In rheumatoid arthritis (RA), B cell depletion occurs in all patients treated with rituximab, but the clinical responses to rituximab are variable. A highly sensitive assay was used to test the hypothesis that B cell depletion is variable, and that incomplete depletion leads to a poorer outcome.

Sixty patients with active RA unresponsive to anti–tumor necrosis factor agents received two 1‐gram infusions of rituximab. B cell numbers were measured by highly sensitive flow cytometry before and after each infusion and at 3‐month intervals thereafter. A reduction in B cell levels below 0.0001 × 10⁹/liter was defined as complete depletion (compared with 0.05 × 10⁹/liter by conventional cytometry). Clinical responses were measured using the European League Against Rheumatism (EULAR) criteria.

At 6 months, 92% of patients had a moderate‐to‐good clinical response according to the EULAR criteria. B cells were detected in 63% of patients after the first infusion of rituximab (median level 0.0009 × 10⁹/liter [range <0.0001–0.0015 × 10⁹/liter), and these patients had poorer clinical outcomes than patients with complete depletion. At 9 months, 82% of patients with complete depletion had a moderate‐to‐ good EULAR response, compared with 43% of those with partial depletion (P = 0.01). At 12 months, 59% of complete responders had a moderate‐to‐good EULAR response, compared with 21% of those with partial depletion (P = 0.01). Patients in whom B cells were depleted only after the second infusion did no better than those in whom depletion was never complete and had poorer clinical outcomes than those in whom depletion was initially complete.

This study is the first to show, using a highly sensitive analysis, that rituximab therapy is associated with variable diminution in B cell numbers. A lack of complete depletion of B cells after 1 infusion was associated with a poorer outcome.