A Phase 1 study of KITE-753 or KITE-363 in patients with relapsed/refractory B-cell lymphoma: Initial safety and preliminary efficacy of KITE-753 and updated results of KITE-363

Abstract Introduction: KITE-363 and KITE-753 are dual-targeting, autologous chimeric antigen receptor (CAR) T-cell therapies designed to overcome heterogeneity in CD19/CD20 tumor antigen expression and potentially prevent relapses due to antigen loss. They use the same bicistronic construct comprised of an anti-CD19 CAR with a CD28 co-stimulatory domain and an anti-CD20 CAR with a 4-1BB co-stimulatory domain. KITE-753 uses a rapid manufacturing process that reduces turnaround time and preserves a greater proportion of juvenile (naive and early memory) T cells in the product than KITE-363, which was associated preclinically with antitumor activity at a >25-fold lower dose (Murakami JL, et al. ASH 2024. Abstract 3481). Here we report the results from an open-label, multicenter, umbrella Phase 1 study of KITE-753 or KITE-363 in patients with relapsed/refractory (R/R) B-cell lymphoma. Initial safety and preliminary efficacy of KITE-753 and updated results of KITE-363 are described. Methods: Eligible adults with B-cell lymphoma R/R after ≥2 lines of therapy (patients with large B-cell lymphoma [LBCL] could have second-line primary refractory disease) were enrolled in dose escalation (1A) and expansion (1B; LBCL only) cohorts. Following leukapheresis and lymphodepletion, patients received dose level (DL) 1, 2, or 3 of KITE-753 (3.0×104, 1.0×105, or 2.0×105 CAR T cells/kg, respectively) or KITE-363 (0.5×106, 1×106, or 2×106 CAR T cells/kg, respectively). Primary endpoints were incidence of dose-limiting toxicities (DLTs; Phase 1A) and investigator-assessed objective response rate (ORR; per Lugano; Phase 1B). Results: As of 03/18/25, 59 patients were enrolled; 14 received KITE-753 (3 in DL1, 8 in DL2, 3 in DL3), and 37 received KITE-363 (3 in DL1, 8 in DL2, 26 in DL3). Median age of patients treated with KITE-753 at the time of analysis was 58.5 years (range, 25-84), 71% had LBCL and 29% had indolent non-Hodgkin lymphoma, 86% had Stage III/IV disease, 14% had bulky disease, 14% were primary refractory (20% in LBCL), and 43% had prior CAR T-cell exposure. Baseline characteristics of patients treated with KITE-363 were previously reported. No DLTs occurred with KITE-753 in dose escalation. Grade ≥3 adverse events (AEs) occurred in 79% of patients (100% in DL3), primarily cytopenias. Serious AEs occurred in 36% of patients (DL3, 0%). Grade 1/2 cytokine release syndrome (CRS) occurred in 3 patients. One Grade 3 CRS (LBCL; DL2) and no Grade 4/5 CRS occurred. Median onset of CRS was 9.5 days with median duration of 6.5 days. Two Grade 1/2 immune effector cell-associated neurotoxicity syndrome (ICANS; DL2) and no Grade ≥3 ICANS occurred. Median onset of ICANS was 12.5 days with median duration of 6.5 days. Three patients died (DL2; 2 due to infections determined by the investigator to be unrelated to KITE-753 [1 respiratory syncytial virus and 1 aspergillus, on days 213 and 23, respectively] and 1 due to progression). At 4.4-months median follow-up with KITE-753, all patients in DL3 had a complete response (CR; 2 CAR naive, 1 CAR exposed). In DL1/2 (n=11), ORR was 64% and CR rate was 45% (CAR naive [n=6]: ORR, 67%; CR rate, 67%). No DLTs or pausing criteria occurred with KITE-363 on study. Grade 3 CRS occurred in 1 patient; Grade 3 ICANS occurred in 3 patients; no Grade ≥4 CRS/ICANS occurred. At 11.1-months median follow-up with KITE-363 in DL3, ORR among CAR-naive patients (n=23) was 87% and CR rate was 78%; median duration of CR (DOCR) was not reached (95% CI, 5.2-not estimable) and the 6-month DOCR rate was 71.4%. Among all treated patients (N=37), 9 died (8 due to progression). Lymphodepletion was reduced and immune reconstitution was improved with KITE-363 compared to axicabtagene ciloleucel (axi-cel; Locke FL, et al. N Engl J Med. 2022). B-cell aplasia and recovery was comparable to axi-cel despite more pronounced CAR T-cell expansion. CAR T-cell expansion with KITE-753 was comparable to axi-cel even at a 10-fold lower dose. Conclusions: KITE-363 results showed high response rates with encouraging durability in DL3 and a manageable safety profile. KITE-753's rapid manufacturing process preserving a juvenile product phenotype, in addition to its bicistronic CAR design with dual co-stimulation, was associated with low incidence of CRS and ICANS, mostly Grade 1/2. The preliminary efficacy and safety profiles of KITE-753 DL3 appeared promising; the expansion phase with DL3 in LBCL (n=17) is ongoing and will be presented.