Researchers at the University of Hong Kong's LKS Faculty of Medicine have developed CardiOmicScore, an artificial intelligence tool that analyzes thousands of circulating proteins and metabolites from a single blood test to predict the onset of six major cardiovascular diseases up to fifteen years before clinical manifestationmanifestation/ˌmænɪfɛˈsteɪʃən/L3表现；显现；明显迹象the act of showing or demonstrating something clearly, or a clear sign of something. Published in Nature Communications, this multiomics approach integrates genomics, proteomics, and metabolomics data from the UK Biobank, examining 2,920 proteins and 168 metabolites to capture a dynamic snapshot of an individual's current biological state. Unlike conventional polygenic risk scores, which remain immutableimmutable/ɪˈmjuːtəbəl/L3不可改变的；永恒不变的not capable of changing or being changed after birth, CardiOmicScore reflects real-time changes induced by diet, exercise, aging, and environmental exposures, thereby offering a more nuanced and contemporaneouscontemporaneous/kənˌtɛmpəˈreɪniəs/L3同时期的；同时发生的existing or occurring at the same time risk assessment. The system demonstrated substantially superior predictive accuracy for conditions including coronary artery disease, stroke, heart failure, atrial fibrillation, peripheral artery disease, and venous thromboembolism, particularly among high-risk individuals.
Cardiovascular diseases remain the leading global cause of mortality, accounting for approximately 19.8 million deaths in 2022, yet conventional risk assessment—relying on age, blood pressure, and smoking history—often fails to detect the earliest biological perturbations. Genetic risk scores, while useful, are fundamentally limited by their static nature, incapable of incorporating the proteomic and metabolomic fluctuations that precede symptomatic disease. Consequently, many patients are identified as high risk only after the optimal window for preventive intervention has narrowed. CardiOmicScore addresses this lacuna by decoding complex molecular signals that reflect subtle alterations in immune activity, metabolism, and vascular health long before clinical onset.
Professor Zhang Qingpeng, Associate Professor in the Department of Pharmacology and Pharmacy at HKUMed, elucidated that while genes delineate baseline health risk, proteins and metabolites provide a dynamic portrait of current physiological state. The AI tool is designed to translate these intricate molecular data into personalized risk estimates, enabling clinicians and patients to implement timely lifestyle modifications and preventive strategies. This paradigmparadigm/ˈpærədaɪm/L3范式；典范；模式a typical example or pattern of something; a model or framework shift from reactive treatment to proactive prediction aligns with broader trends in precision medicine, wherein multiomics profiling facilitates earlier, more targeted interventions. Indeed, a single blood sample could eventually yield a comprehensive risk profile spanning multiple cardiovascular pathologies, affording a critical temporal advantage for disease mitigationmitigation/ˌmɪtɪˈɡeɪʃən/L3缓解；减轻；缓和the action of reducing the severity, seriousness, or harmful effects of something.
