This article is more than 5 years old. Information may no longer be current.
Lomustine, bevacizumab fail to improve OS in progressive glioblastoma
The addition of lomustine to bevacizumab failed to confer a survival advantage for the treatment of progressive glioblastoma, according to results of a phase 3 EORTC clinical trial.
“EORTC was unable to confirm the conclusion of phase 2 trials that the addition of bevacizumab to lomustine improves survival [among] patients with progressive glioblastoma,” Wolfgang Wick, MD, medical director of the department of neuro-oncology at Heidelberg University Hospital, and colleagues wrote. “The effect on PFS was not associated with an increase in OS, and combination therapy was associated with increased toxicity.”
Results of the phase 2 BELOB trial suggested the combination of bevacizumab (Avastin, Genentech) and lomustine (Gleostine, Corden) improved survival for patients with progressive glioblastoma.
“The BELOB trial provided a noticeable survival signal, which prompted the EORTC 26101 phase 2 trial to be transformed into a full phase 3 study,” Wick told HemOnc Today. “BELOB demonstrated improved OS at 9 and 12 months for combined therapies using bevacizumab and lomustine vs. therapies based on either substance alone.”
The EORTC trial expanded to a phase 3 trial to determine the safety and efficacy of the combination among a larger number of patients.
Researchers randomly assigned 437 patients to a median of six treatment cycles with 10 mg/kg bevacizumab every 2 weeks plus 90 mg/m2 lomustine every 6 weeks (n = 288), or 110 mg/m2 lomustine alone every 6 weeks (n = 149).
If no adverse events higher than grade 1 occurred among the combination group in the first treatment cycle, lomustine dose increased to 110 mg/m2.
OS served as the primary endpoint. Secondary endpoints included PFS, landmark analyses for PFS and OS, toxicity, response rates according to the Response Assessment in Neuro-Oncology criteria, clinical or neurologic deterioration-free survival, glucocorticoid use, and health-related quality of life of both patients and health care proxies, among others.
Researchers reported 329 (75.3%) OS events.
Locally assessed PFS was 2.7 months longer for patients in the combination group than the monotherapy-treated group (4.2 months vs. 1.5 months; HR for progression or death = 0.49; 95% CI, 0.39-0.61).
Despite this, median OS was 9.1 months in the combination group and 8.6 months in the monotherapy group, an indication that the combination regimen did not provide an OS advantage over lomustine alone (HR for death = 0.95; 95% CI, 0.74-1.21).
Further, the combination regimen yielded no significant effects on health-related quality of life or neurocognitive function, the researchers noted.
In the combination group, 63.6% experienced grade 3 to grade 5 adverse events compared with 38.1% in the monotherapy group.
Bevacizumab may still be relevant for prolonging symptom-free survival, Wick said.
“In the United States, the 2009 approval [of bevacizumab] has led to the wide use of the compound at first progression,” he said. “In the European Union, bevacizumab is more commonly used in a palliative setting, rather than mainly with the intent to promote survival.
“In countries without approval, the drug is administered to improve symptoms and reduce steroid use,” he added. “Bevacizumab may still have uses in glioblastoma, but not as a means of prolonging OS.” – by Melinda Stevens
For more information:
Wolfgang Wick, MD, can be reached at the Neurology Clinic and National Center for Tumor Disease, University of Heidelberg and German Cancer Research Center, Im Neuenheimer Feld 400, D-69120 Heidelberg, Germany; email: wolfgang.wick@med.uni-heidelberg.de.
Disclosures: Wick reports a consultant role with Bristol-Myers Squibb, Celldex Therapeutics and Merck Sharp & Dohme; lecture fees from Bristol-Myers Squibb; and grant support and drugs for the EORTC trial provided by Boehringer Ingelheim and Roche. Please see the study for all other authors’ relevant financial disclosures.
Perspective
Back to Top
Stephanie E. Weiss
There is no gold standard treatment for recurrent glioblastoma. This trial compares outcomes for patients with recurrent glioblastoma who received a combination of commonly used second-line systemic agents lomustine and bevacizumab, against single-agent lomustine.
