Absolute lymphocyte count (ALC) at apheresis and infusion as a predictor of safety and efficacy for multiple myeloma and B cell malignancies post CAR-T therapy

Abstract

Abstract Background: Host immunity at time of apheresis and infusion represent distinct periods which may influence response to CAR-T therapy and the incidence of side effects such as cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS). Therefore, absolute lymphocyte count (ALC) at these timepoints may have value as a predictor of response to therapy and adverse events. This study performed a retrospective analysis of patients receiving B-cell maturation antigen (BCMA) and cluster of differentiation (CD)19 directed CAR-T products to identify whether ALC at apheresis and infusion was associated with incidence of CRS/ICANS and efficacy in patients with non-Hodgkins lymphoma (NHL), multiple myeloma (MM), and acute lymphoblastic leukemia (ALL). Methods: 315 patients receiving CAR-T Therapy for NHL, MM, and ALL at the University of Kansas Medical Center from Dec 2017-Dec 2023 were included in the study. ALC was measured and recorded on day of apheresis and day 0 of CAR-T infusion (only when ≥ 0.1 cells/µL). ALC was analyzed as a continuous variable and Wilcoxon rank-sum tests were performed to evaluate whether ALC at apheresis and infusion differed significantly by CRS status, ICANS/CRES status, and combined status (where both CRS and ICANS/CRES = 0 vs. either > 0). Overall survival (OS) and progression-free survival (PFS) at 1-year was used to compare status between groups. All statistical tests were conducted at a 0.05 significance level and analyses were performed using R software. Results: Patients with NHL comprised 69.8% of the study population (n=220) and had a median age of 66 years (range: 29-92). 72% of the population had diffuse large B-cell lymphoma (DLBCL) and 65% of the cohort received axicabtagene ciloleucel. Patients with MM comprised 24.4% of the study (n=77) and had a median age of 69 years (range:45-83). 68% of this group had IgG MM and 52% of the cohort received idecabtagene vicleucel. Patients with ALL comprised 5% of the study (n=18) and had a median age of 39 years (range:21-70). 56% of the ALL cohort received brexucabtagene autoleucel. The most common causes of non-relapse mortality in the study were CRS/ICANS (n=16) and infection (n=6). For the whole study population including patients with NHL, MM, and ALL, ALC at infusion showed statistically significant differences across all three status variables: CRS (p= 0.029), ICANS (p=0.002), and combined CRS and ICANS (p=0.0004). ALC at apheresis did not show an association with CRS (p=.460), ICANS (p=.749), or combined CRS and ICANS (p=.449). In the NHL cohort, patients who had progression had a lower mean ALC at apheresis compared to patients who did not progress (Mean ALC= 0.85 ± 0.72 versus 0.97 ± 0.57 cells/µL). There was a statistically significant difference in ALC at apheresis (p=0.012) and ALC at infusion (p=0.038) in NHL patients with versus without progression. There was no statistically significant difference in survival status based on ALC at apheresis (p=0.213) or at infusion (p=0.093). In the MM cohort, patients who had progression had a lower mean ALC at apheresis compared to patients who did not progress (Mean ALC=1.08 ± 0.86 versus 1.16 ± 0.86 cells/µL). There was a significant difference between ALC at infusion and overall survival status (p=0.024). There were no significant associations between ALC at apheresis for progression (p=.772) or survival status (p=0.609), or for progression status and ALC at infusion (p=0.095) in the MM cohort. For the whole study population including NHL, MM, and ALL, there was a statistically significant difference between ALC at apheresis (p=0.021) and infusion (p=0.013) with progression status, and with ALC at infusion and overall survival (p=0.008). There was no significance between ALC at apheresis and survival (p=0.094). Conclusion: ALC at time of apheresis and infusion may influence progression/survival status and incidence of CRS/ICANS. ALC at apheresis may be predictive of the quantity of lymphocytes being collected and ALC at infusion may effect CAR-T expansion by impacting the cellular immune environment. These findings were more pronounced in the NHL cohort than in the MM or ALL cohorts, which may reflect the effect of disease specific characteristics as well as the confounder of circulating leukemic cells registering as lymphocytes. Further analysis should investigate the relationship between low versus high ALC and its impact on CAR-T therapy outcomes.