Résumé
Introduction Autologous hematopoietic stem cell transplantation (auto-HSCT) and chimeric antigen receptor (CAR) T-cell therapy are both effective options for patients with large B-cell lymphoma (LBCL) who achieve complete remission (CR) after salvage therapy. However, the optimal strategy for this specific population remains unclear. A recent study from the CIBMTR (Shadman et al., Blood Cancer J. 2024) reported longer progression-free survival (PFS) and lower relapse incidence with auto-HSCT compared to CAR T-cell therapy, warranting further investigation.
Methods We performed a retrospective, registry-based, multicenter analysis using data from the European Society for Blood and Marrow Transplantation (EBMT). Adult patients with LBCL who underwent CAR T-cell therapy between 2019 and 2023 or auto-HSCT between 2019 and 2022, while in CR, were included. Patients who underwent auto-HSCT after a single prior treatment line were excluded.
The primary endpoint was PFS from the time of infusion. Secondary endpoints included overall survival (OS), relapse incidence (RI), and non-relapse mortality (NRM). To reduce baseline imbalances between groups, we applied a propensity score (PS) matching approach. Variables included in the PS were selected through univariate and multivariate regressions. Final selection included age, sex, presence of secondary or transformed disease, number of prior treatment lines, time from diagnosis to infusion, and year of treatment.
Results Among the 2,734 patients included in the analysis, 349 received CAR T-cell therapy (220 axicabtagene ciloleucel [axi-cel], 125 tisagenlecleucel, and 4 lisocabtagene maraleucel), and 2,385 underwent auto-HSCT. Males comprised 59% of the cohort. At infusion, 29% of patients had a Karnofsky performance status below 90%, 15% had a hematopoietic cell transplantation comorbidity index (HCT-CI) of ≥3, and 82% had an International Prognostic Index (IPI) of ≥2. The median time from diagnosis to infusion was 19 months (IQR, 11–38), with no significant differences between the two groups for these characteristics.
However, patients treated with CAR T-cell therapy were older, with a median age of 64 years (range, 19–82), compared to 59 years (range, 19–79) in the auto-HSCT group (p<0.001). CAR T-cell patients had more prior lines, with 11% having received one prior line, 44% two lines, and 63% at least three lines, compared to 78% of auto-HSCT patients who had received exactly two lines and 22% at least three lines (p<0.001). CAR T-cell recipients more frequently had transformed disease (23% vs. 15%; p<0.001) and were treated more recently (median year 2022 vs. 2020; p<0.001).
After a median follow-up of 2.0 years (95% confidence interval [CI], 1.5–2.1) for CAR T-cell patients and 2.2 years (95% CI, 2.1–2.3) for auto-HSCT patients, univariate analysis showed that CAR T-cell recipients had inferior 2-year PFS (58.3% vs. 66.3%, p=0.002) and OS (70.6% vs. 80.0%, p<0.001) compared to those who underwent auto-HSCT. Additionally, 2-year RI was higher in the CAR T-cell group (33.5% vs. 27.9%, p=0.018), as was 2-year NRM (8.2% vs. 5.8%, p=0.017).
In the CAR T group, 2-year PFS was 55.5% after one prior treatment line, 61.4% after two, and 57.9% after ≥3, while in the auto-HSCT group it was 67.4% after two prior lines and 61.7% after ≥3. In a subgroup analysis restricted to axi-cel recipients, 2-year PFS, RI, and NRM were not significantly different from those observed after auto-HSCT. However, 2-year OS remained lower in the axi-cel group (72.0% vs. 80.0%, p=0.004).
Finally, in the multivariate Cox model applied after PS matching in the overall cohort, no significant differences were observed between CAR T-cell therapy and auto-HSCT for any of the evaluated outcomes, including PFS, OS, RI, and NRM.
Conclusions Auto-HSCT continues to represent a valid therapeutic option for selected, fit, transplant-eligible LBCL patients who achieve complete remission after salvage therapy. Both auto-HSCT and anti-CD19 CAR T-cell therapy are associated with favorable outcomes in this setting. While expanding access to CAR T-cell therapy should remain a priority, considering auto-HSCT in this specific subset of patients remains a reasonable approach.