BY MARK FUERST
CHICAGO -- Adding bevacizumab to standard chemoradiation does not benefit patients with newly diagnosed glioblastoma, according to the results of a randomized Phase III trial reported as Abstract 1 at the American Society of Clinical Oncology Annual Meeting here.
Patients treated with bevacizumab plus temozolomide had more side effects than those treated with temozolomide alone, with no improvement in overall survival (OS), suggesting that bevacizumab should not be a part of first-line therapy for these patients with glioblastoma, said lead author
Mark R. Gilbert, MD, Professor of Neuro-oncology at the University of Texas MD Anderson Cancer Center.
Glioblastoma is the most common malignant primary brain tumor, with an average survival of less than 18 months. “Angiogenesis is a hallmark feature of glioblastoma, he explained. “A large area of tortuous vessels is characteristic of the disease. This makes an anti-angiogenesis strategy very appealing in the treatment of glioblastoma.”
He noted that bevacizumab, a monoclonal antibody that blocks the growth of tumor blood vessels, is currently approved for use in patients with recurrent glioblastoma. Despite a lack of clear evidence, the drug has been used off-label as first-line therapy in certain patients, in hopes of increasing the benefit to the patient.
“Unless we can identify a group of patients who clearly benefit from early use of bevacizumab, it appears that it should not be used in the first-line setting,” said Gilbert. “Bevacizumab remains an important part of our armory against glioblastoma, but in most situations it should be reserved as a salvage regimen.”
In the multi-institutional double-blind, placebo-controlled randomized trial, 637 patients with newly diagnosed glioblastoma were randomly assigned to treatment with chemoradiation, which included temozolomide and radiation, plus placebo (317 patients) or chemoradiation plus 10 mg/kg of intravenous bevacizumab (320 patients).
Before being randomized, the patients received three weeks of standard chemoradiation therapy. Following treatment, patients continued to receive temozolomide for 12 cycles and placebo or bevacizumab every two weeks. At disease progression, treatment was unblinded and patients were allowed to cross over to the bevacizumab arm.
There was “great balance” between the two arms in terms of age, gender, race, and performance status, he said. About 80 percent of patients had had intensity-modulated radiation therapy. About 40 percent of patients on placebo crossed over to bevacizumab, and 20 percent on bevacizumab stayed on the drug after disease progression.
Overall and progression-free survival (OS, PFS) were the co-primary endpoints of the study. Gilbert explained that since a P value of 0.05 signifies statistical significance, the investigators divided the value among the two co-primary endpoints, with a P value of 0.046 being required for statistical significance for median OS and a P value of 0.004 required for significance in PFS.
After patients were followed for a median of 21 months, there was no statistically significant difference in the median overall survival between the two groups (16.1 months with placebo vs. 15.7 months with bevacizumab).
The median PFS was longer in the bevacizumab group (10.7 months) compared with the placebo group (7.3 months), but again, the difference did not reach statistical significance.
A subgroup analysis based on molecular markers -- MGMT promoter methylation and a 9-gene expression signature -- did not identify a group of patients who demonstrated benefit from first-line use of bevacizumab. In fact, those patients with MGMT promoter methylation and a favorable nine-gene signature showed a strong trend towards a worse outcome.
Overall, there were more side effects in the bevacizumab group, particularly low platelet counts, blood clots, and high blood pressure. However, Gilbert said that “overall, the therapy was well tolerated. As expected, there were more adverse events with bevacizumab.”
QOL, Symptom Burden, Neurocognitive Function
In separate abstracts, the researchers assessed patients’ quality of life, symptom burden, and neurocognitive function, which also favored the group of patients who received chemoradiation alone. Patients treated with bevacizumab experienced an increase in symptom burden, a decline in neurocognitive function, and a worsening of multiple components of health-related quality of life (Abstract 2004).
In patients with newly diagnosed glioblastoma, molecular profiling appears to predict a response to first-line bevacizumab (Abstract LBA2010). However, this requires validation from further studies, Gilbert cautioned. “Until a patient subgroup has been defined, these results do not support the use of bevacizumab for newly diagnosed glioblastoma.”
In conclusion, he said, “First-line bevacizumab did not improve OS for patients with glioblastoma. PFS was longer, but did not reach the study-specified target. Symptom burden, neurocognitive function, and quality-of-life data show an overall decline. Bevacizumab remains an important therapy for patients with glioblastoma, but can be reserved as a later treatment.”
