Genotype-driven therapeutics in DEE and metabolic epilepsy: navigating treatment efficacy and drug resistance

Abstract

Neonatal intensive care unit (NICU), particularly in treating developmental and epileptic encephalopathy (DEE) and metabolic epilepsy (ME), requires a deep understanding of their complex etiologies and treatment responses. After excluding treatable cases such as infectious or autoimmune encephalitis, our focus shifted to a more challenging subgroup of 59 patients for in-depth genetic analysis using exome sequencing (ES). The ES analysis identified 40 genetic abnormalities, significantly including de novo variants. Notably, we found structural variation as duplications in regions 2q24.3, including SCN1A and SCN2A were observed in 7 cases. These genetic variants, impacting ion channels, glucose transport, transcription regulation, and kinases, play a crucial role in determining medication efficacy. More than one-third (34.2%) of patients with DEE had an unfavorable response to anti-seizure medications (ASMs) in the chronic phase. However, since the ketogenic supplementary diet showed a positive effect, more than three-quarters (80%) of these drug-resistant patients improved during a 3-month follow-up. In contrast, the ME had a lower adverse reaction rate of 9.1% (2/22) to specialized medications, yet there were 5 fatalities and 10 cases with unidentified genetic etiologies. This study suggests the potential of categorizing drug-resistant variants and that a ketogenic diet could be beneficial in managing DEE and ME. It also opens new perspectives on the mechanisms of the ketogenic diet on the discovered genetic variants.

Keywords: Clinical sequencing; DEE; Drug response; Ketogenic diet; Metabolic epilepsy.

Fig. 1

Flowchart outlining the study of pediatric NICU patients with seizures. A total of 172 patients underwent tests for autoimmune and infectious encephalitis. Positive patients were treated accordingly, while those testing negative were referred for genetic analysis via exome sequencing (ES), splitting them into ME and DEE groups. The results of ES are briefly represented as gene names with accompanying numbers in the following order: the first number indicates variants in drug-responsive cases, the second number specifies variants marked with an asterisk (*) for therapeutic resistance or ‘d’ for decreased response situations. Long-term outcomes, including seizure control, resistance, and mortality, were monitored. Some patients receive a KD as a supplement, especially those with a poor response to ASMs. Treatment success was assessed in relation to the patients’ genetic profiles.

Fig. 2

Variant types across genes for DEE and ME group with SCN1A shows the highest number of missense variants, while CDKL5 has a substantial number of both missense and nonsense variants. PGM1 showing mixed missense/nonsense and POLG having a notable frameshift presence.

Fig. 3

Overall treatment response by genetic variation. The Sankey diagram maps genetic variants to treatment responses, including ASMs and a KD. The left panel lists genes associated with seizure conditions. The middle panel shows ASM response categories to treatments: seizure-free, partial, resistant, suboptimal, and dead. The right side shows how these responses might change with the addition of a KD, with categories such as partial – KD and resistant – KD, suggesting partial or full resistance to treatment despite the diet. The width of the bands represents the number of patients exhibiting each treatment response per genetic variant.