Association between a single-nucleotide polymorphism in the promoter of the human interleukin-3 gene and rheumatoid arthritis in Japanese patients, and maximum-likelihood estimation of combinatorial effect that two genetic loci have on susceptibility to the disease

Genetic variants of interleukin-3 (IL-3), a well-studied cytokine, may have a role in the pathophysiology of rheumatoid arthritis (RA); but reports on this association sometimes conflict. A case-control study was designed to investigate association between RA and a single-nucleotide polymorphism (SN P) in the IL-3 promoter region. Comparison of cases of RA versus control in dividuals yielded a chi (2) value of 14.28 (P = .0002), with a genotype odd s ratio of 2.24 (95% confidence interval [95% CI] 1.44-3.49). When female c ases with earlier onset were compared with female control individuals, the SNP revealed an even more significant correlation, with chi (2) = 21.75 (P = .000004) and a genotype odds ratio of 7.27 (95% CI 2.80-18.89). The stron ger association that we observed in this clinically distinct subgroup (fema les with early onset), within a region where linkage disequilibrium was not significantly extended, suggested that the genuine RA locus should locate either within or close to the IL-3 gene. Combined genotype data on SNPs on eight other candidate genes were combined with our IL-3 results, to estimat e relationships between pairs of loci and RA, by maximum-likelihood analysi s. The utility of combining the genotype data in this way to identify possi ble contributions of various genes to this disease is discussed.