A phase II window trial of procarbazine and topotecan in children with high-grade glioma: a report from the children's oncology group.

The role of chemotherapy in the treatment of high-grade gliomas in children is unclear. Early reports were suggestive of improved outcome in children with high-grade glioma with the addition of chemotherapy after surgery and radiation therapy. Subsequent studies did not show similar favorable contribution of chemotherapy to the outcome of these children. Further efforts to identify active chemotherapy agents in children include use of agents that have shown efficacy in adult patients with high-grade glioma and agents that have shown promise in mice bearing human xenografts of brain tumors. A Pediatric Oncology Group (POG 9431) trial tested the activity of two such agents, procarbazine and topotecan in newly diagnosed patients with high-grade glioma who had measurable disease after diagnostic surgery. Neither agent showed efficacy within the confines of the statistical design of the study. This study showed that children with high-grade glioma have an innate resistance to alkylating agents based on mismatch repair deficiency and high levels of alkyguanine transferase (AGT). Future trials should consider strategies to overcome the resistance mechanisms in children with high-grade glioma.

White matter lesions detected by magnetic resonance imaging after radiotherapy and high-dose chemotherapy in children with medulloblastoma or primitive neuroectodermal tumor.

PURPOSE: White matter lesions (WMLs) have been described as a delayed effect of cranial irradiation in children with brain tumors, or a transient subacute effect characterized by an intralesional or perilesional reaction. We report the occurrence of subacute WMLs detected by magnetic resonance imaging (MRI) in children treated for medulloblastoma or primitive neuroectodermal tumor (PNET) and document the associated clinical, radiologic, and neurocognitive findings. PATIENTS AND METHODS: Among 134 patients with medulloblastoma or supratentorial PNET treated prospectively with risk-adjusted craniospinal irradiation and conformal boost to the tumor bed, followed by four high-dose chemotherapy (HDC) cycles with stem-cell rescue, 22 developed WMLs on T1-weighted imaging with and without contrast and/or T2-weighted imaging on MRI. Patients had > or = 12 months of follow-up. Neurocognitive assessments included intelligence quotient (IQ) tests and tests of academic achievement. RESULTS: Twenty-two patients developed WMLs at a median of 7.8 months after starting therapy (range, 1.9 to 13.0 months). Lesions were predominantly in the pons (n = 8) and cerebellum (n = 6). Sixteen patients (73%) had WML resolution at a median of 6.2 months (range, 1.68 to 23.5 months) after onset; two patients developed necrosis and atrophy. Three developed persistent neurologic deficits. Cumulative incidence of WMLs at 1 year was 15% +/- 3%. Patients with WMLs had a significant decline in estimated IQ (-2.5 per year; P = .03) and math (-4.5 per year; P = .003) scores. CONCLUSION: WMLs in medulloblastoma or PNET patients treated with conformal radiotherapy and HDC are typically transient and asymptomatic, and may mimic early tumor recurrence. A minority of patients with WMLs develop permanent neurologic deficits and imaging changes. Overall, the presence of WMLs is associated with greater neurocognitive decline.

Phase 2 study of idarubicin in pediatric brain tumors: Pediatric Oncology Group study POG 9237.

Idarubicin (IDA), the 4-demethoxy analog of daunomycin, has had significant cytotoxicity in many malignancies. In previous reports, the alcohol metabolite of IDA, 4-demethoxydaunorubicinol (idarubicinol, or IDOL), had cytotoxic activity and the ability to penetrate the blood-brain barrier. For this reason, the Pediatric Oncology Group conducted a Phase 2 trial of IDA for children with recurrent or progressive brain tumors. Ninety-one eligible children were entered on this study, with ages ranging from 3 months to 19 years. Patients were stratified by tumor types into 6 categories: stratum 1, low-grade astrocytoma; stratum 2, malignant glioma (glioblastoma multiforme and anaplastic astrocytoma); stratum 3, medulloblastoma; stratum 4, brainstem glioma; stratum 5, ependymoma; and stratum 6, miscellaneous malignant tumors not included in the previous diagnoses. IDA(18 mg/m2) was infused over 4 h and followed by granulocyte colony-stimulating factor (G-CSF) beginning day 5 after infusion of IDA. G-CSF was continued until blood cell count recovery. Cycles were repeated at approximately 21-day intervals until patients developed progressive disease or had completed 6 cycles with stable or improved disease. Pharmacokinetic plasma and cerebrospinal fluid (CSF) samples were collected from a subset of these patients. Response was poor in all strata. Most patients developed progressive disease; however, in 21 patients with medulloblastoma there was 1 partial response, and 6 patients had stable disease (SD) that in 4 patients lasted more than 20 weeks. Grades 3/4 hematopoietic toxicities were the most common toxic events, and 14 infection-related events resulted in hospitalization of patients. Only 1 patient developed reduced cardiac function. The systemic clearance data for IDA and IDOL were nearly identical to those published on patients with leukemia, and the plasma elimination of the IDOL metabolite was substantially longer than that of the parent drug IDA. The peak CSF:plasma ratios of IDA and IDOL were very low. The overall response rates to IDA were disappointing despite periods of prolonged SD in nearly a fourth of the patients. We conclude from this data and from that in nonhuman primates that it is unlikely that IDA, daunomycin, or other related anthracyclines will be useful for treating primary CNS tumors.