Long-term outcomes of dexamethasone prophylaxis after axicabtagene ciloleucel for non-Hodgkin lymphoma: A single-center retrospective study

Abstract

Abstract Background: A small cohort of patients receiving prophylactic dexamethasone (dex) after axicabtagene ciloleucel (axi-cel) within ZUMA-1 demonstrated lower grade 3-4 cytokine release syndrome (CRS) with similar rates of short-term efficacy and CAR-T expansion; however, widespread adoption of this practice has been mixed, in part due to concerns over the long-term effects of early steroid exposure on CAR-T cell expansion and function (Oluwole et al. Br J Haematol. 2021). We previously reported significantly reduced grade 3-4 ICANS in patients receiving dex prophylaxis, without compromise in short-term efficacy (Stack et al. EHA 2024 Abstract #P2180). Here we report longer follow-up of this cohort with a focus on long-term efficacy. Methods: We performed a single-center, retrospective study to compare safety and efficacy outcomes with and without dex prophylaxis (10 mg orally on day 0, 1 and 2) following axi-cel for non-Hodgkin lymphoma from 10/1/2018 to 11/20/2023. Statistical analysis utilized the Mann–Whitney test and Fisher's exact test for continuous and categorical variables, respectively. Survival analysis used the Kaplan–Meier method. Results: In the final analysis, 28 patients received dex prophylaxis and 16 patients received no prophylaxis. The Dex vs No-Dex groups were similar in terms of median age (63 vs 62 years), proportion of indolent lymphomas (29% vs 25%) and utilization of bridging therapy (both 75%), respectively. Most patients in the Dex group received bendamustine lymphodepletion (27/28), while the No-Dex group primarily received fludarabine + cyclophosphamide (15/16). Median follow-up for the current analysis was 21.4 months and 32.5 months for the Dex and No-Dex groups, respectively. With prolonged follow-up, overall response rate (ORR) was 89% vs 75% (p=0.21) and complete response rate (CR) was 79% vs 69% (p=0.47) in the Dex vs No-Dex groups, respectively. 2-year progression-free survival (PFS) was 49.9% vs 60.6% (p=0.28) and 2-year overall survival (OS) was 72.2% vs 67.7% (p=0.67) for the Dex vs No-Dex groups, respectively. We further investigated the effect of dexamethasone prophylaxis on absolute lymphocyte count (ALC) following CAR-T. Peak ALC was significantly higher in the Dex group at 750 cells/µL vs 500 cells/µL in the No-Dex group (p=0.01). Additionally, median time to peak ALC was shorter in the Dex group at 9.5 days vs 17 days in the No-Dex group (p=0.08). Incidence of CRS and ICANS is unchanged from the prior report. CRS occurred in 71% vs 69% (p=1.00) in the Dex vs No-Dex groups, respectively, with no grade 3-4 events. There was a trend toward delayed onset of CRS in the Dex group (median 4 days [range 1–10]) vs 2 days [range 1–16] in the No-Dex group (p=0.07). All-grade ICANS in the Dex vs No-Dex groups did not statistically differ at 21% vs 38%, respectively (p=0.30); however, there was a significantly lower rate of grade 3-4 ICANS in the Dex group at 0% vs 25%, respectively (p=0.01). Conclusion: In this retrospective analysis of patients receiving axi-cel with longer follow-up, dex prophylaxis was associated with a reduced rate of grade 3-4 ICANS and similar ORR, CR, 2-year PFS and 2-year OS compared to no dex prophylaxis, suggesting that dexamethasone prophylaxis after axi-cel may be used to mitigate inflammatory toxicity without compromising long-term efficacy. Dex prophylaxis was also associated with increased and earlier peak lymphocyte expansion. Interpretation is limited by a relatively small sample size and differences in lymphodepletion regimens between the Dex and No-Dex groups.