Lipid biomarkers in statin users with coronary artery disease annotated by coronary computed tomography angiography

Molecular markers are suggested to improve the diagnostic and prognostic accuracy in patients with coronary artery disease (CAD) beyond current clinical scores based on age, gender, symptoms and traditional risk factors. In this context, plasma lipids are emerging as predictors of both plaque composition and risk of future events. We aim to identify plasma lipid biomarkers associated to CAD indexes of stenosis severity, plaque lipid content and a comprensive score of CAD extent and its risk. We used a simple high performance liquid chromatography-tandem mass spectrometry method to identify 69 plasma lipids in 132 subjects referred to Coronary Computed Tomography Angiography (CCTA) for suspected CAD, all under statin treatment. Patients were stratified in groups using three different CCTA-based annotations: CTA-risk score, lipid plaque prevalence (LPP) ratio and the coronary artery disease-reporting and data system (CAD-RADS). We identified a common set of lipid biomarkers composed of 7 sphingomyelins and 3 phosphatidylethanolamines, which discriminates between high risk CAD patients and controls regardless of the CAD annotations used (CTA score, LPP ratio, or CAD-RADS). These results highlight the potential of circulating lipids as biomarkers of stenosis severity, non calcified plaque composition and overall plaque risk of events.


MS analyses: direct infusion and HPLC-MS/MS conditions
Direct infusions of seven STD solutions were performed in a 5500 QTrap mass spectrometer (SCIEX, Concord, Ontario, Canada) equipped with a Turbo V ESI (electrospray ionization) source. Working conditions were set as follows: positive polarity, mass range 100-1000 m/z (both for full scan and CID (collision induced dissociation) spectra), curtain gas 20 psi, ion spray voltage 5 kV, probe temperature 200 °C, ion source gas 1 and gas 2 10 psi, declustering potential 100 V, entrance potential 10 V. CID experiments were performed by using N2 as collisional gas, the collision cell exit potential was 19 V, while the optimized collision energies (CEns) for each STD and the relative product ions chosen are reported in Table S1. Analyst Software 1.6.3 (SCIEX) was used for acquisition.
The mass spectrometer was set in positive ion mode and the operation conditions were the following: curtain gas 30 psi, ion spray voltage 5 kV, probe temperature 200 °C, ion source gas 1 and gas 2 30 psi, declustering potential 100 V, entrance potential 10 V, collision cell exit potential 19 V, collisional gas N2. The optimized CEns for each STD/lipid species, and the relative product ions chosen, are reported in Table S1. Selected reaction monitoring data acquisition were accomplished using the Scheduled MRM (multiple reaction monitoring) Algorithm in Analyst Software 1.6.3 for 69 lipid species (SCIEX, Concord, Ontario, Canada) with a fixed cycle time of 1.5 sec.   c The three decimal places were necessary in order to differentiate these transitions from those of the corresponding STDs.    CHD, coronary heart disease; LDL-C, low-density lipoprotein cholesterol; HDL-C, high density lipoprotein cholesterol. Table S4. LPP (lipid plaque prevalence) ratio (non-calcified + mixed plaques /total number of plaques): classification and patient characteristics.
For variable definitions and relative p-values see Table S3 heading and text below.   Table S5. CAD-RADS: classification and patient characteristics.

No CAD
For variable definitions and relative p-values see Table S3 heading and text below.  Table S6. Means and standard deviations (SD) (µM) for 69 quantified lipid species among groups of patients defined by CTA score annotations.

No CAD
The degrees of variation evaluated as log2 trasformed fold change (FC = Class mean /No CAD mean) together with their significance after correction using the Benjamini-Hochberg procedure are shown.   Volcano plots showing the differentially expressed lipids (adj. p-value threshold = 0.05) among CAD-RADS annotation Classes relative to No CAD subjects.
Lipid species are colored according to lipid class (CE, Cer, PC, PE, SM, TG). The panels on the left report in details lipids species which resulted significantly dysregulated in comparison Class 2 vs No CAD and Class 3 vs No CAD.