Summary:Cyclosporin A (CsA) has been used clinically to induce graft-versus-host disease following autologous bone marrow transplantation in an attempt to destroy residual leukemia cells and reduce relapse. To analyze the antitumor potential of murine syngeneic graft-versus-host disease (SGVHD), C3H/HeN mice were lethally irradiated, reconstituted with T cell-depleted syngeneic bone marrow (ATBM) and treated with CsA for 21 days. Graft-versus-leukemia activity was assessed by challenging groups of olive oil-treated control ATBM (OO-ATBM) and CsA-treated (CsA-ATBM) mice 1 week after CsA therapy with graded doses of the syngeneic 38C13 B cell lymphoma. Following CsA treatment, up to 70% of CsA-ATBM developed SGVHD and more than 70% of the animals injected with 500 38C13 cells exhibited long-term survival (MST Ͼ80 days). In contrast, none of the OO-ATBM control mice developed SGVHD, and more than 75% of these mice died following injection of 500 38C13 tumor cells (MST = 34 days). Long-term survivors were not resistant to tumor challenge suggesting that tumor-specific immunity did not develop. Finally, class II negative 38C13 cells cultured in IL-4 or IL-10 were not inducible for MHC class II molecules, demonstrating that class II-independent antitumor mechanisms exist in SGVHD mice. Keywords: syngeneic bone marrow transplantation; cyclosporin A; graft-versus-host disease; graft-versus-tumor Bone marrow transplantation (BMT) has been used as a first-line therapy for many patients with leukemia or lymphoma. 1,2 A major cause of morbidity and mortality following allogeneic BMT is graft-versus-host disease (GVHD), the development of which correlates with the presence of alloreactive T cells present in the donor cell inoculum. [1][2][3][4][5] Such cells initiate a multifactorial set of specific and nonspecific immune responses which result in the induction of the two forms of GVHD: acute and chronic. It has been shown in animal models 6 and in retrospective analyses of clinical studies of BMT for leukemias and lymphomas, that the development of GVHD following allogeneic BMT may be beneficial and correlate with a graftversus-leukemia (GVL) response. 7,8 Allogeneic BMT recipients who develop acute or chronic GVHD have lower relapse rates than patients transplanted with autologous or syngeneic BM. 9,10 Furthermore, when patients undergoing allogeneic transplants are given T cell-depleted BM to eliminate GVHD, increased relapse rates occur, suggesting that T cells present in the donor graft are necessary for inhibition of leukemic relapse. 11 Cytotoxic T lymphocytes (CTL), 12 lymphokine-activated killer cells, 13 and natural killer cells (NK) 14 are all thought to participate in the GVL process. Recently, much effort has been expended to discriminate beneficial GVL responses from the harmful GVHD effector mechanisms. In murine models of GVHD, it has been possible to separate GVHD immune mechanisms from GVL responses. 7 However, such findings have not yet been extrapolated to man.Interestingly, rodents treated with the immun...