Evaluating peripheral eosinophilia as a biomarker post CAR-T cell therapy.

Abstract

e19085 Background: Cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) are early complications, while late immune effector cell-associated hematotoxicity (ICAHT) is a delayed side effect of chimeric antigen receptor T-cell (CAR-T) therapy. Currently, no reliable predictive models exist to identify high-risk patients. This study investigates whether peripheral eosinophilia after CAR T-cell has any effect on risk of CRS, ICANS, ICAHT and infection. Methods: This multicenter, retrospective study used TriNetX database to analyze patients with B-cell lymphoma who had received CD19 CAR-T cell therapies—Liso-Cel, Axi-Cel, Brexu-Cel, or Tisa-Cel—and developed eosinophilia (>500/µL) within first month after CAR T-cell administration (Cohort 1) vs those who didn’t develop eosinophilia (Cohort 2).The outcomes of interest were the risk of CRS, ICANS, late ICAHT (after day 30), and risk of infection (after day 30) after CAR-T therapy administration. Cytopenia was defined by anemia (HgB < 8 g/dL), thrombocytopenia (PLT < 50 x 10 3 ) and neutropenia (ANC < 500/µL). Kaplan-Meier analysis, hazard ratio (HR), risk ratio (RR), and 95% CI were used to assess the outcomes. Results: A total of 3,304 patients were identified, with 524 in the eosinophilia cohort and 2,780 in the non-eosinophilia cohort. Propensity score matching (PSM) was performed using forty unique variables, including demographics (age, sex, ethnicity), baseline hemoglobin, WBC, and platelet count. Following PSM, there were 521 patients in each cohort. The eosinophilia group had a significantly lower risk of CRS (42% vs 62%, p < 0.001), ICANS (17% vs 24%, p = 0.002), and late ICAHT (49.9% vs 65.8%, p < 0.001). The risk of anemia (29.9% vs 38.8%, p = 0.003), neutropenia (37.4% vs 55.7%, p < 0.001), and thrombocytopenia (32.6% vs 44.9%, p < 0.001) were all significantly lower in the eosinophilia cohort. Infection-free survival probability was higher in the eosinophilia group (49% vs 44%, p = 0.03), with a numerically lower risk of bacterial, viral, and fungal infections. Overall survival was similar in both groups and was not significant (72% vs 71%, p = 0.59). Conclusions: This study suggests that absence of peripheral eosinophilia may identify patients at higher risk for CRS, ICANS and late ICAHT. The potential explanation for this could be that eosinophils act as potent immune effectors and immune modulators in the tumor microenvironment post-CAR-T cell therapy. These findings could guide decisions on monitoring and administration of alternative therapies. Further prospective clinical research is needed to assess eosinophilia as a prognostic marker in CAR-T therapy recipients. Eosinophilia & CAR T-cell outcomes. Cohort 1 Cohort 2 P value # of patient 521 521 - Survival Probability* 72% 71% 0.59 CRS 42% 62% <0.001 ICANS 17% 25% 0.002 ICAHT 50% 66% <0.001 Infection-free survival* 49% 44% 0.03 *At 1-year post CAR T-cell.