A search of the literature did not find strong evidence to support the use of bevacizumab as a single agent in the treatment of glioblastoma.
The evidence supporting the use of bevacizumab as a single agent in the recurrent setting is based on single arm uncontrolled studies and a meta-analysis by Wong et al.r However, due to analysing single arm uncontrolled studies and significant heterogeneity of the trials, this meta-analysis has been highly criticised.
A systematic review of the role of anti-angiogenic therapy in patients with high grade glioma in both the adjuvant and recurrent setting has been published in the Cochrane database.r The review included 7 randomised controlled trials with bevacizumab involving a total of 2502 participants. Despite evidence of improved progression-free survival, the authors concluded that the use of anti- angiogenic therapy did not improve survival and that the impact on quality of life is still unclear. There appears to be a progression free survival advantage in patients with recurrent glioblastoma who have significant mass effect and/or oedema.
Furthermore, studies have shown that the addition of chemotherapy to bevacizumab does not add any clinical benefit and causes an increase in toxicity. Therefore, bevacizumab should be given as monotherapy.
There is also no high-level evidence to suggest any benefit in continuing bevacizumab treatment beyond progression.r
Source |
Study & year published |
Supports use? |
Is the dose and regimen consistent with the protocol? |
Comments |
Systematic review |
Ameratunga et al 2018r |
N/A |
Yes |
7 RCTs (adjuvant and recurrent) with bevacizumab included. Studies in the recurrent setting involved bevacizumab alone or in combinations |
Meta-analysis |
Wong et al. 2011 r |
Yes |
Yes |
15 single arm and uncontrolled studies included |
Phase II trials |
Friedman et al. 2009r |
Yes |
Yes |
Bevacizumab alone vs bevacizumab + irinotecan |
Kreisl et al. 2009 r |
Yes |
Yes |
Bevacizumab alone |
Field et al. 2015 r |
Yes |
Yes |
Bevacizumab alone vs bevacizumab + carboplatin |
Taal et al. 2014r |
Yes |
Yes |
Bevacizumab + lomustine |
Taal et al. 2014r |
No |
Yes |
Bevacizumab alone |
Observational studies |
Chamberlain et al. 2010r |
Yes |
Yes |
- |
Guidelines |
Date published/revised |
Supports use? |
Is the dose and regimen consistent with the protocol? |
Comments |
NCCN |
V.2 September 2022 |
Yes |
Yes |
- |
BCCA |
July 2022 |
Yes |
Yes |
Recommended after second relapse following metronomic dosing temozolomide or first relapse if not eligible for metronomic dosing temozolomide / with symptoms related to marked cerebral oedema that require high doses of corticosteroids |
CCO |
December 2021 |
Yes |
Yes |
- |
ESMO |
April 2014 |
N/A |
N/A |
- |
Efficacy
Inhibiting VEGF signalling can decrease tumor enhancement without an intrinsic antitumor effect and response rates may not be a true surrogate of tumour response.r The RANO (Response Assessment in Neuro-Oncology) criteria have been developed to address some of these limitations.r
Improvements in neurocognitive function, decreased cerebral oedema and a steroid sparing effect have been associated with bevacizumab therapy.rr However, there is evidence of rebound oedema after discontinuation of bevacizumab and caution is required in using these measures as endpoints as they have not been validated.
Radiologic responses following bevacizumab should therefore be interpreted with caution and follow-up scans, clinical correlation and discussion with an expert in the field are recommended.
Study Description |
6-month PFS survival rates |
Median OS |
Other end points |
Wong et alr
(n=548) |
45% |
9.3 months |
6-month OS 76% |
Friedman et alr
Bev alone group (n=85) |
42.6% |
9.2 months |
ORR 28.2% |
Kreisl et alr
(n=48) |
29% |
7.75 months (31 weeks) |
6-month OS 57% |
Chamberlain et alr
(n=50) |
42% |
8.5 months |
Partial radiographic response 42% |
Taal et alr
Bev alone group (n=51) |
16% |
8 months |
12-month OS 26% |
Taal et alr
Bev+lomustine (n=52) |
42% |
12 months |
12-month OS 48%
The dose of lomustine was reduced from 110 mg/m2 to 90 mg/m2 due to unacceptable toxicity. |
Toxicity
Bevacizumab is generally well tolerated with lower rates of serious adverse events when used a single agent. The most common adverse events include low grade bleeding (i.e. epistaxis), hypertension, impaired wound healing, and proteinuria.r
Adverse events r
Grade > 3 |
Bevacizumab alone group
(n=94) % |
In selected patients |
Hypertension |
8.3 |
Haemorrhage (overall) |
0 |
Haemorrhage (cranial) |
0 |
Wound healing complications |
2.4 |
Venous thromboembolism |
3.6 |
Arterial thromboembolism |
2.4 |
Proteinuria |
0 |
GI perforation |
0 |
RPLS |
0 |
In all patients |
Aphasia |
3.6 |
Confusional state |
2.4 |
Convulsion |
6.0 |
Deep vein thrombosis |
2.4 |
Diarrhoea |
1.0 |
Fatigue |
3.6 |
Hypertension |
8.3 |
Pneumonia |
1.2 |
Pyramidal tract syndrome |
1.2 |
Somnolence |
1.2 |
Hypokalaemia |
3.6 |
Lymphopenia |
2.4 |
Neutropenia |
1.2 |
© adapted from Journal of Clinical Oncology 2009