T lymphocytes contained within an allogeneic stem cell transplant for leukemia are a force for both good and evil. T cells significantly decrease the incidence of disease relapse by killing leukemic cells (graft-versus-leukemia effect, or GVL) while simultaneously causing often life-threatening graft-versus-host disease (GVHD). Whether these processes are fundamentally different enough that GVHD can be abrogated while GVL is preserved, by such approaches as depletion of T-cell subset and manipulation of cytokines thought to induce (eg, TNF) or inhibit (eg, IL-11) the inflammatory GVHD response, has been uncertain. Schmaltz and colleagues (page 2886) now show that distinct T-cell effector pathways can preferentially cause GVL, but not GVHD. Using a mouse parent–into–F1 bone marrow transplantation model, congenic strains deficient in FasL or the granule protein perforin, and 2 myeloid leukemic cell lines, they find that the Fas/FasL pathway is critical to the appearance of GVHD but not to antileukemic T-cell killing. Even though both leukemic lines express Fas and can be killed by cell lines expressing FasL in vitro, GVL does not require Fas/FasL in vivo. Perforin-deficient T cells, in contrast, provide little detectable GVL but still cause GVHD.
These findings, if validated in other stem cell transplantation models, suggest that therapies aimed at blocking the Fas/FasL pathway after T-cell–replete stem cell transplantation might provide sufficient therapeutic index to blunt GVHD without preventing GVL. Whether such treatments would need to be continued for months or years and whether there would be any unexpected infectious or dysplastic side effects caused by blockade of T-cell FasL surveillance would need to be explored as well. Schmaltz and colleagues' studies now illuminate the effector phase of the immune response as a novel opportunity for decisive interventions in the battle to increase the safety and efficacy of stem cell transplantations.
