RATIONALE: Drugs used in chemotherapy use different ways to stop tumor cells from dividing so they stop growing or die. Combining more than one drug may kill more tumor cells.

PURPOSE: This randomized phase III trial is studying two different combination chemotherapy regimens and comparing how well they work in treating children with low-grade astrocytomas or other residual tumors of the brain.

OBJECTIVES:

• Compare the event free survival as a result of treatment with carboplatin and vincristine versus thioguanine, procarbazine, lomustine, and vincristine in children with progressive brain tumors.
• Estimate tumor response rates to each regimen of chemotherapy in these patients.
• Determine toxic effects and quality of life of children treated with each regimen of chemotherapy.
• Investigate biological and clinical factors which may predict tumor response and early progression (tumor size, location, pathologic subtype, cytogenetics, and proliferative index by MIB-1 (Ki67)) in these patients.
• Investigate factors contributing to neuropsychological and endocrine status of children with brain tumors treated without irradiation.

OUTLINE: This is a randomized study. Patients are stratified according to site of disease, status at entry, and pathology. Patients are randomized to one of two treatment arms. Patients with neurofibromatosis are nonrandomly assigned to arm II.

• Arm I: Patients receive induction with carboplatin and vincristine for 10 weeks followed by 2 weeks of rest. Induction is followed by 8 courses of maintenance beginning on day 84 of induction or upon hematopoietic recovery. Each course consists of 4 weekly doses of carboplatin and 3 weekly doses of vincristine (given concurrently with the first 3 weeks of carboplatin), followed by 2 weeks of rest.
• Arm II: Patients receive oral thioguanine, procarbazine, and lomustine on days 0-4, followed by vincristine IV on days 14 and 28. Treatment continues every 6 weeks for a maximum of 8 courses.

PROJECTED ACCRUAL: A total of 280-340 patients will be accrued over 4 years.

DISEASE CHARACTERISTICS:

• Pathologically confirmed low grade residual astrocytomas or other eligible residual tumors of the brain interpreted as low grade (WHO grades I and II) such as the following:

  • Glial Tumors

    • Astrocytic tumors

      • Low grade astrocytoma (variants: fibrillary, protoplasmic, gemistocytic)
      • Pilocytic astrocytoma
      • Pleomorphic xanthoastrocytomas
      • Subependymal giant cell astrocytoma
      • Infantile desmoplastic astrocytoma

    • Low grade oligodendroglial tumors

      • Low grade oligodendroglioma

    • Low grade mixed gliomas

      • Oligo-astrocytoma

  • Neuronal Tumors

    • Ganglioglioma (excluding tumors with anaplastic astrocytic components)
    • Infantile desmoplastic ganglioglioma

  • Chiasmatic-hypothalamic tumor without histologic confirmation

• All of the following diagnostic tests (radiological or clinical evidence of progression, surgery, or confirmatory MRI) must be carried out within 6 weeks of enrollment into this study

• Progressive disease following surgical excision based on clear radiological or clinical evidence of progression, or an incomplete excision (less than 95% or greater than 1.5 cm2) with necessity to begin treatment because of a risk of neurologic impairment with progression

• Chiasmatic lesions that have contiguous extensions of tumor into other regions of the visual pathways demonstrated on contrast MRI will be eligible for study without histopathological confirmation

• Patients with neurofibromatosis who have radiographic diagnosis of chiasmatic-hypothalamic tumor are eligible for the study, without requiring a biopsy confirmation of tumor histology, but not unless tumor progression is documented radiographically

• No intrinsic brain stem tumors of the pons or isolated optic nerve tumors without definitive involvement of the optic chiasm

PATIENT CHARACTERISTICS:

Age:

• Under 10

Performance status:

• Not specified

Life expectancy:

• Not specified

Hematopoietic:

• Absolute neutrophil count greater than 1,000/mm^3 (arm II)
• Platelet count greater than 100,000/mm^3 (arm II)

Hepatic:

• Not specified

Renal:

• Creatinine less than 1.5 times upper limit of normal for age OR
• Creatinine clearance or radioisotope GFR greater than 70 mL/min or equivalent GFR as determined by the institutional normal range

PRIOR CONCURRENT THERAPY:

Biologic therapy:

• Not specified

Chemotherapy:

• No prior chemotherapy for the tumor

Endocrine therapy:

• Prior corticosteroid therapy allowed

Radiotherapy:

• No prior radiotherapy for the tumor

Surgery:

• See Disease characteristics

Other:

• Prior diuretic therapy allowed