New approaches are needed for maximizing specific responses against tumor cells resistant to chemotherapy. While cytokine therapy may amplify natural resistance against minimal residual disease, more robust anti-leukemia reactivity can be provided by allogeneic bone marrow transplantation (BMT) in conjunction with myeloablative, hence hazardous, conditioning, at the cost of graft-versus-host disease (GVHD). Documentation of the capacity of donor lymphocyte infusion (DLI) given late post BMT, when patients were off immunosuppression, in early 1987, with successful reversal of relapse and cure of patients fully resistant to maximally tolerated doses of chemoradiotherapy, with many patients alive and well >10-15 years later, indicated two important facts. First, resistant tumors are unlikely to be cured with higher doses of chemoradiotherapy that may harm the patient but not eliminate all his clonogenic tumor cells. Second, that under condition of tolerance to donor alloantigens, DLI may provide a cure to otherwise resistant patients. These observations paved the road for clinical application of non-myeloablative stem cell transplantation (NST), in the early 90s, based on a two-step procedure, first involving induction of transplantation tolerance to donor alloantigens by engraftment of donor stem cells, following safe lymphoablative rather than myeloablative conditioning. Second, use of donor lymphocytes for elimination of residual tumor or otherwise abnormal hematopoietic cells by immune-mediated graft-versus-host effects inducible by mobilized blood stem cell allografts containing larger inocula of donor T cells, or supported by post-grafting DLI when patients were off immunosuppressive modalities.