Background The clinical response of rituximab (RTX) is related to the degree of B cell depletion, although flow cytometry analysis of peripheral blood leukocytes after RTX therapy has shown some effects on other cell types in addition to B lymphocytes. Currently, however, data on NK cells changes after RTX are lacking.
Objectives The primary objective was to investigate the changes in lymphocyte sub-populations in real-life rheumatoid arthritis (RA) patients treated with RTX, with particular attention to NK cells and their subpopulations. The secondary objective was to analyze the correlations between lymphocyte changes and the therapeutic response of RA patients treated with RTX.
Methods In 33 RA patients diagnosed according the 1987 ACR criteria peripheral blood B and T lymphocytes and NK cells subtypes were counted by flow cytometry before and 3, 6 and 12 months after RTX administration. Patients received a 1,000-mg infusion of RTX preceded by a 100-mg methylprednisolone intravenous pulse, at baseline and at week 2. Twenty-two patients (67%) were treated with another course of RTX after 6 months from baseline. Clinical assessment was performed with the disease activity score in 28 joints (DAS28) with C-reactive protein (CRP) and the EULAR response criteria at each visit. Natural killer (NK) cells were counted using APC-CY7-conjugated anti-CD16 and PE-conjugated antiCD56.
Results RTX significantly increased CD56+3– and CD56dimCD16+ NK cells at 3 months (27% [interquartile range, IQR =46] and 33%  respectively, p≤0.001). The punctual increase persisted both at month 6 and month 12 only for CD56dimCD16+ NK cells (24% , p=0.011 and 41% , p=0.002, respectively). The 22 patients who were treated with a second course of RTX showed an increase respect month 6 values in CD56dimCD16+ cell counts, at month 12 (nonretreatment vs retreatment: - 18%  vs 24% ; p=0.002). CD56dimCD16+ absolute counts were correlated with DAS28 values at baseline (ρ=0.355, p=0.043), month 3 (ρ=0.353, p=0.044) and month 6 (ρ=0.526, p=0.002), and were higher in nonresponders than in responders (n.s.). The CD56bri16– NK cells were unaffected from RTX. No significant changes in CD3+, CD4+ and CD8+ cells were observed during the whole study period.
Conclusions In RA patients, RTX treatment is associated with a persistent increase in CD56dimCD16+ NK cells, although a trend to restoration to baseline values is seen after 12 months. A second course could mantain the initial increase. Therefore, homeostasis of CD16+ cells is somehow influenced by anti-CD20 therapy. A correlation between CD16+ NK cell changes and disease activity after RTX is probable, although the association with clinical response remains to be proved.
Disclosure of Interest None declared