Abstracts

Rationale:
Pathogenic variants in the SCN1A gene can result in a spectrum of seizure disorders ranging from mild genetic epilepsy with febrile seizure plus (GEFS+) to severe Dravet syndrome (DS) associated with intellectual disability and significant comorbidities. In this study, we evaluate whether a newly developed prediction model using clinical and genetic data (1) is able to predict specific features of SCN1A-related epilepsies, such as disease severity and developmental outcome in a large cohort of individuals with pathogenic SCN1A variants that have been followed-up over a period of 10 years.
Method:
We analysed a cohort of 255 individuals with pathogenic SCN1A variants from our institutions genetic testing and clinical research database. Clinical information on disease characteristics such as age at epilepsy onset, seizure types, developmental status and presence of a movement disorder was available. A subset of 140 individuals had completed structured questionnaires in 2009 detailing epilepsy severity, comorbidities and quality of life data, with additional 10-year follow-up data available from 2019 (2;3). The newly developed prediction model includes age at seizure onset and a novel SCN1A-specific genetic score combining paralog conservation of the mutated amino acid position with the physicochemical properties of the observed substitution. Using the new model, we applied an ordinal univariate logistic regression model to predict clinical outcome measures such as epilepsy severity, learning disability, comorbidities and quality of life, each adjusted for the age at assessment. Results241 DS and 14 GEFS+ patients were included in this study. The new model predicts the severity of learning disability in the entire study cohort (p = 0.002), as well as in the group of DS patients only (p = 0.044) and performs better than individual predictor components such as ‘age at epilepsy onset’ (p = 0.319) or ‘mutation type’ (truncating vs non-truncating; p = 0.981, table 1). The new model predicted epilepsy severity at initial (p < 0.001) and 10-year follow-up assessment (p = 0.033) as well as quality of life (p = 0.038) and the presence of a movement disorder (p = 0.027). The predictive power of the new model was superior compared to previously established individual predictors such as ‘age at epilepsy onset’ or ‘mutation type’ alone (table 1).
Conclusion:
Using a novel clinico-genetic prediction model allows better prediction of disease outcomes in individuals with SCN1A-related epilepsies, assisting prognostic counselling and patient management. References •Brunklaus et al. (2020) A clinico-genetic prediction model facilitates early diagnosis of Dravet syndrome. AES 2020 abstract ID: 913520 •Brunklaus et al. (2012): Prognostic, clinical and demographic features in SCN1A mutation positive Dravet syndrome. Brain 2012;135(8):2329-36 •Brunklaus et al. (2011): Comorbidities and predictors of health-related quality of life in Dravet syndrome. Epilepsia 2011;52(8):1476-82
Funding:
:None