1Universitätsklinikum Tübingen Innere Medizin III, Kardiologie und Angiologie Tübingen, Deutschland; 2Eberhard Karls Universität Tübingen Institute of Pharmaceutical Sciences Tübingen, Deutschland
Background
Coronary artery disease (CAD) often leads to adverse events resulting in significant disease burdens. Pathophysiological cascades and underlying risk factors often remain inapparent prior to disease incidence and the cardiovascular (CV) risk is not exclusively explained by traditional risk factors. Platelets inherently promote atheroprogression and enhanced platelet functions are associated with adverse events. Beyond reliable risk factors, distinct platelet lipids are associated with disease severity in patients with CAD.
Methods
Lipidomics data were acquired using liquid chromatography coupled to mass spectrometry. Clinical and platelet lipidomics data were then processed applying machine learning to model estimates of an increased CV risk including platelet lipidomics in a large-scale consecutive CAD cohort (n=595) (Figure 1A).
Results
By training machine learning models on CV risk measurements including platelet lipids, stratification of CAD patients resulted in a phenotyping of risk groups (Figure 2A). We found that characteristic changes occurred in patients with adverse cardiovascular events and patients with altered platelet lipids and platelet hyperreactivity are at elevated risk to develop adverse events (Figure 2B). The most prominent upregulated lipids in patients with cardiovascular events primarily belong to the class of phospholipids (mainly monoacylglycerophosphoethanolamines (LPE)) and fatty acyls (acylcarnitines (CAR) and unsaturated fatty acids (FA)). In addition, specific alterations of the platelet lipidome, including LPE and CAR are associated with modulation of in vitro platelet functions and might contribute to the development of adverse CV events by promotion or inhibition of platelet functions (Figure 2C).
Further, distinct platelet lipid species are associated with an increased CV or bleeding risk and independently predict adverse events (Figure 2D). Notably, the addition of platelet lipids to conventional risk factors resulted in an increased diagnostic accuracy of patients with adverse CV events (Figure 2E). Thus, the estimation of an elevated CV risk by machine learning of platelet lipid species outperformed conventional risk stratification in patients with CAD (Figure 2D).
Conclusion
Our results unveiled that patients with aberrant platelet lipid signatures and platelet functions are at elevated risk to develop adverse CV events. Further, machine learning combining platelet lipidome data and common clinical parameters demonstrated an increased diagnostic value in patients with CAD and might improve early risk discrimination and classification for CV events (Figure 1B).