A phase 1b trial of the EZH2 inhibitor tazemetostat combined with CAR T cell therapy in B cell lymphomas

Abstract

Abstract Background: Chimeric antigen receptor (CAR) T cells have remarkable efficacy in B cell lymphomas, but most patients eventually relapse. Pre-clinical studies have shown that inhibition of the histone methyltransferase EZH2 prevents T cell exhaustion and modulates T cell activity by promoting a memory phenotype and sensitizes lymphoma to immunotherapy by inducing lymphoma immunogenicity. These effects are independent of the lymphoma EZH2 mutation status, thereby supporting the potential of EZH2 inhibition to improve CAR T cell therapy across B cell lymphoma subtypes. Methods: In this single-center phase Ib clinical trial, we administered the EZH2 inhibitor tazemetostat in patients receiving standard of care CAR T cells for diffuse large B cell, follicular, or mantle cell lymphoma (DLBCL, FL, MCL). CAR T product was collected and manufactured as standard of care; selection of CAR T product was investigator's choice. Patients received tazemetostat for a minimum of 7 days prior to CAR T collection and continued during CAR T cell manufacturing until the start of lymphodepletion. Following lymphodepletion therapy and receipt of standard of care CAR T cells, patients resumed tazemetostat at count recovery and continued for up to 6 months in patients with complete remission and up to 12 months for patients with partial remission. The primary endpoint was safety and feasibility of manufacturing CAR T cells; secondary endpoints included overall and complete response rate, progression-free and overall survival (ORR, CR, PFS, OS). Exploratory studies included immune profiling of the peripheral blood and CAR T cell product, and CAR T expansion. Results: 13 patients were enrolled, including 7 with DLBCL, 5 with FL, and 1 with MCL. Median age was 68 (range 41-84), median prior lines of therapy were 2 (range 1-4), and 85% were refractory to the last line of therapy. EZH2 mutations were present in only 2 patients (15%). 4 patients had TP53 mutation or deletion (31%), and of the 7 DLBCL patients, 4 had high-grade myc translocation (57%) and 3 were transformed from indolent lymphomas (43%). Following tazemetostat pre-treatment and bridging, 5 patients received axi-cel (38%), 5 liso-cel (38%), 2 tisa-cel (15%), and 1 brexu-cel (8%). All patients (13/13) were able to successfully complete CAR T manufacturing and receive cells. ORR was 100%, including 77% with CR for the entire population. For patients with DLBCL, the CR rate was 71% (5/7). 100% of patients with FL achieved a CR (5/5), and the patient with MCL had a PR. At a median follow up of 11 months, 54% remain progression-free and 77% remain alive. There was no signal for excess toxicity with the addition of tazemetostat. 10 patients experienced CRS (77%), with only 1 grade 3 CRS in a patient receiving brexu-cel (8%). 6 patients had immune effector cell-associated neurotoxicity syndrome (ICANS) (46%), all grades 1-2. No patients developed immune effector cell hemophagocytic syndrome (IEC-HS). 10 patients (77%) had grade 3+ neutropenia after day 30; 2 had grade 3+ anemia or thrombocytopenia (15%). 9 patients had infections (69%), including 3 grade 3 infections requiring IV antibiotics (23%). Only one patient permanently discontinued tazemetostat due to AEs. The most common AEs associated with tazemetostat were related to gastrointestinal symptoms and were almost exclusively grade 1-2. Tazemetostat had no negative impact on CAR T transduction efficiency, activation, or expansion in correlative studies. Flow cytometry of T cells from patient samples before and after the initiation of tazemetostat treatment showed significant increases in naïve CD8 T cells and memory:effector ratios, as well as a significant decrease in terminally differentiated CD4 and CD8 T cells, which are factors known to correlate with clinical efficacy and durability of CAR T remissions. Conclusion: Addition of the EZH2 inhibitor tazemetostat to the CAR T pre-collection, bridging, and post-infusion periods is feasible, and without early evidence of adverse impact on safety. Initial efficacy data is encouraging but requires additional patients and follow up. Preliminary correlative findings suggest a positive impact of tazemetostat oral administration on the endogenous immune system, tumor microenvironment, and T cell product in patients. Based on these results, additional studies of tazemetostat in combination with T cell immunotherapies are ongoing and in development.