Abstracts

Addressing Epilepsy Variability Through Seizure Frequency Forecasting and Virtual Clinical Trials in STXBP1-Related Disorders

Abstract number : 2.168
Submission category : 4. Clinical Epilepsy / 4D. Prognosis
Year : 2023
Submission ID : 540
Source : www.aesnet.org
Presentation date : 12/3/2023 12:00:00 AM
Published date :

Authors :
Presenting Author: Julie Xian, . – Children's Hospital of Philadelphia

Kim Marie Thalwitzer, MD – University Hospital Heidelberg; Jillian McKee, MD, PhD – Children's Hospital of Philadelphia; Katie Rose Sullivan, MS, LCGC – Children's Hospital of Philadelphia; Elise Brimble, LCGC – Ciitizen, Invitae; Eryn Fitch, MD – Children's Hospital of Philadelphia; Jonathan Toib, MD – Children's Hospital of Philadelphia; Michael Kaufman, MS – Children's Hospital of Philadelphia; Danielle DeCampo, MD, PhD – Children's Hospital of Philadelphia; Kristin Cunningham, MS, OTR/L – Children's Hospital of Philadelphia; Samuel Pierce, PT, PhD, NCS – Children's Hospital of Philadelphia; James Goss, PhD – STXBP1 Foundation; Charlene Son Rigby, MBA – STXBP1 Foundation; Steffen Syrbe, MD – University Hospital Heidelberg; Michael Boland, PhD – University of Pennsylvania; Ben Prosser, PhD – University of Pennsylvania; Nasha Fitter, MBA – Ciitizen, Invitae; Sarah Ruggiero, MS, LCGC – Children's Hospital of Philadelphia; Ingo Helbig, MD – Children's Hospital of Philadelphia

Rationale: Seizure variability limits precise prognostication and clinical care for individuals with epilepsy, even in individuals with shared genetic etiologies. Furthermore, the heterogeneity in the natural history of epilepsy decreases statistical power when evaluating treatment effect in clinical trials. Accordingly, addressing seizure variability is necessary to overcome the bottleneck for precision medicine approaches and will accelerate clinical trial readiness in the epilepsies.

Methods: First, we developed a longitudinal seizure frequency forecasting model to characterize the predictability and range of epilepsy trajectories in STXBP1-related disorders, among the most common genetic epilepsies. We assessed phenotypic signatures in the first year of life that were indicative of high seizure variability later in life. Then, we modeled virtual clinical trials using a synthetic control approach, leveraging real world data across 765 patient-years of epilepsy histories to analyze the probability of clinical trial success across the life span. We identified windows during which a treatment effect that decreases seizure burden would have the highest probability of being detected.

Results: We demonstrate that seizure trajectories can be forecasted better than chance in up to 87% of individuals with STXBP1-related disorders. Characterization of seizure variability revealed stratification by early-life epilepsy markers – refractory seizures throughout the first year of life were indicative of highly unpredictable trajectories later in life, while early infantile seizures that resolved within the first year were indicative of predictable trajectories and seizure freedom. Virtual clinical trials using seizure frequency as the primary outcome resulted in wide range of trial success probabilities in STXBP1, with the most optimal age range for detecting a treatment effect in early childhood between one and three and a half years. The probability of trial success varied when assessing different seizure types, with an optimal window starting at three months for infantile spasms and between two to three and a half years for focal-onset seizures.

Conclusions: Addressing seizure variability is a prerequisite for clinical trial readiness. Understanding epilepsy trajectories will enable personalized risk stratification and provide a prognostic baseline score for inclusion and exclusion of robust subgroups for future trials. Here, we describe the range of seizure trajectories and identified optimal age windows for clinical trials in STXBP1-related disorders, providing a foundation to interpret future therapeutic strategies and inform rational trial design across genetic epilepsies.

Funding: Children's Hospital of Philadelphia; National Institute of Neurological Disorders and Stroke

Clinical Epilepsy