PCN255 ESTIMATION OF THE POTENTIAL IMPACT OF THE FRENCH CRC SCREENING PROGRAM BASED ON REAL-WORLD PARTICIPATION RATES

Abstract

Organized screening for Colorectal Cancer (CRC) has been set up in 2009 in France. Individuals from 50 to 74 with medium risk for CRC (i.e. no familial history (FH), inflammatory bowel disease (IBD) or Lynch syndrome/Familial Adenomatous Polyposis (FAP)) are eligible for screening. Screening is based on fecal occult blood testing with an immunological test. This study presents estimations of the epidemiological impact of the French CRC screening program based on observed screening rates using a validated microsimulation model. A microsimulation model was constructed, calibrated and validated for the French context. The model simulates the natural history of CRC and includes a screening module based on real world screening processes. The model was used to estimate the impact on CRC incidence and mortality reduction of the current CRC French screening program given the current participation of 29.1%. This impact was extrapolated to the French population to give estimates of the absolute reduction in disease burden. A sensitivity analysis explores the impact of lower and higher participations. The results are based on the simulation of 4,000,000 individuals aged 50-74y at model initiation and representative of the French population. The French current screening program with a participation rate of 29.1% could reduce annual CRC incidence by 5% (2,200 cases) and annual mortality by 14% (2,600 deaths) compared with no screening. Reaching a 45% and 65% participation rate could further reduce the incidence by 3% and 7% (1,300 and 3,500 additional cases avoided) and mortality by 8% and 22% (1,400 and 4,000 additional deaths avoided). Results show that incidence and mortality reduction are linearly related to participation rates with no participation threshold needed for the screening program to be effective. These results support the effectiveness of the current screening program in France despite a relatively low adherence.