Sunitinib to Treat Recurrent Brain Cancer

Background: Solid tumors have multiple mechanisms for stimulating angiogenesis with the vascular endothelial growth factor (VEGF)-kinase insert domain receptor (KDR) axis being only one of them. Sunitinib, through its multiple tyrosine kinase receptor targets, represents an attempt to capitalize on the concept of targeting multiple mechanisms responsible for glioma-associated angiogenesis. Sunitinib inhibits platelet derived growth factor receptor (PDGFR) and c-kit (stem cell factor (SCF) receptor) at nanomolar concentrations. The combination blocks all three known major glioma-mediated angiogenic mechanisms (VEGF, c-kit, PDGF). Based on this scientific rationale, the promising anti-glioma activity of sunitinib in preclinical models, and the promising clinical data in patients with gliomas treated with other VEGF inhibitors, we are now proposing a phase II trial of sunitinib in patients with recurrent malignant gliomas. Objectives: To evaluate the anti-glioma activity of sunitinib in patients with recurrent malignant gliomas who are either naive or resistant to prior bevacizumab therapy. Eligibility: Patients with recurrent malignant glioma are eligible for this study. Design: This is a phase II study with a target enrollment of 64 (32 with glioblastoma multiforme (GBM) and 32 with anaplastic glioma (AG)) patients who have not progressed on prior treatment with anti-VEGF therapy, and 64 (32 with GBM and 32 with AG)patients who have progressed on prior bevacizumab therapy. Sunitinib will be self-administered orally at 37.5 mg daily, with dose adjustments allowed for toxicity and concomitant drug interactions. The primary endpoint is six-month progression free survival for both arms of the study. We performed an interim analysis (see below) and after consulting with the pharmaceutical company (Pfizer) we believe that ending the study is the most appropriate course of action to take. In the original design of the trial, for the AG stratum, the agent will be considered effective if at least 14 patients have not progressed by 6 months. Now, we have observed one out 10 Bev-naïve who haven't progressed by 6 months. Then given one patient in the first 10 patients whose PFS \>=6, the conditional probability of observing at least 14 out of 32 patients who are 6 month-PFS is only 43% even if the true 6 months-PFS is 55%. Since in the first 10 patients, we only observed one patient with PFS\>=6 months, it is unlikely that the true 6 month-PFS is 55%. The conditional power would be close to zero, if the true 6 month-PFS close to 10% as observed so far. Based on the conditional probability, the chance of a positive study is small, and hence the trial should be considered stopped for futility. For Bev-resistant patients, because in the first 12 patients none had PFS \>= 6 months the conditional probability of declaring the agent being effective is even lower and equals 12%, if the true PFS at 6 months is 55%. Essentially there is little chance that the agent can be declared effective.