4 February 2021 in Arteriosclerosis, Thrombosis, and Vascular Biology
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.
15 April 2025 in Journal of the American Heart Association
Background: Cardiovascular disease (CVD) is a leading cause of death in women with systemic lupus erythematosus (SLE) due to accelerated atherosclerosis that is not predicted by established CVD risk scores. This study aimed to develop, validate, and test a female‐focused predictive atherosclerosis risk signature based on serum metabolites in patients with SLE. Methods and Results: Female patients with SLE were assessed for the presence (SLE‐P; n=18) or absence (SLE‐NP; n=26) of subclinical atherosclerosis using vascular ultrasound for carotid/femoral intima‐media thickness. CVD risk was assessed using QRISK3 (which includes SLE diagnosis as a risk factor) and Framingham Risk Score. Serum metabolomics (n≥250) was performed and analyzed using machine learning pipelines. Despite having subclinical atherosclerosis, 44.8% to 100% of patients with SLE‐P had low CVD risk according to QRISK3/Framlingham Risk Score scores. Using a lipid‐focused metabolomic analysis, an improved atherosclerosis risk predictive signature was developed comprising 35 metabolites/5 clinical traits that classified patients with SLE‐P and outperformed CVD risk assessment tools, lipid profiles measured in routine care, and clinical features alone. This “atherosclerosis risk signature” was validated in a second adult female SLE cohort (n=98) that predicted plaque status with moderate accuracy (area under the receiver operating characteristic curve, 0.79). The signature was then refined into a 5‐feature subclinical plaque‐predictive score that not only stratified the combined SLE‐P/SLE‐NP cohorts (n=142; area under the receiver operating characteristic curve, 0.84) but also predicted 3‐year atherosclerosis progression in female postpubertal patients with juvenile‐onset SLE (n=36; area under the receiver operating characteristic curve, 0.79). Finally, the 5‐feature score identified distinct high and low subclinical atherosclerosis risk subgroups in a “real‐world” setting of unscanned adult patients with SLE (n=38). Conclusions: This atherosclerosis risk score could improve CVD risk assessment/management in female patients with SLE across age. Validation in non‐SLE and healthy cohorts could further substantiate these findings.
12 August 2020 in medRxiv
Background: Patients with systemic lupus erythematosus (SLE) have an increased risk of developing cardiovascular disease (CVD) and 30-40% have sub-clinical atherosclerosis on vascular ultrasound scanning. Standard measurements of serum lipids in clinical practice do not predict CVD risk in patients with SLE. We hypothesise that more detailed analysis of lipoprotein taxonomy could identify better predictors of CVD risk in SLE. Methods: Eighty patients with SLE and no history of CVD underwent carotid and femoral ultrasound scans; 30 had atherosclerosis plaques (SLE-P) and 50 had no plaques (SLE-NP). Serum samples obtained at the time of the scan were analysed using a lipoprotein-focused metabolomics platform assessing 228 metabolites by nuclear magnetic resonance spectroscopy. Data was analysed using logistic regression and five binary classification models with 10-fold cross validation; decision tree, random forest, support vector machine and lasso (Least Absolute Shrinkage and Selection Operator) logistic regression with and without interactions. Results: Univariate logistic regression identified four metabolites associated with the presence of sub-clinical plaque; three subclasses of very low density lipoprotein (VLDL) (percentage of free cholesterol in medium and large VLDL particles and percentage of 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 SLE-P and SLE-NP with greatest accuracy (0.800). Notably, percentage of free cholesterol in large VLDL particles and age were identified by all models as being important to differentiate between SLE-P and SLE-NP patients. Conclusion: Serum metabolites are a promising biomarker for prediction of sub-clinical atherosclerosis development in SLE patients and could provide novel insight into mechanisms of early atherosclerosis development.
5 October 2019 in medRxiv
Juvenile-Onset systemic lupus erythematosus (JSLE) is an autoimmune rheumatic disease characterised by systemic inflammation and organ damage, with disease onset often coinciding with puberty. JSLE is associated with more severe disease manifestations and a higher motility rate compared to adult SLE. Due to the heterogeneous clinical and immunological manifestations of JSLE, delayed diagnosis and poor treatment efficacy are major barriers for improving patient outcome. In order to define a unique immunophenotyping profile distinguishing JSLE patients from age matched healthy controls, immune-based machine learning (ML) approaches were applied. Balanced random forest analysis discriminated JSLE patients from healthy controls with an overall 91% prediction accuracy. The top-ranked immunological features were selected from the optimal ML model and were validated by partial least squares discriminant analysis and logistic regression analysis. Patients could be clustered into four distinct groups based on the top hits from the ML model, providing an opportunity for tailored therapy. Moreover, complex correlations between the JSLE immune profile and clinical features of disease were identified. Further immunological association studies are essential for developing data-driven personalised medicine approaches to aid diagnosis of JSLE for targeted therapy and improved patient outcomes.
15 October 2017 in bioRxiv
A previous study of exome-sequenced schizophrenia cases and controls reported an excess of singleton, gene-disruptive variants among cases, concentrated in particular gene sets. The dataset included a number of subjects with a substantial Finnish contribution to ancestry. We have reanalysed the same dataset after removal of these subjects and we have also included non-singleton variants of all types using a weighted burden test which assigns higher weights to variants predicted to have a greater effect on protein function. We investigated the same 31 gene sets as previously and also 1454 GO gene sets. The reduced dataset consisted of 4225 cases and 5834 controls. No individual variants or genes were significantly enriched in cases but 13 out of the 31 gene sets were significant after Bonferroni correction and the “FMRP targets” set produced a signed log p value (SLP) of 7.1. The gene within this set with the highest SLP, equal to 3.4, was FYN , which codes for a tyrosine kinase which phosphorylates glutamate metabotropic receptors and ionotropic NMDA receptors, thus modulating their trafficking, subcellular distribution and function. In the most recent GWAS of schizophrenia it was identified as a “prioritized candidate gene”. Two of the subunits of the NMDA receptor which are substrates of FYN are coded for by GRIN1 (SLP=1.7) and GRIN2B (SLP=2.1). Of note, for some sets there was a substantial enrichment of non-singleton variants. Of 1454 GO gene sets, 3 were significant after Bonferroni correction. Identifying specific genes and variants will depend on genotyping them in larger samples and/or demonstrating that they cosegregate with illness within pedigrees.