Retrospective discrimination of PNES and epileptic seizure types using blood RNA signatures.

Objectives: The ability to differentiate epileptic- and non-epileptic events is challenging due to a lack of reliable molecular seizure biomarker that provide a retrospective diagnosis. Here, we use next generation sequencing methods on whole blood samples to identify changes in RNA expression following seizures.
Methods: Blood samples were obtained from 32 patients undergoing video electroencephalogram (vEEG) monitoring. Blood samples were collected in PaxGene tubes at baseline (admission) and following a seizure event (4-6 h and 24 h later or discharge). EEG and video of clinical events were reviewed by the clinical team and study epileptologist and were classified as epileptic seizure, psychogenic nonepileptic spell (PNES), or other. RNA was extracted from blood and RNA expression was determined using RNA-sequencing.
Results: We show significant differences in RNA profiles between patients that did or did not experience an epileptic seizure. Compared to baseline patients with PNES show large increases in RNA expression 4-6 h and 24 h post seizure. Conversely, genes that changed following epileptic seizure showed more modest changes associated with a decrease in immune system function. Transcript usage was changed between patients with PNES and epileptic seizure at all three time points examined. Lists of genes differentially expressed following PNES or epileptic seizure vs. all baseline samples were used as classifiers for prediction. Models generated using random forest and radial support vector machine algorithms were 100% accurate at predicting both PNES and epileptic seizures.
Significance: These data suggest that blood gene expression changes may have utility to retrospectively identify patients who have suffered a seizure or seizure-like event as a cause of transient loss of consciousness.
Competing Interests: Declarations. Conflicts of interest: None to report. We confirm that we have read the Journal’s position on issues involved in ethical publication and affirm that this report is consistent with those guidelines. Monica Dhakar is an employee at Xenon Pharmaceuticals Inc. and has consulted for Ceribell Inc, none of which are relevant to this work. None of the other authors has any conflict of interest to disclose. Ethical approval: Included- see Methods. Patient consent: Included-see Methods.
(© 2024. The Author(s).)