Plasma Proteomic Profiles Predict Individual Future Osteoarthritis Risk
Authors' conclusion
Our study identified the role of plasma proteomics in predicting future OA risk, which could contribute to preemptive measures.
Study context
- Population
- UK Biobank participants without baseline osteoarthritis (OA); plasma proteins (Olink Explore, 1,463 proteins)
- Participants
- 45,307 participants
- Outcome
- Incident OA
Methodological detailsproteomic analysis, machine learning (LightGBM), SHAP
- Proteomic analysis
- Machine learning (LightGBM)
- SHAP
Main finding
As reported in the abstract
Incident OA, most significant protein predictors (adjusted)
COL9A1, HR
1.54
95% CI 1.48–1.61
CRTAC1, HR
1.65
95% CI 1.54–1.78
Observational data: associations, not effects.
Further findings
Predictive model (LightGBM, proteins + clinical covariates), AUC
- 5-year OA prediction AUC
- 0.729
- 10-year prediction AUC
- 0.721
- All incident OA AUC
- 0.723
- Hip OA, 5-year AUC
- 0.820
- Knee OA, 5-year AUC
- 0.803
Temporal trajectories of COL9A1 and CRTAC1
- Levels began to deviate from normal for more than a decade before OA onset
Original publication
Plasma Proteomic Profiles Predict Individual Future Osteoarthritis Risk. Arthritis Rheumatol. Published 20 March 2025. doi:10.
Composed by AI from the published abstract and checked against it. How Briefs are made →