EEG Findings in Chimeric Antigen Receptor T-Cell (CAR-T) Related Encephalopathy Syndrome

Objective: To describe the clinical and electrophysiologic characteristics of CRES (CAR-T related encephalopathy syndrome) Background: Chimeric Antigen Receptor T-Cell (CAR-T) treatment is an emerging cellular antineoplastic therapy. CAR-T-cell related Encephalopathy Syndrome (CRES) is a commonly observed side-effect, characterized by disorientation, altered consciousness, language dysfunction and seizure. The incidence of seizure in CRES is not fully known, and diagnostic role for electroencephalography (EEG) is not yet established. Design/Methods: An IRB-approved protocol identified patients at Stanford Medical Center who received CAR-T cell therapy for aggressive non-CNS lymphoma between 2/2016–9/2018. Neurotoxicity was identified by systematic chart review. EEGs were analyzed to determine associated electrographic features. Results: 28 of 53 patients (53%) treated with CAR-T developed CRES a median of 5 days after infusion. Of 28 patients with CRES, the median duration of symptoms was 7 days (range 1–14+). Eleven patients had prolonged (>7 days) encephalopathy, with 4 experiencing incomplete recovery at the time of hospital discharge. 17(60.7%) CRES patients presented with aphasia. Twenty-six patients with CRES had EEG, with average recording time of 33.8 hours (median 31). Seizure was diagnosed clinically in 3 patients prior to EEG and captured in 1 patient on EEG with clinical correlate. While 3 patients (10.7%) had evidence of non-convulsive seizure on EEG (of whom 2 were deemed non-convulsive status epilepticus), 16 patients (60%) with EEG had evidence of ictal-spectrum discharges. Eighteen (69.2%) patients had diffuse slowing on EEG, 5(19.2%) had both diffuse and focal slowing, and 3(11.5%) were normal. Conclusions: Although aphasia was a common presentation of CRES suggesting focal origin, EEG in these patients most commonly showed diffuse slowing. EEG was useful in detecting non-convulsive seizure in this population. While ictal-spectrum periodic discharges were found in 60%, further clarification is needed to isolate clinical significance of ictal-spectrum discharges, and to identify optimal treatment given confounding effects of polypharmacy and concurrent illness. Disclosure: Dr. Satyanarayan has nothing to disclose. Dr. Markert has nothing to disclose. Dr. Hovsepian has nothing to disclose. Dr. Post has nothing to disclose. Dr. Muffly has nothing to disclose. Dr. Miklos has nothing to disclose. Dr. Thomas has nothing to disclose. Dr. Scott has nothing to disclose.