Background Persistency of arthritis is the hallmark of Rheumatoid Arthritis. Thus far the mechanisms underlying arthritis persistency are largely unknown. In other diseases, it has been shown that longstanding inflammation may induce changes in DNA methylation, that DNA methylation occurs very early in disease development and that DNA methylation is associated with the disease outcome. Since aberrant DNA methylation is passed over to cells it can provide “epigenetic memory”. We hypothesized that changes in DNA methylation occur early in arthritis development and are associated with persistency of arthritis. The present study aimed to evaluate the latter hypothesis.
Objectives To identify differentially methylated positions (DMPs) associated with arthritis persistency in patients presenting with early undifferentiated arthritis (UA).
Methods 74 patients with UA were studied for persistency of arthritis during a median follow-up duration of 42 months (IQR 36-56). Persistent arthritis was defined as the absence of DMARD-free sustained remission (no arthritis for at least one year after cessation of DMARD-therapy). Peripheral Blood Mononuclear Cells (PBMCs) were extracted at the baseline visit (median symptom duration 19 weeks (IQR 9-31)) when no treatment was yet started. The use of these early samples reduced the risk that the methylation status was influenced by effects of longstanding inflammation or of treatment. Methylation was determined using the Illumina 450K methylation array. None of the samples were excluded after quality control. Analyses were done using Cox proportional hazard models. To account for multiple testing the false discovery rate (FDR) was controlled at 5%.
Results During the median follow-up duration of 42 months (IQR 36-56) 50 patients (67.6%) had persistent arthritis and 24 patients (32.4%) achieved DMARD-free sustained remission after a median duration of 31 months (IQR 14-38). After correction for multiple testing, no DMP was significantly associated with arthritis persistency. The lowest p-value was 3.14×10-6 for a difference of 2.2% in methylation levels between patients with and without persistency; this concernced cg17440386 that is located in between SHANK2 and DHCR7.
Conclusions In this first EWAS on arthritis persistency, no significant association was observed. This may indicate that the present study was not large enough, or that potential signals were blurred though heterogeneity in cell compositions, or that DNA methylation does not play a major role in arthritis becoming chronic. In addition, more advanced analysis approaches utilizing methylation patterns and genome annotations may prove to be more powerful than our current GWAS-like focussing on DMPs. Larger studies in selected cell populations are needed to determine the value of methylation in persistency of arthritis.
Disclosure of Interest H. van Steenbergen: None declared, R. Luijk: None declared, B. Heijmans: None declared, T. Huizinga: None declared, R. Shoemaker Employee of: R. Shoemaker is an employee of Ignyta, Inc., A. van der Helm-van Mil: None declared