Five-Year ZUMA-5 Follow-Up: Durable Remissions and Potential for Cure with Axicabtagene Ciloleucel in Relapsed/Refractory Follicular Lymphoma

Highlight

– Axicabtagene ciloleucel (axi-cel) produced an overall response rate of 90% and a complete response rate of 75% at a median follow-up of 64.6 months in ZUMA-5.
– Median duration of response was 60.4 months and median progression-free survival was 62.2 months; overall survival and time to next treatment were not reached.
– Durable remissions correlated with robust early CAR T-cell expansion and a naïve T-cell phenotype in the manufactured product.
– Long-term safety was manageable; late toxicities were infrequent and mostly unrelated to axi-cel.

Background

Follicular lymphoma (FL) is the most common indolent non-Hodgkin lymphoma and typically follows a relapsing course. Despite effective initial responses to chemoimmunotherapy, most patients will relapse and require multiple lines of treatment. Relapsed/refractory (R/R) FL that has progressed after alkylator- and anti-CD20–containing regimens represents a population with significant unmet need: subsequent therapies provide variable durability and cumulative toxicity, and a subset of patients develop early progression associated with poor survival.

Chimeric antigen receptor (CAR) T-cell therapy targeting CD19 has transformed outcomes in aggressive B‑cell lymphomas. ZUMA-5 evaluated axi-cel, an autologous anti-CD19 CAR T-cell product with a CD28 costimulatory domain, in R/R indolent non-Hodgkin lymphoma (iNHL), including FL and marginal zone lymphoma (MZL). The five-year follow-up reported clinically meaningful long-term outcomes that raise the question of durable disease control — and possible cure — for a subset of patients with FL.

Study Design

ZUMA-5 is a single-arm, multicentre, phase 2 trial conducted across centres in the United States and France (ClinicalTrials.gov NCT03105336). Key inclusion criteria were age ≥18 years, histologically confirmed indolent NHL (FL or MZL), relapsed/refractory disease after ≥2 prior lines of therapy (including an anti-CD20 monoclonal antibody and an alkylating agent), and ECOG performance status 0–1.

Patients underwent leukapheresis for autologous T-cell collection. Lymphodepletion consisted of cyclophosphamide 500 mg/m2/day and fludarabine 30 mg/m2/day on days −5, −4 and −3, followed by a single infusion of axi-cel at 2 × 106 CAR T cells/kg on day 0. The primary endpoint for activity was overall response rate (ORR; complete response [CR] + partial response [PR]) assessed by an independent review committee using Lugano criteria.

The primary activity analysis used a per-protocol population with prespecified follow-up times. Safety analyses included all patients who received an axi-cel infusion.

Key Results — Efficacy

The updated long-term analysis included 159 enrolled patients with R/R iNHL (127 with FL, 31 with MZL) and a median follow-up of 64.6 months. Key efficacy outcomes were:

• Overall response rate: 90%
• Complete response rate: 75%
• Median duration of response (DOR): 60.4 months
• Median progression-free survival (PFS): 62.2 months
• Median time to next treatment: not reached
• Overall survival (OS): not reached
• At data cutoff, 55% of patients were alive without receiving subsequent anticancer therapy.

In the FL subgroup, median lymphoma-specific PFS was not reached; by the analysis, 34% of patients experienced progression or death related to lymphoma or study treatment. Notably, events (progressions or lymphoma-related deaths) after approximately 30 months post-infusion were uncommon, suggesting a plateau in the risk of relapse for many patients.

Interpretation of the efficacy signal

These outcomes represent a markedly prolonged PFS and DOR compared with historical controls for multiple lines of therapy in R/R FL, where sequential treatments typically yield shorter remissions. The high CR rate (75%)—combined with a durable median DOR of about five years—implies that a substantial subset of patients achieves long-term disease control with axi-cel. The observation that late relapses beyond ~30 months are uncommon supports the hypothesis that axi-cel may provide curative potential in a fraction of patients with FL, although definitive statements about cure require longer follow-up and broader experience.

