Anti-angiogenic efficacy of Bevacizumab alone and in combination with a dual PI3K/mTOR inhibitor in an orthotopic model of malignant glioma: A multimodal neuro-imaging approach

The treatment of glioblastoma represents one of the main oncological challenges of the 21st century. Despite intensive therapeutic efforts, the median survival remains 15-18 months.

Initially, we conducted a single center retrospective study, at the National Neurosurgical Center in Beaumont Hospital, examining the use of bevacizumab in an homogenous Irish glioblastoma population. We demonstrated the importance of tumour location, MGMT (O6-methylguanine- methyl-transferase gene) methylation, IDH (Isocitrate Dehydrogenase) and extent of resection (EOR) as prognostic factors in the setting of recurrent glioblastoma. Our findings also supported the hypothesis that bevacizumab should not be withheld in elderly patients of good performance status. However, no phase III trials have demonstrated an overall survival benefit primary or recurrent glioblastoma setting, treated with bevacizumab. Current evidence suggests that glioblastoma cells are able to circumvent anti-angiogenic therapy, such as bevacizumab (Bev) and develop resistance to targeted monotherapy via activation of complex molecular escape pathways such as PI3K/mTOR pathway, thereby leading to a paradoxical increase in tumour cell invasion. As a result, it is important to assess bevacizumab based combination treatment strategies in order to improve therapeutic outcomes and enhance our molecular understanding of tumour dynamics.

Implementing advanced clinical neuro-imaging techniques (MRI, PET) we mechanistically interrogated the anatomical, physiological, biochemical and vascular properties of glioblastoma in response to a novel drug treatment strategy, in a reproducible orthotopic model. Our working hypothesis was that the combination of a novel dual PI3K/mTOR inhibitor (BEZ235) and bevacizuamb would convey potent anti-tumour effects in comparison to monotherapy strategies.

Treatment with bevacizumab resulted in a pronounced decrease in tumour volume (T2-weighted MRI). No further reduction on tumour volume was observed with the BEV/BEZ235 combination compared with BEV monotherapy. The proliferation index markers (Ki67 & [18F]FLT) supported the observations. Using ΔR2* and ΔR2 values, the bevacizuamb/BEZ235 combination significantly reduced tumour blood volume and tumour microvessel volume in comparison to bevacizuamb alone. Microvessel density index was further reduced in animals treated with the combination, supported by von Willebrand factor (vWF) immunohistochemistry. [18F]FET uptake was reduced following treatment with bevacizumab alone, but the addition of BEZ235 in the combination group did not further reduced [18F]FET uptake. vWF immunohistochemistry demonstrated that the mean tumour vessel size was increased in all groups.

Conclusion:

Advanced translational neuroimaging techniques provided information on mechanism of action of the drug combination and its potential clinical responses. We have demonstrated that treatment with a BEV/BEZ235 combination could reduce peritumoral oedema reducing the requirement for corticosteroids. Translational studies, similar to this, may help to predict more accurately the clinical potential of the bevacizuamb/BEZ235 combination regimen as a novel therapeutic approach in neuro-oncology.