BACKGROUND AND AIMSPopulation-based observational studies suggest that endoscopic screening may reduce upper gastrointestinal cancer mortality. We aimed to quantify the effect of endoscopy screening.METHODSThis is a community-based, multicenter, cluster randomized clinical trial conducted in both high-risk and non-high-risk areas of China. Randomization and recruitment occurred between 2015 and 2017, with follow-up conducted until 2022. The intervention was an invitation to receive endoscopic screening, as opposed to receiving usual care (unscreened). In non-high-risk areas, only participants assessed as high-risk by risk scores in the screening group were invited for endoscopic screening. The primary outcome was the cumulative risk of death from upper gastrointestinal cancer, adjusted for baseline characteristics and cluster effects.RESULTSA total of 234,635 participants were included in the intention-to-screen analysis, with a median age of 52 years. In high-risk areas, 64,836 individuals from 81 clusters were randomized to the screening group, and 59,379 individuals from 82 clusters were randomized to the control group. In non-high-risk areas, 58,367 individuals from 92 clusters were randomized to the screening group, 52,053 individuals from 90 clusters were randomized to the control group. Among high-risk areas, 480 (adjusted cumulative risk, 0.77%) died due to upper gastrointestinal cancers within 7.5 years in the screening group vs 545 (0.99%) deaths in the control group (risk ratio, 0.78; 95% confidence interval, 0.66-0.91). Among non-high-risk areas, adjusted risk was 0.26% (146 deaths) in the screening group and 0.30% (149 deaths) in the control group (risk ratio, 0.86; 95% confidence interval, 0.65-1.13).CONCLUSIONSAn invitation to endoscopic screening reduced upper gastrointestinal cancer mortality in high-risk areas. In non-high-risk areas, an invitation to endoscopic screening based on risk scores did not significantly decrease upper gastrointestinal cancer deaths, but longer follow-up time was required. (Chinese Clinical Trial Registry Identifier: ChiCTR-EOR-16008577.).