Bone Marrow T Cells Are More Effective Than Peripheral T Cells in Inducing Chimeric Conversion Following MHC-Mismatched Nonmyeloablative Bone Marrow Transplantation

Background & Objectives: Recently, T cells in BM have attracted renewed interest because they are now known to have different surface phenotypes, subsets, and activation states from those in the periphery. Memory T cells undergo extensive migration from the blood to the BM and vice versa. The BM plays an important role in preferential homing and extensive proliferation of memory T cells, and contributes considerably to the longlived memory T cell pool. BM T cells are more activated than their splenic counterparts and have a higher rate of local proliferation. Although BM-T (NK1.1– CD4+ or CD8+) cells did not induce lethal GVH disease, even at high cell numbers, BM-T cells mediated vigorous graft-versus-tumor activity and facilitated engraftment of hematopoietic progenitor cells. These studies suggested that BM-T cells could be a useful cellular source for adoptive immunotherapy following ABMT, instead of peripheral T cells. Non-myeloablative bone marrow transplantation (NMT) and allogeneic mixed chimerism can provide an environment adequate for diminishing susceptibility to DLI-mediated GVHD and an immunological platform for DLI in both mouse and human models. In patients treated with DLI, a successful GVL effect is often associated with conversion to complete donor chimerism, supporting the concept of a graft-versus-host (GVH) response as part of the GVL effect. Thus, a quiet chimeric conversion following DLI is desirable to reach an optimal DLI-mediated GVL effect, without the occurrence of GVHD. Although in a mouse model, the administration of non-tolerant donor spleen cells to established mixed chimeras has been shown to convert mixed hematopoietic chimerism to full donor chimerism, without the concomitant development of GVHD, DLI in humans frequently results in serious GVHD and life-threatening complications. However, the use of BM-T cells, as compared with spleen T cells (SP-T), as the DLI source has not been investigated in allogeneic mixed chimerism prepared with NMT. In this study, we evaluated the beneficial alloreactivity of DLI using cryopreserved BM-T cells, a by-product obtained during the T cell depletion (TCD) procedure in BM grafting, to effectively induce chimeric conversion without the occurrence of GVHD in MHC-mismatched NMT.

Methods: Cells were prepared using established procedures. During the T cell depletion (TCD) procedure in BM grafting, BM-T cells were obtained as a by-product and then cryopreserved for subsequent DLI using BM-T cells 21 days after the bone marrow transplant.

Results: The administration of 5–10 × 10⁵ BM-T (Thy1.2+) cells in mixed chimeras resulted in complete chimeric conversion, with self-limited graft-versus-host disease (GVHD) and no pathological changes. However, the administration of 5–10 × 10⁵ SP-T (Thy1.2+) cells resulted in persistent mixed chimerism, with pathological GVHD signs in the liver and intestine.

Conclusion: Our results suggest that DLI using BM-T cells, even in small numbers, could be more potent for inducing chimeric conversion in mixed chimerism than DLI using SP-T cells. Further study is needed to determine whether cryopreserved BM-T cells are an effective cell source for DLI to consolidate donor-dominant chimerism in clinical practice, without concerns about GVHD.