Background/Purpose: B and T cells contribute to tissue injury in rheumatoid arthritis (RA). Rituximab (RTX), an anti-CD20 monoclonal antibody, is used for the treatment of RA and results in rapid B-cell depletion. However, the therapeutic effect of RTX may not be entirely contingent on depletion of B cells; rather, we hypothesized a contributing factor is the dampening of their interplay with the T-cell drivers of their maturation in secondary lymphoid organs. The latter follicular helper T (Tfh) cells are represented in the blood as a circulating (cTfh) pool. Thus, we examined effects of RTX on cTfh cells in RA patients treated either for the first time (“naïve”) or on maintenance therapy (i.e.; received at least once prior; “maintenance”).

Methods: Blood was drawn from 5 naïve and 11 maintenance patients prior to the first infusion, at 2 weeks (prior to 2^nd infusion, if given) and 12 weeks. Peripheral blood mononuclear cells were stained with antibodies conjugated with fluorochromes for flow cytometry—CD4, CD3, CD45RA, CXCR5, PD-1, ICOS, CCR7, CD127, CD25, TCRβ, CD19, IgD, CD38, CD27, PSGL-1, CD138, and CD5—to determine naïve, activated, and memory CD4 T cells, cTfh and cTfh subsets (cTfh1, cTfh2, cTfh17) and conventional CD4 T helper cell subsets (Th1, Th2, Th17). Naïve and memory B cells, and plasma cells were also enumerated. Data were analyzed using FlowJo® software. Plasma was assessed for anti-citrullinated peptide antibody (IgG) titers (ACPA) using QuantaLite® ELISA kit. Disease activity (DA) was determined by chart review.

Results: Naïve compared to maintenance patients had more cTfh cells (CXCR5⁺PD-1^hiICOS^hi; 5.65 ± 0.68% vs. 2.67 ± 0.36%; among CD4 T cells), with a significant linear relationship between the percentage of cTfh and that of CD19⁺ B cells (Pearson’s correlation coefficient 0.52); more cTfh17 cells (CXCR3^–CCR6⁺; 29.5 ± 2.79% vs. 21.42 ± 1.16%; within the CD4⁺CD45RA^–CXCR5⁺ population); and more PD-1^lo memory cTfh17 cells (35.42 ± 3.32% vs. 21.88 ± 1.27%), although their percentages and ACPA titers did not change over the course of therapy. Th17 cells were differentiated from cTfh17 cells by their expression of CXCR5^– rather than CXCR5⁺. Naïve compared to maintenance patients also had more Th17 cells (CD4⁺CXCR5^–CXCR3^–CCR6⁺; 19.35 ± 1.54% vs. 14.27 ± 1.08%), in both the PD-1^hi effector (16.38 ± 1.82% vs. 11.89 ± 0.87%) and PD-1^lo memory subsets (20.77 ± 1.72% vs. 13.91 ± 1.07%). All naïve patients had moderate-to-high DA at baseline as compared to only 27% of maintenance patients; post-RTX, all naïve patients had either low DA or clinical remission as compared to 82% of the maintenance patients.

Conclusion: We found a higher percentage of cTfh cells in naïve compared to maintenance RA patients treated with RTX, consistent with prior studies showing that B-cell help is needed for Tfh-cell development. We also found a higher frequency of cTfh17 and Th17 cells in naïve compared to maintenance patients and correspondingly higher DA at baseline in naïve patients, suggesting that B-cell depletion results in reduced cTfh and Th17 cells, as DA improves. Future studies are needed to better understand these cTfh subsets and their correlation to B-cell depletion and DA in RA, as well as their potential as therapeutic targets.

« Back to 2018 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/t-follicular-helper-cell-phenotype-in-ra-patients-receiving-rituximab/