EEG-based grading of immune effector cell-associated neurotoxicity syndrome

Abstract

CAR-T cell therapy is an effective cancer therapy for multiple refractory/relapsed hematologic malignancies but is associated with substantial toxicity, including Immune Effector Cell Associated Neurotoxicity Syndrome (ICANS). Improved detection and assessment of ICANS could improve management and allow greater utilization of CAR-T cell therapy, however, an objective, specific biomarker has not been identified. We hypothesized that the severity of ICANS can be quantified based on patterns of abnormal brain activity seen in electroencephalography (EEG) signals. We conducted a retrospective observational study of 120 CAR-T cell therapy patients who had received EEG monitoring. We determined a daily ICANS grade for each patient through chart review. We used visually assessed EEG features and machine learning techniques to develop the Visual EEG-Immune Effector Cell Associated Neurotoxicity Syndrome (VE-ICANS) score and assessed the association between VE-ICANS and ICANS. We also used it to determine the significance and relative importance of the EEG features. We developed the Visual EEG-ICANS (VE-ICANS) grading scale, a grading scale with a physiological basis that has a strong correlation to ICANS severity (R = 0.58 [0.47-0.66]) and excellent discrimination measured via area under the receiver operator curve (AUC = 0.91 for ICANS ≥ 2). This scale shows promise as a biomarker for ICANS which could help to improve clinical care through greater accuracy in assessing ICANS severity.

Background

Immune effector cell-associated neurotoxicity syndrome (ICANS) is a common, potentially severe complication of CAR-T cell therapy. EEG is frequently abnormal in ICANS but is graded subjectively. An interpretable, EEG-feature-based severity score could standardize assessment and complement the clinical ICANS grade.

Software Description

Code + de-identified data to reproduce the VE-ICANS ordinal grading model. ImagesDataNewFeatures_deI.xlsx (committed): expert-rated visual EEG features (X) + ICANS grade/ranks (y) for 316 CAR-T EEG images, surrogate subject IDs, no PHI. Raw name-containing files are excluded.

Technical Implementation

Expert-rated visual EEG features (frequency content, periodic/rhythmic patterns, voltage, continuity) were extracted from EEG images of CAR-T patients and paired with clinical ICANS grades. An interpretable ordinal / learning-to-rank model with coefficients constrained to a clinically usable grid (VE-ICANS) was fit and evaluated with grouped cross-validation (grouping by patient). See the code repository for the full pipeline.

Installation and Requirements

Python 3.10+: pip install -r requirements.txt (scikit-learn, mord, numpy, pandas, scipy, openpyxl).

Usage Notes

Code: https://github.com/bdsp-core/VE-ICANS. Verified 2026-07-07: python fit_model.py fits the model on the committed de-identified features. See REPRODUCE.md / DATA_SOURCE.md.

Release Notes

Version 1.0.0 - initial release accompanying Jones et al., Sci Rep 2022 (doi:10.1038/s41598-022-24010-1).

Ethics

Retrospective analysis of de-identified EEG data from CAR-T patients. [AUTHOR CHECK: IRB approval language.]

Acknowledgements

[AUTHOR CHECK: funding/acknowledgements from the manuscript.]

Conflicts of Interest

M. Brandon Westover is a co-founder of, advisor/consultant to, and has equity in Beacon Biosignals. [AUTHOR CHECK: co-author disclosures.]