LAUSR

Steroids are the first line therapy for acute GVHD but 50% of patients develop steroid refractory GVHD (SR-GVHD). The development of therapies for SR-GVHD is limited by incomplete understanding of its pathophysiology. Here we generated SR-GVHD murine model by developing an MHC matched, multiple minor histocompatibility antigens (miHAs) mismatched BMT and determined whether donor T cells plays an important role in mediating SR-GVHD. To this end, we utilized C3H.swaB6 model. Recipient B6 animals were lethally irradiated and transplanted with splenic T cells along with BM cells from either syngeneic or allogeneicdonors. Dexamethasone (DEX) was administered from day +7 to day +21 at a dose of 0.1mg/kg. We observed three phenotypes in response to DEX: 1) steroid refractory (REF), characterized by progressive GVHD with severe peak GVHD clinical scores, progressive weight loss and early mortality; 2) stable GVHD (ST), characterized by moderate GVHD clinical scores, body weight loss, and minimal late mortality; 3) steroid responsive (RES) GVHD, characterized by near complete remission of GVHD clinical scores, weight loss, and no mortality. These criteria clearly separated treatment animals into three groups and resulted in significantly higher day +21 GVHD specific histopathological scores of GVHD target organs in REF relative to RES animals. We next sought to determine whether steroid refractoriness can be modeled at a later time-point, to account for any contribution from radiation toxicity. To achieve this, we utilized the same model except that DEX was administered from day +21 to +35. We again obtained similar overall survival and GVHD specific histopathological scores in target organs as in our early steroid treatment model. We next determined whether the steroid refractoriness correlated with changes in donor T cells. However, we found no difference in the proportion of donor activated, effector or memory T cells among groups. In addition, there was no difference in absolute number or percent of donor splenic T cells or their subsets including Tregs and exhausted T cells. Furthermore, there were no differences in serum inflammatory cytokines on day 7 or 14 of DEX treatment in the early and late steroid models, respectively. Because target tissue intrinsic mechanisms are critical for initiating GVHD, we tested whether gut homeostasis related proteins were altered in SR-GVHD and found that the expression of CD324 in gut epithelial cells is significantly increased in SR-GVHD. In conclusion, we established clinically relevant SR-GVHD murine models. Donor T cell characteristics were not significantly different between groups. However, CD324 expression in gut epithelial cells is dramatically increased in refractory animals. These data suggest that donor T cell independent mechanisms may also contribute to steroid refractoriness more so than was previously considered.