Predicting Severe CRS During CAR T Cell Therapy for Leukemia Is Critical

C himeric antigen receptor T-cell (CAR T) therapy showed tremendous potential in treating hematological malignancies such as leukemia in recent years; however, it is linked with serious adverse effects that can be lifethreatening, such as severe cytokine release syndrome (CRS), explains Zhenyu Wei, PhD candidate. Past research indicated between 54% and 91 % of patients may manifest CRS during treatment, and if CRS is grade 3 or greater, it is considered serious.
To help predict the occurrence and development of severe CRS, Wei and colleagues analyzed clinical data with artificial intelligence in a study published in Frontiers in Immunology. Although CAR T cell therapy has been used in about 4% of people living with leukemia, that percentage is expected to rise with the FDA approvals of six CAR T therapies since 2017. Also, studies have shown that CAR T therapy leads to significant clinical benefits for patients with leukemia with poor prognosis.

Decision Tree Predicts Timing of Severe CRS

To better understand the development of sCRS and how to predict its occurrence, Wei and colleagues analyzed clinical data of 140 adult patients with B-cell acute lymphoblastic leukemia (ALL) targeting CD19, a cell membrane marker for diagnosing B-lineage ALL that has proven reliable for antibody-targeting leukemia treatment. They assessed blood routine, coagulation and biochemical indexes, cytokines, and a series of other biomarkers in patients during treatment. This enabled the study team to develop novel studies and broaden its understanding of CRS, with the goal of enhancing clinical practice. “By analyzing the heat map of clinical factors and comparing them between severe and non-severe CRS, we can identify significant differences among these factors and understand their interrelationships,” the study authors wrote. They then used a decision tree to predict the timing of severe CRS in patients, taking into consideration variables that included initial, day before, and same day (Figure).

25 Clinical Factors Linked With Severe CRS

All patients in the study received CAR T therapy followed by the measurement of cytokines and clinical biomarkers. Researchers observed that following CAR T cell infusion, peak levels of 25 clinical factors, including IFN-γ, IL6, IL10, ferritin, and D-dimer, were significantly correlated with severe CRS. “Using the decision tree model, we were able to accurately predict which patients would develop severe CRS consisting of three clinical factors, classified as same-day, day-ahead, and initial value prediction,” the study authors noted. “Changes in serum biomarkers, including C-reactive protein and ferritin, were associated with CRS, but did not alone predict the development of severe CRS.”

When comparing their CRS rating scale with other rating scales, the researchers noted that their grading system showed satisfactory similarities and enabled them to identify patients who had the potential to experience life-threatening complications of CRS.

Wei and colleagues noted that their study represents a complete analysis of the biological and clinical manifestations of CRS for patients with B-cell ALL following CAR T cell therapy. They successfully analyzed and identified clinical factors related to severe CRS and biomarkers that predict the development of severe CRS in a timely manner. “The emergence of these models will enable patients with severe CRS to be closely monitored and have the opportunity to start an active support treatment,” the study authors concluded. “At the same time, predicting the potential development of severe CRS will prevent unnecessary cytokine intervention. Therefore, the model we produced, which was based on the use of biomarkers to predict severe CRS, has direct clinical and therapeutic significance.”