Improved survival after transplantation of more donor plasmacytoid dendritic or naïve T cells from unrelated-donor marrow grafts: results from BMTCTN 0201

Abstract

Purpose: To characterize relationships between specific immune cell subsets in bone marrow (BM) or granulocyte colony-stimulating factor-mobilized peripheral blood (PB) stem cells collected from unrelated donors and clinical outcomes of patients undergoing transplantation in BMTCTN 0201.

Patients and methods: Fresh aliquots of 161 BM and 147 PB stem-cell allografts from North American donors randomly assigned to donate BM or PB stem cells and numbers of transplanted cells were correlated with overall survival (OS), relapse, and graft-versus-host disease (GvHD).

Results: Patients with evaluable grafts were similar to all BMTCTN 0201 patients. The numbers of plasmacytoid dendritic cells (pDCs) and naïve T cells (Tns) in BM allografts were independently associated with OS in multivariable analyses including recipient and donor characteristics, such as human leukocyte antigen mismatch, age, and use of antithymocyte globulin. BM recipients of > median number of pDCs, naïve CD8(+) T cells (CD8Tns), or naïve CD4(+) T cells (CD4Tns) had better 3-year OS (pDCs, 56% v 35%; P = .025; CD8Tns, 56% v 37%; P = .012; CD4Tns, 55% v 37%; P = .009). Transplantation of more BM Tns was associated with less grade 3 to 4 acute GvHD but similar rates of relapse. Transplantation of more BM pDCs was associated with fewer deaths resulting from GvHD or from graft rejection. Analysis of PB grafts did not identify a donor cell subset significantly associated with OS, relapse, or GvHD.

Conclusion: Donor immune cells in BM but not PB stem-cell grafts were associated with survival after unrelated-donor allogeneic hematopoietic stem-cell transplantation. The biologic activity of donor immune cells in allogeneic transplantation varied between graft sources. Donor grafts with more BM-derived Tns and pDCs favorably regulated post-transplantation immunity in allogeneic hematopoietic stem-cell transplantation.

Conflict of interest statement

Authors' disclosures of potential conflicts of interest and author contributions are found at the end of this article.

© 2014 by American Society of Clinical Oncology.

Figures

Fig 1.

Significant association between content of cells in bone marrow (BM) and granulocyte colony-stimulating factor–mobilized peripheral blood (PB) stem-cell grafts with overall survival (OS) among those undergoing allogeneic transplantation with unrelated donor. Multivariable analysis showing hazard ratios (HRs) for OS based on content of viable nucleated cells (VNCs), CD34+ cells, B cells, natural killer (NK) cells, T cells (T), and dendritic cells (DCs) as well as calculated ratios of cell subsets in graft. (A) Multivariable analysis of BM graft constituents. (B) Multivariable analysis of PB graft constituents. CD4Tn, naïve CD4+ T cell; CD8Tn, naïve CD8+ T cell; mDC, myeloid dendritic cell; pDC, plasmacytoid dendritic cell; Tn, naïve T cell; Treg, T regulatory cell.

Fig 2.

Larger numbers of donor plasmacytoid dendritic cells (pDCs) in bone marrow (BM) graft are associated with improved survival and decreased treatment-related mortality (TRM) after allogeneic transplantation with unrelated donor. (A) Estimated probability of 3-year overall survival (OS), stratifying 143 evaluable recipients of BM grafts by median number of transplanted donor pDCs; (B) incidence of TRM; (C) incidence of relapse; (D) incidence of grade 3 to 4 acute graft-versus-host disease (aGvHD); (E) incidence of chronic GvHD; and (F) multivariable analysis showing hazard ratio of content of pDCs in BM graft with aGvHD, chronic GvHD, TRM, relapse, disease-free survival, and OS. P values in (A) to (E) represent log-rank tests for patients undergoing transplantation with number of pDCs < or > median value of 0.3 × 106 cells/kg. P values in (F) represent results of multivariable analysis, as described in Patients and Methods.

