DLI: Where are we know?

Abstract

Allogeneic stem cell transplantation has induced lasting remissions in patients with leukaemia, lymphoma and myeloma otherwise refractory to chemoand radiotherapy. Patients with acute and chronic graft-versus-host disease (GVHD) have a lesser risk or relapse of disease than those without GVHD. Patients given T-cell depleted grafts have a lower risk of GVHD and a higher risk of rejection of the graft and of relapse. The dilemma of either risk of GVHD or leukaemia relapse can be solved in animal models by transfusion of donor lymphocytes at a time when chimerism and tolerance is established. Here the evidence for a graft-versus-leukaemia (GVL) effect is reviewed and ways of improving the GVL effect and overcoming mechanisms of immune escape are discussed. Donor lymphocyte transfusions (DLT) have been introduced in the treatment of leukaemia relapse after allogeneic stem cell transplantation in the late 1980’s and the best results have been observed in patients with chronic myelogenous leukaemia (CML). The first patients are stil alive and in remission. In acute myeloid leukaemia (AML) and myelodysplastic syndrome (MDS) results were inferior and in acute lymphatic leukaemia (ALL) results were generally poor. In multiple myeloma DLT could induce remissions, but the duration of the responses has generally been limited. There have been reports on the successful treatment of low grade lymphoma, chronic lymphocytic leukaemia and Hodgkin’s disease with DLT, but in general the results are controversial. In most patients of these disease groups conditioning treatment was not myeloablative and patients received grafts of mobilized blood cells containing a high proportion of T-cells. Relapses may have already escaped immune mechanisms of T-cells when they occurred. The hypothesis that dendritic cells of leukaemia origin are formed in CML and AML has been proven by in vitro culture of leukaemia cells in the presence of GM-CSF and interferon-alpha or IL-4. Dendritic cells had the leukaemia karyotype. Consequently treatment of recurrent CML with DLT was combined with GM-CSF and IFN-a in patients with advanced relapse or not responding to DLT alone. Responses were observed in 6 out of 9 patients treated. In AML the combination of low dose cytosine arabinoside (LD-AraC) with mobilized blood cells and GM-CSF followed by DLT in cases without GVHD induced remission of more than 4 and 8 years. The long-term responses were limited to patients that responded to a short course of LD-AraC. The use of GM-CSF is currently studied also in other diseases as multiple myeloma and lymphoma with persistence or relapse. Another possibility to activate T-cells in the GVL reaction is the use of bispecific antibodies. BI20 is a bispecific antibody of rat and mouse origin with one binding site to CD3 and the other to CD20, the Fc part binds to macrophages and dendritic cells inducing phagocytosis of the lymphoma cell. This antibody has been shown to induce CTL against CLL in vitro and depletes leukaemia cells in the blood. However the induction of cellular immunity has not been shown yet in vivo. Finally unlike T cells NK cells are able to kill leukemia and lymphoma cells not expressing HLAantigens. They are inhibited by their own HLAantigens respectively, HLA-associated killer inhibitor receptor ligands. In the HLA-haploidentical situation they are not inhibited, if the recipients leukaemia cells do not express the KIR Ligands of the donor. We have designed a treatment protocol with unmodified mar-