Abstract Background The association between coronary atherosclerosis progression or regression and long-term prognosis remains poorly defined. We assessed the association of atherosclerosis progression or regression with long-term mortality and factors that promote angiographic progression or regression of coronary atherosclerosis in patients with angiographically-proven coronary artery disease. Methods The study included 605 patients with coronary artery disease who underwent coronary angiography at baseline and at 2 years later. Pan-coronary artery tree quantitative coronary angiography was performed. Of 6259 coronary segments (10.3 lesions per patient) analyzed, 1790 non-stented segments with ≥25% diameter stenosis at baseline were included. Atherosclerosis progression or regression was defined as a decrease or increase in the mean minimal lumen diameter (MLD) of the non-stented segments of ≥0.2 mm in the 2-year angiography compared to baseline angiography. The primary outcome was all-cause mortality. Results Based on the change in mean MLD between baseline and 2-year angiography, patients were divided into 3 groups: the group with progression of atherosclerosis (n=53; 8.8%), group with no progression or regression of atherosclerosis (n=472; 78.0%) and the group with regression of atherosclerosis (n=80; 13.2%). There were 126 deaths over 8-year follow-up: 17 deaths among patients with progression, 103 deaths occurred among patients with no progression/regression and 6 deaths among patients with regression (Kaplan-Meier estimates of mortality, 37.5%, 25.2% and 8.9%, respectively; adjusted hazard ratio=1.16, 95% confidence interval 1.05 to 1.29, P=0.004 for 0.1 mm reduction in mean MLD). Conclusions Progression or regression of coronary atherosclerosis in non-treated coronary segments was significantly associated with 8-year mortality.