Abstracts

Rationale: Pathogenic variants in CHD2 are associated with developmental and epileptic encephalopathy, as well as neurodevelopmental disorders with or without epilepsy, including intellectual disability (ID) and autism spectrum disorder (ASD). Large cohort studies have characterized clinical signs and symptoms of CHD2-related disorders, but we lack an understanding of disease course over time, which is essential for clinical trial readiness when novel targeted therapies emerge. To address this gap, we characterized the phenotypes in detail of a cohort of patients with CHD2-related disorders seen at our institution.


Methods: We identified patients with CHD2 variants detected between 2013 and 2023 through an internal database search and through the epilepsy genetics clinical program. Data collection included genetic reports, clinical notes (including neurology, neuropsychology, and developmental medicine), as well as EEG and MRI reports and raw data (when available). We used institutional-based artificial intelligence tools to aid in the synthesis and presentation of data.


Results: We identified 21 patients with variants in CHD2. These were detected through epilepsy gene panels (8), clinical whole exome sequencing (WES, 5), research WES with targeted CHD2 variant confirmation (4), autism spectrum disorder or intellectual disability panels (2), chromosomal microarray with deletion including CHD2 (1), and research gene panel at an outside institution (1). At most recent review, 10 were considered pathogenic, 7 likely pathogenic, and 4 of uncertain significance. The latter 4 were then excluded from this analysis.

Of the remaining 17 patients, 16 had seizures. Mean age at first seizure was 3.8 years (+/- 2.96 years). Mean age at genetic diagnosis was 8.2 years (+/- 5.3 years), and mean age at last follow-up was 12.3 years (+/- 6.2 years). 13 developed drug-resistant epilepsy including 10 with status epilepticus.

The most common seizure types were GTC (13) and myoclonic (12; including five with eyelid myoclonia with absences). Nine had three or more types. The most common seizure triggers were fatigue (7), light and sunlight (7), and eye closure (3). All had abnormal EEG findings, including nine with a photoparoxysmal and seven a hyperventilatory response.

15 trialled three or more ASMs. At last follow-up 12 took two or more ASMs, most often VPA (10), LTG (7), and CLB (5). 5 trialled the Ketogenic Diet (KD) and 2 the Modified Atkins Diet – none remained on dietary therapy at last follow-up. 11/16 patients had a stably refractory epilepsy phenotype, with only a minority either worsening (n=1) or improving (n=4) significantly over time.

Eight were diagnosed with ID, eight with ASD, and three with both. 16 had at least one MRI, but there were no consistent radiologic trends.

Conclusions: Our retrospective study identifies similar characteristics to previously reported cohorts of patients with CHD2-related disorders. Additionally, we find that the epilepsy phenotype overall appears to be stable in most patients. Ongoing work includes evaluating changes in neuropsychological testing over time in this cohort. Ultimately this work emphasizes the importance of a prospective natural history study.


Funding: N/A