Abstracts

Clinical Validation of SCN1A Variant-Phenotype Association

Abstract number : 3.441
Submission category : 4. Clinical Epilepsy / 4D. Prognosis
Year : 2023
Submission ID : 1426
Source : www.aesnet.org
Presentation date : 12/4/2023 12:00:00 AM
Published date :

Authors :
Presenting Author: Ross Carson, MD – Boston Children's Hospital

Christelle Moufawad El Achkar, MD – Boston Children's Hospital; Jack O’Keeffe Donohue, MD – Boston Children's Hospital; Ivan Ruiz, BA – Boston Children's Hospital; Rozalia Valentine, MS CGS – Boston Children's Hospital

Rationale:
Diagnosis of a pathogenic variant in the sodium channel alpha 1 subunit (SCNA1A) gene early in patients with epilepsy has great implications due to treatment and prognostic implications. The two most common phenotypes associated with SCN1A variant  can include Dravet Syndrome (DS), a developmental epileptic encephalopathy with difficult to control seizures and status epilepticus, developmental regression, and autism spectrum disorder, and Genetic Epilepsy with Febrile Seizures Plus (GEFS+), a milder phenotype not associated with development or cognitive impairment and less severe epilepsy. Age of onset can be a helpful clue as patients with seizures before age one year are more likely to develop DS vs. GEFS+. However, even patients with DS do not present with any developmental delays until after 18-24 months of age, making this early prognostic prediction more difficult for clinicians and caregivers. A prediction model was published in Neurology by Brunklaus et al. in 2022 using age of seizure onset and a novel SCN1A gene variant scoring tool. The goal of this project is to validate this tool in our patient population and examine if additional clinical information may be useful to further refine this tool.

Methods:
We performed a retrospective review of 121 patients with SCN1A pathogenic or likely pathogenic variants and epilepsy at Boston Children’s Hospital. This initial analysis included 57 patients who had both a documented clinical diagnosis by a clinician in their chart and genetic information available (n=50 Dravet and n=7 GEFS+).

Results:
We compared the prediction model score to the clinical diagnosis and found a statistically significant association between the clinical diagnosis with both the prediction score by t-test and predicted diagnosis by chi square analysis. Interestingly the prediction model was more accurate at identifying Dravet than GEFS+ in this subset of our population. Additional clinical variables were evaluated, including first seizure type (generalized vs. focal onset), presence of status epilepticus before 24 months of age, presence of myoclonic seizures, presence of hemiclonic seizures, and fever/immunization documented as a trigger. The presence of myoclonic seizures reached statistical significance while the presence of hemiclonic seizures trended towards significance. None of the other variables were found to be significantly associated.

Conclusions:
Next steps will include analyzing our entire cohort of 121 patients and refining the clinical diagnosis based on the Delphi criteria for DS and assessing the sensitivity and specificity of the tool based on several variables (independent and in combination).

Funding:
This project was unfunded.



Clinical Epilepsy