Siltuximab versus tocilizumab for the management of CAR T-cell associated cytokine release syndrome

Abstract Background Chimeric antigen receptor (CAR) T-cell therapy is a promising treatment for hematological malignancies but is associated with cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS). Tocilizumab, an IL-6 receptor antibody, is the standard for CRS management but can paradoxically increase circulating IL-6. A phase 2 study (Narkhede et al, ASH 2024) demonstrated the efficacy of siltuximab, a direct IL-6 antibody, for CRS/ICANS. Here, we compare results from the siltuximab trial to a contemporaneous cohort of real-world patients treated with tocilizumab at the same institution under identical eligibility criteria. Methods We analyzed two independent cohorts. The siltuximab group included patients from a prospective phase 2 trial (NCT04975555). The tocilizumab group was a retrospective cohort of patients treated at the same institution (Nov 19-Jun 23) outside the trial in a real-world setting. Primary endpoint of the prospective trial was response rate to anti-IL-6 therapy for CRS, which was compared to the retrospective cohort utilizing Fisher's exact test. Non-response was defined as requiring >2 doses of anti-IL-6 therapy, escalation of grade of CRS after 2 doses, or persistence of any grade CRS after 14 days from initial treatment. Secondary endpoints included time to response (TTR: time from completion of first anti-IL-6 infusion to time of first recorded normal vital sign persisting for 24 hours indicating resolution of CRS), CRS relapses and incidence of ICANS after anti-IL-6 therapy. A propensity score matching (PSM) was done for disease type, age, ferritin, LDH, CRP, bridging therapy, and costimulatory domain to compare response rates. The TTR distributions were analyzed utilizing the Mann-Whitney U test and as time to event endpoint with Peto-Peto modification of the Gehan-Wilcoxon test to account for heavy early distribution of events. Results The siltuximab (n=20) and tocilizumab (n=55) groups were compared. Baseline characteristics (siltuximab vs. tocilizumab, respectively) included median age [59 (38-79) vs. 64 (35-80) years]; disease subtypes [LBCL (15 vs 28), FL (2 vs 4), MCL (1 vs. 6), MM (2 vs 13)]; and co-stim domain [4-1BB (5 vs 28), CD28 (15 vs 27)]. Baseline mean (±SD) biomarkers were: CRP (21.20 ± 33.37 vs. 33.48 ± 69.75), LDH (238.20 ± 168.40 vs. 250.96 ± 266.69), and ferritin (379.75±393.38 vs. 552.22±489.60). For the primary endpoint analyzed for patients who developed CRS and received anti-IL-6 therapy, 15/19 pts (79%) in siltuximab group responded compared to 37/51 pts (73%) (p=0.79). One pt in siltuximab group did not develop CRS and 4 pts in tocilizumab group had CRS resolution prior to tocilizumab infusion. Median TTR was substantially shorter with siltuximab (1 hour, 12 minutes [Range 00:01–104:06) compared tocilizumab (6 hours, 14 minutes [Range 00:32–101:52) with p-value 0.0204. CRS relapse rates were similar between groups (tocilizumab 12/55, 22%, siltuximab 4/19, 21%). The incidence of grade 3/4 ICANS after anti-IL-6 therapy was also similar with tocilizumab (13/51, 25%) or siltuximab (4/20, 20%). PSM matched all variables other than ferritin and costimulatory domain and included 15 patients in each cohort. A higher proportion of patients treated with siltuximab (12/15, 80%) responded to treatment compared to those treated with tocilizumab (8/15, 53%), however the difference was not statistically significant (p=0.13, 95%CI 0.049, 1.361). Conclusion Siltuximab was associated with a higher rate of CRS resolution compared to tocilizumab, although this difference did not reach statistical significance, potentially due to limited sample size. However, siltuximab had nearly five times faster time to CRS resolution compared to tocilizumab with no notable differences observed in CRS relapse rates or the incidence of higher grade ICANS rates when anti-IL-6 therapy was administered after onset of CRS. Our limitations include our small prospective sample size and the use of a retrospective comparator group, with one cohort treated under a clinical trial protocol and the other reflecting real-world practice. These findings, along with promising results from a prospective trial, suggest siltuximab is a good option for CRS management. By directly targeting IL-6, siltuximab may accelerate recovery, and these preliminary data support the need for a prospective, randomized trial to evaluate its superiority for CRS response and time to resolution.