Objectives: Genome-wide association studies have proposed susceptibility variants for rheumatoid arthritis in the TRAF1-C5 locus and 6q23 region. Furthermore, additional independent studies have investigated the same or highly linked polymorphisms in the same regions. We meta-analyzed the available evidence on these proposed associations.
Methods: Data were synthesized for four polymorphisms: rs3761847 (n=12 datasets) and rs2900180 (n=9 datasets) in the TRAF1-C5 locus, and rs10499194 (n=5 datasets) and rs6920220 (n=7 datasets) in the 6q23 region. We also performed meta-analyses for subgroups defined by anti-CCP and RF status.
Results: The polymorphism rs6920220 reached genome-wide statistically significance with p=7.9x10-17 and an allelic odds ratio of 1.24 (95% CI: 1.18-1.30) and no between-study heterogeneity (I2 =0%). The risk was significantly stronger in patients with anti-CCP antibodies and in patients with rheumatoid factor (RF). The other three variants showed large between-study heterogeneity across datasets (I2 range 74-82%); rs10499194 was nominally statistically significant after exclusion of the discovery data. Two variants had genome-wide statistical significance in subgroups defined by the presence of RF (rs3761847 and rs6920220) or anti-CCP (rs6920220).
Conclusions: Genetic markers in the 6q23 region and TRAF1-C5 are associated with rheumatoid arthritis, in particular with positive anti-CCP and rheumatoid factor profile. With the exception of rs6920220 that shows highly consistent results, other proposed markers have high between-study heterogeneity that may reflect unrecognized phenotypic or genetic variability (e.g. gene environment interactions) within rheumatoid arthritis. Furthermore, these markers may not be the true causative loci but rather be in linkage disequilibrium with the true ones.