Introductory paragraph The adoptive transfer of anti-CD19 chimeric antigen receptor (CAR)-engineered T cells has shown impressive clinical responses in patients with refractory B-cell malignancies1–7. However, the therapeutic effects of CAR-T cells targeting other malignancies have not yet resulted in significant clinical benefit8–11. Although inefficient tumor trafficking and various immunosuppressive mechanisms can impede CAR-T cell effector responses, the signals delivered by the current CAR constructs may still be insufficient to fully activate antitumor T cell functions. Optimal T cell activation and proliferation requires multiple signals, including T cell receptor (TCR) engagement (signal 1), costimulation (signal 2), and cytokine engagement (signal 3)12. However, CAR gene constructs currently being tested in the clinic contain a CD3z (TCR signaling) domain and a costimulatory domain(s) but not a domain transmitting signal 313–18. Here, we have developed a novel CAR construct capable of inducing cytokine signaling upon antigen stimulation. This new generation CD19 CAR encodes a truncated cytoplasmic domain of IL-2Rβ and a STAT3-binding YXXQ motif together with CD3z and CD28 domains (28-ΔIL2RB-z (YXXQ)). The 28-ΔIL2RB-z (YXXQ) CAR-T cells showed antigen-dependent JAK-STAT3/5 pathway activation, which promoted their proliferation and prevented terminal differentiation in vitro. The 28-ΔIL2RB-z (YXXQ) CAR-T cells demonstrated superior in vivo persistence and antitumor effects in both liquid and solid tumor models compared with CAR-T cells with a CD28 or 4-1BB domain alone. Taken together, these results suggest that our new generation CAR has the potential to demonstrate superior antitumor effects with minimal toxicities in the clinic. Clinical translation of this novel CAR is warranted.

No abstract

Adoptive immunotherapy can induce sustained therapeutic effects in some cancers. Antitumor T cell grafts are often individually prepared in vitro from autologous T cells, which requires an intensive workload and increased costs. The quality of the generated T cells can also be variable, which affects the therapy's antitumor efficacy and toxicity. Standardized production of antitumor T cell grafts from third-party donors will enable widespread use of this modality if allogeneic T cell responses are effectively controlled. Here, we generated HLA class I, class II, and T cell receptor triple knockout (tKO) T cells by simultaneous knockout of the B2M, CIITA and TRAC genes through Cas9/sgRNA ribonucleoprotein electroporation. Although HLA deficient T cells were targeted by natural killer cells, they persisted better than HLA sufficient T cells in the presence of allogeneic peripheral blood mononuclear cells (PBMC) in immunodeficient mice.When transduced with a CD19 chimeric antigen receptor (CAR) and stimulated by tumor cells, tKO CAR-T cells persisted better when cultured with allogeneic PBMC compared with TRAC and B2M double-knockout T cells. The CD19 tKO CAR-T cells did not induce graft-versus-host disease, but retained antitumor responses. These results demonstrated the benefit of HLA class I, class II, and TCR deletion in enabling allogeneic-sourced T cells to be used for off-the-shelf adoptive immunotherapy. SynopsisAntitumor T cell grafts concurrently ablated of HLA class I, class II, and TCR molecules evade allogeneic T cell responses and can be used as a universal T cell source for adoptive cancer immunotherapy.