Contrasting Genetic Architectures in Multiple Sclerosis: Why Disease Risk and Disease Severity Are Not the Same
Multiple sclerosis (MS) is a chronic inflammatory and neurodegenerative disorder marked by striking clinical heterogeneity: some patients remain minimally disabled for decades, whereas others accumulate substantial impairment much earlier. The reviewed article by Sorella and colleagues addresses a central question in contemporary neurogenetics: do the same genetic variants that increase the risk of developing MS also determine how severe the disease becomes after onset? The authors argue that these two domains—susceptibility and outcome—are genetically related only to a limited extent, and that this distinction has major implications for prognosis, mechanism discovery, and therapeutic development. Their synthesis positions MS not simply as an immune-mediated disease of initiation, but as a biologically layered disorder in which progression may reflect distinct processes involving tissue resilience, repair, and neuroaxonal vulnerability.
Established Success in the Genetics of MS Susceptibility
Over the past two decades, genome-wide association studies have transformed understanding of MS susceptibility. The article notes that more than 200 common variants have now been associated with disease risk, with the strongest effects mapping to the major histocompatibility complex, especially HLA-DRB1*15:01. These findings have consistently implicated innate and adaptive immune pathways and have supported applications ranging from risk stratification to causal inference and genetically informed drug development. Yet, as the authors emphasize, the utility of susceptibility genetics in established disease depends on an assumption that has often gone untested: that variants driving disease onset also govern subsequent disability progression. This is precisely the assumption the review challenges. Even though susceptibility loci are robust, replicable, and mechanistically informative for disease initiation, they do not appear to translate straightforwardly into markers of long-term outcome.
Age at Onset as the Main Point of Overlap
One area where susceptibility genetics does appear to influence phenotype is age at onset. The review summarizes evidence that a greater burden of MS risk variants is associated with earlier disease manifestation. In particular, each copy of HLA-DRB1*15:01 is associated with an onset roughly 8 to 9 months earlier, and individuals with high non-MHC genetic risk may develop disease several years sooner than those with low risk burden. This pattern is consistent with a liability-threshold model, in which individuals carrying more susceptibility alleles cross the threshold for clinical disease earlier in life. Importantly, however, this overlap does not extend strongly beyond onset timing. The article therefore draws a careful distinction: susceptibility alleles may influence when MS begins, but that does not mean they substantially dictate the trajectory of neurological decline after the disease is established.
Why Susceptibility Variants Fail to Explain Clinical Severity
The evidence linking known MS risk variants to clinical disability outcomes is notably weak. The article reviews studies showing that HLA-DRB1*15:01 and aggregate genetic risk scores generally do not show robust associations with severity metrics such as the Multiple Sclerosis Severity Score or related disability measures after correction for multiple testing. Earlier positive reports have often emerged from small cohorts with limited power, raising concerns about instability and false-positive findings. The authors also highlight an important methodological issue: outcome phenotypes are much harder to measure than case-control status. Clinical disability scales such as the Expanded Disability Status Scale are nonlinear, heavily weighted toward ambulation, and relatively insensitive to domains such as cognition and upper-limb function. As a result, weak or time-dependent genetic effects may easily be obscured by measurement noise. This means the lack of overlap between susceptibility and severity is likely biological, but it is also amplified by phenotypic imprecision.
Neuroimaging Reveals Complexity Rather Than Simple Genetic Continuity
Neuroimaging has been explored as a more quantitative window into disease progression, but the findings remain mixed. The review describes inconsistent associations between susceptibility variants and MRI outcomes such as T2 lesion burden and brain atrophy. Some longitudinal studies suggest that HLA-DRB1*15:01 or polygenic risk may relate to faster accumulation of tissue loss over long periods, while shorter studies have failed to detect meaningful effects. This temporal inconsistency suggests that any genetic influence of susceptibility loci on imaging phenotypes is likely modest, region-specific, or cumulative over many years. Particularly informative is the figure on page 5, which contrasts tissue enrichment for susceptibility and severity heritability: susceptibility is enriched in immune and lymphoid tissues, whereas severity shows enrichment restricted to central nervous system tissues such as hypothalamus, spinal cord, cortex, and limbic structures. This visual summary encapsulates the article’s central thesis that the biology of progression is compartmentally distinct from the biology of disease initiation.
Direct Genetic Studies of Severity Point Toward CNS Resilience
The most important advance described in the review is the direct genome-wide study of MS severity rather than the indirect testing of known susceptibility loci. In a large multicenter analysis of more than 12,000 cases, SNP-based heritability for long-term clinical severity was estimated at 13%, establishing that common genetic variation contributes meaningfully to outcome. Crucially, severity-associated heritability was enriched in CNS tissues rather than immune compartments. The lead genome-wide significant signal, rs10191329 in the DYSF–ZNF638 locus, was associated with faster disability accumulation and earlier attainment of EDSS 6.0 by a median of 3.7 years in homozygous carriers. The article further notes that this variant has been linked to greater brain atrophy, accelerated retinal thinning, elevated neurofilament light chain before worsening, and neuropathological features including greater cortical demyelination, neuronal loss, and inflammatory lesion activity. These convergent findings support a model in which progression genetics may reflect vulnerability to neuroaxonal injury and impaired CNS resilience, rather than classical immune susceptibility alone.
Implications for Precision Medicine and Future Research
The broader significance of the article lies in its redefinition of what genetic translation in MS should mean. The authors caution that risk genetics cannot be assumed to predict prognosis, treatment response, or progression biology. This has consequences for polygenic scores, causal inference, and drug target prioritization in patients who already have MS. It also reshapes interpretation of modifiable exposures: vitamin D appears genetically and epidemiologically linked to susceptibility, yet not clearly to long-term severity, whereas obesity may influence both onset and outcome. The review additionally surveys emerging outcome genetics in treatment response and imaging-based biomarkers, while emphasizing persistent barriers such as limited sample size, heterogeneous phenotyping, short follow-up, and underrepresentation of non-European ancestries. Overall, the article presents MS as a genetically heterogeneous disorder whose progression biology is only beginning to be mapped. Future progress, the authors argue, will likely depend on integrating genetics with environmental exposures, comorbidities, longitudinal biomarkers, and refined phenotypic subgroups to produce a more mechanistic and clinically actionable classification of the disease.
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:
Sorella, M. Y., Ding, Y., Garcia, A. M., Ygonia, M., Wang, L., Jacobs, B. M., & Harroud, A. (2026). Contrasting genetic architectures of multiple sclerosis susceptibility and outcome phenotypes. Revue Neurologique.
