Publication Full Title: Naïve CD4+ T-cells and disease status at CAR T infusion correlate with clinical outcomes in real-world large B-cell lymphoma patients receiving second-line CAR T therapy.
Publication Link: https://doi.org/10.1038/s41467-026-71710-7
Comment by: Stefan Corradini, Department of Haematology, Comprehensive Cancer Centre, King’s College London, London, United Kingdom
Summary
Schneider et al. retrospectively analysed 64 consecutive patients with primary refractory or early-relapse (R/R) large B-cell lymphoma (l-BCL) treated with second-line axi-cel (n=35) or liso-cel (n=29). At 25 months median follow-up, the overall and complete response rates were 66% and 55%, while 24-month progression-free survival, overall survival and duration of response were 46%, 76% and 68%, respectively. Outcomes were similar between products despite older age, higher IPI scores, more frequent systemic bridging and longer vein-to-vein times among liso-cel recipients. Disease control before infusion, achieved in 73% of enrolled patients, independently predicted day-90 complete response and corresponded to markedly superior 24-month progression-free and overall survival. Advanced stage, extranodal involvement, bulky disease, elevated LDH and higher metabolic tumour volume at infusion also reduced durable response. Severe CRS and ICANS each occurred in 3% of enrolled patients.
At apheresis, higher CD3+ and CD4+ T-cell counts were associated with response, with CD4+ counts remaining independently predictive after adjustment. This effect was strongest in axi-cel recipients, where higher naïve CD4+ counts correlated with prolonged progression-free survival. At day 7, favourable outcomes were associated with greater CAR+ expansion, enrichment of naïve CD8+ CAR+ cells and lower PD-1 expression. Overall, the study links pre-infusion disease control and T-cell fitness with durable second-line clinical CAR T benefit for l-BCL patients.
Scientific and Clinical Commentary
Following ZUMA-7 and TRANSFORM, second-line CAR T has been established for R/R l-BCL. The unresolved question is why durable benefit remains uneven despite treatment earlier in the disease course. Here, the gap is addressed by examining clinical and immune determinants across the treatment pathway, linking T-cell composition at apheresis, disease status immediately before- and early post-infusion expansion. Highlighting how tumour biology and cellular fitness may converge to shape response.
In this context, the immunophenotypic analyses offer a plausible account of where variation in response may arise. Higher naïve CD4+ T-cell counts at apheresis were associated with improved outcome, specifically within the axi-cel cohort, suggesting that the fitness of the starting cellular compartment may influence subsequent CAR T expansion. Patients with controlled disease had markedly better survival than those with progressive lymphoma, although the heterogeneous use of bridging therapy means that disease control should be interpreted as prognostic rather than evidence for a particular bridging approach. The day-7 analyses extend this biological interpretation, with favourable outcomes associated with greater CAR+ expansion, enrichment of naïve CD8+ cells and lower PD-1 expression. Strengths in the study include the use of consecutive real-world enrolment, 25 months of follow-up, serial sampling and the integration of PET burden with spectral flow cytometry. Nevertheless, product assignment was physician-selected, 89% of patients received bendamustine lymphodepletion and, the axi-cel and liso-cel groups differed substantially at baseline. Detailed immune profiling was restricted to a small axi-cel subgroup, limiting inference across products. Internally derived ROC thresholds, multiple nominal tests and limited phenotypic resolution between naïve and related stem-like populations further increase the risk of over-interpretation.
Ultimately, this study illustrates CAR T therapy not as a single therapeutic event, but as a continuum in which disease control, starting T-cell fitness and early expansion shape outcome. Its value lies in identifying where intervention may improve benefit, rather than whom treatment should exclude.