The available data suggest that the antileukemic, immunosuppressive and toxic effects of TBI are highly dose dependent and if TBI is given as single daily fractions at 5-10 cGy/minute, an optimal dose for the treatment of most leukemias is somewhere between 12 and 16 Gy. Giving radiation as a single dose rather than fractionating it increases both its immunosuppressive as well as its toxic effects. The relative anti-leukemic effects of single vs. fractionated irradiation are less clear, but available data support the view that fractionation does not substantially reduce the effects of TBI on myeloid tissue. Increasing the dose rate increases toxicities and, although data are not yet complete, likely increases both immunosuppressive and anti-tumor effects. The impact of total dose, dose fractionation and dose rate are highly interdependent and how best to manipulate these 3 factors to lead to an optimal combination is as yet unknown. Directing radiotherapy to sites of leukemia using monoclonal antibodies or other carriers such as growth factors is feasible and, although there are many aspects of this approach which have yet to be worked out, targeted radiotherapy may prove to be the best way to achieve the therapeutic goals of increased tumor ablation and immunosuppression without increased toxicities.