ASCO: No Benefit From Avastin in Newly Diagnosed Glioblastoma

CHICAGO-About a decade ago, it was discovered that bevacizumab (Avastin) added to irinotecan (Camptosar) dramatically pruned tumor vessels and reduced vascular permeability in patients with recurrent glioblastoma, as well as significantly improved their quality of life (QOL). Confirmatory phase II research led to accelerated US Food and Drug Administration approval of bevacizumab for relapsed glioblastoma in 2009, and today the agent is widely used in this setting. These impressive results in an otherwise deadly disease with few treatment options led researchers and clinicians to wonder whether bevacizumab might also be helpful to patients with newly diagnosed glioblastoma. In 2008 and 2009, two phase III trials-the National Cancer Institute–funded Radiation Therapy Oncology Group (RTOG) 0825 study in the United States, and the industry-funded AVAglio trial in Europe-were opened to test this hypothesis. Results of both trials were presented this year at the annual meeting of the American Society of Clinical Oncology (ASCO); unfortunately, the results do not clearly support the use of bevacizumab in the setting of newly diagnosed disease.

Dr. Mark Gilbert of the MD Anderson Cancer Center presented the results of RTOG 0825. In this study, 637 newly diagnosed patients were randomly assigned to receive either standard chemoradiation (with temozolomide [Temodar]) plus bevacizumab, or standard chemoradiation plus placebo. At disease progression, patients were unblinded and those on the placebo arm were allowed to receive bevacizumab. The primary endpoints of the trial were overall survival and progression-free survival. Most patients were evaluated for symptom burden, QOL, and neurocognitive functioning; QOL and adverse events were among the secondary objectives of the trial.

Progression-free survival was significantly improved in the bevacizumab arm: 10.7 months vs 7.3 months for placebo. However, overall survival was slightly (although not significantly) worse in the bevacizumab arm: 15.7 months vs 16.1 months for placebo. In addition, patients in the bevacizumab arm had a greater symptom burden and worse neurocognitive functioning, and they scored worse on several measures of health-related QOL than did patients who received only standard therapy.

The AVAglio trial, which had a study design very similar to that of RTOG 0825 (including provision for crossover) and which involved 921 patients, also showed an improvement in progression-free survival (10.6 months in the bevacizumab arm vs 6.2 months in the placebo arm) but virtually identical overall survival (16.8 months vs 16.7 months, respectively). However, the European trial, reported by Dr. Warren Mason of Princess Margaret Hospital in Toronto, did show more favorable QOL outcomes than did RTOG. In the bevacizumab arm, time to definitive deterioration was extended by several months on five prespecified health-related QOL measures (compared with the placebo arm), and time to the initiation of corticosteroid treatment to manage adverse effects was also significantly longer in the patients who received bevacizumab (a median of 12.3 months vs 3.7 months for placebo). Dr. Gilbert suggested that one reason for the discrepant QOL results between the two trials might be that the AVAglio trial did not incorporate a measure of neurocognitive functioning into its evaluation of QOL outcomes.

This QOL discrepancy, however, was overshadowed by the failure of both trials to show an overall survival benefit for bevacizumab in the upfront setting. Several theories for these disappointing results were advanced, both by Dr. Gilbert and by Dr. Howard Fine of New York University (who commented on the presentation of the RTOG results). One hypothesis is that the sizable number of patients who crossed over in the two trials may have obscured any overall survival benefit. Dr. Fine also suggested that perhaps differences in tumor biology between newly diagnosed glioblastoma and recurrent disease (which has a more mesenchymal signature and thus is more angiogenic) might be at work.

It is also possible, Dr. Gilbert suggested, that we will be able to identify a subset of patients in whom upfront bevacizumab would increase survival. He pointed to the work of his MD Anderson colleague, Dr. Erik Sulman, whose umbrella study in specimens submitted by patients in RTOG 0825 has been measuring the degree of gene enrichment for a panel of genes known to be involved in cancer cell invasion and establishment of new blood supplies. Dr. Sulman and colleagues have identified a gene signature in the specimens of patients who received bevacizumab that distinguished those who benefited from the drug from those who did not. Further analysis is now required, which then needs to be followed up with a prospective trial. Dr. Gilbert noted, however, that this research has the potential to be a real game changer: It might one day enable clinicians to reliably identify those patients who should receive bevacizumab upfront and those for whom it should be reserved as a salvage therapy.

Despite the somewhat disappointing results of AVAglio and RTOG 0825-especially the latter-Dr. Gilbert underscored that bevacizumab remains an important part of the clinical armamentarium for the treatment of glioblastoma. What we need now, he said, is “to identify better drugs to use with it.”