Chimeric Antigen Receptor Therapy in Lymphoma

Introduction

Clinical trials of CD19 chimeric antigen receptor (CAR) T-cells have demonstrated promising survival outcomes and response rates in relapsed and/or refractory (r/r) non-Hodgkin lymphoma (NHL), providing a potentially curative treatment option for these heterogeneous, aggressive malignancies. There are currently four U.S. Food and Drug Administration (FDA)–approved commercially available CAR T-cell therapies for highly aggressive B-cell NHL, all directed toward CD19. Axicabtagene ciloleucil (axi-cel), tisagenlecleucel (tisa-cel), and lisocabtagene maraleucel (liso-cel) were approved by the FDA in 2017, 2018, and 2021, respectively, for r/r large B-cell lymphoma (LBCL) after two lines of systemic therapy. Brexucabtagene autoleucel (brexu-cel) gained approval for r/r mantle cell lymphoma (MCL) in 2020. In addition, axi-cel received approval for r/r follicular lymphoma (FL) after two lines of systemic therapy in March 2021. This section will summarize the clinical outcomes of the pivotal clinical trials leading to FDA approval of these therapies, as well as real-world data to explore clinical uptake and outcomes for these cellular therapies. This will be followed by a discussion of patient selection and recommendations for CAR T-cell therapy in B-cell lymphoma.

ZUMA-1: Axi-Cel for Relapsed/Refractory Aggressive Large B-Cell Lymphoma

Axi-cel was the first CAR T-cell product to gain FDA approval for r/r LBCL after at least two systemic therapies have failed. The ZUMA-1 clinical trial was a single-arm, multicenter, open-label phase I/II trial demonstrating remarkable efficacy of a single infusion of axi-cel (2 × 10 ⁶ CAR T-cells/kg on day 0) following 3 consecutive days of cyclophosphamide (500 mg/m ² ) and fludarabine (30 mg/m ² ) on days -5, -4, and -3. Systemic BT was not allowed with the exception of corticosteroids, as long as they were not administered within a week of leukapheresis and/or administration of axi-cel.

A total of 111 patients were enrolled. Axi-cel was manufactured for 110 and administered to 101 patients (91%). The baseline characteristics were as follows: median age was 58 years (range: 23–76 years), 85% had stage III–IV disease, 48% had a high risk IPI score of ≥3, 69% had received three or more prior therapies, 26% had primary refractory disease, and 53% were refractory to two consecutive lines. This was a poor-risk population expected to have a median overall survival (OS) of 6 months or less with standard-of-care options at the time of enrollment. The histology of participants in the study included diffuse large B-cell lymphoma (DLBCL, 76%), transformed FL (tFL, 16%), or primary mediastinal B-cell lymphoma (PMBCL, 8%) (see Table 26.1 ).

The objective response rate was 82% (95% confidence interval [CI], 73%–89%). Over half (54%) of the patients that received axi-cel achieved a complete response. The median time to response was rapid, 1 month, and the median duration of response was 8.1 months. The median progression-free survival (PFS) was 5.9 months. These early reports were following 15 months of median follow-up. At that time, the median OS had not been reached. With more mature follow-up, the median OS is 25.8 months, far exceeding the expectations for this cohort of patients at the time.

In terms of acute toxicity, cytokine release syndrome (CRS) of any grade occurred in 93% of patients, with 13% experiencing grade ≥3 CRS (per Lee criteria ). All CRS events resolved except for one patient who developed grade 5 hemophagocytic lymphohistiocytosis (HLH). Neurologic events occurred in 64% of patients, with 28% experiencing grade ≥3 neurologic toxicities according to the Common Terminology Criteria for Adverse Events (CTCAE) grading system. All neurologic events resolved except in four patients in which neurologic toxicity was ongoing at the time of death (none of these deaths were related to neurologic toxicity). Eighty percent of patients had grade ≥3 neutropenia. Further, more than half (55%) of patients continued to have cytopenia of any grade past day 30 of CAR T-cell infusion.

Overall, the robust response in ZUMA-1 set the precedent for the FDA approval of subsequent CAR T-cell therapies for various NHLs. The major questions remaining following this landmark trial were the durability of efficacy, application of additional treatment during the waiting period (BT) to avoid death or progression precluding cell infusion, and whether the eligibility criteria of this trial accurately captured all patients that could benefit from cellular therapy. Specifically, the outcomes of older patients, high-risk patient populations including those with secondary central nervous system (CNS) involvement, and more diverse B-cell histologic subgroups. The logistics surrounding cellular therapy and the requirement for referrals to highly specialized tertiary centers left skeptics questioning whether this therapy would be available to the 40% of patients with r/r LBCL in desperate need for novel therapy. Can you reproduce the promising results of this phase II study outside the tightly controlled environment of a small prospective study? Real-world evidence was desperately needed to address these valid concerns. In a later section, will explore clinical outcomes with commercial CAR T-cell therapy including description of patient demographics, safety, and efficacy outcomes.

JULIET: Tisa-Cel for Relapsed/Refractory Large B-Cell Lymphoma

Tisa-cel has a 4-1BB intracellular costimulatory domain, which has shown to result in longer persistence in mouse models. Tisa-cel was the first CAR T-cell product to demonstrate efficacy in malignancy, achieving high and durable responses in children and young adults with r/r acute lymphoblastic leukemia (ALL), with a manageable safety profile. This set a milestone for CAR T-cell therapy overall, and tisa-cel was investigated for r/r LBCL in the subsequent single-arm phase II, open-label, international JULIET trial. The JULIET trial established the feasibility of a global cellular therapy trial.

