Patients receiving commercially available CAR T-cell therapy for relapsed/refractory follicular lymphoma in routine practice respond similarly to those in clinical trials despite less selective eligibility criteria, a real-world analysis shows.

But a finding of shorter progression-free survival in the real-world setting warrants further attention, with the US researchers postulating that the choice of conditioning chemotherapy could be to blame for the inferior result.

Their study – the first to offer real-world data from multiple centres – provided “important context to inform patient counselling and clinical decision-making”, given the differences between real-world and clinical trial populations, they wrote in Blood Advances [link here].

The study involved 136 patients with R/R FL undergoing axicabtagene ciloleucel (axi-cel) or tisagenlecleucel (tisa-cel) at 13 cancer centres. Patients treated with tisa-cel (26% of the cohort) were older than those receiving axi-cel (median age 68 vs 60) and more likely to be female (47% vs 24%).

Overall (ORR) and complete (CRR) response rates for the overall cohort were 92% and 84%, respectively, after a median follow-up of 14.4 months.

Compared with tisa-cel, axi-cel was associated with:

• A higher ORR (96% vs 80%; P = .007)
• A higher CRR (88% vs 71%; P = .024)
• Longer median PFS (30.5 months vs 11.9 months; P = .021)

Median progression-free survival (PFS) was longer at 30.5 months for axi-cel compared with 11.9 months for tisa-cel (P = .021). The 2-year PFS rate was 59% and 36% for axi-cel and tisa-cel, respectively.

Median overall survival (OS) was not reached for the cohort, and was not statistically different between CAR-T products (not reached vs 23.6 months, P = .061). The 2-year OS rate was 83% and 48% for axi-cel and tisa-cel, respectively.

Importantly, efficacy outcomes remained largely similar after inverse probability of treatment weighting to mitigate the effects of potential confounders, although most were no longer significantly different between the therapies.

Regarding real-world toxicity, cytokine release syndrome rates were comparable (75% vs 75%; P = .99), whereas ICANS occurred more frequently with axi-cel than with tisa-cel (42% vs 17%; P = .008).

An accompanying editorial [link here] by French haematologists suggested the “broadly consistent” response rates confirmed that “high-quality CAR T-cell therapy outcomes are achievable beyond highly selected trial populations”.

However, they also questioned the “unexpectedly short” median PFS observed with tisa-cel (12 months) compared with the 53 months recorded in the pivotal trial.

“The high ORR suggests adequate initial tumour reduction; therefore, the shortened PFS implies early relapse after response in a substantial fraction of patients,” they said. “The frequent use of bendamustine as lymphodepletion in the tisa-cel group emerges as a plausible contributor (36% vs 5% in the ELARA trial) may deserve careful consideration.”

“Long-term follow-up remains essential in this indolent lymphoma, where the true measure of therapeutic impact lies not only in initial response depth but in sustained disease control.”