Preclinical evaluation demonstrates in vitro seizure liability profiling across drug candidates, highlighting its utility for early central nervous system risk detection.
Central nervous system (CNS) liabilities remain a significant cause of attrition during small molecule drug development. We previously developed the in vitro seizure liability assay (iLA seizure ), a new approach methodology (NAM) comprising a human induced pluripotent stem cell neuron microelectrode array (MEA) assay coupled with a panel of human ion channels associated with CNS perturbation and especially seizure, evaluated by electrophysiology. Here, we present a series of illustrative case studies demonstrating the potential applications of iLA seizure . Amoxapine (shows clinical seizures) gave concentration‐dependent changes in MEA parameters and inhibited seizure‐related ion channels with high potency (IC 50 < 30 μM) at the human therapeutic concentration range (0.57–1.91 μM). Comparison of seizure risk in five compounds in a lead series showed clear ranking, with no changes in any of the MEA parameters for one of the five compounds. In a second case study, there were dog‐specific seizures in good laboratory practice (GLP) toxicology: this was explained by a clear seizurogenic phenotype at 100 μM in MEA for a dog‐specific metabolite and a hit on the ion channel panel at NMDA1/2A with high potency (IC 50 of 19.3 μM). In the third case study, one clinical candidate showed no changes in MEA parameters at the clinical C max of 1 μM, whereas the other clinical candidate showed a potent seizurogenic phenotype with changes in many MEA parameters at 3–10 μM, correlating with the clinical C max of 5 μM. Overall, iLA seizure offers new opportunities to predict, understand, and avoid CNS side effects such as seizures in drug discovery and development, before animals, resources, and time have been wasted. This work should be considered within the context of recent proposed updates to safety pharmacology guidelines, such as ICH S7A. However, further work is needed to validate assay performance and clarify the context of use.
No takes yet. Share an insight, caveat, or question.
Rockley et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: