Blockade of IL-23 Signaling Results in Targeted Protection of the Colon and Allows for Separation of Graft Versus Host and Graft Versus Leukemia Effects.

Abstract 231

Separation of graft versus leukemia (GVL) and graft versus host (GVH) reactivity has been a longstanding but elusive goal in allogeneic bone marrow transplantation (BMT). The difficulty in dissociating these two responses stems in part from the fact that alloreactive donor T cells are equally capable of trafficking to sites of disease as well as GVHD target organs such as the colon, liver and skin. Recent studies from our laboratory have shown that donor APC-derived IL-23 secretion has a critical and selective role in mediating pathological damage in the colon during GVHD (Das et al, Blood, 2009). These results demonstrate the feasibility of regional GVHD protection and raise the question as to whether such localized protection might be a strategy to separate GVH and GVL responses. To directly address this question, we performed studies using a Balb/c-derived luciferase-transfected A20 leukemia cell line (A20-luc) in GVL experiments. Lethally irradiated Balb/c mice were administered 1 × 10⁵ A20 cells and then transplanted with C57BL/6 (B6) BM alone or BM and spleen cells from B6 or IL-23^−/− donors. Transplantation with BM and spleen cells from IL-23^−/− donor mice resulted in prolonged survival when compared to mice reconstituted with similarly composed marrow grafts from B6 animals (63% versus 26% at day 70, p=0.03). A five-fold escalation of the A20 cell dose produced similar protective results in recipients of IL-23^−/− marrow grafts indicating that the effects were not leukemia dose dependent. Studies were then conducted using an IL-23 specific (p19) antibody to block IL-23 signaling as a more clinically relevant strategy. To confirm that antibody blockade would also attenuate GVHD, lethally irradiated Balb/c mice were transplanted with B6 BM and spleen cells to induce GVHD. Mice that were administered p19 antibody (once/week for four weeks) had selective protection from colonic GVHD and a significant reduction in colonic proinflammatory cytokine production when compared to isotype control-treated mice, confirming results observed with use of transgenic IL-23^−/− donors. Administration of p19 antibody on a similar schedule to A20-luc-bearing animals resulted in increased survival relative to isotype control antibody-treated mice demonstrating that antibody blockade also resulted in preservation of the GVL response. Serial bioluminescence imaging (BLI) studies and pathological examination confirmed that animals treated with p19 antibody had no evidence of leukemia recurrence. Since the colon is preferentially protected after blockade of IL-23 signaling, we considered that this site might serve as a sanctuary site for residual leukemia cells. BLI studies revealed that A20-luc cells migrated to the liver, lung and nodal sites but were not detectable in the colon. For that reason, we examined this question using a novel, clinically relevant murine model of chronic myelogenous leukemia (CML). In this model, FVB animals that have the bcr/abl oncogene under the control of a tetracycline-inducible repressor are used as donor animals (i.e. CML mice). Lethally irradiated normal FVB mice were transplanted with equivalent numbers of T cell depleted BM from B6 and FVB CML animals. Withdrawal of tetracycline from the drinking water induces expression of bcr/abl oncogene, the development of neutrophilia, and leukemic infiltration in the colon, liver and spleen. Real time PCR studies also demonstrated a 600-fold increase in bcr/abl mRNA levels in the colon of these mice, relative to controls, indicative of CML in this organ. To determine whether the colon would serve as a sanctuary site and limit the effectiveness of the GVL response, lethally irradiated FVB mice were transplanted with the same BM inoculum plus 3.5 × 10⁶ T cells from wild type or IL-23^−/− donors. Whereas mice that received adjunctive B6 T cells succumbed from GVHD, animals transplanted with T cells from IL-23^−/− mice had significantly prolonged survival and no evidence of leukemia by either blood counts or pathological examination. Real time PCR studies also demonstrated absence of the bcr/abl oncogene in the colon of mice indicating that this organ did not serve as a reservoir for subsequent leukemic relapse. In conclusion, these studies demonstrate that targeting of IL-23 is able to reduce GVHD without loss of the GVL effect and suggest that this approach may be a viable clinical strategy for the dissociation of GVH and GVL responses.