Autologous stem cell transplantation (auto-SCT) remains a standard of care for multiple myeloma patients and prolongs progression-free survival. A small cohort of patients achieve long-term control of disease, however the majority of patients ultimately relapse, and the mechanisms permitting disease progression remain unclear. In this study, we used a preclinical model of autologous SCT for myeloma where the disease either progressed (MM-relapsed) or was controlled. In the bone marrow (BM), inhibitory receptor expression on CD8⁺ T-cells correlated strongly with myeloma progression after transplant. In conjunction, the co-stimulatory/adhesion receptor CD226 (DNAM-1) was markedly downregulated. Interestingly, DNAM-1^- CD8⁺ T-cells in MM-relapsed mice had an exhausted phenotype, characterized by upregulation of multiple inhibitory receptors, including immunoglobulin and ITIM domains (TIGIT), programmed death 1 (PD-1) with decreased T-bet and increased Eomesodermin expression. Immune checkpoint blockade using monoclonal antibodies against PD-1 or TIGIT significantly prolonged myeloma control after SCT. Furthermore, CD8⁺ T-cells from MM-relapsed mice exhibited high interleukin-10 (IL-10) secretion that was associated with increased TIGIT and PD-1 expression. However, while donor-derived IL-10 inhibited myeloma control post-SCT, this was independent of IL-10 secretion by, or signaling to T-cells. Instead, the donor myeloid compartment, including CSF-1R-dependent macrophages and an IL-10-secreting dendritic cell population in the BM, promoted myeloma progression. Our findings highlight PD-1 or TIGIT blockade in conjunction with SCT as a potent combination therapy in the treatment of myeloma. Key points Myeloma promotes CD8⁺ T-cell exhaustion and IL-10 secretion from dendritic cells. PD-1 and TIGIT-targeted checkpoint inhibition is effective after SCT.

T _R 1 cells are the major regulatory population generated after allogeneic bone marrow transplantation.

Natural regulatory T cells (nTregs) play an important role in tolerance; however, the small numbers of cells obtainable potentially limit the feasibility of clinical adoptive transfer. Therefore, we studied the feasibility and efficacy of using murine-induced regulatory T cells (iTregs) for the induction of tolerance after bone marrow transplantation. iTregs could be induced in large numbers from conventional donor CD4 and CD8 T cells within 1 wk and were highly suppressive. During graft-versus-host disease (GVHD), CD4 and CD8 iTregs suppressed the proliferation of effector T cells and the production of proinflammatory cytokines. However, unlike nTregs, both iTreg populations lost Foxp3 expression within 3 wk in vivo, reverted to effector T cells, and exacerbated GVHD. The loss of Foxp3 in iTregs followed homeostatic and/or alloantigen-driven proliferation and was unrelated to GVHD. However, the concurrent administration of rapamycin, with or without IL-2/anti–IL-2 Ab complexes, to the transplant recipients significantly improved Foxp3 stability in CD4 iTregs (and, to a lesser extent, CD8 iTregs), such that they remained detectable 12 wk after transfer. Strikingly, CD4, but not CD8, iTregs could then suppress Teff proliferation and proinflammatory cytokine production and prevent GVHD in an equivalent fashion to nTregs. However, at high numbers and when used as GVHD prophylaxis, Tregs potently suppress graft-versus-leukemia effects and so may be most appropriate as a therapeutic modality to treat GVHD. These data demonstrate that CD4 iTregs can be produced rapidly in large, clinically relevant numbers and, when transferred in the presence of systemic rapamycin and IL-2, induce tolerance in transplant recipients.