复发性/难治性弥漫大 B 细胞淋巴瘤 CART-19 治疗中淋巴细胞清除方案的比较分析

Patients with relapsed/refractory (r/r) diffuse large B-cell lymphoma (DLBCL) undergoing chimeric antigen receptor T cell therapy targeting CD19 (CART-19) receive preinfusion lymphodepletion (LD), most commonly with fludarabine/cyclophosphamide (flu/cy). During the fludarabine shortage, bendamustine was used as an alternative. There are only limited data on how outcomes with flu/cy and bendamustine LD compare.

We aimed to compare outcomes and toxicity profiles in r/r DLBCL patients lymphodepleted with flu/cy or bendamustine prior to CART-19 therapy.

We conducted a retrospective single-center study comparing DLBCL patients who received axicabtagene ciloleucel (axi-cel; 74. 4%) or lisocabtagene maraleucel (liso-cel; 25. 6%) following flu/cy or bendamustine LD. A propensity score-matched data set was created to balance baseline covariates between the LD groups (45 patients per LD group). Covariates addressed included age, stage, bridging therapy, number of prior therapies, CART-19 product, and pre-LD C-reactive protein (CRP) and lactate dehydrogenase (LDH).

Outcomes, including response rates, overall survival (OS), progression-free survival (PFS), cytokine release syndrome (CRS), immune effector cell neurotoxicity syndrome (ICANS), and hematologic toxicities, were evaluated. The 3-month overall and complete response rates (ORR; CR) were similar in the flu/cy (64. 4% and 60%) and bendamustine groups (64.

4% and 51. 1%; ORR: P = 1; CR: P = . 40). At 6 months, no differences in PFS (flu/cy 62. 2% vs bendamustine 58. 2%; P = . 37) or OS (flu/cy 86. 7% vs bendamustine 79. 9%; P = . 83) were noted. Incidences of CRS and ICANS were comparable in the flu/cy (CRS, 79. 1%; ICANS, 42. 2%) and bendamustine cohorts (CRS, 71. 1%; ICANS 28. 9%; all P > . 05).

However, the median time to peak CRS (3 vs 5 days; P = . 002) and ICANS (6 vs 9 days; P = . 034) occurred earlier in the flu/cy cohort. The flu/cy cohort also had lower median absolute neutrophil, platelet and hemoglobin nadirs (all P ≤ . 001).

Moreover, flu/cy was associated with higher rates of severe (G ≥ 3) neutropenia (100% vs 73%; P < . 001), anemia (54. 5% vs 24. 4%; P = . 012), and thrombocytopenia (57. 5% vs 29. 7%; P = . 019). Within the limitations of the size of our study and its retrospective nature, flu/cy and bendamustine LD resulted in similar ORR, PFS and OS; however, bendamustine had less hematologic toxicity. Bendamustine can be a viable alternative LD agent for CART-19 therapy for DLBCL, especially in patients with a compromised performance status.