2-Year Follow-up and High-Risk Subset Analysis of Zuma-1, the Pivotal Study of Axicabtagene Ciloleucel (Axi-Cel) in Patients with Refractory Large B Cell Lymphoma

Background: Axi-cel is a US FDA-approved, autologous anti-CD19 chimeric antigen receptor (CAR) T cell therapy for the treatment of patients with relapsed or refractory large B cell lymphoma with ≥ 2 prior systemic therapies. This report focuses on high-risk patient populations as well as long-term durability of response, and B cell recovery.

Methods: In ZUMA-1, eligible patients with refractory large B cell lymphoma (diffuse large B cell lymphoma, primary mediastinal B cell lymphoma, transformed follicular lymphoma) underwent leukapheresis and received low-dose conditioning followed by a target dose of 2 × 10⁶ anti-CD19 CAR T cells/kg (Neelapu and Locke et al. NEJM. 2017; NCT02348216). Outcomes in patients with double-expressor B cell lymphoma (MYC [≥ 40%] and BCL-2 [≥ 50%] protein expression by immunohistochemistry [IHC]) or high-grade B cell lymphoma (HGBCL), defined as double- or triple-hit (MYC+ and BCL2+ and/or BCL6+ by fluorescence in situ hybridization) or not otherwise specified (MYC- and > 70% Ki67 by IHC) were examined by independent pathology review. Exploratory analyses, including normal B cell levels in peripheral blood over time and frequency of use of safety interventions of interest, were also performed. A long-term follow-up analysis will be conducted with a data cutoff of August 11, 2018 for all 108 patients, including the HGBCL subgroup.

Results: As of August 11, 2017, all 108 patients had at least 1 year of follow-up, with a median follow-up of 15.4 months. The objective response rate (ORR) was 82% with a complete response (CR) rate or 58%. The CR rate was 53% (29/55) in patients with disease refractory to ≥ 2 consecutive prior lines of therapy and 72% (18/25) in patients who had relapsed within 12 months after autologous stem cell transplantation. High-risk genetics were assessed in the 47 evaluable pre-treatment tumor samples: 37 patients (79%) had HGBCL or double-expressor B cell lymphoma and had an ORR of 89% (33/37) including a CR rate of 68% (25/37). Forty-two percent of patients overall had ongoing responses with a median follow-up of 15.4 months including 49% (18/37) of patients with high-risk genetics.

To investigate the relationship between B cell recovery and ongoing response, B cell levels were assessed over time. Overall, of the 87 evaluable patients, 47% had no detectable B cells at baseline, and the remainder had levels close to or below the lower level of quantification of the assay. In patients with ongoing responses at 12 months post axi-cel infusion, 19 of the 35 (54%) patients with evaluable samples had detectable B cells at 12 months. This suggests B cell recovery in some patients with ongoing response as only 6 of 40 (15%) patients with evaluable samples had detectable B cells at 3 months after axi-cel infusion.

Grade ≥ 3 cytokine release syndrome (CRS) and neurologic events (NE) were generally reversible and reported for 12% and 31% of patients, respectively. Tocilizumab (45%) and/or steroids (28%) were used for the management of CRS and NE. The use of vasopressors, dialysis, and intubation were minimal, used in only 17%, 3%, and 3% of patients, respectively.

Thirty-four percent of patients (37/108) had either no CRS/NE (6%) or only Grade 1 CRS with (12%) or without (16%) Grade 1 NE. For these 37 patients, the median onset of CRS and NE was 2 and 7 days after infusion, respectively. Patients with Grade 0 - 1 CRS/NE had similar efficacy (ORR, 86%; CR, 65%) but lower peak/AUC CAR T cell levels vs the overall population.

Updated safety and efficacy results will be presented with a minimum follow-up of 2 years and a median follow-up of 27.1 months.

Conclusion: High rates of durable response were observed in patients with HGBCL and double-expressor B cell lymphoma, with approximately half of these patients (18/37) maintaining CR at ≥ 1 year. Efficacy in these high-risk populations was comparable to the overall patient population in ZUMA-1. At 12 months, B-cell recovery was observed in over half the patients with ongoing remission. About one-third of patients had only Grades 0 - 1 CRS/NE yet showed comparable efficacy with the overall patient population. High-grade CRS/NE were largely managed with tocilizumab and steroids with very low use of intensive or invasive interventions. Updated outcomes with a minimum of 2 years of follow-up will be presented.