Steroid Prophylaxis in Patients Receiving CAR T-Cell Therapy for Relapsed and Refractory Non-Hodgkin Lymphoma

Introduction: Anti-CD19 CAR T-cell therapy (CART) is now the standard of care for relapsed/refractory (R/R) non-Hodgkin lymphoma (NHL). CART associated toxicities including cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) remain prevalent. The use of prophylactic (ppx) corticosteroids to mitigate CRS has increased based on improvement in the rates of severe CRS seen in cohort 6 of ZUMA-1. However, this study lacked a control arm and real-world data on use of ppx steroids is limited. There remains concern that steroids may impact the efficacy of CART given their potential to hinder CART expansion. We evaluated the use of ppx corticosteroids for patients with R/R NHL receiving axicabtagene-ciloleucel (axi-cel) or brexucabtagene-autoleucel (brexu-cel) and their outcomes.

Methods: We conducted a single-institution retrospective analysis of adults ≥18 years old who received axi-cel or brexu-cel, with and without steroid ppx, for treatment of NHL at The Ohio State University between January 2018 to April 2023. Steroid ppx was defined as dexamethasone 10 mg daily for three days starting on the day of infusion. Log-rank test was applied to compare the Kaplan-Meier survival probabilities between the two groups. Univariable and multivariable analyses were tested by Cox model for overall survival (OS)/progression-free survival (PFS) hazard ratios (HR), and by logistic regressions for the odds ratios (OR) of objective/complete response rates (ORR, CRR).

Results: A total of 137 patients were treated with axi-cel (n=118) or brexu-cel (n=19). Disease subtypes included diffuse large B-cell lymphoma/high-grade B-cell lymphoma (81%), mantle cell lymphoma (13.9%), and follicular lymphoma (5.1%), with 48.9% of patients having primary refractory disease. Twenty-seven percent of patients received steroid ppx. There were no significant differences in baseline disease and patient characteristics including age, performance status (ECOG), disease stage, ferritin, CRP and LDH for those who did or did not receive steroid ppx. The incidence of any grade CRS was 89.2% in ppx patients and 93% in non-ppx patients (p=0.488). Any grade of ICANs incidence was 48.7% with ppx compared to 61% without ppx (p=0.194). The incidence of grade 3-4 CRS was 2.7% vs 12% (p=0.185), and the incidence of grade 3-4 ICANS was 24.3% vs 32% (p=0.384) in the ppx and non-ppx groups, respectively. In the ppx group, the median onset of CRS was 4 days (range: 0-11) post-infusion vs 3 days (0-9) in the non-ppx group (p=0.005). There was no difference in time to onset of ICANS with a median of 5 days (2-12) in the ppx group vs 7 days (0-12) in the non-ppx group (p=0.831). Median time to initiation of steroids (5 days, p=0.080) and tocilizumab (4 days, p=0.656) was the same for both groups. Survival outcomes in ppx vs non-ppx cohorts including OS [NR (8.54-NR) vs NR (38.14-NR), p=0.402] and PFS (median: 13.21 months vs 19.84 months, p=0.898) were not significantly different.

On univariate analyses, the use of steroid ppx had no effect on OS, PFS, ORR, or CRR. Of note, grade 3-4 CRS and grade 3-4 ICANS negatively affected OS (HR 2.59, p=0.009; HR 2.53, p=0.0006), PFS (HR 2.58, p=0.0038; HR 2.09, p=0.0015), ORR (OR 0.15, p=0.0027; OR 0.32, p=0.0047) and CRR (OR 0.20, p=0.0195; OR 0.43, p=0.0246). After adjusting for sex, product type, and ppx use on multivariate analysis, the effect of severe CRS and ICANS on PFS and OS remained unchanged.

Of 13 patients who developed grade 3-4 CRS, 0/1 patients died in the ppx group and 8/12 patients in the non-ppx group. Causes of death included disease progression (n=3), CRS/ICANS (n=3), and infections (n=2). Of 41 patients who developed grade 3-4 ICANS, 66.7% of patients died in the ppx group (disease, n=4; CRS/ICANS, n=2) and 59.4% patients in the non-ppx group (disease progression, n=11; infection, n=4; CRS/ICANS, n=4).

Conclusion: Data from this single-institution experience suggest the use of prophylactic steroids accompanying CART during treatment of R/R NHL may delay onset of CRS without adversely affecting overall clinical outcomes. Larger cohort studies and/or randomized control trials will be required to confirm these findings.