Treatment for acute lymphocytic leukemia (ALL) has been successful in producing cures for some children with that disease. It has become increasingly important therefore to develop treatment regimens for curable ALL that are effective for disease control but carry minimal risks of late detrimental consequences. For other children current therapy is still inadequate and treatment failure occurs. For these children altered treatment strategy must be developed that includes intensification of some treatment components, since improved methods of disease control are the primary consideration. In this paper the clinical and laboratory methods by which risk groups can be defined at diagnosis have been described. We have increasingly effective means by which the expected treatment outcome with current therapy can be defined at diagnosis. This assignment to a specific risk group is important to provide children with ALL with the most appropriate therapy for their expected clinical response. These studies of clinical utility also pose biological questions concerning the relationship of such things as white blood count, leukemic cell karyotype, and initial proliferative activity to the likelihood of drug resistance emerging. During the coming decade there will no doubt be an increasing emphasis on a biological characterization of leukemic cell populations. Answers must be developed to such questions as the mechanism for leukemic cell growth regulation. Biological information developed in this model tumor cell system in humans will likely produce information of value in understanding human cancer in general.