Abstract
PURPOSE: Symptomatic epilepsy is a common complication of glioblastoma and requires pharmacotherapy. Several uncontrolled retrospective case series and a post hoc analysis of the registration trial for temozolomide indicated an association between valproic acid (VPA) use and improved survival outcomes in patients with newly diagnosed glioblastoma. PATIENTS AND METHODS: To confirm the hypothesis suggested above, a combined analysis of survival association of antiepileptic drug use at the start of chemoradiotherapy with temozolomide was performed in the pooled patient cohort (n = 1,869) of four contemporary randomized clinical trials in newly diagnosed glioblastoma: AVAGlio (Avastin in Glioblastoma; NCT00943826), CENTRIC (Cilengitide, Temozolomide, and Radiation Therapy in Treating Patients With Newly Diagnosed Glioblastoma and Methylated Gene Promoter Status; NCT00689221), CORE (Cilengitide, Temozolomide, and Radiation Therapy in Treating Patients With Newly Diagnosed Glioblastoma and Unmethylated Gene Promoter Status; NCT00813943), and Radiation Therapy Oncology Group 0825 (NCT00884741). Progression-free survival (PFS) and overall survival (OS) were compared between: (1) any VPA use and no VPA use at baseline or (2) VPA use both at start of and still after chemoradiotherapy. Results of Cox regression models stratified by trial and adjusted for baseline prognostic factors were analyzed. The same analyses were performed with levetiracetam (LEV). RESULTS: VPA use at start of chemoradiotherapy was not associated with improved PFS or OS compared with all other patients pooled (PFS: hazard ratio [HR], 0.91; 95% CI, 0.77 to 1.07; P = .241; OS: HR, 0.96; 95% CI, 0.80 to 1.15; P = .633). Furthermore, PFS and OS of patients taking VPA both at start of and still after chemoradiotherapy were not different from those without antiepileptic drug use at both time points (PFS: HR, 0.92; 95% CI, 0.74 to 1.15; P = .467; OS: HR, 1.10; 95% CI, 0.86 to 1.40; P = .440). Similarly, no association with improved outcomes was observed for LEV use. CONCLUSION: The results of this analysis do not justify the use of VPA or LEV for reasons other than seizure control in patients with newly diagnosed glioblastoma outside clinical trials.