Multiple Sclerosis Severity Genetics: Scientific Promise, Clinical Utility
Multiple sclerosis is a chronic inflammatory and neurodegenerative disease of the central nervous system, characterized by substantial heterogeneity in clinical presentation, relapse frequency, disability accumulation, and long-term prognosis. Although more than two hundred genetic loci have been associated with susceptibility to multiple sclerosis, the genetic determinants of disease severity have remained far more elusive. This distinction is clinically important: a variant that increases the risk of developing multiple sclerosis does not necessarily determine whether a patient will later experience rapid disability progression. The study by Kreft and colleagues addresses this unresolved question by examining whether recently reported severity-associated genetic variants can predict real-world disease course in a large, deeply phenotyped multiple sclerosis cohort.
From Susceptibility Genetics to Severity Genetics
Historically, multiple sclerosis genetics has been dominated by immune-related susceptibility loci, particularly those connected to antigen presentation, lymphocyte regulation, and peripheral immune function. This is consistent with the inflammatory biology of early multiple sclerosis and with the mechanism of action of currently licensed disease-modifying therapies. However, disability progression in multiple sclerosis increasingly reflects neurodegenerative mechanisms, including axonal injury, central nervous system resilience, and failure of repair. Recent genome-wide association studies have therefore generated considerable interest by identifying candidate variants associated with disease severity rather than disease onset. Among these, rs10191329A emerged as a genome-wide significant variant in a large International Multiple Sclerosis Genetics Consortium study, raising the possibility that genetic testing might one day contribute to individualized prognostic assessment.
A Real-World Cohort Designed for Clinical Relevance
Kreft and colleagues tested this possibility using the South Wales multiple sclerosis registry, a prospective cohort initiated in 1985 with standardized longitudinal clinical data. The investigators included 1,455 individuals with multiple sclerosis who had available genotyping and detailed clinical follow-up. The study excluded non-Caucasian individuals because the relevant genome-wide association findings had been validated only in people of European ancestry, and it excluded clinically isolated syndrome without subsequent evidence of multiple sclerosis. Importantly, disability outcomes were based on physician-assessed Expanded Disability Status Scale measurements obtained during physical examinations, rather than telephone- or questionnaire-derived estimates. This methodological decision strengthened the clinical validity of the disability data, particularly at lower EDSS ranges where neurological examination is essential.
Testing rs10191329A Across Multiple Disease Outcomes
The central finding of the study was that rs10191329A did not meaningfully predict multiple sclerosis severity in this independent real-world cohort. The investigators found no association between rs10191329 genotype and age-related multiple sclerosis severity score, age at onset, sex distribution, annualized relapse rate, anatomical localization of onset symptoms, localization of subsequent relapses, incomplete recovery after the first demyelinating event, time to EDSS 4, time to EDSS 6, time to EDSS 8, or time to secondary progressive multiple sclerosis. The authors also performed survival analyses and propensity score matching comparing homozygous risk carriers with homozygous non-risk carriers, yet these analyses similarly failed to demonstrate a clinically relevant difference in disability trajectory. This broad negative result is important because it argues against immediate clinical use of this variant for patient counseling or treatment escalation decisions.
Polygenic Scores and the Limits of Current Genetic Prediction
The study also examined whether broader genetic risk measures could explain severity. The authors calculated a weighted genetic risk score for multiple sclerosis susceptibility loci, an HLA genetic burden score, and a severity-weighted genetic risk score based on variants reported in the prior severity genome-wide association study. None of these scores was associated with age-related multiple sclerosis severity, first disability measurements, last disability measurements, age at onset, relapsing versus progressive onset, or time to major EDSS milestones. This finding reinforces a central concept in complex disease genetics: genetic architecture for disease susceptibility may be biologically and clinically distinct from genetic architecture for disease progression. In multiple sclerosis, immune-mediated susceptibility variants do not appear to provide a reliable map of long-term neurodegenerative disability.
Partial Replication of Other Severity Signals
Although rs10191329A was not replicated, the study did provide support for two suggestive severity variants previously identified by the MSBase consortium. Specifically, rs7289446G and rs868824C showed associations with long-term disability outcomes, although their effect sizes were modest. The authors also assessed HLA-DRB1*1501, a well-established multiple sclerosis susceptibility allele. Consistent with prior observations, HLA-DRB1*1501 was associated with younger age at onset, but not with long-term disability severity in this cohort. These findings suggest that some genetic signals related to disease severity may be reproducible, but their individual predictive value remains limited. The results therefore support continued biological investigation while cautioning against premature clinical implementation.
Clinical and Research Implications
The principal message of this article is that multiple sclerosis severity genetics is scientifically promising but not yet clinically actionable. The identification of variants associated with progression may illuminate mechanisms of central nervous system resilience, neurodegeneration, and repair failure; however, current variants have small effect sizes and inconsistent replication across cohorts. For clinicians, this means that genotyping for rs10191329A or related severity variants should not currently guide treatment selection, prognosis, or patient counseling. For researchers, the study highlights the need for larger independent replication cohorts, standardized disability measures, longitudinal imaging and biomarker integration, and more refined phenotypes that capture cognition, upper-limb function, fatigue, and silent progression beyond EDSS alone. Ultimately, the future of prognostic precision medicine in multiple sclerosis will likely depend not on a single variant, but on integrated models combining genetics, clinical history, imaging, treatment exposure, and molecular biomarkers.
Disclaimer: This blog post is based on the provided research article and is intended for informational purposes only. It is not intended to provide medical advice. Please consult with a healthcare professional for any health concerns.
References:
Kreft, K. L., Uzochukwu, E., Loveless, S., Willis, M., Wynford‐Thomas, R., Harding, K. E., ... & Robertson, N. P. (2024). Relevance of multiple sclerosis severity genotype in predicting disease course: a real‐world cohort. Annals of Neurology, 95(3), 459-470.
