Contributions
Abstract: EP495
Type: E-Poster Presentation
Session title: Aggressive Non-Hodgkin lymphoma - Clinical
Background
In Cohorts 1+2 (C1+2; N=101) of ZUMA-1, the pivotal Phase 1/2 study of axicabtagene ciloleucel (axi-cel) in patients with refractory large B-cell lymphoma, rates of Grade ≥3 cytokine release syndrome (CRS) and neurologic events (NEs) were 13% and 28% at the 6-month primary analysis (N Engl J Med. 2017;377:2531), and the objective response rate (ORR) was 82% (54% complete response [CR] rate). With a median follow-up of 51.1 months, median overall survival (OS) was 25.8 months and the 4‑year OS rate was 44% (Blood. 2020;136[suppl, abstr]:40-42). In safety management Cohort 4 (C4; N=41), with its earlier corticosteroid use than C1+2, rates of Grade ≥3 CRS and NEs were 2% and 17%, and the ORR was 73% (51% CR; Blood. 2019;134[suppl, abstr]:243). In Cohort 6 (C6), the effect of prophylactic corticosteroid use on CRS and NEs was evaluated.
Aims
To present the primary analysis of C6, alongside those of C4 and C1+2 for context.
Methods
Eligible patients were leukapheresed, could receive optional bridging chemotherapy (in C4 and C6 but not C1+2), and received conditioning chemotherapy before axi-cel (target dose, 2×106 chimeric antigen receptor [CAR] T cells/kg). Patients in C6 received dexamethasone 10 mg orally on Days 0 (before axi‑cel), 1, and 2 and earlier intervention with corticosteroids and tocilizumab. Primary endpoints were incidence and severity of CRS and NEs.
Results
As of 6/16/20, 40 C6 patients had received axi-cel, with median follow-up of 8.9 months. The median age was 64 years (50%, ≥65 years); 58% of patients were male; 55% had Eastern Cooperative Oncology Group performance status of 1; 65% had disease stage III/IV, and 53% received bridging. C6 patients had median tumor burden (by sum of product diameters post-bridging, if given, but before conditioning) lower than C1+2 and C4 (C6, 1184 mm2; C1+2, 3723 mm2; C4, 2100 mm2). Median baseline lactate dehydrogenase level was similar in C6 and C4 but lower than C1+2 (C6, 236 U/L; C4, 262 U/L; C1+2, 356 U/L).
Among patients who received corticosteroids, median cumulative corticosteroid use (cortisone-equivalent dose) in C6 (1252 mg [n=40]) was greater than that in C4 (939 mg [n=30]) but less than that in C1+2 (6388 mg [n=24]). In C6, there was no Grade ≥3 CRS; 13% of patients had Grade ≥3 NEs, and there were no Grade 5 CRS or NEs. Onset of CRS appeared to be delayed in C6 (median, 5 versus 2 days in C4 and C1+2), and 68% of patients had no NEs or CRS within 72 hours of axi-cel. ORR was 95% (80% CR), and 63% of patients had ongoing responses at the data cutoff.
Median peak CAR T-cell levels trended higher in C6 versus C1+2 and C4 (C6, 64 cells/µL; C1+2, 42 cells/µL; C4, 53 cells/µL), which was corroborated by median CAR area under the curve in blood (C6, 526 cells/µL × day; C1+2, 462 cells/µL × day; C4, 511 cells/µL × day). Median peak serum levels of biomarkers associated with CAR T-cell treatment-related adverse events (eg, C-reactive protein, interferon-γ, granulocyte-macrophage colony-stimulating factor, interleukin-2) were lower in C6 versus C4 and C1+2.
Conclusion
Prophylactic corticosteroid use appears to reduce the rate of severe CRS and NEs and delay onset of CRS, without affecting median peak CAR T-cell levels or responses. Differences in cohort sizes and baseline characteristics should be considered when comparing cohorts. Longer follow-up is required to determine the impact on duration of response and survival.
