The combination of the CAR T-cell therapy axicabtagene ciloleucel and the PD-L1 inhibitor atezolizumab was highly active with a manageable safety profile in patients with refractory DLBCL enrolled in the phase I/II ZUMA-6 trial.
Frederick L. Locke, MD, a medical oncologist at Moffitt Cancer Center in Tampa, Florida
Frederick L. Locke, MD
The combination of the anti‑CD19 chimeric antigen receptor (CAR) T-cell therapy axicabtagene ciloleucel (axi-cel; Yescarta) and the PD-L1 inhibitor atezolizumab (Tecentriq) was highly active with a manageable safety profile in patients with refractory diffuse large B-cell lymphoma (DLBCL) enrolled in the phase I/II ZUMA-6 trial.1
The overall response rate (ORR) in the 9-patient phase I portion of the trial was 89% (n = 8), including a complete response (CR) rate of 56% (n = 5).
“The toxicities associated with axi-cel were similar to what was seen in the pivotal ZUMA-1 clinical trial—similar rates of cytokine release syndrome (CRS) and neurologic toxicity,” lead ZUMA-6 author Frederick L. Locke, MD, said in an interview with OncLive at the 2017 ASH Annual Meeting, where the initial ZUMA-6 findings were presented.
“When we evaluated safety, we tried to determine if the combination increases toxicity rates, and we did not see that. The patients who had severe CRS and neurologic toxicity actually had it before the infusion of the atezolizumab, so it was really not associated with the combination,” added Locke.
The ZUMA-6 trial is aimed at building on the success of the ZUMA-1 trial, in which the CR rate was 40% with a median follow-up of 15.4 months for patients with refractory, aggressive non-Hodgkin lymphoma (NHL) treated with axi-cel.2 Forty-two percent of patients remained progression-free and 56% remained alive at the 15.4-month follow-up. The 18-month progression-free survival rate was 41% and the 18-month overall survival rate was 52%. Based on initial data from the ZUMA-1 trial, the FDA approved axi-cel in October 2017 as a treatment for adult patients with relapsed or refractory NHL.
ZUMA-6 is accruing patients with refractory DLBCL who have progressive or stable disease as their best response to their most recent chemotherapy or disease progression/recurrence within 12 months of autologous stem cell transplant. Patients with CD-20—positive tumors must have received an anti-CD20 monoclonal antibody and all patients must have had prior anthracycline-based chemotherapy. All patients must also have an ECOG performance status of 0 or 1.
Patients in the phase I part of the study were divided into 3 cohorts. All patients received conditioning chemotherapy with 500 mg/m2 of cyclophosphamide and 30 mg/m2 of fludarabine. This was followed in all 3 arms by an axi-cel infusion of 2 X 106 CAR T-cells/kg. Atezolizumab was started at 21, 14, and 1 day(s) after axi-cel in cohort 1, 2, and 3, respectively. The dose of the PD-L1 inhibitor was the same for all 3 cohorts: 1200 mg every 21 days for 4 doses.
The average age across the 3 cohorts was 52 years (range, 29-66). There were 6 male and 3 female patients. Four patients had stage IV disease, 2 had stage III, and 3 had stage II. The average number of prior therapies was 3 (range, 2-4). Regarding baseline PD-L1 status, the following results were determined with tumor cell/immune infiltrate staining: 3+/3+ (1 patient), 3+/2+ (3 patients), 2+/2+ (1 patient), 0/3+ (1 patient), and samples not available (3 patients).
The primary endpoint was the occurrence of adverse events (AEs) defined as dose-limiting toxicities (DLTs). Secondary/exploratory endpoints included ORR, biomarker analyses, and safety.
In cohort 1, there were 2 partial responses (PRs), as well as 1 CR in which the patient experienced a PR to CR conversion. In cohort 2, there was 1 CR, 1 CR that was a conversion from a PR, and 1 patient with stable disease as the best response. In cohort 3, there were 2 CRs and 1 PR.
“Looking at the biomarkers, we found that there is a trend toward an increase in the number of CAR T-cells in the peripheral blood following the combination of CAR T-cell therapy and atezolizumab,” noted Locke.
Across the study population, grade 3/4 treatment-emergent AEs occurring in at least 3 patients included anemia (n = 7), encephalopathy (n = 5), neutropenia (n = 4), hyponatremia (n = 3), decreased lymphocyte count (n = 3), decreased neutrophil count (n = 3), pyrexia (n = 3), and thrombocytopenia (n = 3). Additionally, grade 3 hypophosphatemia, hypoxia, blood bilirubin increase, and lung infection occurred in 2 patients each.
Combining axi-cel with atezolizumab did not result in an increased use of tocilizumab or steroids. One patient had a DLT related to cytopenias that lasted for more than 30 days. There were no patient deaths.
Moving forward, Locke said, “The next step will be to expand it into a phase II trial. So, we really want to find the safe schedule to combine these 2 therapies. Once we’ve settled on a dosing schedule, we will expand with another 22 patients and see if it improves the efficacy as compared to [the efficacy with axi-cel alone] in the DLBCL patients who were treated on the ZUMA-1 trial.”
Axi-cel—related AEs occurred in all 9 patients and 4 patients experienced atezolizumab-related AEs. Overall, 8 patients experienced grade 3/4 AEs. There were 3 grade 3 and 4 grade 4 AEs related to axi-cel, and 1 grade 4 AE related to atezolizumab.