Background - Antiseizure medicine discovery has historically relied on the demonstration of anticonvulsant activity in several historical rodent models of seizure and epilepsy. While there are many different animal seizure models to employ during early lead testing and IND-enabling studies, the specific models that are most predictive of human clinical epilepsy are harder to anticipate for any given investigational compound. To fill this knowledge gap, we performed a concordance analysis between published rodent seizure model efficacy and tolerability response and human focal onset seizure efficacy. This study provides critical early-stage evidence to determine which seizure tests make the most sense in any early screening campaign and how this data may translate to human focal onset epilepsy.
Approach - Using published preclinical response data for the 32 currently marketed Food and Drug Administration-approved antiseizure medicines, the therapeutic response was established between median tolerability (as measured by a fixed-speed rotarod test) and median antiseizure efficacy for each rodent seizure test. These historical rodent models included the maximal electroshock test, 6 Hz (at both 32 and 44 mA current intensities), subcutaneous pentylenetetrazol, mouse audiogenic seizure model, and corneal kindling mouse tests. The responses of each rodent seizure test to each antiseizure medicine was compared to the response in human patient datasets. The alignment between each animal seizure test and the human data was compared on a scale from 100% concordance to 0% concordance to develop a translational framework to determine which rodent seizure models offer the most value to inform clinical direction.
Results - Using this framework, the rodent maximal electroshock seizure (MES), mouse audiogenic seizure, mouse 6 Hz (32 mA), and rat amygdala kindling models demonstrated the greatest data availability and translational relevance to human focal onset epilepsy. It should be noted, however, that these are the most well-established and widely used models. A wide variety of antiseizure medicines have been tested in these models, making their data availability the most well-established, which gives greater confidence in their likelihood for clinical response.
Outcomes - This study provides evidence that the most cost-effective and reliable acutely evoked early screening models for human focal onset epilepsy are the mouse maximal electroshock test and mouse 6 Hz (32 mA) test. These models consistently exhibit high predictive validity for human focal onset epilepsy. Additionally, the corneal kindled mouse model is a strong additional screening assay but there is less published literature with this model.
© 2026 Neurostrategix LLC.