BACKGROUND: Juvenile-onset systemic lupus erythematosus (JSLE) is an autoimmune disorder characterised by immune dysregulation, chronic inflammation and increased cardiovascular risk (CVR). Cardiovascular disease is the leading cause of mortality in JSLE patients not attributable to disease flares. However, it is not possible to predict those patients at greatest risk using traditional CVR factors. METHODSerum metabolomic analysis was performed using a nuclear magnetic resonance spectroscopy-platform in 31 JSLE patients. Data was analysed using cluster, linear regression and receiver operating characteristic analysis. Results were validated in a second cohort of 31 JSLE patients and using data from a cohort of adult-onset SLE patients with known pre-clinical atherosclerotic plaque. RESULTS: Unbiased hierarchical clustering of metabolomic data identified three patient groups. Group-1 had decreased atheroprotective high density lipoproteins (HDL) and increased atherogenic very low and low density lipoproteins (VLDL/LDL); Group-2 had elevated HDL but reduced VLDL/LDL; and Group-3 had low HDL/VLDL/LDL levels. Notably, apolipoprotein(Apo)B1:ApoA1 ratio, a known CVR marker in adult cohorts, was elevated in Group-1 JSLE patients compared to Groups-2/3. The metabolomic signature was validated in a second JSLE cohort and compared with lipid biomarkers previously associated with pre-clinical atherosclerotic plaque in adult SLE patients. Linear regression analysis accounting for demographics, treatment, disease activity, lupus serological markers and body mass index confirmed that a unique metabolomic profile could differentiate between JSLE patients at high and low CVR. CONCLUSIONS: Patient stratification using ApoB:ApoA1 ratio and lipoprotein signatures could facilitate tailored lipid modification therapies and/or diet/lifestyle interventions to combat increased CVR in JSLE.