Low reticulocyte count at infusion is a risk factor for ICANS in CAR-T cell therapy.

Abstract

Chimeric antigen receptor (CAR)-T cell therapies targeting CD19 have shown efficacy against B-cell malignancies. However, they frequently cause immune effector cell-associated neurotoxicity syndrome (ICANS), which can be life-threatening and require intensive care. Even when it is not severe, ICANS can lead to prolonged hospitalization and limit treatment options, especially for elderly patients. Despite its clinical significance, a reliable, early predictive marker for ICANS has not been identified. To identify risk factors for ICANS, we retrospectively analyzed B-cell lymphoma patients who received tisagenlecleucel (tisa-cel), lisocabtagene maraleucel (liso-cel), or axicabtagene ciloleucel (axi-cel) at Kyoto University Hospital from 2019 to 2024. Among 106 patients, 76 received tisa-cel, 22 liso-cel, and eight axi-cel. Median age at infusion was 63.5 years (interquartile range, 57-69). Eight patients (8%) had a history of central nervous system (CNS) involvement. ICANS occurred in 17 patients (16%), all with prior cytokine-release syndrome (CRS). Reticulocyte counts at infusion were significantly lower in patients who subsequently developed ICANS (1.57 versus 2.80 × 104/μL, P < 0.01). Multivariate analysis identified a low reticulocyte count at infusion (HR, 3.67; 95% CI, 1.23-11.02; P = 0.02), history of CNS involvement (HR 8.37; 95% CI, 3.04-23.04; P < 0.01), and axi-cel (HR, 4.56; 95% CI, 1.88-11.09; P < 0.01) as independent risk factors. Patients divided by the median reticulocyte count at infusion (2.57 × 104/μL) demonstrated significantly higher 30-day cumulative incidence of ICANS in those with lower counts (25.5% versus 7.8% at 30 days; P = 0.018). Reticulocyte counts at infusion, a simple hematological parameter, may help predict ICANS development and guide optimal risk-based management of chimeric antigen receptor (CAR)-T cell therapy.