Bevacizumab May Boost Survival in Glioblastoma

Roxanne Nelson

August 23, 2013

The use of bevacizumab (Avastin, Genentech/Roche) as a viable treatment for patients with glioblastoma has been hotly debated. But a new study, published in Cancer, now provides the strongest evidence to date that bevacizumab therapy improves survival in patients with this aggressive and highly lethal brain tumor.

The study consisted of a population-based analysis that found that glioblastoma patients who died in 2010, which was after the US Food and Drug Administration (FDA) approved bevacizumab for this condition, had survived significantly longer than those who died of the disease in 2008.

The median survival was 8 months for patients who died in 2006, 7 months in 2008, and 9 months in 2010. This difference in survival, note the study authors, was highly significant between 2008 (pre-bevacizumab) and 2010 (post-bevacizumab).

They note that this survival difference is unlikely to be due to improvements in supportive care during this short time interval, because there was no significant difference between those who died in 2006 vs patients who succumbed 2 years later in 2008 (P = .4440).

"The controversy is not so much whether Avastin works, although there is fundamentally some question about that, but it has been very difficult to prove," commented lead author Derek Johnson, MD, a neuro-oncologist at the Mayo Clinic Cancer Center, Rochester, Minnesota.

In May 2009, the FDA granted accelerated approval for bevacizumab for use in patients with glioblastoma that has progressed despite previous therapy.

Bevacizumab has also been approved for use in Japan, but not in Europe, except in Switzerland and Rumania. This is the first drug to be approved for this indication in more than a decade.

Indirect Proof, for Now

However, the clinical trials that led to bevacizumab's approval by the FDA did not include head-to-head comparisons with other agents, explained Dr. Johnson in a Mayo Clinic release. "By definition, in that single arm design — without something to compare it to, we don't know for a fact if we are extending life."

In their study, Dr. Johnson and colleagues evaluated whether bevacizumab had an effect on survival in glioblastoma in the United States, using the Surveillance, Epidemiology, and End Results (SEER) Program, following the FDA approval for this indication.

They identified 5607 adult patients with glioblastoma in the SEER database who met criteria for inclusion in survival analysis. Within this cohort, 1715 patients (30.6%) died in 2006, 1924 patients (34.3%) died in 2008, and 1968 patients (35.1%) died in 2010.

There was no significant improvement in survival between 2006 and 2008 in any age subgroup. However, a significant difference in survival between 2008 and 2010 was observed in every age subgroup, the authors note, except for patients aged 18 to 39 years. In addition, in the subgroup aged 18 to 39 years, the longest-lived 25% of these patients were alive 32 months or longer prior to death in 2008. Conversely, in 2010, the longest-lived 25% were alive 41 months or longer prior to death. This survival difference was statistically significant (P = .0342), when analyzed with the log-rank test.

"Admittedly, it is indirect proof that bevacizumab increased survival," Dr Johnson said, referring to the results of his own study. "We found that patients who died after the FDA approval of bevacizumab lived significantly longer than those who died before the FDA approval."

"It's tough to quantify the exact improvement in survival because we are talking about large groups of people," he added. "But overall, our study found that at the population level, treatment strategies involving bevacizumab prolonged survival in patients with progressive glioblastoma."

It may be that there are certain subpopulations that accrued a great deal of that benefit and there may be some populations in which it did not change survival much at all, Dr. Johnson pointed out.

So although this study does not pinpoint the population most likely to benefit, "it does provide some broader level of support that we really are improving the life expectancy of patients with glioblastoma with this therapy," he noted.

Not Yet Ready for First Line

A further indication, using bevacizumab as first-line treatment of patients with newly diagnosed glioblastoma, has also been investigated, but the results are not convincing. Two studies were presented and discussed at length at this year's annual meeting of the American Society of Clinical Oncology — the Radiation Therapy Oncology Group (RTOG) 0825 study and AVAglio. However, the results from both trials together did not clearly support its use in newly diagnosed glioblastoma.

Principal investigator of the RTOG 0825 study and a professor of neuro-oncology at the University of Texas MD Anderson Cancer Center, Houston, Mark Gilbert, MD, told Medscape Medical News that the latest study suggesting a survival benefit is "interesting and compelling. The paper is absolutely in line with what many of us feel and what I have already suggested, in that bevacizumab is a useful agent for patients with glioblastoma," he said in an interview. "What they claim is the reason for improved survival was the approval of bevacizumab for second-line or salvage treatment."

With the crossover design of the RTOG 0825 study, it was unclear whether there was a benefit up front, or back end, or both, continued Dr. Gilbert, but the results of this current study would argue that it is probably both.

"However, given the extra toxicity, there is very little reason to use it as a first-line therapy," Dr. Gilbert said.

In his own study, the RTOG 0825, there was no survival difference in the 2 arms, Dr. Gilbert continued. "If there's extra toxicity and there's no difference if you use it earlier rather than later, then why not delay the toxicity until later — unless we can define a patient subpopulation either on the basis of a clinical or biomarker that would tell us if this is a group who would benefit from front-line therapy."

He also noted that when research is drawn from a database, "you see real-world experience" with an agent. A database includes patients who might have been excluded from a clinical trial because of their performance status.

"But these patients are in the database, and they probably didn't get bevacizumab," Dr. Gilbert said. "So if you just imagine the subpopulation in the SEER database who did get bevacizumab, and even with all of the dilution of the impact, there still is a survival effect, and that makes it even more compelling — there is something therapeutic about it."

No Compelling Evidence

However, another expert feels that it is difficult to draw conclusions from these findings. "The SEER data, unfortunately, lack data on treatment, which is a major shortcoming of this manuscript," commented Martin van den Bent, MD, PhD, professor of neuro-oncology at the Daniel Den Hoed Cancer Center, Erasmus University, in Rotterdam, the Netherlands.

"In general, even for treatments in which an overall survival benefit has been demonstrated, once introduced in everyday clinical practice, the benefit is often smaller," Dr. van den Bent told Medscape Medical News.

He pointed out that bevacizumab was not used in all patients included in this database, and the most likely recipients were those with recurrent disease. "So despite the probability that many patients were not exposed to bevacizumab, an overall survival benefit is suggested," Dr. van den Bent said. "This means that the overall survival benefit in those patients that were receiving bevacizumab must have been even larger. How does that fit in with the randomized trials that completely failed to observe an overall survival benefit?"

Dr. van den Bent noted that "for me, that is really hard to understand. Not even the BRAIN trial provided compelling evidence of an overall survival benefit."

The BRAIN trial (AVF3708) was one of the 2 trials upon which the FDA based its accelerated approval of bevacizumab for recurrent glioblastoma. Although results showed radiographically defined tumor response, they could not definitively demonstrate that bevacizumab use positively affected patient survival.

He added that there is crossover in randomized trials, which may have affected the overall survival analysis. "But in the absence of treatment data in the SEER database, this does not provide real evidence," he said.

No specific funding was disclosed. The authors have disclosed no relevant financial relationships. Dr. Gilbert reports serving as a consultant or in an advisory role for Novartis, EMD Serono, Genentech, and Merck; receiving honoraria from Novartis, EMD Serono, Genentech, and Merck; and receiving research funding from GlaxoSmithKline, Genentech, and Merck.

Cancer. Published online August 19, 2013. Abstract


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.