Safety

Safety was assessed across the trial. From the primary analysis set and prior reports, the most common grade ≥3 adverse events were cytopenias and infections. Specific findings included:

• Grade ≥3 cytopenias: 70% (104 of 148 patients in the earlier cohort)
• Grade ≥3 infections: 18% (26 of 148)
• Grade ≥3 cytokine release syndrome (CRS): 7% (10 of 148)
• Grade 3–4 neurologic events: 19% (28 of 148)
• Serious adverse events (any grade): 50% (74 of 148)
• Deaths due to adverse events: 3% (4 of 148), with one deemed treatment-related (multisystem organ failure)

In the long-term update, late-onset toxicities were infrequent and largely not attributable to axi-cel. Persistent cytopenias and infectious complications are recognized risks after CAR T-cell therapy and require monitoring and supportive care. Neurotoxicity and CRS were mostly early events, and standard mitigation strategies (prompt tocilizumab for CRS, corticosteroids for severe immune effector‑cell neurotoxicity syndrome) remain important.

Biomarkers and Mechanistic Insights

Investigators reported that durable responses and prolonged survival correlated with two product- and patient-level factors: robust early in vivo expansion of CAR T cells and a naïve T-cell phenotype in the manufactured product. Biologically, a higher proportion of less-differentiated (naïve or central memory) T cells in the apheresis product is associated with superior proliferative capacity and persistence after infusion, which likely underpins sustained antitumour activity.

These insights echo broader CAR T literature linking early expansion and persistence to durable remissions, and they emphasize the role of product composition and patient T-cell fitness in therapeutic outcomes.

Expert Commentary and Limitations

Strengths of ZUMA-5 include its multicentre design, high manufacturing success, and long median follow-up exceeding five years. The magnitude and durability of responses in a heavily pretreated FL population are compelling, positioning axi-cel as a potentially disease-modifying therapy for selected patients.

However, important limitations should temper interpretation:

• Single-arm design: Without a randomized comparator, it is difficult to directly quantify the incremental benefit of axi-cel against contemporary options, including bispecific antibodies, PI3K inhibitors, or stem cell transplant in select patients.
• Generalizability: The MZL cohort was small, limiting conclusions for that histology. Trial eligibility (ECOG 0–1, organ function criteria) may select for fitter patients than many treated in routine practice.
• Long-term safety: Although late toxicities were infrequent in this report, longer observation and post-marketing surveillance are needed to fully define risks such as secondary malignancies, late cytopenias, and chronic immune dysfunction.
• Comparative effectiveness: Emerging therapies for R/R FL, including off-the-shelf bispecific T-cell engagers, will need head-to-head or network-comparative evaluation to define optimal sequencing.

Clinical Implications and Future Directions

ZUMA-5 supports axi-cel as a potent therapeutic option for patients with R/R FL after at least two prior lines of therapy. The durable CRs and plateauing of relapse risk after ~30 months suggest potential for long-term remission and, potentially, cure in a subset of patients. Key practical considerations for clinicians include careful patient selection, management of early toxicities (CRS and neurotoxicity), and anticipatory management of cytopenias and infections.

Future research priorities include identification of predictive biomarkers to select patients most likely to derive long-term benefit, optimization of manufacturing to enrich favorable T-cell phenotypes, head-to-head comparisons with other novel agents and bispecific antibodies, and strategies to extend CAR T benefits to broader, less-fit populations.

Conclusion

The five-year ZUMA-5 analysis demonstrates that axi-cel yields high response rates and durable remissions in relapsed/refractory indolent NHL, particularly follicular lymphoma. Durable median DOR of approximately five years and a median PFS of roughly the same duration mark a substantial advance over historical therapies. While the single-arm design limits comparative claims, the long-term data indicate that axi-cel provides sustained disease control with an acceptable safety profile and has the potential to be curative for a subset of patients with FL. Continued follow-up, biomarker-driven patient selection, and comparative studies will be essential to refine its role in care pathways.

Funding and trial registration

Trial registration: ClinicalTrials.gov NCT03105336.
Funding sources and disclosures are reported in the original publications.

References

1. Neelapu SS, Chavez JC, Sehgal AR, et al. Five-Year Follow-Up Analysis of ZUMA-5: Axicabtagene Ciloleucel in Relapsed/Refractory Indolent Non-Hodgkin Lymphoma. J Clin Oncol. 2025 Nov 20;43(33):3573-3577. doi: 10.1200/JCO-25-00668. Epub 2025 Oct 16. PMID: 41100801; PMCID: PMC12622262.

2. Jacobson CA, Chavez JC, Sehgal AR, et al. Axicabtagene ciloleucel in relapsed or refractory indolent non-Hodgkin lymphoma (ZUMA-5): a single-arm, multicentre, phase 2 trial. Lancet Oncol. 2022 Jan;23(1):91-103. doi: 10.1016/S1470-2045(21)00591-X. Epub 2021 Dec 8. PMID: 34895487.