Fig 3.

Larger numbers of donor naïve T cells (Tns) in bone marrow (BM) graft are associated with improved survival, decreased treatment-related mortality (TRM), and decreased grade 3 to 4 acute graft-versus-host disease (aGvHD) after allogeneic transplantation with unrelated donor. Estimated probability of 3-year overall survival (OS), stratifying 129 evaluable recipients of BM grafts by median number of transplanted donor (A) naïve CD8+ T cells (CD8Tns) or (B) naïve CD4+ T cells (CD4Tns). Incidence of TRM stratified by content of donor (B) CD8Tns or (E) CD4Tns. Multivariable analysis showing hazard ratio of content of (C) CD8Tns or (F) CD4Tns in BM graft with aGvHD, chronic GvHD (cGvHD), TRM, relapse, disease-free survival (DFS), and OS. P values in (A), (B), (D), and (E) represent log-rank tests for survival of patients undergoing transplantation with number of CD8Tns or CD4Tns < or > median value of 1.3 × 106 cells/kg or 2.6 × 106 cells/kg, respectively. P values in (C) and (F) represent results of multivariable analysis, as described in Patients and Methods.

Fig 4.

Overall survival (OS) of bone marrow (BM) transplantation recipients stratified by content of both naïve CD8+ T cells (CD8Tns) and plasmacytoid dendritic cells (pDCs) in BM graft. Estimated probability of 3-year OS, stratifying 129 evaluable recipients of BM grafts by median number of transplanted donor CD8Tns and pDCs into four groups: patients who received > median number of both cell subsets (red line), patients who received > median number of one cell subset and < median number of other cell subset (gold and gray lines), and patients who received < median number of both cell subsets (blue line).

Fig A1.

Flow cytometry dot plots of representative data showing phenotype of CD34+ cells, dendritic cells, T regulatory cells (Tregs), and memory and naïve T-cell (Tn) subsets for single patient in donor peripheral blood (PB) apheresis sample (A to F) and donor bone marrow (BM) from participants in BMTCTN (Blood and Marrow Transplant Clinical Trials Network) 0201 (G to L). Cell populations of interest are shown gated with gold polygons. (A, G) Samples are gated on forward scatter and side scatter (SSC), with percentage of CD34+ stem cells shown as fraction of CD45lo/+ 7AAD− population. (B, H) Percentages of CD11c+ myeloid and CD123+ plasmacytoid dendritic cells as fraction of lineage-negative (CD3, CD14, CD16, CD19, CD20) HLA DR+ population are shown. (C, I) Data shown are gated on CD27+, CD69− PB and BM T cells and show percentage of CD4+ T cells with Treg phenotype CD3+ CD4+ CD27+ CD69− CD25+. (D, J) Percentages of central memory and Tn subsets within total population of CD3+ T cells are shown. (E, F, K, L) Expression of CCR7 and CD127 (interleukin-7 receptor) on central memory and Tn subsets from the previously gated subsets in (D) and (J). (F, L) CD45RA+, CD62L+, and CCR7+ triple-positive Tn populations are shown.

Fig A2.

Overall survival among recipients enrolled onto BMTCTN (Blood and Marrow Transplant Clinical Trials Network) 0201 for whom graft samples were analyzed.

Fig A3.

Larger numbers of donor naïve T cells (Tns) in bone marrow (BM) grafts are associated with decreased incidence of severe acute graft-versus-host disease (GvHD) after allogeneic transplantation with unrelated donor without significant effect on relapse or chronic GvHD. Incidence of grade 3 to 4 acute GvHD stratifying 129 evaluable recipients of BM grafts by median number of transplanted donor (A) naïve CD8+ T cells (CD8Tns) or (B) naïve CD4+ T cells (CD4Tns). Incidence of chronic GvHD stratified by content of donor (B) CD8Tns or (E) CD4Tns. Incidence of relapse stratified by content of donor (C) CD8Tns or (F) CD4Tns.