The clinical efficacy and safety of interleukin-6 (IL-6) receptor blockade have been well studied, but the data on the impact of therapeutic inhibition of IL-6 on B cells are scarce and contradictory. Preliminary reports have shown that B cell function and a humoral immune response may be modulated by an IL-6 receptor inhibitor.Objective: to assess the effect of 12-month tocilizumab (TCZ) therapy on B-cell phenotype and gene expression in RA and to analyze the association between B-cell subsets and RA activity.Subjects and methods. Examinations were made in 24 active RA patients (20 women and 4 men) (median age, 55 [49; 64] years; disease duration, 72 [24; 108] months; DAS28 5.8 [5.3; 6.3]; the patients were seropositive for rheumatoid factor (RF) (100%) and for anti-cyclic citrullinated peptide antibodies (87.3%). The patients received TCZ 8 mg/kg every 4 weeks. After 12 months of therapy, 54% of patients were categorized as good responders, 46% as moderate responders according to the EULAR response criteria. A control group consisted of 29 volunteers (21 women and 8 men; median age, 58.5 [53.0; 62.0] years). Peripheral blood lymphocytes were immunophenotyped at the time of enrollment and after 12 months. The absolute and relative counts of CD19+B lymphocytes, memory B cells (CD19+CD27+), non-switched memory B cells (CD19+IgD+CD27+), switched memory B cells (CD19+IgDCD27+), naive (CD19+IgD+CD27-), double-negative (CD19+IgD-CD27-), transitional (CD19+IgD+CD10+CD38++CD27) B cells, plasma cells (CD19+СD38+), and plasmablasts (CD19+СD38+++IgD-CD27+CD20-) were estimated using multicolor flow cytometry. Results and discussion. The relative and absolute counts of memory B cells (CD19+CD27+) (1.3 [0.9; 1.7]%, 0015 [0.001; 0.003]•109/l), switched memory B cells (CD19+IgD-CD27+) (6.8 [3.6; 11.6]%, 0.01 [0.005; 0.02]•109/l), and the absolute number of transitional B cells (CD19+CD38++CD10+IgD+CD27-) (0.00009 [0; 0.00028]•109/l) were found to be lower in RA patients than in donors: 2.2 [1.1; 3.0]%, 0.003 [0.001; 0.007]•109/l; 12.8 [9.3; 17.0]%, 0.02 [0.01; 0.04]•109/l; 0.0001 [0; 0.0003]•109/l, respectively (p<0.05 for all cases). After 12 months of TCZ therapy initiation, there were decreases in the relative and absolute counts of plasmablasts (CD19+CD38+++CD27+IgD-CD20-) from 0.15 [0.1; 0.3] to 0.1 [0.01; 0.1]% and from 0.0003 [0.00007; 0.004]•109/l to 0.0001 [0; 0.0003]•109/l, respectively (p<0.05). At the same time, the relative and absolute counts of memory B cells (CD19+CD27+) and switched memory B cells (CD19+CD27+IgD-) remained lower in RA patients than in donors: 1.0 [0.7; 1.2] and 2.2 [1.1; 3.0]%; 0.001 [0.006; 0.003]•109/l and 0.003 [0.001; 0.007]•109/l; 3.1 [1.1; 4.2] and 12.8 [9.3; 17.0]%; 0.003 [0.002; 0.006]•109/l and 0.02 [0.01; 0.04]•109/l, respectively (p<0.05 for all cases). Following 12 months of TCZ therapy, the numbers of other B-cell subpopulations were not considerably altered. When included in the study, the patients with RA showed correlations between the absolute count of memory B cells (CD19+CD27+) and the level of C-reactive protein (r=0.50; p<0.05); between the absolute count of plasmablasts (CD19+CD38+++CD27+IgD-CD20-) and the level of RF (r=0.41 and r=0.52; p<0.05). There were no correlations of B cell subsets with clinical and laboratory findings after 12 months of TCZ initiation.Conclusion. Immunophenotyping of peripheral blood B lymphocyte subsets showed the lower relative and absolute counts of memory B cells (CD19+CD27+) and switched memory B cells (CD19+CD27+IgD-) in RA patients than in healthy donors. The found correlations between the counts of memory B cells and plasmablasts and the values of laboratory parameters in patients with high RA activity may suggest that B lymphocytes are involved in the pathogenesis of RA. There was a decline in plasmablast levels after 12 months of TCZ therapy.