Balancing act for Treg immunotherapy

In this issue of Blood, Gaidot and colleagues demonstrate effective suppression of graft-versus-host disease (GVHD) by recipient-specific regulatory T cells without compromising immunity after hematopoietic stem cell transplantation (HCT).¹  Should these findings push the transplant community toward widespread use of Treg immunotherapy to prevent GVHD?

Since Sakaguchi's pioneering findings that regulatory T cells (Tregs) constitutively express CD25,²  an explosion of studies showed their critical role in maintaining peripheral tolerance and immune homeostasis.³  Under inflammatory conditions, Tregs prevent pathogenic tissue damage; however, the suppressor effects of Tregs may limit tumor and microbial immunity. The potential for the dual and opposing beneficial and deleterious effects of Tregs indicate the need for judicious application of targeting Tregs or using Treg cell–based therapy in the clinical setting. This is particularly relevant in HCT, whereby donor T cells, which are a primary target of Tregs, mediate both GVHD and tumor and microbial immunity. Thus, hypothetically, nonselective suppression of donor T cells by Tregs would prevent GVHD, but also diminish the graft-versus-tumor (GVT) or antimicrobial effect, much as one may observe with standard immunosuppressive agents.

Experimental studies in mouse models, however, showed that adoptive transfer of donor Tregs protect major histocompatibility–mismatched transplant recipients from GVHD, and at specific dose ratios with effector T cells, Tregs did not impair GVT responses against transplantable leukemia⁴  or impair primary or recall immune responses against cytomegalovirus infection.⁵  Despite questions that linger about the translatability of these findings to the clinic given the contrived experimental conditions, including the cell doses and ratios and the clinical relevance of transplantable leukemia, these studies provide proof-of-principle for the potential role of Treg cell–based therapy.

Because Tregs comprise only 5%-10% of CD4 T cells, multiple strategies have been developed to purify and expand their population ex vivo to obtain sufficient numbers for adoptive transfer. The major challenge with ex vivo expansion of Tregs has been to maintain their phenotypic and functional purity. Treg cultures are often contaminated with effector T cells that could exacerbate inflammatory responses on adoptive transfer. Furthermore, the ability of expanded Tregs to preserve microbial and tumor immunity while suppressing GVHD has not been well studied compared with their freshly isolated Treg counterpart. Thus, the findings by Gaidot et al represent a significant advance, showing that donor Tregs, expanded in culture with irradiated allogeneic host splenocytes and interleukin-2, henceforth called recipient-specific Tregs (rsTregs), remain phenotypically pure after culture and can suppress GVHD without compromising donor immune reconstitution and function.¹ 

After HCT, immune reconstitution, particularly T cells, results from thymopoieis and peripheral expansion of mature donor lymphocytes by antigenic stimulation and homeostatic cytokines. Because thymopoieis is often delayed or compromised in adults after HCT, T-cell reconstitution, particularly in the early period, primarily relies on peripheral expansion of postthymic donor T cells present in the graft.⁶  To recapitulate this clinical scenario, Gaidot and colleagues developed an allogeneic GVHD model using CD3ϵ deficient donors that do not have T-cell precursors, allowing them to determine the impact of rsTregs on the reconstitution of mature T cells in the graft.

In their model, recipient mice were lethally irradiated and subsequently rescued with allogeneic CD3ϵ-deficient bone marrow cells (BM). To induce GVHD, transplant recipients also received CD3⁺ T cells, with or without rsTregs at a 1:1 cell dose ratio, from wild-type donors that are syngeneic to the CD3ϵ-deficient BM donors. Transplant recipients that received rsTregs had reduced GVHD. They also had a lower proportion of effector memory donor T cells, and in an ex vivo nonspecific restimulation assay, their CD4⁺, but not CD8⁺, T cells had a significantly reduced production of proinflammatory cytokines early after HCT. No differences were seen at the later time points measured. These findings suggest the critical suppressive role of rsTregs early after HCT, when active proliferation of donor T cells leads to a cascade of proinflammatory events that cause acute GVHD.⁴ 

To determine whether the adoptively transferred rsTregs compromised donor lymphocyte immune reconstitution, the authors next challenged transplant recipients with a skin allograft at day 30 and found that rsTreg recipients rejected third-party, but not donor-type, consistent with an intact alloreactive response. Furthermore, when rsTreg recipients were infected with vaccinia virus at day 120 after HCT, they mounted a vaccinia-specific T-cell response that controlled the viral disease. These results indicate the ability of rsTregs to reduce the proliferation but not the cytotoxicity of T cells, as was shown previously in a GVT model using freshly isolated Tregs.⁴  However, because rsTregs are relatively absent late after HCT, it is unlikely that rsTregs have a direct effect on the effector T-cell response against the skin graft or the vaccinia virus. The effect of rsTregs might be because of the prevention of early GVHD-induced damage of lymphoid microenvironment that is critical for homeostatic lymphocyte expansion, survival, and function after HCT.^5,7,8  In rsTreg-protected recipients, Gaidot et al further observed an increasing proportion of donor Tregs (dTregs) from the graft over time. It is not clear how dTregs (and thymically derived Tregs that were not evaluated) impact tolerance and immunity in their model, but dTregs overtake rsTregs in number and proportion with time after HCT and therefore likely play a more direct and dominant role than rsTregs in the later stages of HCT. These findings suggest that the positive impact of rsTregs on immune reconstitution extend beyond their absence and that the timing of the Treg transfer is likely more critical than number of infusions during HCT. Not addressed in these studies, but important to define, would be how adoptively transferred rsTregs impact early immunity against microbes, when Treg expansion is most robust and host immunosuppression is significant.

Translating the findings of Gaidot and colleagues to the clinical setting will be challenging given the complexity of the culture system, including the source of stimulator cells. Their work, however, provides further scientific confirmation for ongoing clinical trials with Treg cell–based therapy for HCT, the most mature of which was presented recently with early congruent results favoring GVHD reduction and improved immune reconstitution.⁹