Highlights
- The Phase 1/2 ZUMA-6 trial evaluated the safety and efficacy of axicabtagene ciloleucel (axi-cel) in combination with the PD-L1 inhibitor atezolizumab in patients with refractory diffuse large B-cell lymphoma (DLBCL).
- The combination demonstrated a manageable safety profile with no new safety signals; Grade 3 or higher cytokine release syndrome (CRS) occurred in 9% of patients, and neurologic events occurred in 32%.
- Efficacy outcomes were robust, with a complete response (CR) rate of 54% and a median overall survival (OS) of 32.2 months at a median follow-up of 56.9 months.
- Despite the theoretical synergy of checkpoint inhibition, the efficacy and CAR T-cell pharmacokinetics were largely consistent with historical data from axi-cel monotherapy (ZUMA-1).
The Challenge of Durable Remission in Refractory LBCL
The advent of CD19-directed chimeric antigen receptor (CAR) T-cell therapy has revolutionized the treatment landscape for relapsed or refractory (R/R) large B-cell lymphoma (LBCL). Axicabtagene ciloleucel (axi-cel), an autologous CD19-directed CAR T-cell therapy, has demonstrated significant clinical benefit, leading to its approval and widespread use. However, the curative potential of CAR T-cell therapy remains limited by resistance and relapse. Approximately two-thirds of patients treated with axi-cel do not achieve or maintain long-term remission.
One proposed mechanism of resistance is the immunosuppressive tumor microenvironment (TME), specifically the programmed death-1 (PD-1)/programmed death-ligand 1 (PD-L1) pathway. Preclinical models and early clinical observations suggested that CAR T-cells may upregulate PD-1 upon activation, leading to exhaustion when they encounter PD-L1-expressing tumor cells. Therefore, combining CAR T-cells with immune checkpoint inhibitors (ICIs) like atezolizumab (an anti-PD-L1 antibody) represents a logical therapeutic strategy to sustain CAR T-cell activity and improve clinical outcomes.
Study Design: The ZUMA-6 Framework
ZUMA-6 (NCT02926833) was a multi-center, Phase 1/2 study designed to assess the feasibility, safety, and efficacy of this combination. The study enrolled patients with refractory DLBCL who had received at least two prior lines of systemic therapy. The treatment protocol involved a standard conditioning regimen of cyclophosphamide and fludarabine, followed by a single infusion of axi-cel at a dose of 2×10^6 CAR T-cells/kg.
Atezolizumab was administered intravenously at a fixed dose of 1,200 mg every 21 days for up to four cycles. In Phase 1, the primary objective was to evaluate dose-limiting toxicities (DLTs) and determine the safety of the combination. Phase 2 focused on the complete response (CR) rate. Secondary endpoints included duration of response (DOR), progression-free survival (PFS), overall survival (OS), and pharmacokinetics/pharmacodynamics of both the CAR T-cells and the PD-L1 inhibitor.
Key Findings: Safety and Efficacy
Safety and Tolerability
Overall, 34 patients received the combination therapy. The safety profile was consistent with the known toxicities of the individual agents. In Phase 1, only one patient experienced DLTs, which included Grade 4 neutropenia and thrombocytopenia. Across the entire cohort, 88% of patients experienced Grade 3 or higher treatment-emergent adverse events (TEAEs). The most common high-grade toxicities were hematologic, as expected with lymphodepleting chemotherapy and CAR T-cell therapy.
Cytokine release syndrome (CRS) and neurologic events (NEs) remain the primary concerns with axi-cel. In ZUMA-6, 9% of patients experienced Grade 3 or higher CRS, and 32% experienced Grade 3 or higher NEs. These rates are comparable to those reported in the landmark ZUMA-1 trial of axi-cel monotherapy, suggesting that the addition of atezolizumab did not significantly compound the acute toxicities associated with CAR T-cell activation.
Clinical Efficacy
The efficacy results were encouraging but did not clearly surpass monotherapy benchmarks. In the final analysis, with a median follow-up of 56.9 months, 15 of the 28 evaluable patients in Phase 2 (54%) achieved a complete response (CR). The overall response rate (ORR) was 75%. While these figures reflect high clinical activity in a refractory population, they are remarkably similar to the long-term results of ZUMA-1, which reported a CR rate of 58%.
