Confirmed real-life efficacy of CAR T-cell therapy in a French DLBCL registry

DESCAR-T is the French national registry for patients with B-cell acute lymphoblastic leukaemia (ALL) or diffuse large B-cell lymphomas (DLBCL), treated with commercial CAR-T cells. With about 50 new patients registered every month, DESCAR-T demonstrates that CAR-T cell therapy has become a key treatment option for relapsed/refractory DLBCL. First analysis of this registry seem to confirm CAR-T cell efficacy in real-life and are in line with previously published data.

Several clinical trials demonstrated that CAR T-cells can provide long-term disease control in relapsed/refractory (R/R) patients with B-cell acute lymphoblastic leukaemia (ALL) or diffuse large B-cell lymphomas (DLBCL). Tisagenlecleucel (tisa-cel) and axicabtagene ciloleucel (axi-cel) were the first manufactured CAR T-cell therapies commercially available. However, marketing authorisations were based on phase II trial results and stakeholders are in need of real-world data. Therefore, the DESCAR-T registry (Dispositive d’Evaluation et de Suivi des CAR-T), created in 2019, aims to investigate the real-life efficacy of CAR T-cell therapy in DLBCL patients.

DESCAR-T

By April 2021, DESCAR-T had enrolled 647 patients with DLBCL from 19 sites across France. Among these patients, CAR-T cells were ordered for 607 patients and 550 patients were subsequently infused. Main reasons for not administering CAR-T cells were disease progression (N= 27) and death before administration (N= 14). Of the treated patients, 200 patients received tisa-cel and 350 patients received axi-cel. The median age of patients in the cohort was 63.0 years with 44% of patients being older than 65 years. Lymphoma subtypes at the time of registration were DLBCL (90%), PMBL (4%), and high-grade B-cell lymphoma (1.7%). Most patients were male (331/550) and the aaIPI score at registration was usually low (45.7% had a score of 0-1 and 48.3% had a score of 2). The median number of prior treatment lines was 3 with 24.5% of patients being previously transplanted. The median time from CAR-T order to the time of infusion was 50 days. Importantly, more than half of the patients (54%) had progressive disease at the time of infusion.

Within ten days post CAR-T infusion, 418 patients (81.2%) suffered cytokine release syndrome (CRS), 184 patients (35.7%) experienced neurotoxicity and 163 patients (31.7%) had grade ≥3 opportunistic or medically significant infections. Overall, 32.6% of patients who experienced toxicity were admitted to the intensive care unit, 65.1% received tocilizumab and 41.2% received corticosteroids.

After a median follow-up of 6.5 months (calculated from CAR-T infusion), the ORR was 60.0%. Best ORR was 74.2%, consisting of 53.0% complete and 21.2% partial responses. With a median follow-up of 8.1 months (calculated from time of CAR-T order), the overall survival (OS) rate at six months in the treated patient population was 83.7%, as compared to 5.5% in patients who were planned for CAR T-cell therapy but did not receive the treatment in the end. The progression-free survival (PFS) rate at six months for all treated patients (N= 550) was 44.5%. For patients who had a response (N= 356), 57.7% were still in response after six months. When assessing PFS and OS in function of the disease status at the time of CAR-T infusion, it should be noted that outcomes for patients with a controlled disease (including complete response, partial response and stable disease) and for those who do not need bridging therapy was superior to what was seen in patients with disease progression at the time of infusion. Nevertheless, 29.8% of patients with progressive disease at the time of infusion were still progression-free at six months with a six month OS rate of 65.5%. These data suggest that disease status at the time of infusion is not the only parameter that can predict disease outcomes. PFS and OS results according to disease status were comparable for tisa-cel and axi-cel.

Conclusion

CAR T-cell therapy has rapidly become standard of care for R/R DLBCL, illustrated by the fact that already 640 patients were treated with this treatment modality in less than two years in France. The DESCAR-T real-life experience demonstrates a response duration, and 6 months PFS or OS rates that mirror the results obtained in clinical trials. No new emerging toxicity signals emerged in real-life. The registry also shows that while patients without at least stable disease at time of CAR T infusion are at risk of an early relapse, still around 30% of these patients was able to obtain long-term disease control. Longer follow-up is needed to confirm the plateau in PFS and OS curves. In addition, also the impact of bridging therapy, disease status and patient’s characteristics needs to be investigated in more detail in the years to come.

Reference

Le Gouill S, et al. First results of DLBCL patients treated with CAR-T cells and enrolled in DESCAR-T registry, a French real-life database for CAR-T cells in hematologic malignancies. Presented at EHA 2021; abstract S216.