OBJECTIVE: To investigate 1 functional (rs17266594) and 2 potentially functional (rs10516487 and rs3733197) BANK1 variants, which were previously identified as systemic lupus erythematosus (SLE) susceptibility markers, to test whether they are associated with rheumatoid arthritis (RA). METHODS: Four different cohorts were included in the study: 1,080 RA patients and 1,368 healthy controls from Spain, 278 RA patients and 568 healthy controls from Sweden, 288 RA patients and 287 healthy controls from Argentina, and 288 RA patients and 288 healthy controls from Mexico. Samples were genotyped for BANK1 single-nucleotide polymorphisms (SNPs) using a TaqMan 5'-allele discrimination assay. Statistical analysis comparing allele and genotype distributions was performed with the chi-square test. RESULTS: We did not find a significant association between RA and the rs10516487 and rs17266594 BANK1 polymorphisms. However, there was an increase in the major alleles among RA patients. Similarly, for rs3733197, there was an increase in the major allele among patients in every cohort. Nevertheless, this skewing reached statistical significance in the Spanish (P = 0.01, odds ratio [OR] 1.17 [95% confidence interval (95% CI) 1.03-1.32]) and Argentinean (P = 0.04, OR 1.31 [95% CI 1.00-1.72]) populations. We found a significant association of rs10516487 (P = 0.005, OR 1.15 [95% CI 1.04-1.28]) and rs3733197 (P = 0.0009, OR 1.17 [95% CI 1.07-1.29]) with RA in the pooled analysis. In a 3-SNP haplotype analysis, we found that the major TGG haplotype was significantly associated with RA (P = 0.005, OR 1.14 [95% CI 1.04-1.25]). In addition, we found a common CAA haplotype that was protective against RA (P = 0.0004, OR 0.82 [95% CI 0.74-0.92]). CONCLUSION: These results suggest that BANK1 SNPs and haplotypes may contribute to RA susceptibility with a low risk.