Characterization and Function of Cryopreserved Bone Marrow from Deceased Organ Donors: A Potential Viable Alternative Graft Source

Abstract

Despite the readily available graft sources for allogeneic hematopoietic cell transplantation (alloHCT), a significant unmet need remains in the timely provision of suitable unrelated donor grafts. This shortage is related to the rarity of certain HLA alleles in the donor pool, nonclearance of donors owing to infectious disease or general health status, and prolonged graft procurement and processing times. An alternative hematopoietic progenitor cell (HPC) graft source obtained from the vertebral bodies (VBs) of deceased organ donors could alleviate many of the obstacles associated with using grafts from healthy living donors or umbilical cord blood (UCB). Deceased organ donor-derived bone marrow (BM) can be preemptively screened, cryogenically banked for on-demand use, and made available in adequate cell doses for HCT. We have developed a good manufacturing practice (GMP)-compliant process to recover and cryogenically bank VB-derived HPCs from deceased organ donor (OD) BM. Here we present results from an analysis of HPCs from BM obtained from 250 deceased donors to identify any substantial difference in composition or quality compared with HPCs from BM aspirated from the iliac crests of healthy living donors. BM from deceased donor VBs was processed in a central GMP facility and packaged for cryopreservation in 5% DMSO/2.5% human serum albumin. BM aspirated from living donor iliac crests was obtained and used for comparison. A portion of each specimen was analyzed before and after cryopreservation by flow cytometry and colony-forming unit potential. Bone marrow chimerism potential was assessed in irradiated immunocompromised NSG mice. Analysis of variance with Bonferroni correction for multiple comparisons was used to determine how cryopreservation affects BM cells and to evaluate indicators of successful engraftment of BM cells into irradiated murine models. The t test (with 95% confidence intervals [CIs]) was used to compare cells from deceased donors and living donors. A final dataset of complete clinical and matched laboratory data from 226 cryopreserved samples was used in linear regressions to predict outcomes of BM HPC processing. When compared before and after cryopreservation, OD-derived BM HPCs were found to be stable, with CD34⁺ cells maintaining high viability and function after thawing. The yield from a single donor is sufficient for transplantation of an average of 1.6 patients (range, 1.2 to 7.5). CD34⁺ cells from OD-derived HPCs from BM productively engrafted sublethally irradiated immunocompromised mouse BM (>44% and >67% chimerism at 8 and 16 weeks, respectively). Flow cytometry and secondary transplantation confirmed that OD HPCs from BM is composed of long-term engrafting CD34⁺CD38^-CD45RA^-CD90⁺CD49f⁺ HSCs. Linear regression identified no meaningful predictive associations between selected donor-related characteristics and OD BM HPC quality or yield. Collectively, these data demonstrate that cryopreserved BM HPCs from deceased organ donors is potent and functionally equivalent to living donor BM HPCs and is a viable on-demand graft source for clinical HCT. Prospective clinical trials will soon commence in collaboration with the Center for International Blood and Marrow Research to assess the feasibility, safety, and efficacy of Ossium HPCs from BM (ClinicalTrials.gov identifier NCT05068401).

Keywords: Bone Marrow; CD34; Cryopreservation; Murine Engraftment; Organ Donor; Vertebral Body.

Figure 1.

Comparison of deceased OD VB-derived OD HPC, marrow and LD-IC-aspirated BM. (A) Viability of total CD45⁺WBCs. (B) Viability of CD34⁺HSPCs. (C) Viability of CD3⁺ T cells. (D) Percentage of TNCs that are CD34⁺ HSPCs. (E) Percentage of TNCs that are CD3⁺ T cells. (F) Numbers of granulocyte macrophage colony-forming unit (CFU-GM) progenitors in 10⁵ whole BM cells. (G) Numbers of total CFU progenitors in 10⁵ whole BM cells. *P < .05, Welch 2-tailed ttest.

Figure 2.

