Prevention of Acute GvHD During MHC Haploidentical BMT: Evaluating the Efficacy of T-Cell Costimulation Blockade Using a Novel Rhesus Macaque Transplant Model.

While hematopoietic stem cell transplantation (HSCT) offers a cure for many hematologic diseases, it remains plagued by often fatal graft-versus-host disease (GvHD). Despite the inadequacy of current GvHD prevention strategies, especially for MHC-mismatched HSCT, the pace of the clinical introduction of novel therapeutics has been slow, likely due to the lack of a suitable translational model to rigorously test the immunologic and clinical impact of novel biologic therapies. Among the most promising of these therapies include those that block T cell costimulation blockade. While they have been used for both autoimmune disease and to prevent rejection of solid organ transplants, costimulation blockade reagents have not yet been evaluated for efficacy in preventing clinical GvHD. Here we describe a novel primate model of MHC-mismatched GvHD, that has allowed us, for the first time, to evaluate the mechanisms controlling GvHD in a primate translational system, and to evaluate the efficacy of costimulation blockade for the prevention of primate GvHD, even across haplo-MHC barriers.

Using DNA microsatellite-based pedigree analysis and MHC haplotype determination, we have developed the first MHC-defined Rhesus macaque HSCT system. MHC haplo-identical transplant pairs were chosen, and recipients prepared for transplant with TBI (8 Gy, as a single dose, with lung shielding to 6 Gy). Animals were either treated with no immunosuppression post-transplant (controls) or with a costimulation blockade-based regimen which included CD28/B7 blockade with abatacept (20mg/kg every 7 days), CD40/CD154 blockade with the 3A8 anti-CD40 monoclonal antibody (maintenance dosing at 5mg/kg twice weekly) as well as sirolimus to maintain serum trough levels between 5-10 ng/mL. Either leukopheresis-derived peripheral blood stem cells or bone marrow was used for transplant (average total nucleated cell dose = 9.3 +/-2.7×10⁸/kg; average CD3+ cell dose = 1.1 +/- 0.88 ×10⁸/kg) Donor engraftment was measured by microsatellite analysis, and GvHD was graded clinically using standard scales. The immune phenotype after transplant was determined by multicolor cell- and serum-based flow cytometric analyses.

Seven haploidentical transplants have been completed. Three controls received no immunosuppression. These animals demonstrated rapid and complete donor engraftment, with donor T cell activation and proliferation occurring within one week of transplant, coincident with the onset of severe clinical GvHD, which predominantly targeted the GI tract. Flow cytometric analysis showed loss of CD127 expression on both CD4+ and CD8+ T cells, consistent with their rapid clonal expansion and differentiation. Multiplexed luminex cytokine analysis demonstrated high-level secretion of the inflammatory cytokines IFNγ, and IL18, as well as the counter-regulatory cytokine IL-1RA. Importantly, no rise in TNF, IL-1b, nor IL17 was measured despite severe GvHD.

In contrast, four treated animals received a haplo-identical BMT in the setting of abatacept/anti-CD40 and sirolimus for GvHD prophylaxis. All of these recipients demonstrated rapid donor engraftment, but, unlike the controls, they were protected against clinical GvHD—they displayed neither the skin rash nor the profuse diarrhea noted in the control animals. Flow cytometric analysis demonstrated maintenance of CD127 expression on both CD4+ and CD8+ T cells. Furthermore, luminex analysis revealed that expression of IFNγ, IL18 and IL-1RA were all normal in the setting of GvHD prophylaxis with costimulation blockade and sirolimus.

We have established a robust model of haplo-identical HSCT and GvHD using an MHC-defined Rhesus macaque colony. This model has allowed us to begin to determine the mechanisms underlying GvHD during primate haplo-identical BMT and to assess the efficacy of novel regimens to prevent this disease. We find that unprotected primate GvHD is characterized by rapid T cell proliferation, with concomitant loss of expression of CD127 on both CD4+ and CD8+ T cells. In addition, it is associated with a cytokine storm, including high level secretion of IFNγ, IL18 and IL-1RA into the serum. Finally, we find that CD28/CD40-directed costimulation blockade in combination with sirolimus can effectively inhibit both the clinical and cellular hallmarks of GvHD during haplo-identical BMT, and thus may deserve close clinical scrutiny as a possible prophylaxis strategy during these high risk transplants.

No relevant conflicts of interest to declare.