Updated safety and survival of patients with relapsed glioblastoma treated with bevacizumab in the BRAIN study

Reporter: Samuel Swisher-McClure, MD
The Abramson Cancer Center of the University of Pennsylvania
Last Modified: June 5, 2010

Presenter: T. Cloughesy for the BRAIN Investigators
Affiliations: University of California, Los Angeles Neuro-Oncology Program, Los Angeles, CA


  • Glioblastoma Multiforme (GBM) is the most common primary malignant brain tumor accounting for 40% of primary CNS Malignancies
  • Primary therapy for GBM typically consists of maximally safe surgical resection, targeted radiotherapy (RT), and chemotherapy.
  • Despite advances in therapy for GBM, the diagnosis is still associated with a poor prognosis and the majority of patients develop recurrent disease within 1-2 years of initial therapy.
  • There is no consensus regarding standard therapy for recurrent GBM.
  • Treatment options may include salvage surgery with Carmustine wafers in selected patients, single agent chemotherapy with irinotecan (IR), temozolomide, CCNU, PCV, or bevacizumab.
  • Observed response rates with chemotherapy in the setting of recurrent GBM are typically 10-15% with median overall survival times of 6-9 months.
  • GBM is known to exhibit extensive tumor necrosis, increased vascular proliferation, and elevated levels of pro-angiogenic factors, notably vascular endothelial growth factor (VEGF).
  • Elevated levels of VEGF in GBM are associated with high grade disease and poor prognosis (Lamszus et al. Clin Cancer Research, 2003).
  • Bevacizumab (BV) is a humanized monoclonal antibody that targets VEGF and has been shown to be beneficial for patients with metastatic colorectal cancer, non-small cell lung cancer, breast cancer, and renal cell carcinoma.
  • The BRAIN study was a phase II multi-institutional, randomized, non-comparative trial designed to evaluate long term safety and efficacy of BV and BV in combination with IR for recurrent GBM
  • The initial results, previously published by Friedman et al. (JCO, 2009), reported promising overall response rates of 28.2% and 37.8%, and median overall survival times of 9.2 and 8.7 months, in the BV and BV + IR groups, respectively.
  • This report updates the safety and survival data of that study with follow-up safety data available through October 2008, and follow-up survival data available through July 2009.


  • The BRAIN study was a Phase II multi-institutional, randomized, non-comparative study designed to evaluate long term safety and efficacy of BV and BV in combination with IR for recurrent GBM.
  • Inclusion Criteria:
  • Patients with histologically confirmed GBM, experiencing first or second relapse.
  • Disease progression confirmed by MRI ? 14 days before the first study treatment.
  • Previous treatment with Radiotherapy (RT) and Temozolomide (TMZ)
  • KPS ? 70
  • Estimated life expectancy > 12 weeks
  • Adequate Hematologic, Renal, and Hepatic Function
  • Exclusion Criteria:
  • Previous treatment with IR, other anti-VEGF agents, or prolifeprospan with carmustine wafers
  • MRI evidence of intracranial hemorrhage
  • Any history of a bleeding diathesis or coagulopathy
  • Clinically significant cardiovascular disease or arterial thromboembolism < 6 months before the first study treatment
  • Uncontrolled hypertension
  • Patients were randomly assigned to receive either BV alone (10 mg/kg IV every other week) or BV at the same dose + IR ( 125 mg/m2 or 340 mg/m2 depending on whether the patient was on an enzyme inducing antiepileptic drug)
  • Patients were stratified by KPS (70-80, 90-100) and by first or second relapse.
  • Patients in both groups were treated for 104 weeks or until disease progression
  • Patients in the BV study arm with progression who met eligibility criteria could enroll in a post progression phase (PP) to receive BV+IR.
  • Primary endpoints were 6-month progression free survival and overall response rates.
  • All patients underwent clinical, laboratory, and MRI evaluations prior to each treatment cycle and radiographic progression or response were assessed by a blinded, independent radiology facility according to WHO response criteria.
  • Survival was defined from randomization to death from any cause.
  • Patients lost to follow-up were censored at the date they were last known to be alive.
  • Median overall survival (OS) with 95% CIs and 12-,18-, 24-, and 30-month (mo) Kaplan-Meier survival rates were calculated.
  • Adverse events (AEs) were graded according to NCI criteria (v 3.0).


  • 167 patients were enrolled in the study between June 2006 and February 2007, with 85 patients assigned to BV and 82 patients assigned to BV + IR.
  • 7.8 % of patients were excluded due to diagnosis other than GBM
  • The majority of patients (80.8%) were receiving therapy for initial relapse of GBM.
  • The median number of BV doses received during the primary treatment phase was 9 and 11 for BV and BV+IR, respectively, and 3 during the post-progression phase.
  • 50 of the 85 patients assigned to the BV arm (58.8%) went on to receive post-progression phase BV + IR as was permitted in the study.
  • Overall response rates were 28.2% with BV alone and 37.8% in the BV + IR arm.
  • 6 month progression free survival rates were 42.6% and 50.3% in the BV and BV + IR arms respectively.
  • Median OS at last follow-up (15 July 2009) was 9.3 months (95% CI 8.2-11.8) in the BV arm and 8.9 months (95% CI 7.9-11.9) in the BV+IR arm.
  • Overall survival rates at 30 months were 11% in the BV arm and 16% in the BV + IR arm.
  • Updated safety data (15 July 2008) for 163 pts who received study drug did not significantly change from previously reported data. Toxicities were relatively infrequent and included hypertension, proteinuria, wound healing complications, and thromboembolic events.

Author’s Conclusions

  • Updated survival data from the BRAIN trial showed that up to 16% of pts remained alive at 30 months.
  • The observed response rates and survival rates compare very favorably with results obtained using other chemotherapeutic agents for recurrent GBM including TMZ, and CCNU.
  • The incidence of selected AEs in the updated safety data was consistent with that previously reported, and no new safety signals were identified.

Clinical Implications

  • Despite aggressive multimodality therapy, the majority of patients diagnosed with GBM eventually develop relapsed disease for which there are no standard treatment options.
  • The current study was a well conducted phase II multi-institutional trial designed to evaluate the long-term safety and efficacy of BV and BV combined with irnotecan in the treatment of recurrent GBM.
  • The current report provides an update of the previously published data with longer term follow up available.
  • The response rates and overall survival rates compare favorably with results that have been previously published using TMZ, or CCNU for recurrent GBM.
  • The treatment appears to be relatively well tolerated with low toxicity rates. There were no significant changes in toxicity rates observed since the author’s initial publication of this data.
  • While these results are encouraging, further randomized studies are needed in order to draw more definitive conclusions regarding optimal treatment for patients with GBM and to hopefully further improve response and survival for these patients.