Cindy Boer

Hand osteoarthritis and Matrix-Gla Protein | 119 3.1 Introduction Osteoarthritis (OA) is the most frequent joint disorder worldwide. An estimated 22% of the adult population has a joint affected by OA and this incidence increases to 49% in in- dividuals over 65 years of age.[1] All synovial joints can be affected by OA, with hand OA as one of the most common forms of OA. Hand OA is characterised by osteophyte forma- tion, bony enlargements of finger joints and cartilage degradation in the joints. One of the factors contributing to cartilage degradation is the increase of calcified cartilage in the joint.[2,3] In addition, hand OA is related to the occurrence of OA at other sites, most notably with knee OA.[4,5] Patients affected by hand OA suffer from pain and disability, impacting their quality of life. OA is a leading cause of chronic disability,[6] yet currently no effective therapeutic treatments against OA are known. It is therefore imperative to dissect the underlying mechanism of disease aetiology as this may enhance effective and targeted drug development. OA has a strong genetic component. Depending on the joint affected, the herita- bility of OA is estimated in the range of 40%–60%,[7,8] with hand OA having the largest heritability, that is, ~60%.[9,10] Therefore, in recent years, several large-scale genetic studies have been performed to identify the underlying genes and pathways leading to OA. Multiple significantly associated loci for OA of the hip and knee have been identi- fied through genome-wide association studies (GWAS).[11–18] However, thus far, only one report has described a robust association with OA of the hand.[19] In this previous report, common variants in the ALDH1A2 and rare variants in chromosome 1p31 were genome-wide significantly associated with hand OA using a discovery cohort of 837 cases and 77,325 controls. In this study, we aimed to identify novel genes and pathways involved in the ae- tiology of OA of the hand by performing a large-scale genome-wide association study (GWAS). We used a semiquantitative measure for OA of the hand in order to increase statistical power. We gathered a large sample size of 12,754 individuals for analysis, by combining data from three studies in the discovery phase and an additional three cohorts for replication. Next, we conducted functional follow-up of our top finding to investigate the underlying mechanism.