Interim results from ZUMA-12: Axi-cel as frontline treatment for high-risk LBCL

Patients with high-risk large B-cell lymphoma (LBCL) are mainly treated with rituximab-based chemoimmunotherapies that do not provide sufficient efficacy, as approximately 50% of these patients will not achieve long-term disease remission.¹ The CD19 targeting CAR T product, axicabtagene ciloleucel (axi-cel), has shown great success in the relapsed/refractory (R/R) LBCL setting and has already been approved both in Europe and the US for the treatment of adult patients with R/R diffuse large B-cell lymphoma (DLBCL), follicular lymphoma (FL), and primary mediastinal B-cell lymphoma (PMBCL) after ≥ 2 lines of prior therapy.¹ The phase II trial, ZUMA-12 (NCT03761056), is currently assessing the efficacy and safety of axi-cel as frontline therapy for adults with high-risk LBCL. During the 62nd American Society of Hematology (ASH) Annual Meeting & Exposition, the interim results from ZUMA-12¹ were presented by Sattva Neelapu and are hereby summarized.

Study design

• A multicenter, open-label, single-arm, phase II trial in adult patients with high-grade confirmed LBCL with MYC and BCL2 and/or BCL6 translocations or an International Prognostic Index (IPI) ≥ 3 at any time prior enrolment. Eligible patients needed to have a positive positron interim emission tomography (PET; Deauville score of 4 or 5) scan after two cycles of anti-CD20 monoclonal antibody plus an anthracycline-containing regimen. The full study design of ZUMA-12 is shown in Figure 1
• Prior to axi-cel infusion, on Days −5, −4, and −3, fludarabine (30 mg/m²), and cyclophosphamide (500 mg/m²) were administered as conditioning chemotherapy
• Non-chemotherapy bridging therapy was optional
• Axi-cel infusion was performed as a single intravenous dose of 2 × 10⁶ CAR T cells/kg on Day 0
• Primary endpoint was complete response (CR) rate as defined by the Lugano classification,² while secondary outcomes included overall response rate (ORR), duration of response (DoR), event-free survival (EFS), overall survival (OS), progression-free survival (PFS), safety, and the number of circulating CAR T cells and cytokines

Axi-cel, axicabtagene ciloleucel; CAR, chimeric antigen receptor; CR, complete response; DoR, duration of response; DS, Deauville score; ECOG PS, Eastern Cooperative Oncology Group performance score; EFS, event-free survival; IPI, International Prognostic Index; iv, intravenous; LBCL, large B-cell lymphoma; mAb, monoclonal antibody; ORR, overall response rate; OS, overall survival, PET, positron emission tomography; PFS, progression-free survival.

Results

• In total, 32 out of 37 enrolled patients received an axi-cel infusion. At data cutoff (August 25, 2020), 32 patients were evaluable for safety with a median follow-up of 9.5 months (range, 0.9–18) and 27 patients were included in the efficacy analysis with a median follow-up of 9.3 months (range, 0.9–18)
• Key patient baseline characteristics are shown in Table 1

Table 1. Key patient baseline characteristics in ZUMA-12¹

ECOG PS, Eastern Cooperative Oncology Group performance score; FISH, fluorescence in situ hybridization; IPI, International Prognostic Index. *Determined by FISH per investigator.
Baseline characteristic	N = 32
Age
Median (range), years	61 (23–86)
≥ 65 years, n (%)	13 (41)
Male, n (%)	23 (72)
Disease stage III–IV, n (%)	28 (88)
ECOG PS ≥ 1, n (%)	21 (66)
One prior systemic line, n (%)	32 (100)
Double-/triple-hit*, n (%)	17 (53)
IPI ≥ 3, n (%)	23 (72)
Deauville score, n (%)
4	16 (50)
5	16 (50)

• The following outcomes were reported:
  • ORR: 85%
  • CR: 74% (n = 20)
  • Partial response (PR): 11% (n = 3)
  • Stable disease (SD): 15%
  • None progressed
• At data cutoff, 19 patients had ongoing responses (70%)
• Median time to CR was 1 month (range, 0.9–6.4), similar to the median time to objective response (1 month [range, 0.9–3.1])
• Five patients (19%) converted from PR (n = 4) or SD (n = 1) to CR
• At a median follow-up of 9.5 months, median DoR, PFS, and OS were not reached
• CAR T-cell expansion was higher in this trial when compared with ZUMA-1, with a median time to peak circulating CAR T cells of 8 days

With regards to safety:

• The most common treatment-related Grade ≥ 3 adverse events were:

• • Encephalopathy (16%)
  • Alanine aminotransferase increase (9%)
  • Neutrophil count decrease (9%)
• One death occurred during treatment due to COVID-19
• All 32 patients developed cytokine release syndrome (CRS) while 22 (69%) experienced neurological events (NE). Details regarding grade, time to onset, duration, and management of these events are shown below in Table 2
• The median levels of serum cytokines, including interleukin 6 (IL-6), IL-5, IL-8, interferon γ (IFN-γ), and granulocyte-macrophage colony-stimulating factor (GM-CSF) among others, were associated with Grade ≥ 3 NE or CRS events

Table 2. CRS and NE events in ZUMA-12¹

CRS, cytokine release syndrome; NE, neurological events. *The two unresolved NE events at data cutoff were Grade 1 tremor and Grade 1 memory impairment.
Parameter	CRS (N = 32)	NE (N = 32)
Any Grade, n (%)	32 (100)	22 (69)
Grade ≥ 3, n (%)	3 (9)	8 (25)
Grade 4, n (%)	0 (0)	2 (6)
Grade 5, n (%)	0 (0)	0 (0)
Most commonly associated any-grade symptoms, n (%)	Pyrexia: 32 (100) Chills: 8 (25) Hypotension: 8 (25)	Encephalopathy: 10 (31) Confusional state: 9 (28)
Management, n (%)
Tocilizumab	17 (53)	0 (0)
Steroids	8 (25)	11 (34)
Median time to onset, days (range)	4 (1–10)	9 (2–44)
Median duration of events, days (range)	6 (1–13)	6 (1–54)
Patients with resolved events, n/N (%)	32/32 (100)	20/22 (91)*

Conclusions

 The phase II ZUMA-12 trial is the first to assess the use of CAR T cells as frontline treatment for high-grade LBCL. The interim data show promising response outcomes, with axi-cel infusion leading to an ORR of 85% and a CR rate of 74%. At data cutoff, 70% of patients had ongoing responses, indicating that these responses were durable. Lastly, axi-cel was well tolerated in this patient subset, with manageable CRS and NE toxicities. The authors also highlighted the higher CAR T expansion rates observed in ZUMA-12 compared with ZUMA-1, where axi-cel was used in the R/R setting, potentially indicating an improved CAR T cell fitness when axi-cel is used as first-line treatment. Further data are anticipated to validate these results.