How “Genetic Risk” Shapes Your Chances of Multiple Sclerosis
Multiple sclerosis (MS) has long been known as a disease that needs both genes and environment to appear. On average, the lifetime risk of MS worldwide is roughly 5 in 1,000 for women and 2 in 1,000 for men, but that average hides a huge amount of individual variation. Some people with a strong family history never develop MS, while others with no known affected relatives are diagnosed in their 20s or 30s. The study by Loonstra and colleagues asks a deceptively simple question: given someone’s genetic profile, how much does their lifetime risk of MS actually change? To address this, they combined modern polygenic risk scores (PRS) with a very unusual cohort: essentially all people with MS in the Netherlands who were born in the same year, 1966.
A Birth-Year Cohort as a Natural Experiment: Project Y
The backbone of this work is “Project Y,” a population-based cohort that tried to track down almost every person with MS born in 1966 in the Netherlands—239,611 live births in total. Researchers used hospital records, MS registries, media campaigns, and even outreach to nursing homes to find these individuals and confirm diagnoses based on the 2010 or 2017 McDonald criteria. They identified 452 people with MS (about 73% women), corresponding to a lifetime risk of ~282 per 100,000 for women and 100 per 100,000 for men in this birth year. For comparison, non-MS controls came from Project Y participants without MS and from the Amsterdam Dementia Cohort, but only if they were born between 1963 and 1969 to keep ages closely matched. Everyone was genotyped on the same array, and rigorous quality control ensured consistent, high-quality genetic data across cases and controls.
Turning Many Small Effects Into a Single Polygenic Risk Score
MS risk is influenced by hundreds of genetic variants, most of which individually have tiny effects. Loonstra et al. started from 215 genome-wide significant susceptibility variants identified in the largest MS genome-wide association study to date, including 15 variants in the HLA/MHC region on chromosome 6, a well-known hot spot for MS genetics. For each participant, they calculated a PRS by summing the number of risk alleles at these sites, each weighted by its effect size (log odds ratio) from the original GWAS. The resulting score was then standardized and, crucially, divided into deciles based on the control group distribution—so “top 10%” really means “top tenth of genetic risk in the general population,” not just among people with MS. MHC alleles were imputed with HIBAG, and non-European ancestry samples were excluded to avoid bias from population structure.
From Odds Ratios to Numbers People Can Understand
Odds ratios, the usual currency of genetic association studies, are not intuitive for patients or clinicians trying to think in terms of “What are my chances?” Loonstra and colleagues tackled this by translating PRS into lifetime risk within their birth cohort. They simulated a population of 100,000 individuals in each PRS decile, separately for men and women, using sex-specific lifetime incidence estimates from the 1966 cohort. They then assigned the same proportion of MS cases to each decile as observed in their real data, effectively asking: “If we could replay the 1966 birth year, keeping the same genetic architecture, how many people in each genetic risk band would end up with MS over their lifetime?” The survival-style curves on pages 4–6 of the article show these cumulative risks across age for each decile, giving a visual sense of how risk spreads out over the lifespan.
How Much Does Genetic Risk Actually Change Your Chance of MS?
The answer: quite a lot, but in absolute terms MS remains a relatively uncommon disease. Among women in the top 10% of PRS (the highest genetic risk decile), about 1 in 92 developed MS—more than 1% lifetime risk. In contrast, in the lowest 30% of genetic risk (the bottom three deciles combined), only about 1 in 2,739 women developed MS. For men, the contrast was similarly striking: about 1 in 293 in the top 10% PRS developed MS, versus roughly 1 in 7,900 in the lowest 30%. These translate to a relative risk of about 29-fold for high- vs low-risk women and 26-fold for men. Yet even in the highest genetic risk group, most people do not develop MS, underscoring the continuing importance of environmental and stochastic factors. The Kaplan–Meier–like plots on pages 5 and 6 clearly show that high-PRS individuals accumulate MS diagnoses earlier and more often, but the curves never approach 100%.
PRS and Disease Course: Risk Is Not Severity
One might expect that people with a high PRS would not only be more likely to develop MS, but also to develop it earlier or progress faster. Interestingly, that is not what the authors found. Using Cox proportional hazards models, they tested whether PRS predicted age at first symptom, age at secondary progression, or time from onset to secondary progression. After correcting for multiple testing, none of these associations were statistically significant in the overall sample or when stratified by sex or PRS decile. In other words, the genes captured by this PRS appear to influence whether you get MS much more than how severe or fast the disease will be once it starts. This echoes previous large GWAS work suggesting that the genetic architecture of MS risk is at least partially distinct from the genetic architecture of MS severity.
What Could This Mean for Patients and Clinicians?
Clinically, these findings suggest that polygenic scores for MS might become a useful supporting tool, especially in tricky diagnostic situations. A very high PRS could add weight to a suspected MS diagnosis in someone with ambiguous MRI or clinical findings. Perhaps more importantly, a PRS in the lowest tail of the distribution might serve as a “red flag”: if a patient’s genetics indicate extremely low MS susceptibility, clinicians may want to think harder about alternative diagnoses that can mimic MS. At the same time, the study has important caveats: it is restricted to people of European ancestry, it relies on current (still incomplete) knowledge of MS genetics, and it does not account for environmental exposures such as EBV infection, smoking, or vitamin D status. For now, PRS is not a screening test to predict MS in the general population, but it is a step toward a future where genetic information and clinical data are integrated to sharpen diagnosis and, eventually, to personalize prevention and treatment strategies.
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:
Loonstra, F. C., Álvarez Sirvent, D., Tesi, N., Holstege, H., Strijbis, E. M., Salazar, A. N., ... & Uitdehaag, B. (2024). Association of Polygenic Risk Score With Lifetime Risk of Developing Multiple Sclerosis in a Population-Based Birth-Year Cohort. Neurology, 103(7), e209663.
