DEVELOPING A CLINICALLY AND PHYSIOLOGICALLY RELEVANT <EM>IN VIVO</EM> MODEL FOR FETAL VALPROATE SPECTRUM DISORDER

Sodium valproate (VPA) is a potent antiseizure mediation (ASM) with therapeutic benefits for many people with epilepsy. However, prenatal exposure to VPA can result in Fetal Valproate Spectrum Disorder (FVSD). FVSD is characterised by both physical malformations (e.g., spina bifida) and neurodevelopmental disorders (e.g., autism). Although VPA prescription to people of reproductive age is declining, there are ~20,000 afflicted children within the UK and VPA is still commonly prescribed in developing nations; therefore, research questions regarding VPA risks and teratogenic mechanisms remain. Current preclinical FVSD models lack clinical relevance as they use a one-time above-therapeutic dose of VPA at mid-gestation, missing critical early windows of development. We developed a new model to study FVSD in which mice are treated daily with low, medium or high clinically relevant VPA doses, administered non-invasively. We have performed a battery of behavioural tests (e.g., ultrasonic vocalisation, social interaction, open field test) to assess development through to P21 and coupled this with ex vivo molecular (RT-qPCR) and electrophysiological (extracellular hippocampal gamma oscillations) readouts at key timepoints. Our protocol allows for chronic dosing throughout gestation at teratogenic concentrations with pharmacokinetics that allow us to maintain therapeutic doses. Preliminary analyses suggest specific impacts on certain behaviours, including USVs, alongside morphological disruption to hippocampal CA1. We will use our model to both characterise teratogenicity following specific VPA exposure profiles and screen newer ASMs. This will provide urgently needed preclinical evidence regarding the potential reproductive risk of epilepsy medications taken during pregnancy.