Infection during the period of bone marrow aplasia remains one of the major risks associated with high-dose chemotherapy and transplantation. Over the past several years, a number of investigators in Europe and North America have evaluated the use of GM-CSF in the setting of autologous bone marrow transplantation. These studies have almost all shown a hastening of myeloid engraftment. This, for the most part, has led to fewer serious infections and a decreased hospital stay for the GM-CSF treated patients. An overall survival advantage has not been noted. There has also not been any consistent multi-lineage effect. Future trials with combinations of sequentially used cytokines may lead to more rapid recovery of red blood cells and platelets in addition to granulocytes.