Journal
ARTERIOSCLEROSIS THROMBOSIS AND VASCULAR BIOLOGY
Volume 41, Issue 4, Pages 1446-1458Publisher
LIPPINCOTT WILLIAMS & WILKINS
DOI: 10.1161/ATVBAHA.120.315321
Keywords
atherosclerosis; lipoprotein; lupus erythematosus; systemic; machine learning; metabolomics
Categories
Funding
- NIHR University College London Hospital Biomedical Research Centre [BRC531/III/IPT/101350]
- UCL & Birkbeck MRC Doctoral Training Programme [MR/N013867/1]
- Lupus UK
- Rosetrees Trust [M409]
- British Heart Foundation PhD Studentship [FS/13/59/30649]
- Versus Arthritis grant [21226]
Ask authors/readers for more resources
This study found that patients with SLE have global changes in their lipoprotein profiles, but standard serum lipid measurements cannot accurately predict their risk of cardiovascular disease. Serum metabolites may be promising biomarkers to uncover the multi-metabolic phenotypes of subclinical atherosclerosis in SLE patients.
Objective: Patients with systemic lupus erythematosus (SLE) have an increased risk of developing cardiovascular disease. Standard serum lipid measurements in clinical practice do not predict cardiovascular disease risk in patients with SLE. More detailed analysis of lipoprotein taxonomy could identify better predictors of cardiovascular disease risk in SLE. Approach and Results: Eighty women with SLE and no history of cardiovascular disease underwent carotid and femoral ultrasound scans; 30 had atherosclerosis plaques (patients with SLE with subclinical plaque) and 50 had no plaques (patients with SLE with no subclinical plaque). Serum samples obtained at the time of the scan were analyzed using a lipoprotein-focused metabolomics platform assessing 228 metabolites by nuclear magnetic resonance spectroscopy. Data were analyzed using logistic regression and 5 binary classification models with 10-fold cross validation. Patients with SLE had global changes in complex lipoprotein profiles compared with healthy controls despite having clinical serum lipid levels within normal ranges. In the SLE cohort, univariate logistic regression identified 4 metabolites associated with subclinical plaque; 3 subclasses of VLDL (very low-density lipoprotein; free cholesterol in medium and large VLDL particles and phospholipids in chylomicrons and extremely large VLDL particles) and leucine. Together with age, these metabolites were also within the top features identified by the lasso logistic regression (with and without interactions) and random forest machine learning models. Logistic regression with interactions differentiated between patients with SLE with subclinical plaque and patients with SLE with no subclinical plaque groups with the greatest accuracy (0.800). Notably, free cholesterol in large VLDL particles and age differentiated between patients with SLE with subclinical plaque and patients with SLE with no subclinical plaque in all models. Conclusions: Serum metabolites are promising biomarkers to uncover and predict multimetabolic phenotypes of subclinical atherosclerosis in SLE.
Authors
I am an author on this paper
Click your name to claim this paper and add it to your profile.
Reviews
Recommended
No Data Available