Recently, Knevel et al 1 reported that interleukin 15 (IL-15) genetic variants are associated with progression of joint destruction in patients with rheumatoid arthritis (RA) enrolled in a European multicohort study. IL-15 is a proinflammatory cytokine that is overexpressed in patients with RA. Overproduction of IL-15 is thought to induce T cell activation and proliferation, possibly contributing to the pathogenesis of RA.2 ,3 Some variants of genetic predisposition factors for RA are restricted to particular ethnic groups, revealing the presence of genetic heterogeneity.4 ,5 We investigated the associations between IL-15 genetic variants and progression of joint destruction in Japanese patients with RA.

We studied 865 patients with RA for whom Sharp/van der Heijde (SHS) scores of the hands were available at a disease duration of 5 years. DNA samples of the subjects were obtained from the Institute of Rheumatology Rheumatoid Arthritis (IORRA) cohort study DNA collection.6 All of the patients who donated DNA samples consented to participate in this study in accordance with the process approved by the Tokyo Women's Medical University genome ethics committee, and satisfied the American College of Rheumatology 1987 revised criteria for the classification of RA. Two of four single nucleotide polymorphisms (SNPs) reported by Knevel et al 1 were selected, rs6821171 and rs7667746, since the other SNPs were not polymorphic in the Japanese population. We genotyped rs6821171 and rs1521761 as a substitute for rs7667746 that is in absolute linkage disequilibrium (r²=1) with rs7667746 according to HapMap phase 3 JPT data. SNP genotyping was performed by using the TaqMan fluorogenic 5′ nuclease assay (Applied Biosystems, Tokyo, Japan). Multivariate regression analyses were performed in order to examine the association between the SNPs and the estimated yearly radiographic progression in the first 5 years after onset of RA, using SHS scores of hands at the 5-year disease duration. Adjustments were made for gender, age of onset, anti-citrullinated peptide antibody status6 and year of disease onset (categorised as table 1). All radiographs were scored by one experienced reader. Interobserver and intraobserver agreements were 0.85 and 0.95, respectively. All SHS scores were log-transformed to obtain a normal distribution.

Demographic, clinical and therapeutic data of 865 patients at 5 years after onset of RA are shown in table 1. Multivariate linear regression analyses revealed that neither of the SNPs was associated with progression of joint destruction in the tested models (table 2). The effect of the SNPs went into the opposite direction only in the additive model of rs1521761.

To date, many RA susceptibility genes have been identified in multiple populations.4 ,7 ,8 However, the genetic predisposition factors for progression of joint destruction remain almost unknown. The IL-15 signalling pathway has been reported to play a relevant role in RA pathogenesis.9 ,10 IL-15 genetic variants are associated with progression of joint destruction in European patients with RA, but we could not replicate those associations in this study. The data sets used in this study were relatively large and were part of a single-institution cohort study. Sample size that is required to have the same length of CIs as those of Leiden, Groningen, Lund and Sheffield study were 3400, 1630, <400 and <400, respectively.1 Thus, our study provides the information more than Lund and Sheffield. A possible limitation is that we used SHS scores of the hands at the first 5 years after onset of RA as a substitute for delta-SHS because baseline radiographs were not available for most patients. Our results indicate that IL-15 exhibits genetic heterogeneity among different ethnic populations. However, it might be worthwhile to investigate genes coding for other molecules in the IL-15 signalling pathway as novel candidate genes.