Background/Purpose:  BAFF is a TNF-like cytokine that supports the survival and differentiation of B cells. The anti-BAFF antibody belimumab is the only new drug that has been approved by the FDA for the treatment of lupus in the last 50 years. While it is clear that BAFF inhibition depletes B cells and can alter selection of the naïve B cell repertoire in mice, especially in the absence of competition from non-autoreactive B cells, some autoreactive VH genes are unaffected by belimumab and can subsequently enter and expand in the germinal center. A single study in humans did not show exaggerated loss of autoreactivity between the transitional and naïve compartments in belimumab treated patients. Furthermore, it is still not certain that altered selection is the mechanism by which BAFF inhibition achieves its therapeutic effect either in mice or humans. The purpose of this study was to determine how belimumab treatment alters the VH repertoire of the naïve B cell repertoire in SLE patients.

Methods:  Blood was collected from 14 SLE patients who had received and responded to belimumab for > 3years and from 10 SLE patients matched for age, ethnicity, disease duration, disease activity and other medications. Blood was also collected from 5 patients before and 6 months after starting belimumab therapy. PBMCs were analyzed by flow cytometry for B cell phenotype and naïve B cells (CD3^–/CD11b^–/CD56^–/CD19⁺/CD27^–/CD10^–) were sorted as pellets. RNA was generated from the isolated cells and libraries were made for next generation sequencing using iRepertoire primers. Multiplexed libraries were sequenced using miSeq.

Results: Naïve B cells were depleted in all chronic belimumab treated patients with up to 90% depletion in treated patients 6 months after treatment initiation. 60-100 x 10³ cells were sorted from belimumab treated patients and 350-700 x 10³ cells were sorted from control SLE patients. 120-600 x 10³reads were obtained from each sample. The number of distinct CDR3 regions was two to three-fold less in belimumab treated patients compared with either SLE or pre-treatment controls. The D50, calculated as a percent of dominant B cell clones that cumulatively account for 50% of the total CDR3s per sample was lower in the chronic belimumab patients compared with the SLE controls (p<0.02) but was not different in the pre-treatment vs. 6 month post-treatment belimumab samples. Each patient had a unique distribution of VH genes. Little difference was observed in the overall distribution of VH genes in patients examined before and 6 months after belimumab treatment. Since VH3-23 and VH4-34 are autoreactive genes it might be expected that naïve B cells expressing these genes would be preferentially deleted by belimumab therapy. However no difference was observed in the frequency of either of these two VH genes in pre vs. 6 month post belimumab treated samples or in chronic belimumab vs. SLE controls. Furthermore, belimumab did not preferentially delete the “naïve activated” B cell population that preferentially expresses VH4-34.

Conclusion:  Treatment with belimumab depletes approximately 90% of naive B cells with a modest decrease in diversity over a long time period. However at 6 months after treatment initiation, when B cell depletion has already occurred, belimumab treatment does not result in significant skewing of the overall VH repertoire. Belimumab does not induce preferential deletion of naïve B cells expressing the autoreactive VH3-23 or VH4-34 genes even after three years of treatment. These data suggest that loss of autoreactivity in the naïve B cell compartment is not a sufficient explanation for the therapeutic efficacy of belimumab.

« Back to 2016 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/analysis-of-the-naive-b-cell-repertoire-in-belimumab-treated-sle-patients/