Lymphocyte kinetics as a predictor of clinical outcomes following CAR-T therapy for non-Hodgkin lymphoma

Abstract

Abstract Introduction - Chimeric antigen receptor (CAR) T cell therapy has become a highly effective salvage option for patients with relapsed or refractory non-Hodgkin lymphoma (NHL); however, despite greater-than-80% response rates, most patients will ultimately relapse. Preliminary research has suggested that absolute lymphocyte count (ALC) or the ratio of ALC/absolute monocyte count (AMC), indicators of immune status, can provide prognostic value in CAR-T treatment. (Mejia, Blood Adv. 2024; Faude, Blood Adv. 2021; Bansal, Clin Lymphoma Myeloma Leuk. 2022). Methods - We performed a single-center, retrospective study evaluating ALC kinetics and associated clinical outcomes of patients with NHL treated with axicabtagene ciloleucel (axi-cel) from October 1, 2018, to November 20, 2023. Results - In total, 45 patients received axi-cel during the study period with a male/female ratio of 0.96 and median age of 63 years (42-80). The ratio of high-grade to low-grade lymphoma was 3:1. Fludarabine-cyclophosphamide lymphodepletion was used in 38% (n=17), while 62% (n=28) received bendamustine lymphodepletion. Dexamethasone for CRS/ICANS prophylaxis was given to 62% (n=28). Median progression-free survival (PFS) and overall survival (OS) for the entire cohort were 19.1 and 22.1 months, respectively. CRS occurred in 71% (n=32) and ICANS occurred in 31% of patients (n=14). Median ALC was 0.6 x 109 cells/L (0.1-2.5 x 109 cells/L) at leukapheresis and 0.1 x 109 cells/L (0-0.7 x 109 cells/L) at time of CAR-T cell infusion. The ALC/AMC ratio at leukapheresis had a median of 1.0 (0.2-9.0). Peak ALC occurred after a median of 10 days following infusion (4-30 days) with a median peak value of 0.7 x 109 cells/L (0.2-7 x 109 cells/L). Median rate of ALC expansion (day 1-10) was 0.04 x 109 cells*day/L (0-0.12 x 109 cells*day/L). When area under the curve (AUC) was calculated for ALC in the first 15 days following CAR-T infusion, the median was 3.2 x 109 cells*day/L (0.6-11.6 x 109 cells*day/L). In terms of clinical outcomes, patients with an ALC above the median at leukapheresis had a significantly prolonged PFS (not reached [NR] vs 10.1 mo, p=0.0179) and OS (NR vs 34.0 mo, p=0.0439) following CAR-T infusion. Patients with a 10-day rate of ALC expansion above the median had a significantly greater complete response (CR) rate (87% vs 59%) (p=0.0472) and a prolonged PFS (NR vs 34.5 mo, p=0.0242). Rate of ALC expansion to peak above the median was also associated with a prolonged PFS (NR vs 9.2 mo, p=0.0487). A time to peak ALC of <10 days was associated with a prolonged PFS (NR vs 8.4 mo) and OS (NR vs 20.75 mo), compared to those with a time to peak >10 days. Patients with a peak ALC above the median within 30 days of CAR-T infusion had a significantly prolonged PFS (NR vs 8.2 mo, p=0.0028) and OS (NR vs 20.8 mo, p=0.0031), compared to those with a peak below the median. A higher median area under the curve (AUC) for ALC in the first 15 days following CAR-T infusion was associated with a numerically prolonged PFS (NR vs 10.1 mo, p=0.0561) and a significantly prolonged OS (NR vs 34.0 mo, p=0.0387). ALC/AMC at the time of leukapheresis had no association with response rates or survival in this cohort. Conclusion - In this retrospective study of patients receiving axi-cel for NHL, multiple measures of lymphocyte kinetics, including ALC at the time of leukapheresis, rate of ALC expansion, peak ALC, and day 1-15 lymphocyte AUC, were consistently associated with improved PFS and/or OS following CAR-T infusion. The rate of ALC expansion within the first 10 days following CAR-T infusion was also associated with better CR rates. Although limited by small sample size, these data support further investigation of ALC as a dynamic and easily accessible biomarker for clinical outcomes following CAR-T cell therapy.