Now that it has been clearly established that tumor-associated antigens exist in acute leukemia in man, as in animals, the possibility of stimulating the patient's immune system to react against them arises. In animal experiments the most effective method of influencing the progress of leukemia after the implantation of living malignant cells has been a combination of nonspecific stimulation of the reticuloendothelial system, with agents such as BCG or Corynebacterium parvum, either with chemotherapy or with specific immunization with irradiated leukemic cells. However, such treatment is only effective if the number of living malignant cells is small as it takes a powerful immune response to overcome even a small number of malignant cells. It is for these reasons that most of the studies in man have been on patients with acute leukemia in remission. Mathé, in 1969, produced evidence that BCG and irradiated allogenic leukemia cells could lengthen the duration of remission in ALL in children. However, later results of intensive combination chemotherapy, together with prophylactic treatment of the central nervous system by Pinkel and his colleagues, were so encouraging that immunotherapy is not felt to be needed and therefore is not being extensively used in this form of leukemia at the moment. The situation in AML, particularly in adults, is completely different. The maintenance of remission with chemotherapy in this type of leukemia is difficult and relapses occur quite rapidly. Various centers have now shown that both remission lengths and overall survival are significantly prolonged by using BCG with or without irradiated allogenic leukemia cells.