Bevacizumab may have a potential role if the right combination is found with other drugs, he said. “There has been a tremendous effort to look at anti-angiogenesis agents, and as data mature, we will look for a survival benefit.”
Second Phase III Trial: AVAglio
A similar Phase III trial, the AVAglio trial (Abstract 2005), also evaluated the efficacy of bevacizumab in patients with glioblastoma. The results from that trial also show only a modest improvement in PFS. The median PFS was 10.6 months in the chemoradiation-plus-bevacizumab group compared with 6.2 months among patients who received only chemoradiation.
An analysis of health-related quality of life in the AVAglio trial found that bevacizumab significantly delayed the time to definitive deterioration compared with patients in the placebo arm. Performance status was stable or improved during PFS in both arms. Physical functioning, social functioning, motor dysfunction, and communication deficit were all improved in patients who received the combination therapy.
Gilbert noted that the methods used to analyze quality of life in the AVAglio trial were different from those used in his study, and this could account for the improvement in quality of life observed in the AVAglio trial.
The Discussant for Gilbert’s study,
Howard Fine, MD, Division Chief of Hematology and Oncology at New York University Langone Medical Center, noted that although bevacizumab received FDA approval for recurrent glioblastoma based on “dramatic” radiographic activity in several Phase II trials, there are now these two multinational randomized Phase III trials showing that although the drug modestly increases progression-free survival, there is no increase in overall survival for newly diagnosed patients.
“Although bevacizumab will clearly continue to have an important role in the treatment of patients with glioblastoma, the timing of its use, the specific subpopulation of patients who benefit the most, and the biological and clinical consequences of chronic VEGF inhibition on glioma and normal cells within the central nervous system all need to be clarified.”
He said there are several possible reasons for the differing quality-of-life results in the two trials, including the statistics involved in subset analyses and the different methods used to analyze the data. He also noted that in the AVAglio study about one-third of patients discontinued bevacizumab treatment as a result of adverse events, which may have contributed to the lack of significance in the survival data.
Fine also offered insights into the biology of glioblastomas managed with bevacizumab. “There are five major subtypes of glioblastoma, and two predominant growth patterns. One growth pattern is driven by angiogenesis and the other is non-angiogenic driven -- that is, is VEGF independent. Generally most glioblastoma patients who recur die of angiogenic, VEGF-driven tumors.”
For patients treated upfront with bevacizumab, when their tumors recur, “the tumors have a different look,” Fine continued. “They have low vascularity and low alkaloids. Further angiogenic therapy won’t make a difference. This may, in part, be the reason that the data do not support the routine use of bevacizumab as part of upfront treatment in most patients with newly diagnosed glioblastoma.”
Scientists can now model whether a tumor is angiogenic-driven or not in mice, he noted.
Fine concluded by saying that he feel strongly that bevacizumab “still represents the single most important new drug in glioblastoma,” but that future trials will have to define markers for response to bevacizumab in the first-line setting and explore bevacizumab in combination with other agents that inhibit anti-vascular endothelial growth factor-associated invasiveness.
One promising combination, he said, appears to be bevacizumab, irinotecan, and radiotherapy, and a study presented at the meeting by German researchers evaluated the use in 182 patients with MGMT-non-methylated newly diagnosed glioblastoma (Abstract LBA2000).
In that randomized, multicenter, open-label trial, patients were randomly assigned to receive bevacizumab/irinotecan (116 patients) or temozolomide (54 patients). All patients received radiation therapy at the standard dose for six weeks. Patients in the bevacizumab/irinotecan arm received four cycles of bevacizumab over six weeks of radiation and then received the two drugs every two weeks until disease progression.
Patients in the temozolomide arm received six courses of therapy. Six-month PFS was the primary endpoint of the study.
At six months. progression-free survival was significantly longer for patients receiving bevacizumab/irinotecan (9.74 months) compared with those receiving temozolomide (6 months). Similarly for overall survival: 16.6 months vs. 14.8 months, respectively.
There were more Grade 3/4 vascular adverse events with bevacizumab/irinotecan and more severe hematotoxicity with temozolomide.
In conclusion, the researchers said the significant and clinically meaningful increase in the primary endpoint suggests that the bevacizumab/irinotecan combination is superior to standard temozolomide therapy in newly diagnosed MGMT-non-methylated glioblastoma patients.