RATIONALE: Immunotoxins can locate tumor cells and kill them without harming normal cells. Immunotoxin therapy may be effective in treating malignant glioma.

PURPOSE: Phase I trial to study the effectiveness of immunotoxin therapy before and after surgery in treating patients who have recurrent malignant glioma.

OBJECTIVES:

• Determine the concentration of interleukin-13 PE38QQR immunotoxin that produces histologic evidence of toxicity to tumor and the corresponding toxic effects of this drug when administered via continuous intratumoral infusion prior to second resection in patients with recurrent resectable supratentorial malignant glioma.
• Determine the toxic effects of this drug when administered via continuous peritumoral infusion, at concentrations determined in objective I, after second resection in these patients.
• Determine any toxic effects of increasing the duration of continuous peritumoral infusion of this drug, at concentrations determined in objective II, after second resection in these patients.
• Determine the time to progression and survival of patients treated with this regimen.

OUTLINE: This is a dose-escalation, multicenter study.

• Pre-resection therapy (initial cohorts of patients only): Patients undergo stereotactic biopsy of brain tumor followed by stereotactic placement of 1 intratumoral catheter on day 1. Patients with histologically confirmed malignant glioma receive interleukin-13 PE38QQR immunotoxin via continuous intratumoral infusion over 48 hours on days 2 and 3.

Cohorts of 3-6 patients receive escalating doses of pre-resection interleukin-13 PE38QQR immunotoxin until the histologically effective concentration (HEC) is reached or maximum tolerated dose (MTD) is determined. The HEC is defined by pathologic observations. The MTD is defined as the dose preceding that at which 2 of 6 patients experience dose-limiting toxicity. After the HEC is reached or MTD is determined, up to 6 additional patients are enrolled at selected dose levels to study safety and tolerability. Subsequent cohorts of patients are not treated with a pre-resection infusion.

• Resection (all patients): Patients undergo maximal resection (en bloc, if feasible) followed by placement of 2-3 peritumoral catheters (4 days after completion of pre-resection infusion for the initial cohorts of patients and at study entry for subsequent cohorts of patients).
• Post-resection therapy (all patients): Beginning on the second day after resection, patients receive interleukin-13 PE38QQR immunotoxin via continuous peritumoral infusion over 96 hours.

Cohorts of 3-6 patients receive escalating doses of interleukin-13 PE38QQR immunotoxin until the previously-defined HEC is reached or MTD is determined, whichever occurs first. If dose-escalation is stopped after HEC is reached, then three additional cohorts of patients receive escalating durations (5, 6, or 7 days) of post-resection infusion. If dose escalation is stopped after the MTD is determined, then the duration of post-resection infusion is not escalated.

Patients are followed every 8 weeks.

PROJECTED ACCRUAL: A total of 25-50 patients will be accrued for this study.

• Biological: cintredekin besudotox
• Procedure: adjuvant therapy
• Procedure: conventional surgery
• Procedure: neoadjuvant therapy

DISEASE CHARACTERISTICS:

• Histologically confirmed supratentorial malignant glioma (grade 3 or 4)

  • Anaplastic astrocytoma
  • Glioblastoma multiforme
  • Mixed oligoastrocytoma
  • Malignant astrocytoma, not otherwise specified
• Prior first resection of brain tumor
• Prior cranial radiotherapy with tumor dose of at least 48 Gy

• Radiographic evidence of recurrent or progressive supratentorial tumor

  • In patients who have received external beam radiotherapy or localized radiotherapy (e.g., gamma-knife or brachytherapy) within the past 12 weeks, progression must be confirmed by metabolic imaging (magnetic resonance spectroscopy or positron-emission tomography)
• Must be a candidate for second resection
• No signs of impending herniation
• No midline shift greater than 1 cm
• No multifocal disease or subependymal tumor spread

PATIENT CHARACTERISTICS:

Age:

• 18 and over

Performance status:

• Karnofsky 70-100%

Life expectancy:

• Not specified

Hematopoietic:

• Absolute neutrophil count at least 1,500/mm^3
• Platelet count at least 100,000/mm^3
• Hemoglobin at least 9 g/dL

Hepatic:

• PT and PTT no greater than upper limit of normal

Renal:

• Not specified

Other:

• No uncontrolled seizures
• No other neurologic condition that would interfere with study evaluation
• Not pregnant or nursing
• Negative pregnancy test
• Fertile patients must use effective barrier contraception during and for 60 days after study

PRIOR CONCURRENT THERAPY:

Biologic therapy:

• Not specified

Chemotherapy:

• At least 4 weeks since prior cytotoxic therapy (2 weeks for vincristine or 6 weeks for nitrosoureas)

Endocrine therapy:

• Concurrent steroids allowed
• No tapering of steroids during or immediately after study infusion

Radiotherapy:

• See Disease Characteristics
• At least 4 weeks since prior radiotherapy

Surgery:

• See Disease Characteristics

Other:

• Recovered from prior therapy
• At least 4 weeks since prior investigational agents
• At least 2 weeks since prior non-cytotoxic agents
• No other concurrent antitumor therapy