Polygenic Risk Scores: A New Tool for Understanding and Managing Rheumatic Diseases
If you’ve ever wondered why some people develop rheumatic diseases like rheumatoid arthritis or ankylosing spondylitis while others don’t, genetics plays a bigger role than you might think. For decades, scientists have studied how genes contribute to these conditions—and a new tool called polygenic risk scores (PRS) is changing the game for diagnosis, treatment, and prevention.
What Are Polygenic Risk Scores?
PRS are like a “genetic report card” that sums up thousands of tiny genetic variations (called single-nucleotide polymorphisms, or SNPs) to estimate someone’s risk of developing a disease. Unlike single-gene tests (which look for one big mutation, like the BRCA gene for breast cancer), PRS capture the combined effect of hundreds or thousands of small genetic changes. That’s crucial because most common diseases—including nearly all rheumatic conditions—are driven by many genes working together, not just one.
For example, a PRS for ankylosing spondylitis (a type of inflammatory arthritis that affects the spine) might analyze 1,750 SNPs. Each SNP contributes a tiny bit to risk, but together they paint a clear picture of whether someone is more or less likely to develop the disease than the average person.
How Are PRS Built?
Creating a PRS requires big data—and lots of it. Scientists start with genome-wide association studies (GWAS), which compare DNA from thousands of people with and without a disease. They identify which SNPs are linked to the condition, then weight each SNP by how strong its association is. The more people in the study, the better the PRS performs—because larger samples help distinguish true genetic links from random noise.
Once a PRS is built, it needs to be validated (tested in independent groups) to make sure it works for people outside the original study. For example, a PRS developed using European DNA might not perform as well in people of African or Asian descent—so validation across diverse populations is key.
What Makes a Good PRS?
Several factors affect how useful a PRS is:
- Sample size: Bigger studies (with 5,000+ cases/controls) produce more accurate scores.
- Genetic data quality: Whole-genome sequencing (which reads every part of DNA) is better than SNP microarrays (which only read common variations).
- Clinical homogeneity: Studies using well-defined, similar patients (like people with confirmed ankylosing spondylitis) produce more reliable PRS than broad biobank data.
- Ethnic diversity: Most PRS are based on European DNA—so we need more research on non-European populations to ensure equity.
PRS in Rheumatology: Changing the Way We Care for Patients
Rheumatic diseases are a perfect fit for PRS because they’re highly heritable (genes explain 30–60% of risk) but rarely caused by a single gene. Here’s how PRS are already making an impact:
1. Predicting Risk Early
PRS are stable from birth—your DNA doesn’t change, so your score never does. That means PRS can predict risk decades before symptoms start. For example, a PRS for rheumatoid arthritis could flag a child with a high genetic risk, so their doctor can monitor them closely and start treatment the moment symptoms appear (which drastically improves outcomes).
2. Improving Diagnosis
Rheumatic diseases often have overlapping symptoms—making diagnosis tricky. PRS can help “triage” patients by ruling out unlikely conditions. A 2020 study tested a PRS suite called G-PROB on patients with undiagnosed inflammatory arthritis. The score was able to:
- Rule out one disease in 100% of patients
- Rule out two or more diseases in 84% of patients
- Rule out three or more diseases in 40% of patients
With a 98% negative predictive value (meaning if the score says you don’t have a disease, it’s almost always right), G-PROB could save doctors hours of guesswork.
3. Personalizing Treatment
PRS can also help doctors choose the right treatment. For example, a PRS for ankylosing spondylitis is more informative than the common HLA-B27 test (which only looks for one gene). And since PRS capture the full genetic picture, they might one day predict who will respond to biologic drugs (like TNF inhibitors) or have side effects.
4. Reducing Unnecessary Testing
For people at low genetic risk of diseases like systemic lupus erythematosus (SLE) or rheumatoid arthritis, PRS could mean fewer unnecessary autoantibody tests. That saves patients time, money, and anxiety—and frees up healthcare resources for people who need them.
How Well Do PRS Work?
You might be wondering: Are PRS accurate enough to use in real life? The answer is yes—especially when paired with clinical information. For example:
- A PRS for ankylosing spondylitis has an area under the curve (AUC) of 0.78. AUC measures how well a test distinguishes between people with and without a disease (0.5 = no better than a coin flip; 1.0 = perfect). That’s better than the HLA-B27 test (AUC 0.87–0.90) and similar to widely used tests like C-reactive protein (CRP, AUC 0.7 for ankylosing spondylitis) or MRI (AUC 0.62–0.89 for back pain screening).
- In high-risk groups (like people under 45 with chronic back pain), the PRS for ankylosing spondylitis has a positive predictive value (PPV) of 93%—meaning 93 out of 100 people with a high score will have the disease. That’s nearly diagnostic on its own.
Limitations and What’s Next
PRS aren’t perfect yet. The biggest gaps are:
- Ethnic diversity: Most PRS are based on European populations. The PGS Catalogue (a public database of PRS) has scores for over 800 diseases—but only 132 are for East Asian, 79 for South Asian, and 149 for African ancestries. We need more diverse research to ensure PRS work for everyone.
- Multiomic integration: PRS work best when combined with other data (like blood tests, imaging, or medical history). Future scores will likely merge genetic risk with clinical and biomarker data to be even more useful.
- Cost: SNP microarrays (used to calculate PRS) cost less than £20 per person—and direct-to-consumer tests (like 23andMe) already provide the data needed. So PRS could soon be affordable for most people.
The Future of PRS in Healthcare
PRS are on the brink of becoming a routine part of care. Programs like Our Future Health (a UK study of 5 million people) are testing how PRS perform in real-world settings—and early results are promising. As more data becomes available (and more PRS are developed for non-European populations), PRS could:
- Help doctors diagnose rheumatic diseases faster
- Guide personalized treatment plans
- Enable preventive care for high-risk people
- Reduce healthcare costs by cutting down on unnecessary tests
Conclusion
Polygenic risk scores aren’t a “crystal ball”—but they’re the closest thing we have to predicting and managing rheumatic diseases with precision. For patients, this means earlier diagnosis, better treatment, and less uncertainty. For doctors, it means more tools to make informed decisions. And for scientists, it’s a step closer to unraveling the genetic mysteries of these complex conditions.
As PRS become more common, the biggest challenge will be ensuring they’re used equitably—so everyone, regardless of ethnicity, can benefit. But the potential is enormous: a world where your DNA helps you stay healthy, not just tell you what you’re at risk for.
This article is based on research by Matthew A. Brown (Guy’s & St Thomas’ NHS Foundation Trust and King’s College London) and Zhixiu Li (Queensland University of Technology), published in the Chinese Medical Journal (2021).
doi.org/10.1097/CM9.0000000000001845
Was this helpful?
0 / 0