Association of CAR T cell kinetics with outcomes after standard of care brexucabtagene autoleucel for relapsed/refractory Mantle Cell Lymphoma

Abstract

Abstract Purpose/Objectives: Brexucabtagene autoleucel (brexu-cel) is an autologous anti-CD19 chimeric antigen receptor T-cell (CAR-T) therapy approved for the treatment of relapsed/refractory mantle cell lymphoma (MCL). Grade 3 or higher (G3+) cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) were seen in 15% and 30% of patients (pts) in Zuma-2. We aimed to correlate toxicities following standard of care (SOC) brexu-cel with CAR T-cell kinetics and inflammatory markers. Materials/Methods: We conducted a retrospective review of 26 pts who received standard of care brexu-cel between 11/2022 and 2/2024 and were assessed for CAR T cell expansion using a CLIA-certified flow cytometry-based assay (Lionel et al; ASH 2024). Bridging therapy (BT) was given at the discretion of the treating physician. Response was assessed using Lugano criteria. CRS and ICANS were assessed using the American Society for Transplantation and Cellular Therapy criteria. Laboratory values, including inflammatory markers (ferritin, C-reactive protein), were abstracted from the medical record. Results: Twenty-two of 26 pts were male (85%), 24 (92%) identified as white, and 4 (15%) as Hispanic or Latino. Eleven pts (42%) had blastoid or pleomorphic morphology at diagnosis. Bone marrow involvement was confirmed in 21 pts (81%) at diagnosis, with GI involvement confirmed in 10 pts (38%). Sixteen pts (62%) had progression within 24 months of frontline therapy, with a median of 4 prior lines of therapy prior to CAR-T (IQR 3-5). Seventeen pts (65%) had prior bendamustine at a median of 48 months prior to brexu-cel, and eleven pts (42%) had been on multiple Bruton tyrosine kinase inhibitors (BTKi). The median age at apheresis was 66 years (IQR 61-72). At the time of apheresis, 4 pts (15%) had documented transformation to blastoid/pleomorphic disease morphology, which was associated with mutations in TP53 (p = 0.02). Four pts (15%) had CNS involvement at apheresis, and 5 had bulky disease >10cm (19%). Twenty-three pts (88%) received BT between apheresis and infusion, which included systemic therapy (11 pts,42%), BTKi (10 pts, 38%), and radiotherapy (8 pts, 31% with a median dose of 4 Gy in 2 fractions). The median duration from apheresis to lymphodepleting chemotherapy (LDC) was 29 days (IQR 24-46). The median duration of hospitalization was 19 days (IQR 11-30). Twenty-three (88%) pts developed CRS, which was grade 3+ in 4 pts (15%). The median onset of CRS was 3 days (IQR 2-3) with a duration of 2 days (IQR 2-5). ICANS was seen in 17 pts (65%) and was grade 3+ in 8 pts (31%). The median onset of ICANS was 5 days (IQR 4-9), with a median duration of 5 days (IQR 2-19). (Peak ferritin)/(ferritin on day 0 (D0) was associated with grade 3+ CRS (p = 0.013) and grade 3+ ICANS (p = 0.026). However, ferritin on D0 was not associated with grade 3+ CRS or ICANS (p = 0.23, p = 0.24, respectively). LDH on D0 was not associated with G3+ CRS or G3+ ICANS (p = 0.78, p =0.74, respectively). A CAR-T-cell peak after D10 was associated with grade 2+ ICANS (7/20 pts with a peak after D10 developed G2+ ICANS, compared to 0/5 pts without, p = 0.009); however, this was not associated with G3+ ICANS (p = 0.53), G2+ CRS (p = 0.55) or G3+ CRS (p = 0.62). CNS involvement at apheresis was not associated with G2+ (p = 0.49) or G3+ ICANS (p = 0.37), or ICU admission (p = 0.14). On D30 PET/CT, 19 pts (73%) experienced a CR. There was a trend towards higher absolute CAR-T cell count in pts who experienced a CR at D30 (median 222 (IQR 79-845) vs. 50 in those who did not (IQR 25-292)), though this was not statistically significant (p = 0.17). At last follow-up (median 17 mo (IQR 5-25 months)), 9/26 pts were deceased (35%). The median progression-free survival (PFS) after brexu-cel was 13.5 months (IQR 2-22 months). Median overall survival (OS) after brexu-cel was 17 months (IQR 5-24 months). Conclusion: In a high-risk, heavily pre-treated cohort of pts with MCL, CAR-T therapy was associated with high response rates and toxicity consistent with previously published series. A relative increase in ferritin after D0 was associated with both high-grade CRS and ICANS, and late CAR-T cell expansion was associated with neurotoxicity. If confirmed in larger studies, early interventions to mitigate inflammatory response and dampen the CAR-T cell peak can be explored to improve the therapeutic index of brexu-cel.