Recovery of Circulating T-Follicular Helper Cells after Hematopoietic Stem Cell Transplantation

T-Follicular Helper Cells (TFH) interact with B-cells in the germinal center to support B-cell proliferation, differentiation and immunoglobulin gene class switch. TFH have been reported to alter B-cell homeostasis in autoimmune diseases (Morita et al, Immunity 2011) and to contribute to murine chronic graft-vs.-host-disease (GVHD) (Flynn et al, Blood 2014). However, the role of TFH in immune reconstitution and GVHD after human hematopoietic stem cell transplantation (HSCT) has not been studied. Although TFH function primarily in lymphoid tissues, these cells have also been identified in peripheral blood. We analyzed fresh blood samples from 98 patients obtained 2 to 97 months after HSCT, and 11 Healthy Donors (HD). We defined circulating TFH (cTFH) as CD4⁺CD45RA^-CXCR5⁺ T-cells. cTFH were then subdivided into Th1, Th2, Th17 subsets (Th1: CXCR3⁺CCR6^-, Th2: CXCR3^-CCR6^-, Th17: CXCR3^-CCR6⁺) as previously described (Morita et al).

Analysis of patient peripheral blood revealed a progressive increase in the cTFH absolute count from 2 months to 2 years after HSCT, which then reached a plateau. Nevertheless, cTFH counts remained below normal values when compared with HD throughout this period, and only reached normal levels 2 years after HSCT in parallel with recovery of total CD4+ T cells (Figure 1). In contrast, the fraction of cTFH within the CD4 T cell compartment reached normal values within 6 months after HSCT and remained stable.

To study the role of cTFH in cGVHD, we studied 71 patients (excluding patients receiving prednisone >30mg/day, rituximab or IL-2). In patients with active cGVHD, cTFH frequency (median 8.78% vs. 10.5% of CD4+ T-cells, p=0.0065) and cTFH absolute counts (48.75 vs. 38.3/΅L, p=0.07) were decreased compared with patients without cGVHD. cTFH subsets are reported to have different functional capacities and this was confirmed by purifying and culturing each subset with naive B-cells. In HD, we observed that only Th2 and Th17 subsets of cTFH promoted B-cell IgG and IgM secretion (Figure 2). In cGVHD patients, we observed a selective decrease of the Th1 compartment in the cTFH subset (16.2% and 10.9% for No vs. cGVHD, p=0.034), and relative increase of both Th2 and Th17 cTFH (Th2+Th17= 71.15 and 78.1% respectively, p=0.07) leading to a greater (Th2+Th17)/Th1 ratio (p=0.044) (Figure 3). We did not find any differences in cTFH subsets when comparing Mild vs. Moderate and Severe cGVHD. To further characterize cTFH, we analyzed BCL-2 and CD95 apoptosis pathways. Th2 and Th17 cTFH subsets appeared to express lower levels of CD95 and higher levels of BCL-2 compared to Th1 cTFH.

Increased representation of Th2 and Th17 cTFH subsets that are relatively resistant to apoptosis may be a mechanism that promotes B-cell deregulation and pathologic antibody production in patients with cGVHD.  Targeting Th2/Th17 cTFH or TFH-B cell interactions may provide novel therapies in patients with cGVHD.