A Clinical Lipidomics Platform: Development and Validation of a High-Throughput LC-MS Assay for Cardiovascular Disease Risk Assessment

Traditional lipid biomarkers, such as cholesterol, HDL, LDL, and triglycerides, are frequently used to assess cardiometabolic health in clinical practice. However, these measures provide a limited view of the human lipidome, leaving clinically relevant information untapped. Lipidomic profiling using LC-MS/MS enables the measurement of hundreds of individual lipid species, yet low throughput and complex data processing have limited clinical integration. To overcome these limitations, we developed a Clinical Lipidomics Platform (CLP), a high-throughput (6 min runtime) LC-MS/MS assay measuring 270 lipid species (248 analytes + 22 internal standards) from 37 lipid subclasses in human plasma. The CLP incorporates automated data processing and normalization to an external reference material (NIST SRM 1950) to ensure reproducible and accurate data. We validated the CLP using the BioHEART-CT Discovery Cohort (<i>n</i> = 994) and compared lipidomic data to those from our Research Lipidomics Platform (RLP), which used a 16 min LC gradient and manual data processing to measure >800 lipids. CLP and RLP lipid measurements were highly correlated. A Lipidomic Risk Score (LRS), previously developed to model 10-year cardiovascular event risk using lipidomic, clinical, and demographic data, was calculated for each individual. CLP-derived LRS showed a strong correlation with RLP-derived LRS (<i>R</i>² = 0.97). The LRS outperformed traditional risk scores, such as the Framingham Risk Score (FRS), in predicting the coronary artery calcium score (CACS), particularly in intermediate-risk individuals. These findings demonstrate the clinical utility of the CLP for cardiovascular risk assessment and its potential for broader clinical application.