Abstract
Off-the-shelf CAR T cells may offer advantages over autologous strategies, including ease of manufacturing, improved quality control with avoidance of malignant contamination and T cell dysfunction as well as the ability to generate a final product from healthy T cells. While TCR editing can effectively prevent graft-versus-host reactions, the significant host-versus-graft immune response against histoincompatible T cells prevents the expansion and persistence of allogeneic CAR T cells and mitigates the efficacy of this approach. The goal is to achieve improved rates of durable complete remissions by improving allogeneic CD19CAR persistence since it has been shown that autologous CAR T cells have greater durability over years than allogeneic CAR T cells (N Engl J Med. 2021;384(7):673-674).
We describe here the engineering of human immune evasive CAR T cells based on our previously described hypoimmune technology (Nat Biotechnol 2019;37(3):252-258 and Proc Natl Acad Sci U S A 2021;118(28):e2022091118).
A major challenge is that, while HLA deletion can result in adaptive immune evasion, innate reactivity is enhanced by this strategy. Since CD47 overexpression can block both NK cell and macrophage killing (J Exp Med 2021;218(3):e20200839), we hypothesized that T cells would lose their immunogenicity when human leukocyte antigen (HLA) class I and II genes are inactivated and CD47 is over-expressed.
Human T cells from healthy donors were obtained by leukapheresis. To generate hypoimmune CD19CAR T cells, gene editing was used to delete b2m, CIITA, and TCR expression and lentiviral transduction was used to overexpress CD47 and CD19CAR containing a 4-1BB costimulatory domain to generate hypoimmune CAR T cells. Control T cells were unmanipulated except for lentiviral transduction used to overexpress the same CD19CAR and the deletion of the TCR. When transplanted into allogeneic humanized mice, hypoimmune CD19CAR T cells evade immune recognition by T cells even in previously sensitized animals as evidenced by a lack of T cell activation measured using ELISPOT analysis. In contrast, transplantation of non-hypoimmune-edited CD19CAR T cells generated from the same human donor resulted in a significant T cell activation (see figure: mean 59 and 558 spot frequencies for hypoimmune CD19CAR T cells and non-edited CD19CAR T cells, respectively; p<0.0001 unpaired T-test).
In addition to evading T cells, immune cell assays show that CD47 overexpression protects hypoimmune CD19CAR T cells from NK cell and macrophage killing in vitro and in vivo. Relative CD47 expression levels were analyzed to understand the relevance of CD47 for protection from macrophage and NK cell killing. A blocking antibody against CD47 made the hypoimmune CAR T cells susceptible to macrophage and NK cell killing in vitro and in vivo, confirming the importance of CD47 overexpression to evade innate immune clearance.
The hypoimmune CD19 CAR T cells retained their antitumor activity in both the Daudi and Nalm-6 B cell leukemia models, in vitro and in vivo. This indicated that the hypoimmune technology-i.e. isolated CD47 overexpression, deletion of b2m, CIITA, and TCR- did not show any effect on the cytotoxic potential of CD19 CAR T cells (see figure). These studies demonstrate that in vivo clearance of leukemic cells in NSG mice occurs across a range of tumor cell toCD19 CAR T cell ratios in a manner comparable to control, unedited CD19 CAR T cells (see figure). This result was validated using T cells from 3 different donors
These findings show that, in these models, hypoimmune CD19 CAR T cells are functionally immune evasive in allogeneic humanized mouse recipients and have cytotoxic anti-tumor capacity. They suggest that hypoimmune CAR T cells could provide universal CAR T cells that are able to persist without immunosuppression. Furthermore, these data suggest that hypoimmune CD19 CAR T cells can be used in sensitized patients and for re-dosing strategies.
Figure 1 Figure 1.
Disclosures
Hu: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Dao: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. White: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Gattis: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Clarke: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Landry: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Basco: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Tham: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Tucker: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Luo: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Bandoro: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Chu: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Young: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Foster: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Dowdle: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Rebar: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Fry: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company. Schrepfer: Sana Biotechnology: Current Employment, Current equity holder in publicly-traded company.
Novel APC-like properties of human NK cells directly regulate T cell activation