A total of 238 patients were screened for this trial, 165 were enrolled and among these 111 (67%) received tisa-cel infusion (see Table 26.1 ). This highlights the challenges of conducting international prospective studies with cellular therapy in aggressive lymphoma subtypes. The median time from enrollment to infusion was 54 days, and 92% received BT ranging from chemoimmunotherapy to targeted therapy. The baseline characteristics of the 111 patients receiving tisa-cel infusion (median dose 3 × 10 ⁸ CAR T-cells) included a median age of 56 years (range 22–76 years, 23% over age 65 years), advanced stage (III–IV 76%), median of three prior lines of therapy (52% had ≥3), DLBCL (79%), and tFL (19%). Lymphodepletion conditioning was administered using either fludarabine (25 mg/m ² ) and cyclophosphamide (250 mg/m ² ) daily for 3 days (73%) or bendamustine (90 mg/m ² ) daily for 2 days (20%). Among patients with a white blood cell count ≤1000 cells/mm ³ within 1 week before tisa-cel infusion lymphocyte depletion was not required, marking another notable difference between JULIET and ZUMA-1, which used higher-doses fludarabine and cyclophosphamide and no option to omit lymphocyte depletion or use an alternative chemotherapy.

The most common grade 3 or higher adverse events were CRS (22%), neurologic events (12%), cytopenia persisting past 28 days (32%), infections (20%), and febrile neutropenia (14%). CRS and adverse neurologic events of any grade were experienced by 58% and 21% of patients, respectively.

The best overall response rate was 52% (95% CI, 41–62), with 40% achieving a complete response. The median OS was 12 months among patients who received an infusion of tisa-cel, whereas the OS for the entire cohort of enrolled patients was 8.3 months highlighting the poor prognosis of patients that failed while awaiting a product. Among the 165 enrolled, 50 (30%) discontinued study before cell infusion, 12 were caused by failure to manufacture CAR T-cells, 38 caused by other reasons, most likely disease progression, and four were awaiting infusion at the time of data cutoff. This highlights the need to identify barriers that may be minimized with policies and better practices as well as the need to educate primary oncologists to refer patients as early as possible to improve the chance of success.

TRANSCEND: Liso-Cel Relapsed/Refractory Large B-Cell Lymphoma

Liso-cel was the third FDA approved CAR T-cell therapy for r/r LBCL adults who failed at least two lines of therapy. Liso-cel has a 4-1BB costimulatory domain, similar to tisa-cel.

The TRANSCEND trial was a single-arm, open label, U.S. multicenter trial that used a dose escalation with three dose levels: 50 × 10 ⁶ CAR T-cells (one or two doses), 100 × 10 ⁶ CAR T-cells, and 150 × 10 ⁶ CAR T-cells administered as a sequential infusion of two components (CD8 ⁺ and CD4 ⁺ CAR T-cells). Because of the absence of any clear dose-related toxicity, data across all doses were combined for analyses. Three hundred and forty-four patients underwent leukapheresis with the intent to manufacture liso-cel, of whom 269 received liso-cel infusion making this the largest study among the pivotal studies (see Table 26.1 ). The inclusion criteria for the TRANSCEND study was also the most inclusive. The median age was 63 years (range 54–70 years, 42% 65 years or older, 10% 75 years or older). Patients with the following histologic diagnoses were eligible: DLBCL (51%), high-grade B-cell lymphoma (double-hit or triple-hit lymphoma, 13%), DLBCL transformed from any indolent lymphoma (22% tFL; 7% transformed from other indolent NHL), PMBCL (6%), and FL grade 3B (1%). Having secondary CNS lymphoma was allowed in this study and seven such patients were enrolled. In addition, the eligibility criteria allowed for less conservative thresholds for renal function, cardiac function, and performance status.

The overall response rate was 73% and 53% achieved a complete response. PFS and OS estimates at 12 months were 44% and 58%, respectively. The estimated duration of response at 12 months was 55% among patients who had a complete or partial response and slightly higher (65%) among those who achieved a complete response. The most frequent treatment-emergent adverse events were neutropenia in 63% of patients, CRS in 42%, and nausea in 33%. Grade 3 or 4 CRS and neurologic events occurred in 2% and 10% of patients, respectively. We cannot compare toxicity across the pivotal studies for several reasons, notably, different grading systems were used and management including mitigating strategies were different.

TRANSCEND and ZUMA-1 utilized the same grading system for CRS and had markedly different rates of CRS and neurotoxicity ( Table 26.2 ). In ZUMA-1, any grade and severe CRS occurred in 93% and 13% of patients, compared with 42% and 2% in TRANSCEND. Further, neurologic toxicities occurred in 64% (severe 28%) of patients treated with axi-cel compared to 30% (severe 10%) of patients treated with liso-cel. However, it appears that liso-cel has one of the most favorable safety profiles. This is thought to be caused by the difference in costimulatory domains (CD28 vs. 4-1BB) and potentially differences in the baseline characteristics of the patients.