A comprehensive real-world analysis of 809 patients with relapsed/refractory diffuse large B-cell lymphoma (DLBCL) has demonstrated superior efficacy of axicabtagene ciloleucel (axi-cel) compared to tisagenlecleucel (tisa-cel), providing the first evidence of overall survival advantage between these two CAR T-cell therapies in routine clinical practice.
The study, conducted through France's DESCAR-T registry between December 2019 and October 2021, represents one of the largest real-world comparisons of these CD19-targeting CAR T-cell products. All patients had received at least two prior lines of therapy, with a median of three previous treatments.
Efficacy Outcomes Show Clear Advantage for Axi-cel
Using rigorous propensity score matching to balance patient characteristics, researchers compared 209 patients treated with axi-cel to 209 patients treated with tisa-cel. The results revealed significantly superior outcomes across multiple efficacy measures for axi-cel.
Overall response rates reached 80.4% with axi-cel compared to 66.0% with tisa-cel (P < 0.001), while complete response rates were 60.3% versus 42.1% respectively (P < 0.001). These real-world response rates closely mirrored those observed in the pivotal ZUMA-1 and JULIET trials, with ZUMA-1 reporting 82%/58% and JULIET showing 52%/40% for overall/complete response rates.
One-year progression-free survival was 46.6% for axi-cel compared to 33.2% for tisa-cel (hazard ratio 0.61; 95% CI, 0.46-0.79; P = 0.0003). Overall survival also favored axi-cel, with one-year survival rates of 63.5% versus 48.8% (hazard ratio 0.63; 95% CI, 0.45-0.88; P = 0.0072).
Notably, while axi-cel demonstrated superior initial response rates, duration of response showed no significant difference between the two therapies (one-year duration of response: 53.8% for axi-cel versus 41.8% for tisa-cel, P = 0.106).
Safety Profile Reveals Trade-offs
The enhanced efficacy of axi-cel came with increased toxicity across multiple domains. Cytokine release syndrome of any grade occurred in 86.1% of axi-cel patients compared to 75.6% of tisa-cel patients, though severe (grade ≥3) CRS rates were similar between groups.
More concerning was the significant difference in immune cell-associated neurotoxicity syndrome (ICANS). Grade ≥3 ICANS affected 13.9% of axi-cel patients versus only 2.9% of tisa-cel patients (P < 0.001). Any grade ICANS occurred in 48.8% versus 22.0% respectively.
Hematological toxicity proved substantially more frequent and severe with axi-cel. At one month post-infusion, any grade cytopenia occurred in 64.6% of axi-cel patients compared to 39.2% of tisa-cel patients, with grade ≥3 cytopenia affecting 34.0% versus 12.4% respectively. This pattern persisted at three months, indicating prolonged hematological recovery with axi-cel.
Subgroup Analyses Support Broad Applicability
The efficacy advantage of axi-cel remained consistent across patient subgroups. In patients aged 70 years or younger, median progression-free survival was 5.9 months with axi-cel versus 3.1 months with tisa-cel (P = 0.0128). For patients older than 70 years, median PFS was not reached with axi-cel compared to 3 months with tisa-cel (P = 0.0026).
Tumor bulk did not influence the relative efficacy between products. In patients without bulky disease (≤5 cm), median PFS was 7.9 months with axi-cel versus 3.5 months with tisa-cel (P = 0.0164). With bulky disease present, median PFS was 8.2 months versus 2.1 months respectively (P = 0.0023).
Biological Mechanisms Underlying Differences
The observed differences align with known biological characteristics of the two CAR T-cell products. Axi-cel utilizes a CD28 co-stimulatory domain associated with rapid proliferation and higher peak expansion, while tisa-cel employs a 4-1BB domain linked to longer persistence but potentially lower initial activity.
The study findings suggest that much of the progression-free survival difference stemmed from axi-cel's superior ability to achieve responses, particularly complete responses, rather than differences in response durability once achieved.
Clinical Implications and Context
These real-world findings provide important context for recent clinical trial results. The ZUMA-7 trial demonstrated significant event-free survival benefit for axi-cel versus standard of care in second-line DLBCL, while the BELINDA trial found no difference between tisa-cel and standard of care in the same setting.
The current analysis suggests that beyond trial design differences, inherent efficacy differences between the two CAR T-cell products may contribute to these divergent outcomes in second-line therapy.
Study Limitations and Strengths
The retrospective nature of data collection represents a key limitation, though the DESCAR-T registry maintains high quality control standards. Missing data for matching variables could introduce residual bias, but multiple sensitivity analyses using different statistical approaches yielded consistent results.
The study's strengths include its large sample size, comprehensive matching approach encompassing 14 parameters, and multiple sensitivity analyses. E-values above 2 for both progression-free and overall survival indicate that strong unmeasured confounders would be required to explain away the observed differences.
The analysis provides crucial real-world evidence to inform treatment decisions, demonstrating that while axi-cel offers superior disease control, this comes at the cost of increased toxicity. These findings may help clinicians weigh the trade-offs between efficacy and safety when selecting CAR T-cell therapy for individual patients with relapsed/refractory DLBCL.
