Seizures During Radiotherapy for High-grade Gliomas

Gliomas represent the most common type of primary brain tumors and are frequently associated with clinical symptoms including seizures. The majority of patients with high-grade gliomas (grade III or IV according to the classification of the World Health Organization) receive radiotherapy with or without chemotherapy, either after neurosurgical resection or as definitive treatment after biopsy. For patients with grade III or IV gliomas, pre-treatment seizure rates of 29-67% and 9-45%, respectively, were reported. In a study from our group, the prevalence of seizures prior to radiotherapy was 48.8% in patients with grade III gliomas and 21.5% in patients with grade IV gliomas, respectively.

Radiotherapy can improve long-term seizure outcomes in glioma patients. However, due to an acute inflammatory reaction, radiotherapy may lead to acute onset or progression of clinical symptoms including seizures. To our knowledge, no studies are available that focused on the subacute effect of radiotherapy on seizure frequency during radiation treatment in glioma patients. These data would be important to improve the monitoring and, if required, the anti-epileptic treatment of these patients.

The primary objective of this trial is to evaluate the seizure frequency during a course of radiotherapy for high-grade (grade III or IV) gliomas.

The patients keep a seizure diary during the period of radiotherapy and up to 6 weeks following radiotherapy. Every day, the patients document the number (and type) of seizures and intake of anti-epileptic medication. At the end of radiotherapy, the patients are asked to complete a questionnaire regarding their satisfaction with the seizure diary. One a week during the radiotherapy course, the seizure diary will be reviewed by a medical staff member. During the 6 weeks following radiotherapy, the patients are contacted by phone (to minimise the number of visits to the hospital) once a week to obtain the information from the seizure diary regarding number and type of seizures. During and following radiotherapy, the weekly findings of the seizure diary are discussed with a neurologist to initiate or adjust anti-epileptic medication, if necessary. Patient satisfaction with the seizure diary will be assessed at the end of radiotherapy using a questionnaire and subjected to descriptive analysis. In case of a dissatisfaction rate >40%, the seizure diary will be considered not suitable for patients with high-grade gliomas.

To assess the rate of patients with progression of seizure activity during the course of radiotherapy compared to baseline, defined as

• increase of frequency of seizures by more than 50% during the course of radiotherapy as documented in the seizure diary or
• increase of severity of seizures, i.e. increase of generalized seizures by more 50% during the course of radiotehrapy as documented in the seizure diary or
• increase of the dose of anti-epileptic medication by at least 25% or initiation of anti-epileptic medication during the course of radiotherapy.

To obtain an objective assessment of seizure activity in addition to patient reported outcomes, an electroencephalography (EEG) is performed during the first and the sixth week of radiotherapy, and during the sixth week following radiotherapy. Activity typical for epilepsy includes spike waves, sharp waves and/or sharp slow waves and is classified as absent or present. A mean change to baseline (during first week of radiotherapy) by 50% regarding the number of patients with EEG activity typical for epilepsy will be considered clinically relevant.

The main goal of the study is to generate objective data regarding the occurrence, frequency and severity of seizures as well as regarding the use of anti-epileptic medication during the course of radiotherapy for high-grade gliomas. These data are used to evaluate the potential effect of radiotherapy on occurrence of seizures in these patients and generate hypotheses. Therefore, statistical analyses of primary and secondary endpoints focus on descriptive methods. If statistical tests are applied, they are to be interpreted from an exploratory perspective. The seizure frequency at baseline and during the course of radiotherapy will be calculated by adding the seizures during each time period evaluated. The resulting sum will divided by the total duration (days), excluding those days with no available diary data. This figure will be normalized to a weekly rate. The resulting normalized frequencies form the basis for calculating the primary endpoint. The point estimate of the rate of progressors (increased seizure activity) will be presented together with 95% confidence interval. To test whether the rate of progressions is significantly increased beyond 10%, a one-sided binomial test at a one-sided 2.5% significance level will be applied.

Normalized seizure frequencies over time will also be calculated within sequential 3-week intervals, namely weeks 1-3 and 4-6 during radiotherapy, and weeks 1-3 (weeks 7-9 in total) and weeks 4-6 (weeks 10-12 in total) following radiotherapy. These frequencies will be subjected to descriptive statistics as well as graphical presentations.

Thirty-two patients with documented start of radiotherapy and any documented diary data at baseline and after start of radiotherapy should be subjected to statistical analysis. Assuming that 10% of patients do not fulfil these requirements, a total of 35 patients should be enrolled to this trial. This sample size is set at the maximum that is deemed achievable in this single-center trial within the timeframe of the study given the size of the target population. However, with this sample size a one-sample binomial test with a one-sided significance level of 2.5% has a power of 80% to yield statistical significance if the rate of patients with progression of seizure events during the course of radiotherapy compared to baseline is 30% (rate under the alternative hypothesis) assuming a 'natural' background progression-rate of 10% without radiotherapy (null hypothesis). The latter rate was chosen after discussions with neurologists. If the natural course of the disease would lead to a progression-rate of 5% without radiotherapy only, the power increases to 98%.

The recruitment of all 35 patients (32 patients plus 10% drop-outs) should be completed within 12 months. The radiotherapy period will be 6-6.5 weeks, and the follow up period 6 weeks. This equals a total running time for the trial of approximately 15 months.