Phase 1 clinical trial of CRISPR-engineered CAR19 universal T cells for treatment of children with refractory B cell leukemia

Genome editing of allogeneic T cells can provide "off-the-shelf" alternatives to autologous chimeric antigen receptor (CAR) T cell therapies. Disruption of T cell receptor α chain (TRAC) to prevent graft-versus-host disease (GVHD) and removal of CD52 (cluster of differentiation 52) for a survival advantage in the presence of alemtuzumab have previously been investigated using transcription activator-like effector nuclease (TALEN)-mediated knockout. Here, we deployed next-generation CRISPR-Cas9 editing and linked CAR expression to multiplexed DNA editing of TRAC and CD52 through incorporation of self-duplicating CRISPR guide RNA expression cassettes within the 3' long terminal repeat of a CAR19 lentiviral vector. Three cell banks of TT52CAR19 T cells were generated and cryopreserved. A phase 1, open-label, non-randomized clinical trial was conducted and treated six children with relapsed/refractory CD19-positive B cell acute lymphoblastic leukemia (B-ALL) (NCT04557436). Lymphodepletion included fludarabine, cyclophosphamide, and alemtuzumab and was followed by a single infusion of 0.8 × 10⁶ to 2.0 × 10⁶ CAR19 T cells per kilogram with no immediate toxicities. Four of six patients infused with TT52CAR19 T cells exhibited cell expansion, achieved flow cytometric remission, and then proceeded to receive allogeneic stem cell transplantation. Two patients required biological intervention for grade II cytokine release syndrome, one patient developed transient grade IV neurotoxicity, and one patient developed skin GVHD, which resolved after transplant conditioning. Other complications were within expectations, and primary safety objectives were met. This study provides a demonstration of the feasibility, safety, and therapeutic potential of CRISPR-engineered immunotherapy.