How a Common Epilepsy Drug Affects Fetal Brain Development

By VirentaNews Staff — May 23, 2026

In a dimly lit lab at the Karlsruhe Institute of Technology, tiny clusters of cells pulse with faint electrical activity, not unlike the earliest sparks of a developing human brain. These are cerebral organoids—3D models grown from human stem cells that mimic the architecture and function of the fetal brain. For researchers, they are a window into the first weeks of neurodevelopment, a period long hidden from direct observation. Now, these miniature brains have revealed a troubling truth: exposure to valproate, a medication prescribed to millions of women with epilepsy, triggers profound disruptions in neural growth within just 30 days. The findings, drawn from a collaborative study involving KIT, the Heidelberg Academy of Sciences and Humanities, and the Universities of Tübingen and Heidelberg, underscore a long-suspected danger and offer concrete biological evidence of how this common drug interferes with the delicate choreography of brain formation.

Valproate Alters Neural Growth in Lab Models

After 30 days of exposure to therapeutic levels of valproate, human brain organoids showed significant structural and molecular deviations from healthy controls. The drug induced abnormal cell proliferation, disrupted the balance between excitatory and inhibitory neurons, and impaired the formation of cortical layers—critical scaffolding for cognitive function. RNA sequencing revealed dysregulation in over 300 genes linked to neurodevelopment, including those involved in axon guidance and synaptic formation. The organoids also displayed altered electrophysiological patterns, suggesting functional deficits that could mirror the neurodevelopmental disorders observed in children exposed to valproate in utero. Published in Molecular Psychiatry, the study provides the first direct evidence of valproate’s impact on early human neural tissue outside the womb, confirming clinical observations with cellular precision.

The Long History of Valproate and Birth Risks

Valproate, marketed under names like Depakote and Epilim, has been a cornerstone of epilepsy treatment since the 1960s. It is also used for bipolar disorder and migraine prevention. But its legacy is shadowed by decades of epidemiological evidence linking prenatal exposure to autism, intellectual disability, and spina bifida. As early as the 1980s, clinicians noted higher rates of congenital malformations in children born to mothers taking the drug. By 2018, the European Medicines Agency issued strict warnings, advising against its use in pregnant women and those of childbearing age unless strictly necessary. Despite these cautions, valproate remains in use globally due to its unmatched efficacy in controlling certain seizure types. The new organoid study bridges the gap between population data and biological mechanism, showing not just that harm occurs, but how—by derailing the very foundation of brain architecture during its most vulnerable phase.

Scientists Behind the Breakthrough

The research team, led by neuroscientists at KIT and developmental biologists from Heidelberg and Tübingen, combined expertise in stem cell biology, pharmacology, and systems neuroscience. Their goal was not merely to confirm valproate’s toxicity but to model it in a human-relevant system that avoids the limitations of animal studies. “Mice don’t perfectly replicate human brain development,” explained Dr. Lena Hofmann, co-lead investigator at KIT. “Organoids allow us to watch how human neurons respond in real time.” The team’s interdisciplinary approach—merging gene expression analysis, live imaging, and electrophysiology—enabled a comprehensive view of valproate’s effects. Their motivation extends beyond documentation; they aim to use these models to screen safer alternatives and identify molecular pathways that could be protected or bypassed during treatment.

Implications for Patients and Clinicians

For women with epilepsy, the findings reinforce the urgency of preconception counseling and alternative treatment plans. While stopping valproate abruptly can be dangerous, the study supports the need for personalized, risk-aware management during reproductive years. Regulatory agencies may use this data to strengthen prescribing guidelines, particularly in low-resource settings where valproate remains widely prescribed due to cost and availability. For drug developers, the organoid platform offers a tool to test neurodevelopmental safety early in the pipeline. The research also raises ethical questions about the use of psychotropic medications during pregnancy, emphasizing that efficacy must be weighed against invisible, long-term developmental costs.

The Bigger Picture

This study is part of a growing shift toward human-based models in medical research, reducing reliance on animal data and improving translational accuracy. As organoid technology advances, it holds promise for modeling the effects of environmental toxins, infections, and other drugs on fetal development. The valproate findings exemplify how such models can transform public health policy by grounding recommendations in human biology. They also highlight the unintended consequences of medications that cross the placental barrier, urging a reevaluation of how we assess drug safety in pregnancy—a domain long underserved by clinical trials.

What comes next is not just more research, but action. The team plans to test protective compounds that might counteract valproate’s neurotoxic effects without compromising seizure control. Meanwhile, the organoid model itself may become a standard for developmental neurotoxicity screening. For the millions of women navigating epilepsy and family planning, this work offers both a warning and a path forward—a vision of treatments that protect both mother and child without sacrificing one for the other.

Source: MedicalXpress