BACKGROUND: The National Lung Screening Trial (NLST) demonstrated that low-dose computed tomography (LDCT) screening reduces lung cancer mortality in a high-risk U.S. population. A microsimulation model of LDCT screening was developed to estimate the impact of introducing population-based screening in Canada. DATA AND METHODS: LDCT screening was simulated using the lung cancer module of the Cancer Risk Management Model (CRMM-LC), which generates large, representative samples of the Canadian population from which a cohort with characteristics similar to NLST participants was selected. Screening parameters were estimated for stage shift, LDCT sensitivity and specificity, lead time, and survival to fit to NLST incidence and mortality results. The estimation process was a step-wise directed search. RESULTS: Simulated mortality reduction from LDCT screening was 23% in the CRMM-LC, compared with 20% in the NLST. The difference in the number of lung cancer cases over six years varied by, at most, 2.3% in the screen arm. The difference in cumulative incidence at six years was less than 2% in both screen and control arms. The estimated percentage over-diagnosed was 24.8%, which was 6% higher than NLST results. INTERPRETATION: Simulated screening reproduces NLST results. The CRMM-LC can evaluate a variety of population-based screening strategies. Sensitivity analyses are recommended to provide a range of projections to reflect model uncertainty.
BACKGROUND: The National Lung Screening Trial (NLST) demonstrated that low-dose computed tomography (LDCT) screening reduces lung cancer mortality in a high-risk U.S. population. A microsimulation model of LDCT screening was developed to estimate the impact of introducing population-based screening in Canada. DATA AND METHODS: LDCT screening was simulated using the lung cancer module of the Cancer Risk Management Model (CRMM-LC), which generates large, representative samples of the Canadian population from which a cohort with characteristics similar to NLST participants was selected. Screening parameters were estimated for stage shift, LDCT sensitivity and specificity, lead time, and survival to fit to NLST incidence and mortality results. The estimation process was a step-wise directed search. RESULTS: Simulated mortality reduction from LDCT screening was 23% in the CRMM-LC, compared with 20% in the NLST. The difference in the number of lung cancer cases over six years varied by, at most, 2.3% in the screen arm. The difference in cumulative incidence at six years was less than 2% in both screen and control arms. The estimated percentage over-diagnosed was 24.8%, which was 6% higher than NLST results. INTERPRETATION: Simulated screening reproduces NLST results. The CRMM-LC can evaluate a variety of population-based screening strategies. Sensitivity analyses are recommended to provide a range of projections to reflect model uncertainty.
Authors: C L Gauvreau; N R Fitzgerald; S Memon; W M Flanagan; C Nadeau; K Asakawa; R Garner; A B Miller; W K Evans; C M Popadiuk; M Wolfson; A J Coldman Journal: Curr Oncol Date: 2017-12-20 Impact factor: 3.677
Authors: C Popadiuk; C L Gauvreau; M Bhavsar; C Nadeau; K Asakawa; W M Flanagan; M C Wolfson; A J Coldman; S Memon; N Fitzgerald; J Lacombe; A B Miller Journal: Curr Oncol Date: 2016-02-29 Impact factor: 3.677
Authors: William K Evans; Cindy L Gauvreau; William M Flanagan; Saima Memon; Jean Hai Ein Yong; John R Goffin; Natalie R Fitzgerald; Michael Wolfson; Anthony B Miller Journal: CMAJ Open Date: 2020-09-22
Authors: Deirdre A Hennessy; William M Flanagan; Peter Tanuseputro; Carol Bennett; Meltem Tuna; Jacek Kopec; Michael C Wolfson; Douglas G Manuel Journal: Popul Health Metr Date: 2015-09-03
Authors: Jean H E Yong; Claude Nadeau; William M Flanagan; Andrew J Coldman; Keiko Asakawa; Rochelle Garner; Natalie Fitzgerald; Martin J Yaffe; Anthony B Miller Journal: Curr Oncol Date: 2022-03-03 Impact factor: 3.677