Harnessing ADR T cells to enhance allo-HCT

In this issue of Blood, Mo et al¹ show that human T cells engineered to express an OX40 cytotoxic receptor and a CD19-targeted chimeric antigen receptor (CAR) protect animals from fatal xenogeneic graft-versus-host disease (GVHD) and leukemia relapse following allogeneic hematopoietic cell transplantation (allo-HCT).

Allo-HCT is a potentially curative treatment for patients with hematologic malignancies. Yet, morbidity and mortality due to relapse, GVHD, and infectious complications limit its therapeutic efficacy. The current approaches to decrease acute GVHD, including administering immunosuppressive medications,² depleting T cells from the allograft before transplantation,³ and administering posttransplant cyclophosphamide for in vivo depletion of T cells,⁴ are associated with either increased infectious complications or relapse. Hence, a central area of ongoing investigation in the field of allo-HCT focuses on strategies to enhance graft-versus-leukemia activity while limiting GVHD.

Prior data demonstrated the upregulation of a costimulatory receptor, OX40, on activated CD4 T cells and a subset of CD8 T cells.^5,6 OX40 has also been found to mediate expansion of alloreactive T cells. Further work from several groups elucidated that the selective depletion of OX40⁺ alloreactive cells from the allograft or OX40 blockade decreased the expansion of alloreactive T cells while preserving antitumor and antiviral activity.

Thus, Mo et al sought to advance the targeting of OX40 on activated T cells as a strategy to decrease GVHD without compromising beneficial T-cell activity. First, the authors found that OX40 was upregulated on T cells that infiltrated GVHD target organs in a rhesus macaque allo-HCT model. Notably, previous data on OX40⁺ T cells in tissues damaged by acute GVHD were limited. Next, the authors developed a cytotoxic OX40-specific alloimmune defense receptor (OX40.ADR), consisting of an extracellular domain of OX40 ligand with 4-1BB and CD3ζ signaling domains. Human T cells were retrovirally transduced to express the OX40.ADR, and the authors found that the ADR T cells suppressed alloreactivity in vitro. In a xenogeneic mouse allo-HCT model, donor ADR human T cells protected mice from fatal xenogeneic GVHD. The authors also showed that targeting OX40 with ADR did not limit the antiviral activity of T cells in vitro or in vivo. Furthermore, CAR-ADR T cells induced leukemia clearance and decreased GVHD in tumor-bearing mice in a xenograft model of residual disease post-transplantat. The authors’ findings underscore a potential approach to engineer T cells from donor sources to decrease GVHD and relapse following allo-HCT.

Prior work by the authors⁷ as well as other investigators has evaluated the use of donor adoptive cell therapies that target antigens, such as 4-1BB⁷ and CD83. These markers are expressed on alloreactive T cells, and depleting them in the posttransplant context is a strategy that has been assessed to decrease GVHD. Although these approaches eradicate alloreactive T cells, they also ablate activated T cells that may mediate desirable T-cell immune activity. In contrast to these earlier data, the authors found that culture conditions in their model enriched ADR T-cell expression on CD8 T cells. Furthermore, ADR T cells preferentially targeted activated CD4 T cells, which likely facilitated the preservation of T-cell antiviral activity.

Of note, relapse is the leading cause of death following allo-HCT. Preclinical and clinical data suggest that adoptive therapy with donor CAR T cells administered following allo-HCT induces less GVHD compared with unmodified donor T cells. However, preclinical models have demonstrated that the potential for GVHD varies depending on aspects of the CAR, including the number of immunoreceptor tyrosine–based activation motifs within the CD3ζ signaling domain and the costimulatory receptor.^8,9 One preclinical model deciphered the biological basis for the attenuation of GVHD with donor CAR T cells, which suggests ways to make CARs from a donor safer and more effective.⁸ However, to effectively administer donor CAR T cells following allo-HCT to decrease relapse, studies will need to interrogate how to manage GVHD prophylaxis, as these medications may diminish CAR T-cell activity.

Hence, an approach that diminishes GVHD while maintaining antitumor and antiviral activity is a considerable advance for the allo-HCT field. Further research will need to assess the impact of ADR T cells on chronic GVHD, characterize their effect on the function of regulatory T cells, and evaluate approaches to regulate their persistence. Nonetheless, the authors convincingly demonstrate that CAR-ADR T cells are a novel and innovative strategy to enhance leukemia clearance and diminish acute GVHD. These findings have strong translational implications as to how the administration of engineered donor T cells following allo-HCT may improve clinical outcomes.

Conflict-of-interest disclosure: The author declares no competing financial interests.