Towards the further understanding of prognosis and outcomes for patients with high grade gliomas: A randomised phase II study of Carboplatin and Bevacizumab in recurrent glioblastoma

Abstract

Glioblastoma (GBM) is the most common, and the most aggressive, adult brain tumour, affecting over one thousand Australians per year. It carries a high mortality burden and a high social burden to both the cancer sufferer and their family. GBM recurs in nearly all patients within months of treatment, and median overall survival from diagnosis remains less than 18 months. At recurrence, no chemotherapy drug has been associated with reliable and meaningful gains in survival, and no new drugs have been added to the Australian Pharmaceutical Benefits Scheme for GBM since temozolomide over a decade ago. We need better treatments for this devastating disease.

Multiple trials have tested the role of targeted therapies – either alone or in addition to chemotherapy – in both the newly diagnosed and recurrent disease settings. Bevacizumab, a vascular endothelial growth factor inhibitor monoclonal antibody, showed promise in early and small clinical trials in recurrent GBM. We developed and conducted a multi-centre, stratified randomized phase II trial comparing bevacizumab monotherapy with bevacizumab + carboplatin chemotherapy in patients with recurrent GBM - the CABARET (Carboplatin And BevAcizumab in REcurrenT glioblastoma) clinical trial.

CABARET represents a substantial achievement in Australian medical research. This is the largest investigator-initiated brain tumour clinical trial in Australia to date. It was conceptualized, developed and conducted entirely within Australia. The primary objective of CABARET was to determine the effect of bevacizumab plus carboplatin versus bevacizumab monotherapy on progression-free survival in patients with recurrent GBM. The trial found that adding carboplatin to bevacizumab resulted in more toxicity without additional clinical benefit: no significant differences in response rate, progression-free or overall survival.

Several secondary and exploratory endpoints have been investigated. This was the first prospective study to assess the role of continuing bevacizumab beyond progression. Albeit with a limited sample size, no differences in survival outcome were noted after patients with progression on study underwent a second randomization to either continue or cease bevacizumab. We measured time to deterioration in health-related quality of life, finding no difference between the two arms but noting that up to 50% of patients gained some improvements in relevant quality of life parameters while on treatment, suggesting benefit from treatment. We also assessed the role of early MRI after four weeks on study, finding this to be a robust indicator for overall survival in this patient cohort. Finally, we found that using centralized radiology review resulted in significantly earlier time to progression, but the absolute difference for the cohort of 0.3 months was not felt to be clinically relevant.

The study results have significantly improved knowledge regarding the use of bevacizumab in the setting of recurrent GBM. We successfully challenged and refuted the dogma, previously established from either retrospective studies or single arm early phase trials, that bevacizumab should be continued after progression or combined with chemotherapy in GBM. This will have significant impact on both future clinical trial design, and also managing patients with this disease, as we have further insight into anticipated effects and survival in the Australian context.