There is a substantial need for biomarkers that can be used to aid the diagnosis and management of MS. Additionally, the discovery of novel biomarkers for MS may shed light on disease mechanisms and reveal new therapeutic targets. Metabolomics is emerging as a promising tool for identifying small molecule biomarkers of various diseases such as MS.

To identify serum metabolite biomarkers of multiple sclerosis (MS).

A targeted nuclear magnetic resonance spectroscopy assay was used to measure 249 metabolites in serum samples from 47,992 subjects in an institutional biobank repository, including 635 subjects with MS. Remaining subjects had diagnoses including type 2 diabetes, cardiovascular disease, depression, and asthma, as well as healthy controls. Relative concentrations of metabolites were compared across subjects using a linear regression model to account for the effects of age and sex. All reported p-values were corrected for the 249 metabolites investigated using a Bonferroni correction.

The MS cohort was comprised of 75.0% females with an average age of 61.8 and standard deviation (SD) of 12.7 years. Compared to the total biobank cohort, MS subjects had a significant reduction in serum branched-chain amino acids leucine (-0.17 SD, p=0.002), isoleucine (-0.17 SD, p=0.003), and valine (-0.15 SD, p=0.018), and a reduction in aromatic amino acids tyrosine (-0.17 SD, p=0.006) and phenylalanine (-0.19 SD, p=0.0002). Citrate was also significantly lower in MS subjects (-0.16 SD, p=0.012). Both total cholesterol (0.15 SD, p=0.019) and LDL cholesterol (0.16 SD, p=0.024) were elevated in MS subjects. Serum sphingomyelins (0.14 SD, p=0.051) showed a trend toward elevation in the MS cohort.

We identified several serum metabolites as potential biomarkers of MS. Ongoing analyses will characterize the relationships of these biomarkers to MS relapse activity, treatment response, and disability progression.