Background: Neuromyelitis optica (NMO) and multiple sclerosis (MS), two inflammatory demyelinating diseases, are characterized by different therapeutic strategies. Currently, the only biological diagnostic tool available to distinguish NMO from MS is the specific serum autoantibody that targets aquaporin 4, but its sensitivity is low.
Objective: To assess the diagnostic accuracy of metabolomic biomarker profiles in these two neurological conditions, compared to control patients.
Methods: We acquired serum spectra (47 MS, 44 NMO and 42 controls) using proton nuclear magnetic resonance ((1)H-NMR) spectroscopy. We used multivariate pattern recognition analysis to identify disease-specific metabolic profiles.
Results: The (1)H-NMR spectroscopic analysis evidenced two metabolites, originating probably from astrocytes, scyllo-inositol and acetate, as promising serum biomarkers of MS and NMO, respectively. In 87.8% of MS patients, scyllo-inositol increased 0.15 to 3-fold, compared to controls and in 74.3% of NMO patients, acetate increased 0.4 to 7-fold, compared to controls. Using these two metabolites simultaneously, we can discriminate MS versus NMO patients (sensitivity, 94.3%; specificity, 90.2%).
Conclusion: This study demonstrates the potential of (1)H-NMR spectroscopy of serum as a novel, promising analytical tool to discriminate populations of patients affected by NMO or MS.
Keywords: Acetate; biomarkers; metabolomics; multiple sclerosis; neurodegeneration; neuromyelitis optica; nuclear magnetic resonance; scyllo-inositol.