This issue of Immunological Reviews bridges two traditionally separate fields. While the average immunologist may think of “transplantation” as a single entity, it has, for largely practical reasons, been split into two separate fields, namely hematopoietic cell transplantation (HCT) and organ transplantation. In allogeneic HCT, which was originally used as a rescue therapy for the marrow toxicity of high-dose chemotherapy and/or irradiation employed to treat a hematologic malignancy, the recipient’s immune system is severely compromised and the greatest concern has been the attack of immune cells in the donor inoculum on the severely immunocompromised recipient. This attack, which caused the sometimes fatal syndrome known as graft-vs-host disease (GVHD), was so severe when HLA barriers were transgressed that until recently the practice of HCT has been largely confined to the HLA-identical or near-identical donor situation, severely limiting the availability of this curative therapy. Even in the present day, as approaches to performing HLA-mismatched HCT have permitted more widespread transplantation in this setting, the incidence of significant GVHD is in the 30–40% range in the least immunogenic combination involving HLA-identical sibling donors, despite the use of pharmacologic immunosuppression as GVHD prophylaxis (1). Thus, GVHD remains a major cause of morbidity and mortality following HCT.

While it was recognized as early as the 1970s that GVHD was mediated by T cells in the donor graft and could be eliminated by depletion of these T cells, it also became apparent in the same period that this attack was associated with a beneficial effect in patients with malignant disease (2). This effect came to be known as the graft-vs-leukemia/lymphoma (GVL) effect and is mediated in large part by the same cells that cause GVHD, namely donor T cells recognizing recipient alloantigens. While T cell depletion of the donor graft greatly attenuates GVHD, this has been associated with increased relapse rates of several malignancies (3), increased graft failure (4) and, when the intensity of conditioning is increased to promote engraftment, with a high incidence of serious opportunistic infection due to the failure of the thymus in heavily pretreated adults to rapidly generate new T cells from the hematopietic cell graft (5). Thus, the separation of GVHD and GVL has been the central challenge throughout the> 50 years that allogeneic HCT has been practiced clinically. Several advances in our understanding of GVHD and GVL have suggested ways in which this alloresponse can be harnessed for its beneficial GVL effect while minimizing GVHD. A particularly promising approach revolves around controlling the trafficking of alloreactive T cells so that they remain in the lymphohematopoietic system where the hematologic malignancy resides without trafficking into the epithelial GVHD target tissues, namely the skin, gut and liver. This trafficking into epithelial tissues requires inflammation