Cindy Boer

Cartilage Thickness and Hip Osteoarthritis | 69 2.1 Introduction In spite of advances in the understanding of OA, the absence of effective therapeutic targets demonstrates that a better comprehension of its causes and pathophysiological mechanisms is needed. Since genome-wide genetic studies are hypothesis-free and do not suffer from the bias of previous knowledge, they have the potential to identify novel biological pathways involved in OA. The discovery of novel genes has the potential to identify novel treatment options. In addition, more personalized medicine approaches for OA can be explored through prediction of risk for disease as well as classification of disease subtypes. Heritability of hip OA has been estimated to be around 40–60%. However, to date only few genetic variants have been successfully identified[1,2]. The reasons for finding only a modest number of genetic loci associated with hip OA can be attributed partially to relatively modest samples sizes in comparison to other complex diseases, such as myocardial infarction[3]. In addition, phenotype heterogeneity is an important issue in OA genetics as this is well known to reduce power to robustly detect signals. The problem of heterogeneity in genetic association studies of OA has been highlight- ed before[4,5] and is exemplified by the fact that the definition of the phenotype is a combination of bone and/or cartilage features as well as clinical complaints. Moreover, there is growing consensus that OA can be divided into multiple sub-phenotypes each with their own aetiology and risk factors. For example, it has been demonstrated that individuals with hip OA, where only cartilage degradation is involved (atrophic OA form), are linked to a different systemic bone phenotype compared to individuals with OA where bone formation is also present[6]. As a way to overcome this, we examined a quantitative trait, which is one of the structural components of joint health, cartilage thickness, as a distinct phenotype. Joint Space Width (JSW) is considered to be a proxy for cartilage thickness mea- sured on hip radiographs. Minimal JSW (mJSW) has been shown to be a more reliable measure for hip joint health compared to the classical Kellgren & Lawrence score[7]. Previously, we have demonstrated that using only a modest discovery sample size (n = 6,000), we were able to successfully identify a genome-wide significant association of the DOT1L locus with mJSW as well as hip OA[1,8]. We now aimed to perform a more powerful analysis by combining data from five studies in the discovery phase, and sub- sequent replication in seven additional studies, amounting to a total sample size of 21,240 to identify new genes implicated in joint health using mJSW as a proxy for carti- lage thickness. Using whole exome sequence data from 2,050 individuals we screened the discovered genes for potential functional variants. Subsequently we used multiple