Comparison of HSPC populations in the CD34⁺ cell fraction of OD HPC, marrow and living donor BM. (A) Representative gating strategy to define and enumerate HSPCs. (B) Long-term repopulating HSC populations, defined as CD34⁺CD38⁻CD45RA⁻CD90⁺CD49f⁺, in CD34⁺ populations selected from OD-VB and LD-IC BM. (C) Multipotent progenitor (MPP) populations (CD34⁺CD38⁺) in CD34⁺ populations selected from OD-VB and LD-IC BM. (D) CFU progenitor populations in selected CD34⁺ cells from OD-VB (black bars) and LD-IC (gray bars) BM. CFU-E, CFU-erythroid; BFU-E, burst forming unit-erythroid; CFU-GEMM, CFU-granulocyte, erythroid, macrophage, megakaryocyte; CFU-total, the sum of individual progenitors.

Figure 3.

Characterization of fresh (green circles) and post-cryopreserved (blue boxes) OD HPC, marrow. (A) TNC and CD45⁺WBC counts/mL. (B) Total and viable CD34⁺ HSPCs/mL. (C) Viable CD34⁺ cells as a percentage of WBCs. (D) Total and viable CD3⁺ T cells/mL. (E) Viable percentages of CD45⁺ WBCs, CD3⁺ T cells, and CD34⁺ HSPCs. (F) Mean ± SD and range CD34⁺ HSPC counts. Viable CD34⁺ cells/unit is the number of HSPCs in 65 mL stored in 250-mL blood bags. Units/donor is the number of bags obtained from a donor. Viable CD34⁺ cells/donor is the total yield of HSPCs per donor. Grafts/70 kg patient is the number of transplantations at 2 × 10⁶ CD34⁺ cells/kg that can be performed with the total CD34 cells yielded from each donor. (G) Frequency of viable WBCs positive for markers defining the indicated lymphocyte subsets. (H) Fractions of CFU subsets in fresh and cryopreserved HPC, marrow. The average total CFUs per 10⁵ BM cells plated is indicated. *P < .05; **P < .01; ***P < .001; ****P < 0.0001, 2-way analysis of variance with Sidak’s multiple comparison test. The total number of fresh samples was 250, of which 226 were analysed post-thaw. Six samples were analyzed in (G).

Figure 4.

Transplantation of cryopreserved human immunomagnetically selected CD34⁺ cells from OD HPC, marrow recovered from 2 donors (BM1 and BM2). Immunocompromised NSG mice were irradiated at 300 cGy, followed by injection of CD34⁺ cells at a dose of 5 × 10⁵ through the tail vein. (A) Percentage of human CD45⁺ cells in mouse peripheral blood at 8 weeks. (B-F)Analysis of BM at 16 weeks for percentage of cells expressing human surface epitopes for CD45 (B), CD34 (C), CD38 (D), CD33 (E), and CD19 (F). (G and H) Percentage of human CD45⁺ cells in peripheral blood (PB) (G) and spleen (H) at 16 weeks. (I) Comparison of total CFU in cryopreserved selected CD34⁺ cells. (J) Comparison of total human CFU in BM of mice at 16 weeks. (K and L) Secondary transplantations: human CD45⁺ cells in peripheral blood (K) and BM (L) at 16 weeks following irradiation and injection with whole BM from mice receiving transplantation with HPC, marrow CD34⁺ cells (10 NSG mice per group). *P < .05; **P < .01, ***P < .001, analysis of variance with Tukey’s multiple comparisons test.

Figure 5.

Summary results from linear regression analyses showing the donor-related variables found to be significantly associated with variation in OD HPC, marrow composition. A complete list of the donor-related variables tested in the regressions is provided in Supplementary Table S4. Means (intercept values) are represented by vertical lines. The left columns identify the donor-related variables that were significantly associated with processed cell outcomes. The right columns show the values of regression coefficients associated with those donor-related variables. The bars indicate whether a donor-related variable had a negative (red) or positive (green) impact (ie, raised or lowered the slope, respectively) on the outcome. The range is represented by the horizontal bar. Donor-related variables contributed significantly to OD HPC, marrow variation in CD34⁺ cell viability (A), CD3⁺ cell viability (B), and total CD3⁺ cell count/mL (C).