As our knowledge of the immune system and its control expands so does the potential for the development of specific immunotherapeutic agents with few adverse or toxic actions. Selective immunosuppression to prevent graft rejection without the toxic effects of an increased susceptibility to infection or an increased incidence of cancer may be achieved if the specific cell populations responsible for each event can be well defined and then selectively affected by an immuno-therapeutic agent. Adverse toxic effects may also be avoided by carefully optimizing drug dosage and timing to affect specifically the desired immune cell population. Drug delivery to certain cell populations by specific antibody, or structural modifications of the synthetic immunosuppressive agents improve immunotherapy and result in fewer adverse effects. Toxic actions of immunotherapeutic agents such as are seen with levamisole, gold, and D-penicillamine may be related to the drug acting as an antigen and evoking an immune response to itself. Such adverse allergic reactions are much more difficult to contend with in terms of drug modification. Nonetheless, an increased knowledge of the immune system with definition of important control steps will greatly increase the potential for rational design of effective immunotherapeutic agents without immunotoxicology.