Machine Learning-Based Predictive Model for High-Grade Cytokine Release Syndrome in Chimeric Antigen Receptor T-Cell Therapy

Xiaofeng Yu¹Qingqing Wang¹Tangnuran Halimulati¹Jiling Lv¹Kai Zhou²Guilai Chen¹Li Yin¹Yulin Liu¹Jingwang Bi¹Zhuo Xiang^1*Qiang Wang^1*

• ¹Shandong Second Provincial General Hospital, Jinan, China
• ²Army Medical University, Chongqing, China

The development of robust predictive models for high-grade cytokine release syndrome (CRS) in CAR-T recipients remains limited by sparse clinical trial data. Analysis of 496 COVID-19 patients revealed that CRS plays a pivotal role in disease progression and serves as a valuable data source for understanding CRS progression. Among evaluated algorithms (XGBoost, Random Forest, Logistic Regression), XGBoost demonstrated superior performance in high-grade CRS prediction. Feature importance analysis identified SpO₂, D-dimer, diastolic blood pressure, and INR as key predictors, enabling development of a validated risk-assessment algorithm. In an independent CAR-T cohort (n=45), the algorithm achieved impressive predictive performance for high-grade CRS prediction. Using machine learning, we identified key clinical biomarkers strongly associated with high-grade CRS. This tool efficiently predicts progression to high-grade CRS post-onset and shows significant potential for clinical deployment in CAR-T therapy.