Biological disease-modifying anti-rheumatic drugs (bDMARDs) have changed care of patients with rheumatoid arthritis (RA). However, bDMARDs are costly, can lead to serious infections, and induce a sustained remission in only 30% of RA patients. In this study, we sought to determine if the clinical response to treatment with Tocilizumab (TCZ), an IL-6 inhibitor, varied with genetic background. The efficacy of TCZ was assessed using the European League Against Rheumatism (EULAR) response criteria, measured after 3 months of treatment in two samples of French RA patients (TOCI and ROC studies). Single nucleotide polymorphisms (SNPs) in 21 candidate genes were genotyped using KasPar method (LGC-genomics, UK) and then analyzed to determine their contribution to variation in the response to treatment. One hundred twenty-three patients in the TOCI group (79.8%) and 48 patients in the ROC group (80%) experienced good or moderate EULAR response. The clinical response to treatment was associated with SNP genotype in the gene IL6R, with patients with the homozygous AA-genotype for rs12083537 (IL6R) showing a significantly better response than homozygous or heterozygous patients with the G allele [TOCI: 87.5% of responders for AA genotype vs. 72.2% for AG or GG genotype (p = 0.018); ROC patients: 89.2% of responders for AA genotype vs. 65.2% for AG or GG genotype, p = 0.044]. A meta-analysis combining data from the two cohorts confirmed the lower response rate in patients carrying a copy of the G allele (OR (95% CI) = 0.35 (0.16-0.61), p = 0.001). No association was found with any of the other SNPs tested.