Primary analysis of phase I/II of the ZUMA-6 trial – axicabtagene ciloleucel plus atezolizumab for patients with refractory DLBCL

The anti-CD19 chimeric antigen receptor (CAR) T-cell therapy, axicabtagene ciloleucel (axi-cel), was approved in the US and Europe for the treatment of adult patients with relapsed/refractory (R/R) large B-cell lymphoma (LBCL) following two or more prior lines of therapy.¹ The approvals were based on the results from the ZUMA-1 study, and the long-term results can be found on the Lymphoma Hub here.

The programmed death receptor 1 (PD-1) and its ligand (PD-L1) immune checkpoint axis is imperative for physiological immune regulation but is dysregulated in a number of cancers, resulting in decreased immune-mediated tumor destruction.² Upregulation of PD-1 on CAR T-cells and PD-L1 on tumor cells has been observed following CAR T-cell infusion, which may jeopardize treatment effectiveness.³

The phase I/II of the ZUMA-6 (NCT02926833) study is currently investigating the safety and efficacy of axi-cel in combination with the humanized, anti-PD-L1 monoclonal antibody, atezolizumab (atezo), for the treatment of adult patients with R/R diffuse LBCL (DLBCL). The study aims to evaluate the hypothesis that immune checkpoint blockade may augment CAR T-cell activity.³

At the 2020 American Association for Cancer Research (AACR) virtual annual meeting, Caron Jacobson presented the results from the phase II portion of the ZUMA-6 study – below is a summary.³

 ZUMA-6

• Data cutoff: February 21, 2019

• Median follow-up: 10.2 months

• Primary endpoints: Dose-limiting toxicities (DLTs; phase I) and complete response rate (phase II)

Patient eligibility

• R/R disease was defined as no response to the last chemotherapy or relapse ≤ 12 months following autologous stem cell transplant (auto-SCT)
• Prior therapy must have included an anti-CD20 monoclonal antibody and anthracycline
• Patients with DLBCL were screened, and those that met the inclusion criteria were enrolled at the time of T-cell collection

Phase I

• The phase I (n = 12) portion was a 3 × 3 design dose schedule finding study (Figure 1)
• Each patient received four fixed doses of atezo, one every 21 days, which differed by the time of the first atezo dose, with respect to axi-cel infusion
• No DLTs were observed in the first three patients in Cohorts 1 and 2
• One DLT was observed in the first three patients of Cohort 3 due to prolonged cytopenia
  • Three more patients were enrolled, and no DLTs were observed

Phase II

• The dosing schedule employed in the Cohort 3 of phase I (Figure 1) was taken forward to the phase II dose schedule expansion, which enrolled a further 22 patients

• The total number of patients receiving the recommended phase II dosing schedule: 28

Figure 1. Study design for phase I/II portions of the ZUMA-6 study³

Atezo, atezolizumab; axi-cel, axicabtagene ciloleucel; Cy, cyclophosphamide; DLT, dose limiting toxicity; flu, fludarabine

Results

Results presented represent the phase I Cohort 3 and phase II portions of this study, unless stated otherwise.

Patient characteristics

• All patients enrolled across the study (N = 34) received axi-cel and ≥ one dose of atezo
• Ten patients did not receive all four doses of atezo, for reasons including
  • disease progression (n = 3)
  • adverse events (AEs; n = 6)
  • death (n = 1)
• Patient characteristics are presented in Table 1

Table1. Baseline patient characteristics and tumor PD-L1 expression³

ECOG, Eastern Cooperative Oncology Group; IPI, International Prognostic Index *H = intensity score multiplied by the % of cells that were positive for each level. Four patients had missing data; †+1, +2, and +3 refer to weak, moderate, and strong staining for PD-L1 using the VENTANA PD-L1 assay. Four patients had missing data
Characteristic	Phase I Cohort 3 + phase II (n = 28)
Median age, years (range)	58 (42 – 71)
Male, %	57
ECOG performance status, % 0 1	61 39
Disease stage, % II III or IV	21 79
IPI score, % 0 – 2 3 – 4	54 46
Prior therapies, % ≥ 2 ≥ 3	86 50
Primary refractory disease, %	14
PD-L1 expression Median baseline tumor cell PD-L1 H score (range)* PD-L1 immune infiltrate intensity score,† %    +1    +2    +3	40 (0 – 240)    25 25 39

 AEs

• The most common Grade ≥ 3 treatment-emergent AEs (TEAEs) were cytopenias, electrolyte abnormalities and encephalopathy
• Grade ≥ 3 AEs were associated with axi-cel, atezo or both in 39%, 14% and 25% of patients, respectively
• One Grade 5 AE, not believed to be treatment-related, was observed on day +22 following axi-cel infusion, resulting in multiorgan failure and death

• Cytokine release syndrome (CRS) was experienced in 96% of patients with Grade 3 CRS observed in 4% of patients. None of the patients experienced Grade 4 or 5 CRS
  • median time to onset of CRS was 2 days (1–13) with a median duration of 7 days (1–92)
  • all events were resolved by the data cutoff
• Neurotoxicity was experienced in 68% of patients with Grade 3 or 4 neurotoxicity observed in 29% of patients. None of the patients experienced Grade 5 neurotoxicity
  • median time to onset of neurotoxicity was 6 days (1–23) with a median duration of 9 days (2–60)
  • two patients had ongoing events at the data cutoff: one patient with Grade 1 tremor and one patient with hypoesthesia

 Efficacy

• Patient outcomes to axi-cel + atezo are presented in Figure 2 and Table 2
• The median duration of response (DOR), progression-free survival (PFS) and overall survival (OS) were not reached (Table 2)
• Four patients demonstrated late conversions as follows:
  • Partial response – complete response (n = 3)
  • Stable disease – complete response (n = 1)

 Figure 2. Patient response rates³

CR, complete response; ORR, overall response rate; PD, progressive disease; PR, partial response; SD, stable disease

Table 2. Patient outcomes³

CI, confidence interval; DOR, duration of response; NE, not evaluable; NR, not reached; OS, overall survival; PFS, progression-free survival *February 21, 2019
	Phase I Cohort 3 + phase II (n = 28)
Patients with ongoing response at cutoff*, %	46
DOR Median, months    95% CI 6-month estimate, %	NR 1.8 – NE 62
PFS Median, months    95% CI 6-month estimate, %	NR 3.1 – NE 50
OS Median, months    95% CI 6-month estimate, %	NR 12 – NE 71

 Conclusions

The administration of atezo following axi-cel demonstrated, with a manageable safety profile, that it was consistent with that reported in the ZUMA-1 study. Although there has been no statistical comparison between patient outcomes in the ZUMA-6 vs ZUMA-1 studies, the combination of axi-cel with atezo demonstrated similar clinical efficacy to axi-cel alone. Similarly, the pharmacokinetic and pharmacodynamic profiles of axi-cel when administered with atezo were comparable to when the CAR T-cell therapy is administered alone. The timing of atezo administration with respect to CAR T-cell treatment is thought to impact CAR T-cell persistence and requires further investigation. A range of techniques are being employed to further define immune cell subsets and study the tumor microenvironment to better understand the anti-tumor immune response of axi-cel. Furthermore, enhanced knowledge of such parameters may prove prognostically valuable in deciding which patients could benefit from immune-checkpoint blockade alongside CAR T-cell therapy.