Responses to donor lymphocyte infusion for acute lymphoblastic leukemia may be determined by both qualitative and quantitative limitations of antileukemic T-cell responses as observed in an animal model for human leukemia

Objective Donor lymphocyte infusion (DLI) after allogeneic stem cell transplantation is largely unsuccessful in the treatment of acute lymphoblastic leukemia (ALL). To allow identification of the causes of this failure, we established an animal model of DLI in human ALL. Methods NOD/scid mice were inoculated with primary human ALL cells. Cells from five different patients were studied. After engraftment, DLI was performed by infusion of human leukocyte antigen (HLA)-identical donor T cells or HLA-disparate donor T cells. Results DLI resulted in expansion of activated, leukemia-reactive T cells in all donor–patient combinations. After 40 days of expansion, T cells abruptly declined in numbers and displayed loss of cytotoxicity. At this moment, remissions were observed in three of five donor–patient combinations. In animals engrafted with the two unresponsive ALL, remissions could be achieved when HLA-disparate DLI was given. Conclusion Our results indicate that the inefficacy of DLI in ALL may be due to the limitation of the proliferative capacity of ALL-reactive T cells and that the antileukemic efficacy during the limited time span of proliferation depends on the antigenic disparity between the donor and the patient. The model can be used to study whether alternative strategies may result in more sustained antileukemic responses after DLI in ALL.