Abstracts

Rationale: Up to 60% of patients with brain tumors develop epilepsy. Brain tumor patients with epilepsy experience increased morbidity, mortality and a lower quality of life. Meticulous consideration of the combined effects and side effects of both cancer and epilepsy treatment must be undertaken in order to best optimize care. The underlying pathophysiology of tumor-related epilepsy remains poorly understood with even less known regarding pediatric brain tumor-related epilepsy. Here, we described features of epilepsy risk and treatment in pediatric brain tumors. Methods: Six hundred ninety-one patients from June 2010 - March 2018 were seen by pediatric neuro-oncology at University of California San Diego (UCSD)/Rady Children’s Hospital. Of these patients, 607 had brain tumors. Retrospective review of their charts identified 66 patients as having epilepsy. Data collected from this cohort included identification of seizure upon presentation, electroencephalogram (EEG) findings (spiking, slowing, or normal), seizure type (generalized, partial, none), anti-epileptic drug (AED) chosen, number of AEDs utilized, tumor size (small < 3cm vs large >= 3 cm), tumor edema (small < 1 cm vs large >= 1cm), and resection type (gross total, subtotal, near total, none). Patients requiring two or more AEDs for treatment were categorized as having intractable epilepsy. Results: Of the patients consulted by pediatric neuro-oncology at UCSD/Rady Children’s Hospital, 10% (n=66) had both brain tumor and epilepsy.  Of these patients, 59% (n=39) had seizure upon presentation. 83% (n=55) of pediatric brain tumor patients with epilepsy remain on antiepileptic treatment despite cancer treatment. 1^st line antiepileptic drug of choice was typically Levetiracetam (59%) followed by Oxcarbazepine (12%) and then Valproate (5%). 38% (n=25) in this cohort had intractable epilepsy requiring two or more AEDs to treat their seizures. 61% (n=40) had spikes found on EEG. No significant correlation was found between intractable epilepsy and seizure on presentation, seizure type, tumor size, tumor edema, or resection type. No significant correlation was found between seizure on presentation and current use of AED or EEG findings. Smaller tumors (<3 cm in size) were significantly correlated with spikes seen on EEG (p<0.01). Those with intractable epilepsy were less likely to have slowing seen on EEG (p<0.04). Conclusions: In summary, with these results, future families and patients can be educated regarding the possibility of developing epilepsy if seizure is the initial presentation of a brain tumor. Additionally, counseling can be provided regarding the high risk of treatment with continued antiepileptic medications, with some requiring two or more AEDs. Clinicians can be guided with regards to 1st line antiepileptic drug choice with Levetiracetam being frequently chosen. Thus, understanding clinical features of pediatric brain tumor-related epilepsy can help inform on future decision making regarding approaches to epilepsy management. Funding: None