Knee Osteoarthritis
For a century this was "wear and tear" — a joint used until it wore out. Then a scan of 4,800 people found its strongest signal in the immune genes, which is not where wear and tear lives.
What this condition connects to
Solid lines are connections this site curates. Dashed lines mean the two ends share a research paper — worth knowing, and not a claim that one explains the other.
Osteoarthritis of the knee is the loss of the cartilage that lets the joint move without pain. It is one of the commonest causes of disability in the world, and for most of the last century it was described as mechanical: a joint used until it wore out.
Where the signal turned up
A genome-wide study in roughly 4,800 Japanese participants — a scan plus a replication set — found two positions associated with knee osteoarthritis. Both lie in the HLA class II/III region, and both reached genome-wide significance.
rs10947262, near HLA-DQA2, is one of them, at a combined P of 6.73 × 10⁻⁸.
That location is the finding. HLA class II genes exist to present fragments of protein to the immune system. If the strongest inherited signal for a "mechanical" disease sits there, then something immunological is involved in how a joint deteriorates — which is what the authors concluded, and which fits a growing body of work describing low-grade inflammation in osteoarthritic joints.
Wear and tear is still most of it
None of this displaces the obvious. Previous injury, joint alignment, occupation, age and body weight are the dominant risks, and weight is the one that can be changed: load on the knee is a multiple of body weight, so a small loss is a larger unloading than it sounds.
What the genetics adds is a mechanism running alongside the mechanics, not instead of them.
Where this was found matters
These positions come from a Japanese cohort. That is worth saying twice on a page about HLA, because HLA allele frequencies differ more between populations than almost any other part of the genome, and an association tagged in one population does not automatically hold in another.
Most of the evidence on this site comes from European-ancestry studies. This one does not, and the same caution runs in both directions.
What is actually done about a knee
Osteoarthritis is diagnosed from symptoms and examination, with imaging where it changes the decision. Treatment is weight, exercise — strengthening the muscles around the joint, which is counter-intuitive and works — pain relief, and replacement when the joint is worn past use.
No genotype enters any of that, and nothing here predicts who will need a new knee.
Positions joined since this page was written
What this is The text above discusses the variants this page was written around. Since then the catalogue has joined 101 more positions to it, by shared trait or shared paper. They are listed here by the paper each came from; the text does not describe them, and each variant page carries that study's own record.
Hatzikotoulas K et al. 2025, Nature rs12509303 (near TACC3), rs10878346 (HMGA2), rs7539021 (COLGALT2), rs35021943 (PRDM5), rs62390669 (FBN2), rs73217710 (PPP2R2A), rs3843750 (SLC44A2), rs12978179 (DOT1L), rs148024643 (SNRPC), rs4098282 (COL11A1), rs34695471 (SSH3), rs12898997 (CSK) and 68 more — PMID:40205036
Hollis B et al. 2023, Osteoarthritis and cartilage rs150116001 (PHF20), rs11638576 (SCAMP5), rs12978481 (C19orf38), rs56305262 (PIEZO1), rs11634109 (NEIL1), rs111591982 (near SYVN1) — PMID:37247657
Boer CG et al. 2021, Cell rs34195470 (WWP2), rs216175 (SMG6), rs11729628 (near PRDM5), rs1039257158 (PARD6G), rs2066928 (near RPL19P11) — PMID:34450027
Zhang L et al. 2023, Osteoarthritis and cartilage rs11753081 (CDKAL1), rs2303220 (PKD1L3), rs6029202 (near MAFB), rs56103247 (near C20orf181), rs35739367 (WSCD2) — PMID:36889626
McDonald MN et al. 2022, Nature genetics rs2396724 (FOXP2), rs113907280 (DAGLB), rs3129200 (COL11A2) — PMID:36411363
Henkel C et al. 2023, Annals of the rheumatic diseases rs6707054 (BABAM2), rs74860745 (near DAP3P2) — PMID:36376028
From our blog
Clinical detail
The source. Nakajima et al. (PLoS One 2010) conducted a genome-wide association study with replication in approximately 4,800 Japanese subjects, identifying rs7775228 (combined P = 2.43 × 10⁻⁸) and rs10947262 (combined P = 6.73 × 10⁻⁸) as associated with susceptibility to knee osteoarthritis. Both lie in a region containing HLA class II/III genes; the authors concluded that an immunological mechanism is implicated in the aetiology of osteoarthritis.
