Genetic test results alone may not predict severe epilepsy outcomes

Across a group of children with severe childhood epilepsies, including Dravet syndrome, genetic test results did not independently predict long-term seizure or developmental outcomes, according to a new study.

“Although … genetic testing has diagnostic and precision-medicine selection utility in childhood-onset [developmental and epileptic encephalopathies], long-term seizure and developmental outcomes are driven more strongly by age at onset and baseline neurodevelopmental status than by genetic test positivity alone,” the researchers wrote. “These findings emphasize the continued importance of early clinical recognition, aggressive seizure management, and timely developmental interventions alongside genetic diagnostics to improve long-term outcomes in children with [developmental and epileptic encephalopathies].”

Recommended Reading

July 21, 2026 News by Margarida Maia, PhD

First child dosed in study of new Dravet syndrome treatment

Study examines whether genetic findings predict long-term outcomes

The study, “Does genetic diagnosis influence long-term seizure and developmental outcomes in childhood developmental and epileptic encephalopathies? A longitudinal audit,” was published in Seizure: European Journal of Epilepsy.

Developmental and epileptic encephalopathies (DEEs) are a group of severe epilepsies in which development is affected by the underlying disorder, seizure activity, or both.

Dravet syndrome is a specific type of DEE most often linked to disease-causing variants in the SCN1A gene. Like Dravet syndrome, many other DEEs and severe childhood epilepsies are thought to have a genetic cause. But genetic testing does not identify a disease-causing variant in every child, in part because some epilepsy-linked genes and genetic changes may still be unknown or may not be detected by the testing used.

Identifying a pathogenic variant in SCN1A or another epilepsy-linked gene can help guide diagnosis and treatment. The researchers wanted to determine whether children with an identified disease-causing variant had different long-term seizure and developmental outcomes than those without one.

To answer that question, researchers analyzed data from 255 children treated at one tertiary-care epilepsy center in India. All had a DEE or drug-resistant epilepsy of presumed genetic or unknown cause and had undergone next-generation genetic testing. Forty had Dravet syndrome or a Dravet-borderline phenotype. A pathogenic or likely pathogenic variant was identified in 111 children, or 43.5%. Of those, 51, or 46%, received treatment changes informed by the genetic testing.

The researchers used statistical analyses to identify factors associated with two key outcomes: being seizure-free for at least 12 months at the last follow-up and having an adverse developmental outcome.

Early clinical signs emerge as stronger predictors

Specifically, the likelihood of being seizure-free was lower in children who had developmental delays before seizures began and in those without a family history of seizures. An adverse developmental outcome was associated with developmental delay before seizure onset, seizure onset at age 1 or younger, and developmental regression — the loss of skills a child had previously gained.

“Our findings demonstrate that while [genetic testing] yields a genetic diagnosis in nearly half of affected children … genetic test positivity alone does not independently predict long-term seizure freedom or developmental outcomes. Instead, early age at seizure onset, pre-existing developmental delay, absence of family history of seizures and developmental regression emerge as the most robust predictors of adverse outcomes,” the researchers concluded.

In a more detailed analysis of children with specific epilepsy syndromes, the researchers found that those with Dravet syndrome or a Dravet-borderline phenotype who had a negative genetic test were more likely to be seizure-free than those with a positive result. The researchers said this was consistent with evidence that genetically confirmed Dravet phenotypes, most often involving SCN1A variants, tend to be more resistant to current treatments.