Combination of B cell biomarkers as independent predictors of response in patients with rheumatoid arthritis treated with rituximab.

OBJECTIVES

Identification of B cell biomarkers predictive of response prior to therapy with rituximab (RTX) and evaluation of the efficacy of long-term treatment in patients with rheumatoid arthritis (RA).

METHODS

302 RA patients failing one TNFi were treated with two applications of 1000 mg RTX (FIRST study). During the follow-up study (ReFIRST) the patients were treated for up to three more courses if they showed measurable clinical response but RA was still active. In a substudy on 154 RA patients peripheral B cell subsets were determined by flow cytometry before starting RTX. Rheumatoid factor (RF), RF-isotypes and anti-citrullinated protein antibodies (ACPA) were also measured.

RESULTS

Based on multivariate analyses patients with positive RF and normal (>lower limit) levels of CD19⁺ B cells (RF⁺CD19⁺) showed better treatment effects compared to patients who had only one or none of those parameters. Considering the RF status of the patients, analysis of B cell subpopulations yielded a correlation between higher ER rates and "double negative" CD19+CD27⁻IgD⁻ B cells. Lowest ER rates were observed for RF negative patients in combination with low numbers of CD19⁺CD27⁻IgD⁻ B cells as independent risk factors, thus defining a group with lower responses. Conversely, higher CD19⁺CD27⁻IgD⁻ B cells identified a responder group within RF negative patients.

CONCLUSIONS

The data of this large biomarker study suggest that beyond RF positivity, normal levels of CD19⁺ B cells together with increased CD19⁺CD27⁻IgD⁻ B cells predict response to RTX in RA, in particular when all parameters were present.