Preliminary Analysis of Seizure Cluster Periodicity from a Long-term, Open-label Safety Study of Diazepam Nasal Spray
Abstract number :
3.288
Submission category :
7. Anti-seizure Medications / 7B. Clinical Trials
Year :
2022
Submission ID :
2204001
Source :
www.aesnet.org
Presentation date :
12/5/2022 12:00:00 PM
Published date :
Nov 22, 2022, 05:22 AM
Authors :
Nathan Fountain, MD – University of Virginia School of Medicine; Mark Quigg, MD – University of Virginia School of Medicine; Chad Murchison, PhD – Department of Biostatistics, University of Alabama; Adrian Rabinowicz, MD – Neurelis, Inc.; Enrique Carrazana, MD – Neurelis, Inc.; University of Hawaii John A. Burns School of Medicine
Rationale: Seizure clusters are emergencies associated with risk of status epilepticus, emergency room visits, and disruption of daily life. Identifying cluster patterns may help improve patient care. Diazepam nasal spray (Valtoco®) is FDA approved for acute treatment of seizure clusters in patients with epilepsy ≥6 years old. We evaluated the periodicity of seizure clusters treated with diazepam nasal spray.
Methods: Data were assessed from a long-term, open-label, repeat-dose safety study of diazepam nasal spray. Patients with epilepsy aged 6 to 65 years old with frequent seizure clusters were enrolled. Care partners or patients administered age- and weight-based doses of diazepam nasal spray. Start and end time of each treated seizure cluster was recorded. Periodicity of cluster times was investigated using cosinor statistical modeling. The cosinor model assigns a harmonic curve to time series with a consistent repeating pattern. Periodicity and peak phase of clusters were evaluated within fixed time periods of 24 and 12 hours.
Results: In the safety population (n=163), mean trial participation was 528 days. 81.6% of patients participated ≥12 months, and mean number of doses/month was 2.3 (range, 1-10). Clusters that required second doses within 24 hours of the first (12.6% of clusters) were represented in analysis by the time of first cluster. For the cosinor analysis, 139 patients had ≥2 seizure clusters and provided 4217 seizure observations. Plotting seizure events for the entire cohort across a 24-hour day shows a bimodal peak (Figure) corresponding to visual peaks at about 7:00 am and 7:00 pm. For the 24-hour cosinor analysis, 108 patients’ data was modeled and 28 patients (136 observations) had insufficient data. In the 12-hour cosinor analysis, 116 patients were included in the model and 23 (85 observations) had insufficient data. A large majority of subjects had relatively few seizures at any particular timepoint, either because events were dispersed across time or they had few total events. The 12-hour period was a better fit for the cohort-wide analysis than 24 hours. Cosinor fits for individuals across 12 hours observed low amplitudes for most patients, and the acrophase (peak of curve) varied.
Conclusions: Population seizure clusters identified by rescue medication administration have a bimodal distribution of occurrence through the day with peaks every 12 hours, roughly corresponding to sleep-wake transitions. Whether this observation reflects availability of caregivers to witness seizures before/after work or reflects biologic or observational/behavioral factors is not known. No consistent cosinor patterns were observed by individual cases. Ongoing analyses will investigate how patient characteristics affect cluster periodicity. Seizure cluster patterns, and potentially those for specific patient subgroups, may provide important information for improving aspects of patient care ranging from seizure prediction and prophylaxis to quality of life.
Funding: Neurelis, Inc.
Anti-seizure Medications