Infants with acute lymphoblastic leukemia (ALL) have a poor prognosis. Early disease recurrence, rather than excessive toxicity and complications resulting in limitation of therapy is the major factor responsible for this disappointing outcome. The CCG-192P trial was a groupwide pilot study of the Childrens Cancer Study Group for the treatment of ALL in patients at high risk for relapse, which was defined by wbc count greater than 50 X 10(3)/microliters at diagnosis. Because of the recognized poor prognosis, all infants less than 1 year of age were entered in this study regardless of wbc count at diagnosis. Therapy included intensive induction and consolidation followed by a cyclic, sequential maintenance program. The CNS prophylaxis consisted of intrathecal chemotherapy and cranial irradiation, which was deferred until patients were greater than 1 year of age. During the period January 1982 to January 1984, 27 infants ranging in age from 2 days to 11 months who had ALL were entered in this study; 71% had wbc counts greater than 50 X 10(3)/microliters, and 23% presented with CNS leukemia. Complete remission was achieved in 93% of the patients. The median duration of remission is 17 months. With a median follow-up of 43 months, the life-table estimate of event-free survival (EFS) is 36% at 4 years. A recently reported historical control group of infants with ALL who were treated with previous Childrens Cancer Study Group protocols demonstrated a median remission duration of 8 months and an estimated EFS of only 21% at 4 years. Toxicity and therapy-related complications were not observed more frequently in infants than in older patients treated with this protocol. However, EFS of infants was significantly worse than that of patients greater than 1 year of age (P = less than 0.001). All four CNS relapses occurred in patients who had received cranial irradiation. A wbc count less than 50 X 10(3)/microliters at diagnosis demonstrated significance (P = 0.03) as a favorable prognostic indicator in this small patient sample. Although these data are preliminary, they suggest that intensive therapy is reasonably well tolerated by infants and results in prolongation of remission duration and improved EFS.