Real-World CAR T-Cell Therapy Outcomes Match Clinical Trial Results in Lymphoma

Real-world outcomes from second-line treatment with axicabtagene ciloleucel (axi-cel) for large B-cell lymphoma (LBCL) are "remarkably similar" to those seen in clinical trials, researchers report. In an analysis of data on more than 300 patients, more than 85% responded to the CD19-directed autologous CAR T-cell therapy and around three-quarters of patients were alive at one year and more than half had not progressed, despite over 40% requiring holding therapy.

The study, published in HemaSphere, analysed data on 345 LBCL patients (median age 62 years, 21% ≥70 years; 13% from an ethnic minority background) approved for second-line axi-cel between May 2023 and November 2024. Among the cohort, 329 (95%) underwent leukapheresis and 302 (88%) were infused with the treatment.

The overall response rate after infusion was 86% and, after a median follow-up of 16 months, median overall survival (OS) was 24 months and 12-month OS was 74%. In contrast, the median OS of patients who were not infused was just 1.5 months. Median progression-free survival (PFS) was not reached, but 12-month PFS was 52% and was similar in patients aged ≥70 years and <70 years (51% vs 53%).

Response rates, survival outcomes and toxicity were similar for patients aged above or below 70 years. The most common reason for patients not proceeding to infusion was clinical deterioration due to progressive disease (30 of 43 patients). The group who were not infused had a significantly higher proportion with bulky disease, advanced stage, extranodal involvement, elevated LDH, and progressive disease following first-line treatment.

Outcomes were influenced by response to earlier treatment, as patients with stable or progressive disease after first-line therapy had significantly worse PFS than those who achieved a partial response or who relapsed after achieving complete metabolic response (CMR). Depth of initial first-line response (CMR vs. PR) made no difference in the long term, with earlier events in PR patients, but with the curves meeting at six months. The effect remained when adjusting for measures of tumour burden such as LDH and CRP, indicating that 'response to first-line' is largely a surrogate for aggressive disease biology rather than a mere reflection of tumour burden.

Toxicity outcomes were in line with those seen in the trial, with any grade cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) recorded in 98% and 48% of patients, respectively, and grade ≥3 events occurring in 5% and 18% of patients, respectively. Early immune effector cell-associated haematotoxicity (ICAHT) grade ≥3 was seen in 29% of patients at one month. Toxicity rates were similar between age groups, as was non-relapse mortality at 12 months, which had a cumulative incidence of 7%.

The researchers demonstrated that results from the pivotal ZUMA-7 trial can be translated into standard practice, setting a new benchmark for second-line treatment of refractory LBCL patients. With a fast-moving field of novel treatments for LBCL, and availability of different T-cell engaging therapies, optimal sequencing of treatments for individual patients will require detailed knowledge of clinical outcomes in specific subgroups and implications of each treatment pathway.