Keyword(s): CAR-T, Lymphoma, Prophylaxis, Toxicity
Abstract: EP495
Type: E-Poster Presentation
Session title: Aggressive Non-Hodgkin lymphoma - Clinical
Background
In Cohorts 1+2 (C1+2; N=101) of ZUMA-1, the pivotal Phase 1/2 study of axicabtagene ciloleucel (axi-cel) in patients with refractory large B-cell lymphoma, rates of Grade ≥3 cytokine release syndrome (CRS) and neurologic events (NEs) were 13% and 28% at the 6-month primary analysis (N Engl J Med. 2017;377:2531), and the objective response rate (ORR) was 82% (54% complete response [CR] rate). With a median follow-up of 51.1 months, median overall survival (OS) was 25.8 months and the 4‑year OS rate was 44% (Blood. 2020;136[suppl, abstr]:40-42). In safety management Cohort 4 (C4; N=41), with its earlier corticosteroid use than C1+2, rates of Grade ≥3 CRS and NEs were 2% and 17%, and the ORR was 73% (51% CR; Blood. 2019;134[suppl, abstr]:243). In Cohort 6 (C6), the effect of prophylactic corticosteroid use on CRS and NEs was evaluated.
Aims
To present the primary analysis of C6, alongside those of C4 and C1+2 for context.
Methods
Eligible patients were leukapheresed, could receive optional bridging chemotherapy (in C4 and C6 but not C1+2), and received conditioning chemotherapy before axi-cel (target dose, 2×106 chimeric antigen receptor [CAR] T cells/kg). Patients in C6 received dexamethasone 10 mg orally on Days 0 (before axi‑cel), 1, and 2 and earlier intervention with corticosteroids and tocilizumab. Primary endpoints were incidence and severity of CRS and NEs.
Results
As of 6/16/20, 40 C6 patients had received axi-cel, with median follow-up of 8.9 months. The median age was 64 years (50%, ≥65 years); 58% of patients were male; 55% had Eastern Cooperative Oncology Group performance status of 1; 65% had disease stage III/IV, and 53% received bridging. C6 patients had median tumor burden (by sum of product diameters post-bridging, if given, but before conditioning) lower than C1+2 and C4 (C6, 1184 mm2; C1+2, 3723 mm2; C4, 2100 mm2). Median baseline lactate dehydrogenase level was similar in C6 and C4 but lower than C1+2 (C6, 236 U/L; C4, 262 U/L; C1+2, 356 U/L).
Among patients who received corticosteroids, median cumulative corticosteroid use (cortisone-equivalent dose) in C6 (1252 mg [n=40]) was greater than that in C4 (939 mg [n=30]) but less than that in C1+2 (6388 mg [n=24]). In C6, there was no Grade ≥3 CRS; 13% of patients had Grade ≥3 NEs, and there were no Grade 5 CRS or NEs. Onset of CRS appeared to be delayed in C6 (median, 5 versus 2 days in C4 and C1+2), and 68% of patients had no NEs or CRS within 72 hours of axi-cel. ORR was 95% (80% CR), and 63% of patients had ongoing responses at the data cutoff.
Median peak CAR T-cell levels trended higher in C6 versus C1+2 and C4 (C6, 64 cells/µL; C1+2, 42 cells/µL; C4, 53 cells/µL), which was corroborated by median CAR area under the curve in blood (C6, 526 cells/µL × day; C1+2, 462 cells/µL × day; C4, 511 cells/µL × day). Median peak serum levels of biomarkers associated with CAR T-cell treatment-related adverse events (eg, C-reactive protein, interferon-γ, granulocyte-macrophage colony-stimulating factor, interleukin-2) were lower in C6 versus C4 and C1+2.
Conclusion
Prophylactic corticosteroid use appears to reduce the rate of severe CRS and NEs and delay onset of CRS, without affecting median peak CAR T-cell levels or responses. Differences in cohort sizes and baseline characteristics should be considered when comparing cohorts. Longer follow-up is required to determine the impact on duration of response and survival.
Keyword(s): CAR-T, Lymphoma, Prophylaxis, Toxicity