The fundamental obstacle to the successful application of partially HLA-mismatched related donor, or HLA-haploidentical stem cell transplantation, is the strength of the host and donor T-cell response to allogeneic HLA molecules, which results in increased incidences of graft failure, GVHD, and nonrelapse mortality. The holy grail of haplo-SCT is to mitigate host-versus-graft and graft-versus-host responses while preserving immune responses to infection and the patient's malignancy. Two strategies have been taken to achieve this goal. The first strategy is to supplement a T cell-depleted graft with pathogen-specific T cells or populations of T cells in which alloreactivity can be controlled. The second strategy is to eliminate alloreactive T cells selectively from a T cell-replete graft. Substantial progress has been made with both approaches so that the safety of haplo-SCT now approaches that of SCT using grafts of umbilical cord blood or from HLA-matched donors. In light of the rapid and near universal availability of HLA-haploidentical related donors, it should now be possible to identify and mobilize a donor for every patient referred for allogeneic SCT. Prospective comparisons between haploidentical SCT and unrelated donor SCT should be performed to identify the most efficacious approach to alternative donor transplantation.