A Randomized Phase II Trial of Vandetanib (ZD6474) in Combination With Carboplatin Versus Carboplatin Alone Followed by Vandetanib Alone in Adults With Recurrent High-Grade Gliomas

Background: In vivo experiments have documented the ability of vandetanib (ZD6474) to inhibit tumor growth in various preclinical tumor models. Given the pronounced neovasculature associated with malignant gliomas, and abundant published data demonstrating the dependence of glioma growth on the maintenance and proliferation of this neovasculature, vandetanib represents a potentially promising new therapeutic approach to these otherwise refractory tumors. Phase II data of vandetanib for recurrent glioblastomas conducted at the National Institutes of Health showed promising activity but responses were usually short-lasting. Carboplatin has shown activity as monotherapy in the treatment of recurrent malignant gliomas in adults and preclinical data generated at Dr. Fines' laboratory demonstrate additive anti-glioma activity with vandetanib. The safety profile of carboplatin and the preclinical and clinical data supports its use in combination with vandetanib in patients with malignant gliomas. Vandetanib is also an epidermal growth factor receptor (EGFR) inhibitor and it has been demonstrated that the presence of the EGFRvIII mutant and/or the presence of an intact phosphatase and tensin homolog deleted on chromosome 10 (PTEN) and non-phosphorylated protein kinase B (AKT) predict for a higher likelihood of response to the EGFR inhibitors Tarceva and Iressa. Objectives: To establish data regarding the anti-tumor activity of vandetanib in combination with carboplatin and single agent carboplatin and to collect information regarding the spectrum of toxicities. To determine if the presence of the EGFRvIII mutant and/or the presence of an intact PTEN and non-phosphorylated AKT predict for a higher likelihood of response to vandetanib. Eligibility: Patients with histologically proven malignant glioma are eligible for this study. Design: Patients will be randomized (1:1) to one of two groups. Patients in group one will be treated with vandetanib (300 mg daily for patients not on enzyme inducing anti-epileptic drugs (EIAEDs) and 400 mg daily for patients on EIAEDs) and with carboplatin (area under the concentration-time curve at steady-state \[area under curve (AUC)\], 6mg/mL x min) once every 4-week cycle (combination group). Patients in group two will receive carboplatin alone (AUC 6mg/mL x min) once every 4-week cycle. Patients who develop tumor progression or unacceptable toxicity on carboplatin alone (group 2) can then receive single agent vandetanib (300 mg daily for patients not on enzyme inducing anti-epileptic drugs (EIAEDs) and 400 mg daily for patients on EIAEDs) in 4-week cycles (sequential group). A total of 128 evaluable patients will be analyzed. The total accrual ceiling will allow for 74 patients to be enrolled in the glioblastoma multiforme (GBM) stratum and 74 patients in the anaplastic gliomas (AG) stratum (total 148) to factor in replacing those patients who come off treatment prior to cycle 1.