Cindy Boer

164 | Chapter 4.1 Table 1: Summary of the Go consortium GWAS results Genome-wide association study Cases / Controls Signals† Novel Sig- nals† Known Signals† All osteoarthritis* 177,517/649,173 21 8 13 Knee and/or hip osteoarthritis 89,741/400,604 31 9 22 Hip osteoarthritis 36,445/316,943 45 17 28 Knee osteoarthritis 62,497/333,557 24 11 13 Total hip replacement (THR) 23,021/296,016 38 12 26 Total Knee replacement (TKR) 18,200/233,841 10 4 6 Total joint replacement (TJR) 40,887/327,689 37 12 25 Hand osteoarthritis 20,901/282,881 7 5 2 Finger osteoarthritis 10,804/255,814 5 3 2 10,536/236,919 4 2 2 Thumb osteoarthritis 28,3721/3057,578 1 1 0 Spine osteoarthritis 177,517/649,173 21 8 13 Independent signals § 100 Independent novel OA signals § 52 Independent known OA signals § 48 * Cases are defined as any-site Osteoarthritis: hip, knee, hand, finger, thumb and spine. † Signals numbers presented are based per defined OA phenotype, novel/known are thus also based per phenotype, Number represent only genome-wide significant signals p-value≤1.3x10 -08 ‡ Known OA loci defined as known OA signal (regardless of discovery OA phenotype) present within 1Mb upstream or downstream of the identified GWAS signal. § Independence calculated within and across OA phenotypes. Among the 96 previously-reported osteoarthritis-associated SNVs, 91 showed association at nominal significance (p-value≤0.05), and 57 reached genome-wide sig- nificance (p-value≤5x10 -08 ) in at least one osteoarthritis phenotype ( Supplementary Table 2 ). As a sensitivity analysis, we excluded the UK Biobank dataset (up to 68,621 osteoarthritis cases and 247,846 controls) from the meta-analyses, and found that 74 of the lead SNVs continue to demonstrate association at nominal significance (18 at p-value≤5x10 -08 ) ( Supplementary Table 3 ). Signals from 4 osteoarthritis phenotypes (spine, knee, knee and/or hip, and osteoarthritis at any site) included individuals of non-European ancestry (between 0.9 to 2.8% of cases were of East Asian descent). Even though sample sizes in the East Asian cohorts are small, we observed that 62% of the signals have supportive evidence in East Asian ancestry-only meta-analysis, with the same direction of effect, and that 20% of these signals are also nominally significant (p-value≤0.05) (Signals across European and non-European ancestry, Supplementary Information ).