Success of Centralized Manufacturing of CD33 CAR T-Cells (CD33CART) for Children and Young Adults with Relapsed/Refractory AML

Introduction Autologous chimeric antigen receptor (CAR) T cell therapies are remarkably effective in patients with relapsed/refractory (r/r) B-cell malignancies, and development of CAR T cell therapies in r/r acute myeloid leukemia (AML) is urgently needed. Early experience with CAR T-cell manufacturing in patients with AML has been challenging. We describe feasibility of centralized manufacturing for CD33 CAR T-cell products (CD33CART) on the dose-escalation part of a multicenter Phase I/II clinical trial in children and adolescents/young adults (AYA) with r/r AML (NCT03971799). Methods Cryopreserved apheresis products from subjects enrolled at 6 sites were sent to the Biopharmaceutical Development Program at the Frederick National Laboratory for Cancer Research, National Cancer Institute. CD33CART manufacturing was performed on a ClinicMACS Prodigy® under Good Manufacturing Practice (GMP) conditions. Following thaw, apheresis products were enriched for CD4+ and CD8+ T-cells, cultured with 200 IU/mL of IL-2 and CD3/CD28 agonist beads, transduced with the CD33.28z lentiviral vector, and expanded for 7-10 days. CD33CART products were cryopreserved and returned to sites following a minimum 3-day interim sterility testing and after meeting quality-control release criteria such as identity, viability, and transduction efficiency (TE). Manufactured cell dose levels (DLs) were DL1: 3 × 105 CD33 CAR+ T-cells/kg, DL2: 1 × 106/kg, DL3: 3 × 106/kg, and DL4: 1 × 107/kg. Results Across 24 enrolled subjects with median age of 16 years (range 1-34), CD33CART manufacturing was attempted in and generated for 23 subjects. Manufacturing was not performed for one subject with rapid AML progression. CD33CART was infused in 19 subjects in the dose-escalation phase.Amongst infused subjects, apheresis products varied substantially, including high AML blast counts and low CD3+ T cells. The median blast% was 9.9% (range 0-61.4%). Median T cell CD3+ was 43.2% (range 1.4-93.4%) with 36.2% CD4+ (range 4.3-75.3%) and 11.9% CD8+ (range 1.0-57.6%). Despite this heterogeneity, final CD33CART products had median CD3+ of 99.6% (range 96.2-99.9%) with 66.8% CD4+ (range 33.3-92%) and 33.5% CD8+ (range 6.7-66.2%) with CD33+ AML <1.38% (lower limit of quantification) (Fig. A). The median manufacturing time was 7 days (range 7-10) with median TE and fold expansion of 45.4% (range 21.2-59.7%) and 19.9 (range, 9.8-88.7 Fig. B), respectively. In vitro co-culture assays assessing cytotoxicity against CD33+ AML cells demonstrated generally consistent activity across products (Fig. C). Clinical safety and initial activity of CD33CART was reported at the ASH 2023 meeting. Conclusion Despite wide inter-patient heterogeneity of apheresis products, centralized manufacturing of CD33CART was feasible in children and young adults with r/r AML. Correlative analyses characterizing CD33CART products are ongoing.