Lomustine is an alkylating chemotherapy. Bevacizumab — an antibody against vascular endothelial growth factor — is mechanistically distinct and works by pruning and renormalizing tumor-related neovasculature. The FDA approved bevacizumab for use in recurrent glioblastoma in 2009 based, in part, on findings from the BRAIN trial, in which bevacizumab appeared to be a relatively safe drug that produced responses resulting in a meaningful decrease in steroid use, and prolonged PFS. Despite great enthusiasm to study bevacizumab in newly diagnosed patients with glioblastoma, the results were disappointing.
The current trial looks at the safety and effectiveness of adding bevacizumab to lomustine. Importantly, the authors used the Response Assessment in Neuro-Oncology (RANO) criteria to assess progression on imaging. Researchers also administered a battery of tests related to cognitive and functional well-being.
In spite of improved PFS with combined therapy, there was no survival difference between the study arms. This uncoupling of progression-free benefit from a survival benefit has been previously observed in an earlier trial of bevacizumab in newly diagnosed glioblastoma (RTOG 0825). However, in that study, researchers assessed response to therapy — and progression — with older and less nuanced imaging criteria that would miss increasing tumor infiltration even while the main mass appeared controlled. This also would explain the worse cognitive outcomes in the experimental arm despite the appearance of tumor control.
This study used the more appropriate RANO criterion, designed to identify the nonenhancing, infiltrative progression more commonly seen among patients actively on bevacizumab. It is then particularly interesting that no neurocognitive benefit occurred with the use of bevacizumab, as the authors effectively assessed superior PFS — including infiltrating PFS — in that group
Bevacizumab failed to demonstrate steroid-sparing effects, contrary to prior observations. Given the drug’s ability to restore the blood-brain barrier and suppress mass effect, it is frequently used among patients with glioblastoma and subtotal resection to facilitate a steroid taper and avoid their oft-debilitating side effects. This is an important observation because severe — probably drug-specific — side effects were more common among the combined-therapy group.
Finally, the authors observed that their findings were contrary to those of the BELOB trial in which bevacizumab resulted in an OS benefit when added to lomustine. However, unlike the BELOB trial — where bevacizumab was not available off study and thus crossover to the drug was virtually unheard of in the control group — in the EORTC trial, more than 35% of patients on single-arm therapy went on to receive bevacizumab at progression. Thus, the lack of survival benefit may be due to effective salvage.
Ashley L. Sumrall
Wick and colleagues successfully answered a question that has plagued the neuro-oncology community for years. In this pivotal EORTC trial, the addition of bevacizumab to lomustine did not prolong survival for patients with progressive glioblastoma. This is the first phase 3 randomized controlled trial to demonstrate this finding.
As enthusiasm for using bevacizumab to treat patients with glioblastoma has waned over the last few years, this may be the proverbial “nail in the coffin” of this expensive therapy’s role in palliation. In prior phase 2 trials, investigators have touted the glucocorticoid-sparing effects of using this therapy. For example, the BRAIN trial by Friedman and colleagues demonstrated a trend of decreased glucocorticoid usage among patients managed with bevacizumab. This trial evaluated only 167 patients and noted that glucocorticoid-dependent patients required stable or lower doses while on bevacizumab. In EORTC 26101, researchers found no statistically significant benefit between use of glucocorticoids between the two arms.
Numerous questions arose after the release of the phase 2 trial data. Then, when the BELOB trial showed a trend of improved survival, researchers configured the EORTC trial into a phase 3 trial. The neuro-oncology community was excited again, thinking we almost had the answer to the elusive question of what constitutes optimal treatment for recurrent glioblastoma. We expected improved survival in this bevacizumab-based doublet but were incorrect in our assumptions. The treatment conferred no improvement in neurocognitive functioning, amount of glucocorticoid requirement or OS. Sadly, almost two-thirds of patients on combination therapy experienced grade 3 to grade 5 adverse events. For a palliative-intent therapy, this therapy seems to cause more harm than good. A paradigm shift is likely to follow.