Clinical Activity of Tisagenlecleucel in Large B Cell Lymphomas According to the 5th Edition of the Who Classification

Introduction: Anti-CD19 chimeric antigen receptor T cells (CART19) is a standard treatment for patients with relapsed or refractory (r/r) large B-cell lymphomas (LBCL). However, LBCL is a heterogeneous group of diseases, including diffuse large B-cell lymphomas (DLBCL) not otherwise specified (NOS), high-grade B-cell lymphomas (HGBCL), and LBCL arising from indolent lymphomas (tLBCL). These LBCLs share several characteristics, such as CD19 expression and primarily nodal disease, but differ in terms of overall prognosis, tumor microenvironment, and pathogenetic pathways. It has been shown that LBCL subtype affects the outcome of CART19 therapy (Abramson et al. Lancet 2020; Schuster et al. Lancet Oncology 2021). In 2022, the World Health Organization (WHO) released an updated (5^th edition) lymphoma classification which redefined HGBCL as DLBCL/HGBCL with MYC and BCL2 rearrangements regardless of BCL6 rearrangement (DLBCL/HGBL-MYC/BCL2) or HGBCL NOS. HGBCL NOS is a diagnosis of exclusion based on morphology and FISH testing. To our knowledge, no studies have evaluated the efficacy and safety of CART19 in HGBCL NOS according to the new WHO criteria.

Methods: We retrospectively evaluated the outcomes of 166 consecutive LBCL patients treated with tisagenlecleucel (tisa-cel) at the University of Pennsylvania between January 2018 and October 2023. The data cut-off was February 28^th, 2024. Diagnosis were retrospectively re-assessed according to WHO 5^th revision based on the pathology report and FISH test results. Specifically, a diagnosis of HGBCL NOS was based on biopsies with blastoid or Burkitt-like morphology (in the absence of MYC translocation), and without concomitant translocation of MYC and BCL2 by FISH. Response (Lugano 2014 criteria) was evaluated at 3 months and toxicities (ASTCT, CTCAE) within 30 days after tisa-cel infusion. Dichotomous variables were analyzed using the chi-square test. Survival analysis was performed using the Kaplan-Meier method. Multivariate analysis was conducted using Cox regression method.

Results: Of 166 patients included in the analysis, 99 (59.6%) patients had a diagnosis of DLBCL NOS, 6 patients (3.6%) had HGBCL NOS, 44 patients (26.5%) had tLBCL, and 17 patients (10.2%) had a diagnosis of DLBCL/HGBL-MYC/BCL2. Demographic characteristics, including age, sex, more than 2 prior lines of therapy before tisa-cel, and prior treatment with autologous stem cell transplantation were similar across LBCL
subtypes. However, LDH levels above normal limits at the time of tisa-cel infusion were more frequent in HGBCL NOS patients (83.3%) compared to the other patients (39.4%, p=0.032). The complete remission rate at 3 months was 46.5% for DLBCL NOS, 45.5% for tLBCL, 41.2% for DLBCL/HGBL-MYC/BCL2, and 16.7% for HGBCL NOS. No differences in cytokine release syndrome and immune effector-associated neurotoxicity syndrome were observed between to LBCL subtypes (p = 0.278 and 0.196, respectively). We then performed univariate analysis of progression-free survival (PFS) by specific LBCL subtype. At a median follow-up time of 15.2 months, the median PFS for DLBCL NOS 5.4 months, tLBCL 5.8 months, and DLBCL/HGBL-MYC/BCL2 2.8 months, while median PFS for HGBCL NOS was significantly reduced at 0.8 months (p<0.001). We also confirmed that LDH levels at infusion significantly predicted PFS (p<0.001) in the entire LBCL population. Multivariate analysis of PFS included sex, age, prior lines of therapies, prior autologous transplant, LDH at infusion, and diagnosis of HGBCL NOS was also performed. Elevated LDH at infusion and HGBCL NOS diagnosis were both independent factors associated with a higher risk of short PFS after tisa-cel. Lastly, the diagnosis of HGBCL NOS had the highest hazard ratio (HR) for short PFS (HR: 4.75, CI95%: 1.98-11.19; p<0.001), confirming the potential role of this histology as a marker of poor prognosis for treatment outcome with tisa-cel.

Conclusions: In this retrospective analysis, we report the outcome of CART19 according to the new 5^th edition of the WHO classification. We observed that, while most LBCL disease entities respond similarly to tisa-cel, patients with HGBCL NOS histology had a significantly shorter
PFS. This study suggests that patients with HGBCL NOS have poor outcomes with tisa-cel. A larger study with more patients with a diagnosis HGBCL NOS and with other CART19 products is necessary to confirm and extend our observations.