Interleukin-1 region meta-analysis with osteoarthritis phenotypes.
Moxley G., Meulenbelt I., Chapman K., van Diujn CM., Slagboom PE., Neale MC., Smith AJP., Carr AJ., Loughlin J.
OBJECTIVE: Several research groups have examined osteoarthritis (OA) association with Interleukin-1 (IL-1) region markers and haplotypes. The results have been suggestive for hand OA, negative for knee OA, and conflicting for hip OA. DESIGN: Our aim was to address conflicts employing meta-analytical methods on data from 1238 European-descent cases with various OA phenotypes and 1269 European-descent controls from four study centers. We imputed some missing genotype data and reconstructed IL-1 region extended haplotypes. A previously reported 7-marker IL1A-IL1B-IL1RN extended risk haplotype was tested for association with each specific index phenotype. RESULTS: For hip OA, data from three centers showed heterogeneity of extended-risk-haplotype effect, two panels showing trend toward risk and another showing protection, with overall odds ratio (OR) 1.24 (95% Confidence interval (CI) 0.45-3.41, P 0.67). The heterogeneity fell partly along control ascertainment lines, chiefly between controls ascertained as spouses of arthroplasty patients and controls identified through population radiographic survey. For knee OA, the results showed no heterogeneity and no significant extended-risk-haplotype effect. For hand OA, the results showed little heterogeneity and a modest trend toward positive association (summary OR 1.34, 95% CI 0.83-2.17 P 0.23). Using a Bayesian partition modeling approach, the 7-marker extended haplotypes showed no significant effect on any OA phenotype examined. A 3-single-nucleotide polymorphism (SNP) IL1B-IL1RN haplotype rs1143627-rs16944-rs419598 showed a trend toward hand OA association (posterior probability of association 0.72) with the most prominent feature being protection from a specific haplotype representing a partial mirror image of the extended risk haplotype (OR estimated at 0.46). CONCLUSIONS: The meta-analysis data do not confirm but only suggest that some hand and hip OA risk could be associated with the IL-1 region, particularly centered in IL1B and possibly also IL1RN.