Rationale: Glioblastoma multiforme (GBM), a malignant and invasive brain cancer with high resistance to treatment, presents with seizures as one of the first symptoms. These seizures have been particularly resistant to anticonvulsant medications which further increases the rate of mortality. Most of the pre-clinical models of GBM do not represent a clinical scenario of the human disease. The goal of this project was to determine and optimize a randomized pre-clinical translational model of GBM-induced seizures that will allow for testing of potential experimental anti-tumoral or antiseizures therapies for patients with GBM. Methods: U87MG cells (5 x 105 cells in 5 uL serum-free DMEM ) that were marked with firefly luciferase were stereotactically implanted via a microsurgery into the hippocampal region of female BalbC (nu/nu) mice that were approximately six to eight weeks of age. Control mice received 5 uL of serum free DMEM injected into the hippocampal region. Post-operative weights, neurological outcomes, responses to pain, and reflexes were gathered twice a week for the duration of the study to ensure the animal’s wellbeing. Tumor growth was monitored weekly using the IVIS fluorescence imaging system. Approximately 15 days after tumor growth, animals with progressive tumor growth and their respective controls received pentylenetetrazol (PTZ) at sub convulsive doses (10mg/kg) injected intraperitoneally (IP) every five minutes until the animal achieved tonic clonic seizures or reached a total dose of 60mg/kg. Then, animals that showed an increase in seizure susceptibility (Racine’s score = or > 3) compared to controls, without any neurological deficit or cachexia were randomly selected to test the effectiveness of the anticonvulsive drug midazolam (MDZ). On day 17, animals were injected with MDZ one hour before a convulsive dose of PTZ (35mg/kg, IP). Control animals were injected with saline instead of MDZ. Seizure severity and latency to maximal seizure were measured using the Racine scale. Results: Ninety percent of mice with GBM showed no neurological deficits or cachexia. GBM-implanted mice showed progressive growth and a positive correlation between tumor size and seizure susceptibility. Mice with GBM showed an increase of seizure severity (p=0.0015) and latency to the seizure severity compared to controls (p=0.011). There was no significant difference (p=0.079) in the maximum Racine’s score reached among the GBM population. MDZ administration in GBM mice protected against seizures induced by a single convulsive dose of PTZ compared to the vehicle treated mice (p= 0.0003). Conclusions: This protocol allows for the testing of safely and effectively clinically achieved seizures without severe neurological complications induced by the GBM that could interfere with the evaluation of the pathogenesis of the seizure. Seizure susceptibility associated with GBM in this in vivo model could represent an aberrant metabolism or expression of the main components of the excitatory neurotransmission. Currently, further characterization of the evolution of GBM and its association with spontaneous recurrent seizures are under investigation. Funding: Please list any funding that was received in support of this abstract.: UCB
