P. Lacaze, G. Polekhina, A. Bakshi, M. Riaz, A. Owen, A. Franks, Y. Wang, J. Abidi, J. Tiller, S. Orchard, J. Mcneil, F. Cicuttini
2021 Annals of the Rheumatic Diseases  
Background:Osteoarthritis (OA) is a major cause of disability, with treatment options mostly limited to advanced disease when joint replacement is indicated. Recent genome-wide association analysis more than doubled the number of OA-associated variants identified (1, 2), extending the feasibility of identification of high-risk individuals.Objectives:Including these variants, we calculated polygenic risk scores (PRSs) and performed validation in a well-characterised population of older
more » ... of older individuals.Methods:We calculated PRSs for knee and hip OA respectively, using joint replacement surgeries as markers of advanced disease in 12,724 older individuals of European descent in the ASPREE trial. We considered in-trial joint replacement (hospitalizations during median 4.7 years follow-up) and pre-trial joint replacement from self-reported medical history. Multivariable models examined the effect of PRS as a continuous variable (per standard deviation [SD]) and categorical variable (low-risk [0-20%], medium-risk [21-80%], high-risk [81-100%] groups), adjusting for age, sex and BMI.Results:Mean population age at baseline was 75 years and 54.9% of participants were female. In total, 1478 (11.6%) participants had knee replacements and 1324 (10.4%) had hip replacements. Female sex, higher BMI and age were associated with higher risk of knee and hip replacements. PRSs as continuous variables per SD were associated with knee (odds ratio [OR] 1.11, 95% confidence interval [CI] 1.04-1.17) and hip (OR 1.18, 95% CI 1.11-1.25) replacements. We found meaningfully different rates of knee or hip joint replacement occurring between low-, medium- and high-risk PRS groups. Participants in the high-risk PRS group, compared with the low, had a higher risk of knee replacements (OR=1.35, CI 1.12-1.62), and hip replacements (OR 1.66, 95% CI 1.37-2.02). We found no interaction between PRS and sex, and no collinearity between PRS and BMI, suggesting PRS is an independent risk factor for OA.Conclusion:Joint-specific genomic risk scores predict advanced OA in older adults, independent of age, sex and BMI. Stronger associations are observed for hip versus knee OA. Our study provides some of the first evidence of potential clinical utility of genomic risk prediction for OA, which may help identify individuals who would benefit most from targeted clinical management and preventive intervention.References:[1]Tachmazidou I, Hatzikotoulas K, Southam L, et al. 2019. Identification of new therapeutic targets for osteoarthritis through genome-wide analyses of UK Biobank data. Nat Genet, 51, 230-236[2]Zengini E, Hatzikotoulas K, Tachmazidou I, et al. 2018. Genome-wide analyses using UK Biobank data provide insights into the genetic architecture of osteoarthritis. Nat Genet, 50, 549-558Disclosure of Interests:None declared
doi:10.1136/annrheumdis-2021-eular.2918 fatcat:2b67z66ohja4ddvbziv2672lgu