Article Text

A7.15 B-cell lymphopenia is associated with a low response rate to abatacept in patients with rheumatoid arthritis
  1. P Gazeau1,
  2. Y Renaudineau2,
  3. V Devauchelle-Pensec1,2,
  4. P Pochard2,
  5. JO Pers2,
  6. A Saraux1,2,
  7. D Cornec1,2
  1. 1Service de Rhumatologie, CHRU Brest, Best, France
  2. 2EA2216, ESPRI/ERI29, Université de Bretagne Occidentale, Brest, France


Objectives Abatacept (CTLA4-Ig) prevents T-cell activation through the inhibition of B7-CD28 interaction. However, abatacept may also modulate B cells, which express B7 molecules. The objectives of this study were to evaluate if lymphocyte subset distribution before treatment may be associated with the clinical response in patients with rheumatoid arthritis (RA), and to assess the effects of abatacept therapy on the lymphocyte phenotype.

Methods We included 22 patients with RA treated by abatacept in our centre with available data on lymphocyte phenotype at the moment of abatacept initiation. Among them, 16 patients had another lymphocyte phenotyping 6 months later. Patients who previously received rituximab were excluded. Response was defined at 6 months using EULAR definition. Patients were compared to 22 age and sex-matched healthy controls. Lymphocyte phenotyping was performed on peripheral blood by flow cytometry using the following surface markers: CD3, CD4, CD8, CD56, CD19, IgD, CD27, CD38, and CD24. Absolute numbers (expressed as cells/µL) of each lymphocyte subset were computed.

Results Median age of the patients was 60 years (IQR 54–73.5), median disease duration 17 years (6–26.3), and 16 were females. We did not find significant differences in lymphocyte subset levels between RA patients and matched controls. 14 patients were responders at 6 months after abatacept. Responders and non-responders did not differ in terms of age, disease duration, sex ratio, ACPA positivity, previous therapies, concomitant steroids and methotrexate use, and total lymphocyte, T-cell (CD3, CD4 and CD8 subsets) and NK-cell counts. Conversely, responders had a higher CD19+ B cell count at baseline than non-responders: 290 (140–370) vs. 121 (76–154), p = 0.037. Using ROC curve analysis, 250 B cells/µL was the best cut-off to predict the response at 6 months, with a sensitivity of 57% and a specificity of 100%. Switched memory B cells (IgD-CD27+) but not pre-switched memory B cells (IgD+CD27+) were higher in responders than in non-responders. In the 16 patients with baseline and 6-month lymphocyte phenotyping, we did not observe any variation in the levels of total lymphocytes, T cells or NK cells after abatacept. Conversely, a significant decrease of CD19+ B cells was found in responders but not in non-responders.

Conclusion The mechanism of action of abatacept in RA seems to rely largely on the presence of B cells. Further studies are needed to understand how abatacept may modulate B-cell physiology.

  • rheumatoid arthritis
  • abatacept
  • B lymphocyte
  • transitional B cell

Statistics from

Request permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.