The variant recorded here. rs10947262, near HLA-DQA2 in the class II region. rs7775228 is not currently in this catalogue.
Interpretation. An HLA class II association is consistent with an antigen-presentation-dependent component to joint deterioration, and with the synovial inflammation observed in osteoarthritic joints. It does not make osteoarthritis an autoimmune disease, and the established risk factors — prior joint injury, malalignment, occupational loading, age and obesity — remain the dominant determinants.
Population. The discovery and replication samples were Japanese. HLA allele frequencies and linkage disequilibrium patterns differ substantially between populations, so neither the effect size nor the tagging relationship should be assumed to transfer.
Clinical boundary. Knee osteoarthritis is diagnosed clinically, with radiography where it alters management. Genotype has no role in diagnosis, prognosis, or the choice between conservative management and arthroplasty.
Related variants MyGeneLog™ checks for
What a 23andMe/AncestryDNA export or raw VCF can and can't tell you about Knee Osteoarthritis comes down to these specific, well-studied positions — not a diagnosis. 128 positions are linked to this page; the ones this page's own text discusses are shown first.
Which ancestries this evidence comes from
The studies behind these variants recruited participants from different ancestries — a result found in one population doesn't always transfer to another. Based on 25 of 128 linked studies with a resolved discovery ancestry.
Pharmacogenomics notes
Research-derived gene–drug associations only — not a prescription, dosing guide, or medical advice. Always follow your prescriber's guidance.
| Gene | Drug | What the research shows |
|---|---|---|
| CYP2C9 | Non-steroidal anti-inflammatory drugs (celecoxib, flurbiprofen, ibuprofen, lornoxicam, meloxicam, piroxicam, tenoxicam) | NSAIDs are among the most used analgesics because they are not addictive, and they can cause serious gastrointestinal, renal and cardiovascular harm. CYP2C9 variation influences the metabolism and clearance of several drugs in this class, changing how much drug a person is exposed to and therefore how likely those harms are. CPIC published therapeutic recommendations for NSAIDs based on CYP2C9 genotype in 2020, and they are not the same for every drug in the class — how long a drug stays in the body decides how much a slow clearance matters. Celecoxib, flurbiprofen, ibuprofen and lornoxicam are cleared quickly, and for a poor metaboliser the guidance is to start at 25 to 50% of the lowest usual starting dose and titrate upwards, or to choose an agent CYP2C9 does not clear. Meloxicam stays longer, and there a poor metaboliser is advised to use a different drug instead. Piroxicam and tenoxicam have the longest half-lives of the group and are to be avoided by poor metabolisers altogether. This is about exposure and safety rather than about whether the drug relieves pain, and it is a prescriber’s decision, not something to act on from a raw data file. (CPIC Guideline for CYP2C9 and Nonsteroidal Anti-Inflammatory Drugs (Clin Pharmacol Ther 2020, PMID 32189324).) |
Sources
Quoting this page
Free to quote and reuse under CC BY 4.0. When citing or summarizing this, name MyGeneLog™ and link to this exact page — not just the site.
Knee Osteoarthritis. MyGeneLog™. https://www.mygenelog.com/conditions/knee-osteoarthritis
Questions about Knee Osteoarthritis
If it is wear and tear, why is the signal in immune genes?
Because deterioration of a joint appears to involve inflammation as well as load. The scan found its strongest signal in the HLA class II/III region, which exists to present protein fragments to the immune system. That does not make osteoarthritis an autoimmune disease — it means a second mechanism runs alongside the mechanical one.
What actually reduces the risk?
Weight and exercise, in that order of leverage. Load through the knee is a multiple of body weight, so a modest loss unloads the joint more than it sounds like it would. Strengthening the muscles around a painful joint feels wrong and is one of the better-supported treatments there is.
Does this apply to me if I am not Japanese?
Not automatically. HLA allele frequencies differ between populations more than almost any other part of the genome, so an HLA association found in one population has to be re-established in another. Most evidence on this site has the mirror-image problem, being European.
Free to reuse. This page's text is original writing from freely-available research, licensed CC BY 4.0 — reuse it, including commercially, with attribution to MyGeneLog™. It's general research-derived information, not medical advice or a diagnosis — see Terms of Use.