Several single nucleotide polymorphisms (SNPs) within the CTLA-4 gene and elevated serum levels of soluble CTLA-4 (sCTLA-4) have been associated with autoimmunity including rheumatoid arthritis (RA). In this case-control study, we evaluated the relationship between the -319C/T (rs5742909) and CT60 G/A (rs3087243) SNPs and sCTLA-4 levels in 200 RA patients and 200 control subjects (CS) from Western Mexico. Both SNPs were genotyped with the polymerase chain reaction-restriction fragment length polymorphism technique and the sCTLA-4 levels were quantified using an enzyme-linked immunosorbent assay kit. In addition, we performed a haplotype analysis, including our previous data of the +49A/G (rs231775) SNP. The G/A genotype of the rs3087243 SNP was associated with a decreased risk of RA [odd ratio (OR) 0.61, 95% confidence interval (CI) 0.38-0.96, p = 0.024]. This protection was also observed in the negative anti-cyclic citrullinated peptide group of RA carriers of the A allele (OR 0.48, 95% CI 0.22-1.05, p = 0.042). On the contrary, we identified the -319C/+49G/CT60G haplotype of CTLA-4 gene as a risk factor for RA (OR 1.69, 95% CI 1.13-2.52, p = 0.01). The sCTLA-4 levels were not associated with RA (p = 0.377), but were correlated with the functional disability of these patients (r = 0.282, p = 0.012). However, in CS the C/T genotype of the rs5742909 SNP, as well as the G/G and G/A genotypes of the rs3087243 SNP were associated with higher sCTLA-4 levels (p < 0.001). In conclusion, our results suggest that the -319C/+49G/CT60G haplotype of CTLA-4 gene is a genetic marker of susceptibility to RA in Western Mexico, whereas the rs3087243 SNP confers protection against this disease. Moreover, both SNPs showed an effect on the sCTLA-4 production in our control population. However, further studies are required to evaluate the role of sCTLA-4 in RA, as well as the molecular and functional basis of the association between both CTLA-4 gene SNPs and soluble levels of CTLA-4 in CS.