According to study records, IL-2 has not been tested in HIV-infected children. Experience with IL-2 in pediatric populations is extremely limited. Pahwa et al. gave 30,000 units/kg daily IV to a child with severe combined immunodeficiency. This dose was well tolerated and the patient improved clinically as well as immunologically. Part A is necessary to determine the maximum tolerated dose of IL-2 in infected children. Part B will determine the efficacy of the maximum tolerated dose in infected children.

Part A: Children will receive rIL-2 intravenously for 5 days every 8 weeks for 3 cycles. The study will enroll 4 patients in each of 3 dose levels. Dose escalation may occur if all 4 patients in a dose level tolerate therapy without evidence of Grade 3 (or higher) toxicity. If 1 of 4 subjects in any dose level experiences at least Grade 3 toxicity, 2 additional patients will be enrolled in that dose level. If 1 of these 2 additional patients experiences at least Grade 3 toxicity, dose escalation will not proceed. NOTE: Once Part A is completed and the maximum tolerated dose is established, children who participated in Part A and received less than the maximum tolerated dose will be offered additional therapy consisting of 3 cycles of rIL-2 at the maximum tolerated dose.

Part B: Children will receive rIL-2 intravenously at the maximum tolerated dose established in part A. Treatment will be given for 5 days every 8 weeks for 3 cycles. [AS PER AMENDMENT 6/4/98: Children will receive rIL-2 intravenously at the lowest dose for 5 days every 8 weeks for 6 cycles. Patients who received this dose in part A will also be offered this regimen.]