Introduction

The transplantation of hematopoietic stem cells (HSCs) obtained from bone marrow (BM), mobilized peripheral blood (MPB) and umbilical cord blood (CB) has become an established therapy to regenerate the immuno-hematopoietic system in cancer patients following ablative chemo/radiotherapy and in patients with non-malignant hematological disorders. An important question in this field is how to measure the number of hematopoietic stem and progenitor cells (HSPCs) in a cellular product prior to its infusion. This information can be used to assist in graft selection and, ideally, to prospectively predict the likelihood of successful engraftment. While numerous “potency” assays have been adopted for this purpose, investigators disagree about the value of the different methods used, and there remains a critical need for assay standardization. These assays are particularly important for CB transplantation because most allogeneic CB units currently stored in public banks contain low numbers of viable HSPCs and are thus unsuitable for transplantation, particularly of adults.^1,2 Data from the National Marrow Donor Program (NMDP) presented at the 2012 Cord Blood Symposium indicated that ~80% of currently stored CB units fall into this category. The need to improve the quality of the CB inventory should drive adoption of the most biologically informative and accurate potency assays but this will need to be balanced against the cost and practicality of performing each type of assay in a routine lab setting. Cord blood banks are facing increased global competition and cost pressures due to constrained funding. Clearly, continuing to expend increasingly limited resources to bank unsuitable CB units is not sustainable. The choice of assay can be used by CB banks to champion their products to make them more attractive for transplant physicians and parents. This in turn should increase profitability and improve the quality of the worldwide CB repository. This article reviews the assays that are most commonly used to measure the “potency” of hematopoietic cell grafts, with specific consideration given to what these assays do and do not measure, and summarizes key studies that demonstrate a positive correlation between these assay end-points and the rate and success of hematopoietic engraftment.

Hematopoietic Stem Cell Transplantation

All mature blood cells are produced by a small population of multi-potential hematopoietic stem cells that reside primarily in the BM in adults. In the late 1950’s, it was first shown that HSCs could be harvested from the BM of a donor, transplanted into a patient whose own hematopoietic system had been ablated by chemotherapy or radiation, and that the transplanted cells could regenerate the blood and immune systems.³ Stem cell transplantation is now a widely used therapy for a variety of malignant and non-malignant hematological disorders. HSCs can also be collected from “mobilized” peripheral blood following treatment with agents that stimulate the migration of stem/progenitor cells from the BM into the circulation (e.g. granulocyte colony-stimulating factor, or the small molecule drug Mozobil® [Plerixafor]). Umbilical cord blood has become recognized as a rich source of HSCs and is now often cryopreserved for future use by parents who elect to bank their child’s cells for autologous transplantation, or who donate the cells to a public bank for allogeneic transplantation.

Engraftment following HSPC transplantation is typically considered to be successful when the numbers of neutrophils and platelets in the circulation have recovered to a sufficient level that the patient no longer requires treatment with supportive drugs (e.g. human growth factors, antibiotics) or blood products (e.g. red cells, platelets). Typically, this is defined as >500 and 50,000 neutrophils and platelets per µL of blood, respectively. Successful treatment for the underlying hematological disorder (usually cancer) is defined by the duration of disease-free and overall survival.

Hematopoietic cell grafts are typically manipulated in different ways in the laboratory before transplantation. Common procedures for CB-derived grafts involve automated processing to reduce the overall volume to a size that can be more readily stored and infused. Removal of red blood cells is also advantageous as many are lysed during cryopreservation. Infusion of this erythrocyte lysate can increase the likelihood of renal and/or cardiotoxicity in the patient. T lymphocytes can be depleted from allogeneic grafts to reduce the possibility of graft-versus-host disease. Alternatively, T cells can also mediate beneficial graft-versus-leukemia effects that improve survival. Processed hematopoietic cells are often cryopreserved for future use. This procedure and the subsequent thawing and washing steps often result in reduced recovery of viable CD34⁺ cells and functional progenitors. In order to monitor the effects of these different manipulations on HSPCs, it is critical that appropriate assays be used to assess the functional potential of the final cell product for transplantation.

i) Phenotyping Assays:

The discovery in the mid-1980’s that most HSPCs express the cell surface protein CD34 facilitated a more direct measurement of primitive cells that mediate engraftment.⁴ CD34-positive cells typically represent approximately ~1-4% of BM, MPB or CB cells and can be readily enumerated by flow cytometry following staining with a CD34-specific monoclonal antibody. The advantage of the CD34 assay is that it can be completed in a few hours so the number of CD34 cells can be used prospectively to predict the engraftment potential of a stem cell product. The International Society for Cellular Therapy (ISCT; formerly the International Society for Hematotherapy and Graft Engineering [ISHAGE]) has established a protocol for staining cells with anti-CD34 antibodies and standardized kits are available commercially. The gating criteria are complex, however, and there remains widespread variation in how the assay is deployed. This has resulted in a considerable variation in the accuracy of CD34⁺ cell enumeration that is probably not widely appreciated.⁵ Indeed, the coefficient of variation of the CD34 assay is not significantly lower than that of the colony-forming unit (CFU) assay when colonies are counted manually.⁶ Numerous clinical studies have demonstrated a positive correlation between the number of CD34 cells in different hematopoietic cell grafts and the speed of hematological recovery; for an example see.⁷ However, the CD34 assay also has several disadvantages. Foremost, it is a phenotyping assay that measures a physical property of HSPCs as a surrogate assessment of their biological function. The physical properties of primitive hematopoietic cells, like CD34 expression, can change when cells are subjected to different manipulations, for example when they are maintained in culture. Second, in thawed cryopreserved cells, measuring the total number of viable CD34 cells alone will typically over-estimate the number of functional stem and progenitor cells. This is because the CD34 phenotyping assay used routinely in clinical laboratories does not identify cells that have been damaged by this process and that have begun to undergo apoptosis.⁸ Such cells will die within days after thawing and will not contribute to engraftment following transplantation. Viable and functional stem and progenitor cells typically comprise only a minority (~10-20%) of CD34⁺ cells, depending on the tissue.

ii) Functional Assays:

Human HSCs can only be definitively identified using a functional assay that directly measures properties that are relevant to engraftment, such as proliferation and differentiation. In a research lab setting, human HSCs are assayed by transplantation into genetically immune-compromised (e.g. NOD-SCID) mice. HSCs are measured retrospectively when human blood cells of multiple lineages are detected in the recipient mice at least 5 weeks after transplantation. Unfortunately, this type of xenotransplantation assay is expensive and impractical for routine use in a clinical laboratory. In this setting, an in vitro assay that measures the number of functional progenitor cells able to produce colonies of hematopoietic cells in methylcellulose-based culture medium supplemented with stimulatory growth factors offers the best alternative. The CFU assay has several important advantages over the phenotyping assays described above. First, it measures a functional property directly relevant to engraftment; i.e. the ability of a progenitor cell to divide and produce daughter blood cells of different lineages. Second, only viable cells that have not begun to undergo apoptosis are detected in the CFU assay. This ensures that the number of progenitor cells is not over-estimated as can happen with some phenotyping assays. Third, the CFU assay directly measures the number of clonogenic cells. In other words, unlike in the CD34 assay, in which on