Brain metastases is a common complication in lung cancer. It occurs in 25-30% of patients with non-small-cell lung cancer. A Japanese study documented a 0.6% incidence of intracranial hemorrhage in patients with secondary or primary brain tumors and a report of autopsy specimens shows an upper incidence of 14%. Bevacizumab is a recombinant, humanized monoclonal antibody directed against vascular endothelial growth factor, a key factor in tumor associated angiogenesis. Bevacizumab with carboplatin and paclitaxel improved overall survival and disease free survival as first line treatment in metastatic non-small-cell lung cancer compared with carboplatin and paclitaxel alone in an Eastern Cooperative Oncology Group (ECOG) phase III trial. This trial excluded patients with brain metastases, so there is lack of experience in this subset of patients.

The phase II PASSPORT trial addresses the safety of bevacizumab in the treatment of patients with metastatic non-small cell lung cancer and previously treated brain metastases. The primary objective was to assess the incidence of symptomatic grade ≥ CNS hemorrhage for bevacizumab treatment.

This was an open-label, single arm, phase II trial. Eligible patients had non-squamous NSCLC and previously treated brain metastases with whole-brain radiation therapy (WBRT), radiosurgery and/or neurosurgery. Patients received 15 mg/kg bevacizumab every 21 days with either platinum based doublet chemotherapy in the first line setting or single agent erlotinib, docetaxel or pemetrexed in second-line. One hundred and fifteen patients were enrolled; 80% received WBRT for the treatment of brain metastases, 19% radiosurgery and 0.9% neurosurgery. Patients did not receive steroids during the treatment. Distribution of treatments was as follow: first line setting; 48.7% carboplatin/paclitaxel, 10.5% carboplatin/gemcitabine and 10.5% carboplatin/pemetrexed. Among second-line patients, 56.4% single agent pemetrexed and 23.1% erlotinib. One hundred and six patients who received at least one dose of bevacizumab were evaluable for safety. Median treatment duration was 85 days and the median number of bevacizumab cycles received was 5. 

There have been no reports of grade 1 to 5 CNS hemorrhage (95% CI, 0.0% to 3.3%) to date among the 106 patients who received bevacizumab based therapy. Two grade 5 pulmonary hemorrhages did occur. Neither of these two patients had apparent risk factors for bleeding. The most common grade 4 event in the study was pulmonary embolism, which occurred in seven patients.

This phase II trial is the first prospective study designed to specifically assess the safety of the angiogenesis inhibitor bevacizumab for the treatment of non-squamous NSCLC in patients with previously treated non-progressing brain metastases. No grade ≥ 2 CNS hemorrhage events were reported in the cohort of 106 patients who entered the study demonstrating a minimal risk of intracerebral hemorrhage in this population. The overall safety results from this trial strongly suggest that the use of bevacizumab in patients with treated brain metastases has an acceptable safety profile. As a result the NCCN guidelines for this agent permit the use of bevacizumab in eligible patients with treated brain metastases.

Socinski M, Langer C, Huang J. Safety of Bevacizumab in patients with non-small-cell lung cancer and brain metastases – J Clin Oncol 27 – 2009 /10.1200 (published ahead of print)