Multi-Omics and Digital Biomarkers for Dynamic Cardiometabolic Risk Prediction

Cardiometabolic risk changes as metabolic state, vascular injury, behavior and treatment evolve. Multi-omics assays characterize the underlying molecular state, whereas wearable sensors and continuous glucose monitoring capture physiological responses during daily life. This review asks whether combining these measurements improves a defined clinical decision beyond established clinical information in type 2 diabetes, atherosclerotic cardiovascular disease and heart failure. We distinguish biological association, prediction and clinical utility, and define dynamic risk prediction as repeated estimation of future event probability over a specified horizon using only information available at each prediction time. Large cohorts link proteins and metabolites with future disease, and intensive studies connect molecular variation with activity, nutrition and glycemic responses. However, evidence for joint molecular–digital integration remains predominantly observational. Trials of glucose monitoring and structured telemanagement show that monitoring can improve outcomes when it is embedded in an effective care pathway, but they do not establish the incremental value of multi-omics. We therefore organize the evidence around three requirements: temporal alignment of stable genomic information, intermittent molecular assays and dense digital trajectories; validation against a strong clinical comparator; and prospective assessment of patient outcomes, cost and burden. Emerging sensors, dynamical network biomarkers and digital twins may eventually extend this framework, but they remain experimental or conceptual for integrated cardiometabolic event prediction.