Impact of Fludarabine Dosage on Outcomes in Large B-Cell Lymphoma Patients Treated with CAR T-Cell Therapy: A Retrospective Study on Behalf of the Ctiwp of the EBMT

Background

CAR T-cell therapy has significantly advanced treatments for acute lymphoblastic leukemias, B cell lymphomas, and multiple myeloma. However, durable remissions are seen in less than half of the patients, prompting extensive research to understand response and resistance mechanisms and improve treatment efficacy. One strategy is optimizing conditioning regimens, with clear benefits of adding fludarabine. Despite these benefits, pharmacokinetics studies showed high variability in fludarabine exposure among patients, impacting clinical outcomes. Studies on CAR T-cell therapy for acute leukemias also imply that high fludarabine exposure might be linked to an increased CAR T-cell expansion which translates into better progression-free survival (PFS). Moreover, translational studies suggest that optimal fludarabine exposure may enhance the cytokine profile in patients, leading to sustained remissions. Nevertheless, some recently reported conflicting data render the results in this field currently controversial, making it challenging to provide recommendations on the ideal drug dose in particular also within the context that the first two commercial CART tisagenlecleucel (tisacel) and axicabtagene ciloleucel (axicel) used different fludarabine dosing during registration studies. We therefore aimed to study the impact of fludarabine dosage on outcomes in a homogeneous cohort of large B-cell lymphoma patients after CAR T-cell therapy for both tisacel and axicel.

Methods

The Cellular Therapy & Immunobiology Working Party performed a retrospective study based on the European Society of Blood and Marrow Transplantation (EBMT) registry, including all patients receiving tisacel or axicel between 2019 and 2023 for large B-cell lymphomas and for whom outcomes, including complication data, were recorded. We excluded patients with renal insufficiency. The standard fludarabine dose in the conditioning regimen, based on registration studies, was defined as 75 ± 3.75 mg/m² for tisacel patients and 90 ± 4.5 mg/m² for axicel patients. Patients receiving doses outside of ± 5% of the standard were classified into lower or higher dose groups, resulting in three distinct categories (i.e. lower, standard, and higher). The endpoints were overall survival (OS), PFS, and the cumulative incidence of cytokine-release syndrome (CRS) (grade 3-5) and neurotoxicity (grade 3-5) (both considering death as a competing event). Patients were censored at last follow-up. Univariate (Kaplan-Meier and Cumulative incidence) and multivariate analyses using a Cox proportional hazards model were performed.

Results

In total, 1951 patients met the inclusion criteria, with 734 patients receiving tisacel and 1217 patients receiving axicel. The median age was 61.6 years (range: 18.4-89.2), 62.9% of the patients were male, and 90.1% had an ECOG performance status of 0-1. Most patients received CAR T-cell therapy after at least three lines of treatment. Univariate analyses for the axicel cohort showed no significant differences based on log-rank and Gray’s tests in OS (p=1), PFS (p=0.5), CRS (p=0.36), and neurotoxicity (p=0.62) cumulative incidences among the three fludarabine dose categories. Univariate analyses for the tisacel cohort showed an increased incidence of neurotoxicity for higher doses of fludarabine (p=0.025), but no differences in OS (p=0.6), PFS (p=0.6), and CRS (p=0.38) cumulative incidences. The Cox proportional hazards model using a continuous fludarabine dose variable (per steps of 10 mg/m²) as well as sex, age at treatment, year of therapy, time to apheresis to infusion (in days), disease stage at CAR T-cell infusion, previous transplant (yes/no), and ECOG performance score confirmed the negative impact of higher doses of fludarabine on neurotoxicity (HR=1.28; 95% CI: 1.07, 1.53; p=0.007) for patients receiving tisacel.

Conclusion

This extensive retrospective study suggests that administering higher doses of fludarabine does not enhance overall survival or progression-free survival in patients with B-cell lymphomas receiving CAR T-cell therapy. On the contrary, it appears to elevate the risk of neurotoxicity in those treated with tisacel. These results underscore the importance of carefully considering fludarabine dosing to balance efficacy and safety in patients with large B-cell lymphoma receiving CAR T-cell therapy.