A Novel Human OX40L-CAR-Treg Specifically Targets Activated Antigen-Presenting Cells and Effectively Controls T Cell Alloreactivity and Xeno-GvHD

Background: Graft-versus-host disease (GvHD) remains the leading cause of non-relapse mortality after allogeneic hematopoietic stem cell transplantation (HCT). Adoptive transfer of regulatory T cells (Tregs) has demonstrated promising results in controlling GvHD in both animal models and phase I/II clinical trials. However, strategies to increase the stability of ex-vivo expanded Tregs, and to optimize their potency, are needed. Recent efforts to accomplish these goals have focused on engineering Chimeric Antigen Receptor (CAR) Tregs, with the first attempts using CARs targeted against disparate HLA antigens. However, Tregs targeting HLA have several limitations, including the downregulation of these antigens that accompanies some diseases, as well as the lack of utility of this approach in the HLA-matched setting. In this study, we tested a novel GvHD prophylaxis approach consisting of infusing donor Tregs engineered to express an anti-OX40L CAR that, upon engagement with activated OX40L+ host antigen-presenting cells (APCs), increases their suppressor potency to dampen APC-mediated activation of alloreactive T cells.

Methods and Results: We designed a CAR construct containing an scFv targeting OX40L, a CD28 costimulatory domain, along with CD3ζ, as well as the fluorescent reporter NeonGreen, all under the control of a synthetic FOXP3 promoter. This design enabled the selective and sustained expression of the anti-OX40L CAR on CD25+CD127- FoxP3+ Treg cells. We also created Tregs transduced with a non-targeting NeonGreen-only control construct (Neon-Tregs). Engineered OX40L CAR-Tregs maintained expression of the canonical Treg markers FOXP3, CD25, and CTLA-4, and demonstrated robust activation in the presence of OX40L-expressing cells. In co-culture assays of OX40L CAR-Tregs with OX40L-expressing K562 target cells, we demonstrate upregulation of Treg suppressive proteins CTLA-4, LAP, CD71, GARP and LAG3 without inducing expression of pro-inflammatory cytokines (no upregulation of IFNγ, TNFα, IL17a) (Figure 1A). Activation of Tregs via CAR recognition of OX40L+ cells resulted in 2.5-fold enhancement in their potency to suppress anti-CD3/CD28-driven T cell proliferation in vitro compared to Neon-Tregs at a 1:4 Treg-to-Teff ratio (p<0.05). Furthermore, OX40L-CAR-Tregs demonstrated a greater capacity to inhibit monocyte-derived dendritic cell activation, highlighting the superior suppressive activity of these OX40L-CAR-Tregs compared to non-targeting Neon-Tregs.

Importantly, in a human xenograft model of GvHD, induced by transplantation of human PBMC into sublethally irradiated NOD.Cg-Prkdc^scid Il2rg^tm1Wjl/SzJ (NSG) mice, OX40L CAR-Treg demonstrated superior GvHD disease control compared to both PBMC alone and Neon-Tregs. Clinical scores were assessed by a combinatorial system including five parameters: weight loss, posture, mobility, skin, and fur conditions, and curves were compared using a 2-way ANOVA mixed-effects analysis, which demonstrated improvement in clinical xeno-GvHD with the OX40L CAR-Treg > Neon-Treg > PBMC alone over the entire length of analysis (p <0.001). An example of the substantial improvement in the clinical score on Day +14 is as follows: combined GvHD clinical score = 6.5 for PBMC vs 2.5 for Neon-Treg vs 0.15 for OX40L CAR-Treg, p <0.0001 for all comparisons. This control of clinical GvHD resulted in prolonged recipient survival when compared to both PBMC alone and to control Neon-Tregs (MST = 34 days for OX40L-CAR-Treg, 20 days for Neon-Treg, 14 days for PBMC alone, p<0.001, Figure 1B).

Conclusions: Collectively, these results demonstrate a novel and efficacious approach using OX40L-CAR Tregs to enhance Treg suppressive function and to control GvHD. Because OX40L is upregulated on APCs in inflammatory environments, and its expression is limited to APCs and activated endothelial cells, this approach provides a unique strategy for the control of allo-immunity after HCT. Moreover, these OX40L-CAR Treg may be broadly applicable beyond HCT, to control allo-immunity after solid organ transplant and to suppress T cell activation in autoimmune diseases.

Gerdemann:AlloVir: Divested equity in a private or publicly-traded company in the past 24 months, Patents & Royalties. Blazar:Rheos Medicines: Research Funding; Magenta Therapeutics: Membership on an entity's Board of Directors or advisory committees; Tmunity Therapeutics: Current holder of stock options in a privately-held company; Carisma Therapeutics: Research Funding; BlueRock Therapeutics: Membership on an entity's Board of Directors or advisory committees, Research Funding. Kean:Mammoth Biosciences: Current equity holder in private company, Current holder of stock options in a privately-held company; Magenta: Research Funding; Vertex: Consultancy; Novartis: Research Funding; Bluebird Bio: Research Funding; EMD-Serono: Research Funding; Bristol Myers Squibb: Patents & Royalties: clinical trial, Research Funding; HiFiBio: Consultancy.