With the recent remarkable progress in cell processing technology, the number of patients or types of disease that are treated with hematopoietic stem cell transplantation (HSCT) has been rapidly increasing in both autologous or allogeneic settings. Transplantable cells can be harvested from blood or the umbilical cord as well as from bone marrow, and CD34+ cells can be effectively isolated as a pure stem cell source. However, the clinical benefit of HSCT in the treatment of hematological malignancies has remained unclear. As HSCT is principally a measure of stem cell rescue after high-dose chemotherapy with or without radiation therapy, it is difficult to clarify its contribution to the cure of the disease. Given these circumstances, pediatric patients with extremely high-risk features like Ph1 ALL or 11q23 translocation are usually selected for allogeneic HSCT even in the 1st CR. Additionally, HSCT could also reasonably be applied to those in 2nd CR or later. These considerations are based upon the idea that transplant-related mortality (TRM) is still too high in allogeneic settings. Allogeneic HSCT may be the first choice for treatment of such patients, if TRM becomes low and it is scientifically proved to be effective. On the other hand, autologous HSC grafts may be contaminated with cancer cells and lack the potency of allogeneic cell-mediated immune function. To improve the clinical results in this setting, a new type of anticancer agent with lower toxicity and a new strategy including immunotherapy should be established.