Genes That Set the Tempo: How Your DNA May Predict Multiple Sclerosis Activity

Multiple sclerosis (MS) behaves like a genetic kaleidoscope: some people remain stable for years, whereas others accumulate new relapses, MRI lesions and disability despite treatment. The new Molecular Neurobiology paper “Genetic Contribution to Medium-Term Disease Activity in Multiple Sclerosis” asks a precise question: do inherited DNA differences help predict how active MS will be during the first four years on standard first-line therapies?

Study population and follow-up
Researchers analysed two real-world cohorts treated at San Raffaele Hospital in Milan. Cohort 1 (C1) contained 1,086 people who began therapy between 1994 – 2016, while Cohort 2 (C2) added 208 people who started between 2011 – 2016. After applying strict inclusion criteria—relapsing-remitting MS, first-line drug at baseline, and ≥ 4 years of monitoring—they retained 1,294 patients in the genetic study. When four-year outcomes were scored as “no evidence of disease activity” (NEDA-3) or “evidence of disease activity” (EDA), 84 % of C1 and 69 % of C2 were classified as EDA, underlining how often first-line drugs fail to provide lasting control.

Methods in brief
Researchers first performed a genome-wide association study (GWAS) on ~6.9 million single-nucleotide polymorphisms (SNPs) in each cohort, then meta-analysed the results. They next collapsed SNP signals into gene-based p-values using VEGAS2, producing scores for 23,750 genes. Finally, those gene scores were projected onto tissue-specific interaction maps from HumanBase to build brain and lymphocyte gene-gene networks; the dmfind algorithm highlighted subnetworks enriched for activity-linked genes.

Key genetic findings
The GWAS uncovered 23 SNPs with suggestive evidence (p ≤ 1 × 10⁻⁵) of association with four-year disease activity. The strongest cluster sits just upstream of SERPINE2 on chromosome 2, while a second hotspot near PON2/PON3 on chromosome 7 implicates antioxidant pathways.

When signals were aggregated at gene level, ILRUN, ACTBL2 and three other genes crossed the p < 1 × 10⁻⁴ threshold, hinting that innate immune regulators and cytoskeletal actors contribute to medium-term outcomes.

What the networks revealed
Mapping gene scores onto interactomes produced a brain module of 228 genes and a lymphocyte module of 287 genes; remarkably, 167 genes appeared in both, suggesting a shared genetic script across the immune and central nervous systems. In the brain network, the mitochondrial fusion gene OPA1 emerged as a top “connector hub,” whereas the lymphocyte module was anchored by CCT4; both networks shared hubs such as MPHOSPH9, underscoring cross-tissue coherence.

Pathway annotation of these modules highlighted complement and coagulation cascades, circadian rhythm biology and the citrate (TCA) cycle among the most enriched processes.

Biological interpretation
Together, these signals support a model in which mitochondrial resilience, oxidative-stress handling and innate immune amplification shape whether first-line therapy can keep MS quiet. Variants near PON2/PON3 have known roles in defending neurons and glia against oxidative damage; loss of PON2, for instance, disrupts mitochondrial function and inflames the nervous system. Enrichment of complement genes meshes with pathological evidence of complement deposition at the rims of smouldering lesions, a hallmark of ongoing tissue damage. Meanwhile, circadian-clock genes may modulate immune cell trafficking and cytokine surges over the 24-hour day, linking sleep biology to disease activity.

Strengths, limitations and future directions
A major strength is the homogeneous treatment background, reducing confounding by drug potency. Multi-layer analysis—SNP, gene and network—also captures both strong and subtle genetic effects. Yet no variant met the strict genome-wide threshold; findings must therefore be replicated in larger, multi-ethnic datasets. The authors acknowledge that NEDA-3 lumps together several biological processes and that Italian ancestry limits generalisability.

Future work should validate SERPINE2, PON2 and ILRUN variants in other populations, test whether they predict response to stronger disease-modifying therapies, and integrate epigenomics and proteomics to see how these genes behave over time. Take-home messages
Medium-term MS activity appears to be shaped by dozens of small-effect variants rather than a single “bad” gene.

Brain and immune cells share most of the same genetic drivers, pointing to whole-body rather than compartment-specific mechanisms.

Pathways tied to mitochondrial health, complement activation and circadian control rise to the surface, offering new therapeutic angles.

With further validation, a polygenic “activity score” could help clinicians identify who needs early escalation beyond first-line drugs.

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:
Mascia, E., Nale, V., Ferrè, L., Sorosina, M., Clarelli, F., Chiodi, A., ... & Esposito, F. (2025). Genetic contribution to medium-term disease activity in multiple sclerosis. Molecular Neurobiology, 62(1), 322-334.