Opening the black box: defining true-negative outcomes in esophageal cancer screening - a population-based study.

BACKGROUND: Current negativity definition in esophageal cancer screening overlooks the risk heterogeneity between individuals with non-dysplastic Lugol's unstained lesions (ND-LULs) and normal-stained epithelium. We aimed to define the screening negativity by the incidence risk of severe dysplasia and above lesions (SDAs) after chromoendoscopy and ascertain their re-screening interval. METHODS: We enrolled 815 participants with ND-LULs and 14,123 with normal-stained epithelium from the screening arm of a community-based randomized controlled trial (RCT). The control groups included the RCT's unscreened arm (n = 16,473), and population controls matched 1:5 by birth year, sex, and village (4075 controls for ND-LULs group, 70,615 controls for normal-stained group). All participants were followed up for up to 10 years. Poisson regression and the cumulative incidence function compared the cumulative incidence of SDA between groups. RESULTS: Compared with the RCT control, the cumulative incidence of SDAs was significantly higher in the ND-LULs group over 10 years (adjusted IRR = 3.16, 95% CI: 2.02-4.76) but lower in the normal-stained group (adjusted IRR = 0.52, 95% CI: 0.37-0.72). Similar results were achieved using the general population control. The cumulative incidence of SDAs in the ND-LULs group significantly exceeded controls after 2-3 years while remaining consistently lower in the normal-stained group over 10 years. CONCLUSIONS: Individuals with normal Lugol staining should be considered the genuinely screening-negative population for esophageal cancer. A 10-year screening interval is recommended for normal-stained individuals, while 2-3 years are necessary for ND-LULs.