Bevacizumab is well tolerated in the brain tumor population, with the same spectrum of side effects seen in other cancer populations (hemorrhage, thrombosis, hypertension, bowel perforation, impaired wound healing, and proteinuria).12,13 The incidence of life-threatening events, such as significant intracranial hemorrhage (3%) or thromboembolism (2%–12%), is within the expected range for the population and is not clearly increased by ¬bevacizumab.12
Several small series have reported that corticosteroid reductions were feasible in 33%–59% of patients with recurrent GB after bevacizumab treatment, and others have reported average corticosteroid dose reductions of 72% and 59%.11,13 This is an important benefit of bevacizumab, as chronic or high-dose corticosteroid use in patients with glioma is associated with significant morbidity. The ability of bevacizumab to control edema confounds the definition of tumor progression. Figure 2 shows a patient who had early neurologic improvement after initiation of bevacizumab and then remained clinically stable without corticosteroids over 14 months of therapy, despite slow continuous growth of the tumor mass.
Bevacizumab is also tolerated by older patients; there is an intriguing study suggesting that not only do older patients tolerate bevacizumab, they may also have increased benefit over younger patients (age separation: younger than 55 years or 55 years and older).20
Another option for some patients, prior to bevacizumab, is rechallenge with temozolomide at alternative dosing schedules, which result in prolonged exposure to higher cumulative doses than that achieved by standard 5-day dosing.21 Resistance to temozolomide occurs through direct repair of DNA by MGMT; a proposed mechanism to overcome resistance would be to deplete tumor-cell MGMT. Several studies have shown that prolonged exposure of peripheral blood mononuclear cells results in depletion of MGMT, and it has been suggested that this could also occur in glioma cells.22 Other studies suggest that prolonged exposure to temozolomide may be directly toxic to endothelial cells.23 These data provide a rationale for temozolomide rechallenge using alternative dose and dosing schedules that deliver higher culmulative doses over prolonged periods.
Commonly tried temozolomide schedules have been 21 days on/7 days off at doses of 75–100 mg/m2, 7 days on/7 days off at a dose of 150 mg/m2, and continuous daily dosing at 50 mg/m2 (Table 1). These schedules were well tolerated in these pretreated patients, with cumulative leukopenia after several cycles. Results have shown PFS-6 of 23%–48% with a suggestion (not supported by all studies) that best responses are seen in patients who were rechallenged after a treatment-free interval (from standard adjuvant temozolomide).24–27 Responses were also similar in patients with high and low levels of tumor MGMT, suggesting that these regimens may overcome -MGMT-mediated resistance.28
In addition to bevacizumab and temozolomide, lomustine, carmustine, irinotecan, cisplatin, and carboplatin have shown modest efficacy in studies as single agents or in combination regimens.29–31 The populations in these studies usually included both recurrent GB and anaplastic tumors, including oligodendrogliomas, and were carried out prior to standard use of chemoradiation and adjuvant temozolomide. Thus, it is difficult to extrapolate how these results would translate into today’s patient population. Interestingly, in a recent randomized phase III trial of recurrent GB (after prior temozolomide), lomustine was found to be superior to the investigational pan--VEFG receptor inhibitor cediranib.32
As noted, the NovoTTF-100A System was only recently approved, and experience is limited. Despite the need for patients to wear the device for 20 hours a day, and for intermittent adjustments to electrode placement at a clinic site, patient compliance in the study was good, and toxicities were minimal. It is clearly an option for patients with recurrent GB, but additional experience is needed to clarify the optimal time of use. The device is currently under study for the treatment of newly diagnosed GB; results of this study may help clarify the role of NovoTTF in treating this ¬malignancy.
In deciding which of the above strategies to use at first or even subsequent recurrences, we take into consideration the rapidity of tumor growth, extent of edema, mass effect, need for corticosteroids, and symptoms. Furthermore, to date there are no agents that improve outcomes when combined with bevacizumab or used after relapse on bevacizumab. This is often the last treatment regimen many patients receive before palliative end-of-life care. Thus, for a patient with a small, asymptomatic recurrence found on surveillance imaging and in the absence of an available clinical trial, we initiate therapy with any of the above-mentioned standard chemotherapeutic agents. At tumor progression and if chemotherapy is still tolerated, another agent may be tried. In contrast, the patient with a rapidly growing, large, or symptomatic recurrence requiring increasing doses of steroids will usually have an immediate clinical benefit from bevacizumab, which will improve quality of life.