Plasma N-Glycan Profiling Enhances Diagnostic Precision in Multiple Sclerosis, AQP4-Ab NMOSD, and MOGAD.

BACKGROUND AND OBJECTIVES: Differentiating multiple sclerosis (MS) from antibody (Ab)-defined diseases, such as neuromyelitis optica spectrum disorders (NMOSDs), remains challenging, particularly as Ab levels decline. N-glycans play a key role in immunity, with changes in branching and fucosylation linked to T/B-cell function and MS onset while increased N-acetylglucosamine residues correlate with disease progression. Despite growing recognition of glycosylation in neuroinflammation, direct comparisons of the N-glycome between MS and Ab-defined diseases are lacking. This study aims to assess whether plasma N-glycome profiling can effectively differentiate these conditions and their subtypes. METHODS: This cohort study included 120 participants: 30 with relapsing-remitting MS (RRMS), 30 with secondary progressive MS (SPMS), 30 with myelin oligodendrocyte glycoprotein Ab-associated disease (MOGAD), and 30 with aquaporin-4 (AQP4)-Ab NMOSD, recruited from the John Radcliffe Hospital, Oxford University Hospitals National Health System (NHS) Trust. Plasma N-glycans were analyzed using ultra-high-performance (UHPLC) hydrophilic interaction liquid chromatography (HILIC) coupled with high-resolution mass spectrometry. Orthogonal partial least-squares discriminant analysis was applied to identify disease-specific glycomic patterns. RESULTS: Distinct N-glycome profiles were identified across diseases and phenotypes. Plasma N-glycans differentiated MS from Ab-defined diseases with 80.5% accuracy (±1.5%), MOGAD from AQP4-Ab NMOSD with 77.8% accuracy (±3.1%), and RRMS from SPMS with 75.2% accuracy (±3.6%). Key discriminatory features included increased monosialylation (S1; odds ratio [OR] = 2.57, p < 0.0001), trigalactosylation (G3; OR = 2.70, p < 0.0001), highly branched N-glycans (OR = 2.32, p = 0.0002), and antennary fucosylation (OR = 2.89, p < 0.0001), effectively distinguishing Ab-defined diseases from MS, independent of Ab serostatus at the time of sampling. DISCUSSION: These findings underscore the potential of plasma N-glycomics as a diagnostic tool for neuroinflammatory diseases. While further research is needed to clarify the mechanistic links between glycomic alterations and disease pathology, our results suggest that plasma N-glycan profiling could improve disease classification. Given its noninvasive and cost-effective nature, this approach holds promise as a complementary diagnostic tool for CNS demyelinating diseases in clinical practice.