Inclusion Criteria:

- Must have biopsy-proven GBM. First relapse (progression or recurrence) of GBM after
surgery (or biopsy) and treatment with radiotherapy (conventional fractionated
external beam) and chemotherapy (temozolomide- or nitrosurea-based therapy)

- Must have past biopsy samples available for central pathology review

- Must have evidence on Gd-MRI of progressive/recurrent disease

- Must have measurable disease on Gd-MRI obtained within 14 days prior to start of
study treatment

- Must be at least 18 years of age

- Must have a Karnofsky Performance Status score of at least 70

- If being treated with steroids, the steroid dose must be stable or decreasing for 1
week prior to randomization

- If being treated with anticonvulsants, must have no change in the type of
anticonvulsants for 2 weeks prior to randomization

- All acute toxic effects (except for alopecia) of any prior treatment must have
resolved or are no greater than grade 1 (NCI Common Toxicity Criteria, Version 2.0)

- Baseline laboratory data must be within the following limits: absolute neutrophil
count at least 1500; platelets at least 100,000; hemoglobin at least 9.0 g/dL; serum
creatinine no greater than 1.5 mg/dL, total serum bilirubin no greater than 1.5 times
the upper limit of the normal range; SGOT and SGPT no greater than 2.5 times the
upper limit of the normal range; albumin at least 3.0 g/dL, serum or urine pregnancy
test (for females of childbearing potential) negative within 7 days prior to start of
study treatment

- At least 6 weeks must have elapsed since completion of prior nitrosurea therapy; at
least 4 weeks since completion of prior temozolomide therapy

- Must have written informed consent

- Must be able and willing to comply with study procedures

- Must have received prior treatment with radiotherapy (conventional fractionated
external beam) and (neo)adjuvant/concurrent chemotherapy (with a temozolomide- or a
nitrosurea-based containing )regimen for GBM

Exclusion Criteria:

- Must not have received prior treatment (except for surgical debulking) of first
relapse (progression or recurrence) of GBM

- Must not have received prior treatment with another topoisomerase-I inhibitor (e.g.
irinotecan, topotecan, rubitecan)

- Must not have had radiosurgery or radiotherapy within 1 month prior to randomization

- Must not have had prior brachytherapy or chemotherapy wafer implantation

- Must not have had prior high-dose chemotherapy with bone marrow or stem cell support

- Must not receive concomitant treatment with any other investigational agent or
anti-cancer treatment during the study

- Must not be currently enrolled in another therapeutic clinical trial for the
treatment of GBM

- Must not currently (or in the past 5 years) have other malignancies (except for
adequately treated basal cell or squamous cell skin cancer or non-invasive cervical
cancer)

- Must not have any of the following in the past 6 months: myocardial infarction (heart
attack), severe/unstable angina, coronary artery bypass graft, symptomatic congestive
heart failure, cerebrovascular accident (stroke), or transient ischemic attack (TIA)

- Must not have had any of the following in the past 2 months: pulmonary embolus (blood
clot in lungs), deep venous thrombosis (blood clot in veins), or other significant
thromboembolic event

- Must not have an ongoing cardiac dysrhythmia (abnormal heart rhythm) of grade 2 or
higher (NCI Common Toxicity Criteria, Version 2.0)

- Must not have known human immunodeficiency virus (HIV) infection

- Must not be pregnant or breastfeeding. Patients (male and female) must be surgically
sterile (or postmenopausal for females) or must agree to use effective contraception
during the period of study treatment

- Must not be inappropriate for entry into the study, in the judgment of the
investigator