Introduction: Coronary computed tomography angiography (CCTA) is a viable method for monitoring vulnerable plaques; however, noncontrast calcium scoring (CAS) has a limited utility in this context. It is therefore necessary to search for novel noncontrast imaging biomarkers to track high-risk plaques. Objectives: We aimed to investigate possible associations between low-attenuation plaque (LAP) identified on CCTA and noncontrast CAS, pericardial adipose tissue (PAT), and epicardial adipose tissue (EAT) in patients with nonobstructive coronary atherosclerosis undergoing serial CCTA. Patients and methods: We analyzed data from 89 participants (40% women; mean [SD] age, 60 [7.7] y) enrolled in the Dietary Intervention to Stop Coronary Atherosclerosis in Computed Tomography trial. CCTA was performed at baseline and repeated at mean (SD) follow-up of 66.8 (13.7) weeks. CAS and cardiac fat were measured on noncontrast reconstructions and analyzed in relation to contrast-derived LAP. Results: Mean (SD) decrease in LAP burden was 0.23% (1%; P = 0.04), median (interquartile range [IQR]) decrease in PAT volume was 4.2 (–36.9 to 8.3) cm³ (P <⁠0.001), median (IQR) decrease in EAT volume was 2.8 (–12.2 to 6.7) cm³ (P = 0.01), and median (IQR) increase in CAS was 21.2 (1.7–71.7) Agatston units (P <⁠0.001). Positive correlations were found between changes in LAP burden and both PAT and EAT volumes (r = 0.349; P = 0.001 and r = 0.49; P <⁠0.001, respectively), but not with a change in CAS. Multivariable linear regression analysis identified EAT as an independent predictor of LAP regression (95% CI, 0.007–0.032; P = 0.003). Conclusions: Pericardial and epicardial fat volumes and their dynamics are positively associated with LAP burden, while CAS is not. Monitoring EAT volume on noncontrast CT may facilitate the assessment of high-risk plaque progression or regression, overcoming the limitations of CAS.