Survival metrics were also stable over long-term follow-up. The median PFS was 9.0 months, and the median OS reached 32.2 months. For patients who achieved a CR, the responses tended to be durable, highlighting that when the therapy works, it provides long-term disease control. However, the addition of atezolizumab did not appear to “rescue” patients who would have otherwise progressed on CAR T-cells alone.
Translational Insights: Pharmacokinetics and Biomarkers
A critical component of ZUMA-6 was the evaluation of CAR T-cell expansion and persistence. Correlative analyses showed that peak CAR T-cell levels in the blood occurred within the first two weeks post-infusion, mirroring the kinetics seen in monotherapy trials. Furthermore, the levels of key inflammatory cytokines (such as IL-6, IFN-gamma, and TNF-alpha) were similar to those observed in ZUMA-1. This suggests that the PD-L1 blockade did not significantly alter the initial proliferative burst or the systemic inflammatory response of the CAR T-cells.
Biomarker analyses attempted to identify which patients might benefit most from the combination. While PD-L1 expression on tumor cells or immune cells in the TME is often a predictor for ICI success in solid tumors, its role in the context of CAR T-cell therapy for DLBCL remains less clear. The study noted that correlative data might eventually help identify a subset of patients with high T-cell exhaustion markers who could potentially derive more benefit from checkpoint inhibition, but no definitive predictive biomarker was established in this cohort.
Expert Commentary: Interpreting the ZUMA-6 Results
The results of ZUMA-6 provide a nuanced view of combination immunotherapy. From a safety perspective, the study is a success; it proves that PD-L1 inhibitors can be safely administered alongside CAR T-cells without increasing the risk of life-threatening CRS or neurotoxicity. This opens the door for other combinations, perhaps involving different checkpoints or earlier interventions.
However, from an efficacy standpoint, the study raises questions about the timing and biology of checkpoint blockade in DLBCL. Why did the addition of atezolizumab not significantly improve the CR rate or PFS compared to axi-cel alone? Several hypotheses exist:
- Timing of Administration: In ZUMA-6, atezolizumab was started after the axi-cel infusion. It is possible that the “exhaustion” of CAR T-cells happens much faster than the ICI can counteract, or that the ICI should be administered during the manufacturing or lymphodepletion phase.
- Alternative Checkpoints: PD-L1 may not be the primary driver of resistance in all patients. Other markers such as LAG-3, TIM-3, or TIGIT may play more prominent roles in CAR T-cell dysfunction in the lymphoma microenvironment.
- Patient Selection: The study treated an unselected refractory population. It is possible that only a minority of DLBCL cases are truly driven by PD-L1-mediated immunosuppression, diluting the observable benefit in a small Phase 1/2 cohort.
Despite these questions, the 32.2-month median OS is a testament to the power of the axi-cel backbone. For clinicians, the takeaway is that while the combination is safe, it is not yet a new standard of care. Axi-cel monotherapy remains the benchmark, and future research must focus on identifying the specific biological signatures of resistance to tailor these expensive and intensive combinations to the right patients.
Conclusion
The final analysis of the ZUMA-6 trial confirms that combining axi-cel with atezolizumab is feasible and maintains a manageable safety profile in patients with refractory DLBCL. While the efficacy outcomes were robust and durable for many, they did not represent a significant leap forward over axi-cel monotherapy. These results underscore the complexity of the lymphoma tumor microenvironment and suggest that overcoming CAR T-cell resistance will likely require more than a “one-size-fits-all” approach to checkpoint inhibition. Future studies should leverage the correlative data from ZUMA-6 to refine patient selection and explore novel combination strategies.
Funding and clinicaltrials.gov
This study was funded by Kite, a Gilead Company. Atezolizumab was provided by Genentech, a member of the Roche Group. The trial is registered at clinicaltrials.gov as NCT02926833.
References
- Jacobson CA, Westin JR, Miklos DB, et al. Axicabtagene Ciloleucel in Combination With Atezolizumab in Patients With Refractory Diffuse Large B-cell Lymphoma: The Phase 1/2 ZUMA-6 Trial. Clin Cancer Res. 2025. doi:10.1158/1078-0432.CCR-25-0602.
- Neelapu SS, Locke FL, Bartlett NL, et al. Axicabtagene ciloleucel CAR T-cell therapy in refractory large B-cell lymphoma. N Engl J Med. 2017;377(26):2531-2544.
- Locke FL, Miklos DB, Jacobson CA, et al. Axicabtagene ciloleucel as second-line therapy for large B-cell lymphoma. N Engl J Med. 2022;386(7):640-654.