Patients with glioblastoma (GBM) often suffer from symptomatic epilepsy. Older antiepileptic drugs (AEDs) which affect the enzyme system cytochrome P450 have been in extensive use, but there is an increasing focus on interactions with other drugs. This study investigated whether newer AEDs with little or no enzyme effect are increasingly preferred. Previous research has indicated that valproate improves survival in GBM. We investigated the impact of AEDs on overall survival in GBM patients. All GBM patients diagnosed in Norway 2004-2010 were included through a linkage of national registries, and follow-up data on the malignancy and drug usage were analyzed. In a multivariate cox proportional-hazards regression, AEDs were adjusted for each other and for relevant factors. Immortal time bias was eliminated with time-dependent variables. The study population was 1263 patients with histologically confirmed GBM. Carbamazepine was the most frequently prescribed AED to patients diagnosed with GBM during 2004-2006, while levetiracetam was increasingly prescribed to patients diagnosed later. Taking AEDs on a reimbursement code of epilepsy was not beneficial for survival. None of the six AEDs valproate, levetiracetam, carbamazepine, oxcarbazepine, lamotrigine or phenytoin significantly altered overall survival. There has been a shift in the prescriptions of AEDs to GBM patients from older to newer AEDs over time. We found no significant survival benefit in GBM patients neither from treatment with AEDs for epilepsy in general, nor from the usage of six separate AEDs.

Pseudoprogression is a treatment-related effect seen on imaging in high-grade glioma. Enhancement of gadolinium contrast on control MRI can be misinterpreted as tumor recurrence and is also difficult to distinguish from radiation necrosis. Pseudoprogression is seen in up to 30% after standard treatment for glioblastoma multiforme (GBM), which is radiotherapy concurrent with chemotherapy with temozolomide (TMZ) and adjuvant cycles of TMZ. In this article, the current literature on pseudoprogression in high-grade glioma is reviewed by searches in PubMed. We also present two clinical cases, one of which had medullary pseudoprogression. No articles on this subentity of pseudoprogression were found in PubMed. Standard MRI with gadolinium contrast cannot differentiate between pseudoprogression, tumor recurrence and radiation necrosis. More advanced imaging techniques are often not available. Pseudoprogression seems to be related to methylated promoter of the O(6)--methyl-guanine methyl transferase (MGMT) gene, which is associated with improved treatment effect. Discontinuation or change of therapy on the basis of misinterpretation of MRI as disease progression is thus unfortunate. MRI should be interpreted with caution the first 6 months after standard treatment of high-grade glioma. In a GBM patient with contrast enhancement on MRI but few or no new symptoms and/or stable steroid doses, treatment should be continued and control imaging performed after 2-3 months.