo simulation of 100,000 60 year old current and former US smokers ( 55% men, 45% women) undergoing annual CT lung cancer screening for 20 years. The stage distribution of the screened population was estimated as 64%, 18% and 18% for local, regional and distant disease, respectively. In the nonscreened population, the stage distribution was 26%, 32% and 42% for local, regional and distant disease, respectively. The model estimates 8511 lung cancer deaths per 100,000 in the non-screened population and 7731 deaths per 100,000 in the screened population. The lung cancer mortality reduction estimate was 9% at 40 years from the initial screening exam, and varied across time reaching a maximum of 23%.
Conclusion: Our initial modeling efforts indicate that CT lung cancer screening has the potential to shift lung cancer to earlier stages and reduce lung cancer mortality.
Questions about this event email: sylvia.plevritis@stanford.edu
Plevritis PhD, S,
Estimating the Stage Shift and Mortality Reduction from CT Lung Cancer Screening. Radiological Society of North America 2003 Scientific Assembly and Annual Meeting, November 30 - December 5, 2003 ,Chicago IL.
http://archive.rsna.org/2003/3107957.html
