A Functional Assay for Peripheral Blood T Cells to Predict Acute Graft-Versus-Host Disease.

Abstract 4497

Patients undergoing allogeneic hematopoietic cell transplantation (allo-HCT) are at high risk to develop acute graft-versus-host disease (aGVHD) which is caused by alloreactive donor T cells. Early diagnosis of aGVHD remains difficult: there are no efficient methods to identify patients at risk of aGVHD, which could improve disease prevention. Therefore, to potentially predict aGVHD, we asked whether it is possible to detect in vivo activated alloreactive T cells in the peripheral blood immediately before entering aGVHD target tissues.

To address this question, we used the CD107a/b degranulation assay (Betts et al., J Immunol Methods 2003 281:65) to measure cytotoxicity responses of alloractive T cells in a mouse model of aGVHD that allows us to trace alloreactive T cells in different phases of aGVHD by flow cytometry. Utilizing bioluminescence imaging in this model we have previously observed two distinct phases in aGVHD pathophysiology (Blood 2005;106:1113): During the initiation phase until day+3 alloreactive T cells are activated, proliferate in secondary lymphoid organs and acquire appropriate homing receptors to migrate into target tissues during the effector phase.

In preparation for transplantation experiments, we tested T cell receptor transgenic (OT-1) T cells, which recognize the SIINFEKL-H-2b complex as a positive control. We observed degranulation of in vitro activated OT-1 T cells (84.40 ± 1.10%) against SIINFEKL-H-2b⁺ targets in contrast to C57Bl/6 wildtype targets (53.17 ± 2.65%). Unstimulated OT-1 T cells degranulated to a lower extend (22.93 ± 1.05%) against SIINFEKL-H-2b⁺ targets but less against SIINFEKL negative targets (9.47 ± 0.56%) (t-tests p<0.0001).

We then transplanted 1,2×10⁶ C57Bl/6 CD4⁺ and CD8⁺ T cells (H-2^b, Thy1.1⁺) plus 5×10⁶ C57Bl/6 bone marrow (BM) cells (H-2^b, Thy1.2⁺) into myeloablative conditioned allogeneic Balb/c (H-2^d, Thy1.2⁺, 8 Gy) recipients. As syngeneic controls we used C57Bl/6 (H-2^b, Thy1.2⁺, 9 Gy) recipients. To prove specificity of the functional assay we transplanted 1,2×10⁶ TCR transgenic OT-1 T cells (H-2^b, Thy1.2⁺) plus 5×10⁶ C57Bl/6 BM into conditioned C57Bl/6 (H-2^b, Thy1.2⁺, 9 Gy) mice that expressed the SIINFEKL-H-2b⁺ complex ubiquitously.

On day+2 and day+5 after HCT we analyzed the peripherial blood of the transplanted mice. As expected we did not detect donor T cells in the peripheral blood on day+2 during the GVHD initiation phase. However, by day+5 at the transition to the aGVHD effector phase large numbers of T cells had entered the circulation. Employing the degranulation assay revealed that peripheral blood CD8⁺ T cells from allogeneic recipients (C57Bl/6 □ Balb/c) degranulated against allogeneic targets (29.08 ± 4.46%), antigen specific CD8⁺ OT-1 T cells against SIINFEKL expressing targets (43.10 ± 3.78%) but less against SIINFEKL negative targets (9.78 ± 2.64%) (ANOVA p<0.0001). Reactive T cells in syngeneic controls were negligible (< 3%).

Importantly, subsequent histopathological analysis of the same allogeneic recipients (where alloreactive T cells had degranulated) revealed aGVHD of the intestinal tract (grade 2-3) and the liver (0-2). No signs of GVHD were observed in mice where T cells had not degranulated (syngeneic controls). These preclinical data from our in vivo experiments encourage translation of this predictive test to patients undergoing allo-HCT.