Perfecting Adoptive Cellular Therapy for Graft-Versus-Host Disease: Alloreactive Induced T Regulatory Cells

Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a therapeutic option for hematologic malignances. Donor T effector cells (Teffs), recognizing mismatched major and minor histocompatibility antigens, are both beneficial and detrimental, as they mediate anti-tumor response (GVL) and graft-versus-host disease (GVHD), the latter of which remains a significant complication arising after allo-HSCT. Adoptive cellular therapy using polyclonal natural T regulatory (nTreg) has been shown to be a promising strategy in the prevention and treatment of GVHD. However, clinical application of nTreg therapy requires high doses and expanded nTregs have sparse availability, limited expansion capacity, and potential contamination with Teffs. Induced T regulatory cells (iTregs) could solve these limitations given their robust proliferative potential and increased availability. Moreover, iTregs can be easily educated with alloantigen prior to clinical infusion, increasing their selectivity and potency, which may result in a more targeted and effective cellular therapy compared to polyclonal nTregs.

In the current work, we tested the hypothesis that alloreactive iTregs can be readily generated and possess high potency to suppress GVHD with better selectivity. By stimulating CD4⁺CD25^- T cells with allogeneic dendritic cells (DCs) in the presence of IL-2, TGF-β, and retinoic acid, high percentages of CD25⁺Foxp3⁺ iTregs (60-80%) were induced in 5-day culture. These alloreactive iTregs were highly suppressive in T-cell responses to the same allogeneic stimulation, and were 16-32 fold more suppressive than those polyclonal iTregs generated by αCD3/αCD28-stimulation in vitro. Using a well-defined, fully MHC-mismatched B6 (H2^b) to Balb/c (H2^d) murine model of allogeneic BMT, we demonstrated that co-transfer of donor iTregs (H2^b) generated with recipient DCs (H2^d) significantly suppressed GVHD mortality, whereas the donor iTregs (H2^b) generated with a third-party DCs (H2^k) were significantly less effective. In mechanistic studies, we found that iTregs reactive to recipient alloantigens (H2^d) expanded more dramatically and were significantly more stable in terms of maintaining Foxp3 expression as compared to the iTregs reactive to third-party alloantigens (H2^k) in H2^drecipients. As a consequence, recipient-reactive iTregs were significantly more effective than third-party reactive iTregs in inhibiting activation, expansion, and migration of Teffs into GVHD target organs.

Although alloreactive CD4 iTregs were effective in alleviating GVHD, high dose and multiple infusions were necessary to prevent GVHD lethality for the majority of recipients. Published studies indicate that CD8 iTregs are generated in allogeneic recipients after allo-BMT in vivo and also contribute to the regulation of GVHD development. Aiming to improve Treg efficacy, we hypothesized that alloreactive CD8 iTregs can be generated in vitro and used for singular or combinational cell therapy. To this end, we established a protocol to induce alloreactive CD8 iTregs, and evaluated their suppressive activity. While these CD8 iTregs were potently suppressive of alloresponse in vitro, they were only partially protective in vivo in attenuating GVHD. Given we also observed that CD8 iTregs were superior to CD4 iTregs in the suppression of CD8 Teff responses to alloantigens in vitro, we further hypothesized that a combinational therapy with CD4 and CD8 iTregs would be more effective in the prevention of GVHD than either alone. Indeed we found the combined CD4 and CD8 iTreg therapy to be significantly more potent in attenuating GVHD and reducing clinical score than either single therapy alone.

In conclusion, alloreactive CD4 iTregs are significantly more effective than polyclonal counterparts in suppressing alloresponse in vitro and preventing GVHD in vivo. These alloreactive CD4 iTregs prevent GVHD by inhibiting CD4⁺ Teff expansion and migration through an antigen-dependent manner. For the first time, to our knowledge, we showed that alloreactive CD8 iTregs were highly potent in suppressing alloresponse in vitro and moderately potent in preventing GVHD in vivo. Furthermore, we provided evidence indicating that perfecting adoptive iTreg therapy may require a combination of CD4 and CD8 iTregs to achieve maximum efficacy in the prevention of GVHD.