|Home | About | Journals | Submit | Contact Us | Français|
Bone marrow-derived CD34+ cells are a well-characterized population of stem cells that have traditionally been used clinically to reconstitute the hematopoietic system after radiation or chemotherapy. More recently, CD34+ cells have also been shown to induce therapeutic angiogenesis in animal models of myocardial, peripheral, and cerebral ischemia. The mechanism by which CD34+ cells promote therapeutic angiogenesis is not completely understood, although evidence supports both direct incorporation of the cells into the expanding vasculature and paracrine secretion of angiogenic growth factors that support the developing microvasculature. Phase I and phase II clinical trials have explored the usefulness of CD34+ cells in the treatment of ischemic conditions in human patients. As the population of patients diagnosed with some form of ischemic cardiovascular disease expands, the need for more effective treatments also grows, especially in patients who are refractory to standard pharmacologic or revascularization treatment. As phase III trials begin, CD34+ cells will be definitively tested as a novel treatment for myocardial and peripheral ischemia. This review will discuss what is known about the CD34 antigen and the cells that harbor it, the preclinical evidence supporting the therapeutic potential of CD34+ cells in ischemic models, and, last, the current evidence for the clinical usefulness of CD34+ cells in the treatment of human ischemic disease.
The American Heart Association has estimated that cardiovascular disease (CVD) is present in 1 out of every 3 residents of the United States and is the primary cause of death in this country. The estimated cost of treating and managing this epidemic in 2010 was $503 billion (U.S.).1 Without a clear solution to this growing problem, the increasing demand for intervention has necessitated the development of new and more effective therapeutic approaches. One relatively novel approach has been to use multipotent, autologously derived stem cells to treat the various conditions that manifest themselves in patients who have CVD. One such stem cell, the bone marrow-derived CD34+ cell, is being evaluated as a means to repair the damage associated with CVD. This review will discuss 1) what is known about the bone marrow-derived CD34+ cell, 2) some of the preclinical studies evaluating CD34+ cells for CVD treatment, and 3) the usefulness of CD34+ cells in the treatment of cardiovascular disease in human beings.
The CD34 cell-surface antigen was first identified by using monoclonal antibodies that were targeted to a cell-surface marker common to many hematopoietic progenitor cells.2,3 The identification of CD34 on both circulating and resident bone marrow hematopoietic cells evolved into a convenient and relatively simple method for purification of the cells from human beings and thus opened the door to their development as a novel therapeutic strategy to treat conditions as divergent as cancer, diabetes mellitus, autoimmune disorders, and, most important for our discussion, ischemic CVD. The belief that the CD34+ cell could be useful in the treatment of CVD arose primarily from the fact that both “endothelial progenitor cells” and fully differentiated endothelial cells were found to express the CD34 antigen.4–9 The likelihood that these cells could eventually adopt an endothelial role suggested that CD34+ cells might be helpful in combating CVD by forming or contributing to the formation of new blood vessels from existing vascular structures (that is, angiogenesis), primarily in ischemic tissues. It was also believed that certain bone marrow-derived stem cells had the ability to terminally differentiate into cardiomyocytes and thus regenerate the myocardium after cell death associated with myocardial infarction (MI) or heart failure.10 Although current evidence supports both a role of transdifferentiation of CD34+ cells to cardiomyocytes11–14 and their ability to fuse with existing cardiomyocytes,15,16 it is fairly well established that CD34+ cells also have the ability to differentiate into endothelial lineage cells.4,17,18 The primary focus of research in our own laboratory has been on the known role of these cells in supporting the development of vascular structures as a means to improve blood flow in ischemic regions,19–22 and the balance of this review will focus on their use in the treatment of clinical conditions involving ischemia.
The CD34 protein is actually the canonical member of the CD34 family of proteins, which also includes podocalyxin and endocalyxin on the basis of conserved structural domains and genomic organization (reviewed by Nielsen and McNagny23 and by Furness and McNagny24). Although these proteins exhibit similar structure, CD34 is the only member of the 3 routinely used clinically for the identification of stem cells; therefore, our discussion will be limited exclusively to the structure, function, and clinical usefulness of CD34-expressing cells. Structurally, CD34 is a single transmembrane helix-containing protein whose extracellular N-terminal domain is much larger than its intracellular C-terminus. Of the 3 proteins in the family, CD34 happens to be the smallest. The N-terminal domain contains many serine, threonine, and proline residues that are heavily O-glycosylated and sialylated; in addition, it contains putative N-glycosylation sites. The extracellular portion of the protein has a cysteine-containing globular domain, while the C-terminus of CD34 contains many consensus phosphorylation sites and a PDZ-binding domain. The extensive post-translational modifications described above account for a much higher observed molecular mass (~90 kDa) than that predicted for the protein (~35kDa).23 Currently, there is no evidence to suggest that the CD34 protein dimerizes with itself or any of the other CD34 family proteins, although the details regarding its oligomerization with other molecules will surely expand once a more complete picture of its signaling capabilities is brought to light.
Despite their well-defined structure and their usefulness in identifying hematopoietic stem cells, the actual cellular functions of the CD34 antigen have remained relatively elusive. Of the many purported roles of the CD34 protein, evidence most strongly supports the following:
Although the specific function of the CD34 protein has not been well described, its expression profile in adults is known. Of interest is the fact that CD34 is highly expressed on vascular endothelium in most organs and is especially robust in small capillaries.6,9,32–41 Its well-established expression on hematopoietic progenitor cells has enabled their immunopurification from both blood and bone marrow, and that has permitted their expanding use in the clinical setting.
As mentioned above, the CD34 antigen was first characterized as a protein by identifying hematopoietic progenitor cells found both in the blood and in the bone marrow.2,3,42 Early in the investigation, it was shown that isolated CD34+ cells could effectively reconstitute the hematopoietic systems of lethally irradiated baboons7 and rhesus monkeys,43 which suggested that at least some of the CD34+ cells were multipotent stem cells. After a follow-up study showed that effective engraftment also occurred in human breast cancer and neuroblastoma patients,44 CD34+ became the major cell type used for human patients in need of hematopoietic reconstitution after myeloablative therapy.45
One of the major early drawbacks to the therapeutic use of CD34+ cells was their relatively low basal density in the circulation. Fortunately, the discovery of agents that help to mobilize these cells from their resident bone marrow niche into the systemic circulation (where they can be collected and purified) considerably strengthened the field. Treatment of human patients with granulocyte-colony stimulating factor (G-CSF), which is now considered the standard mobilizing agent,46 increases the number of CD34+ cells in the peripheral blood (Fig. 1). Another agent, granulocyte-macrophage CSF, has been tested for mobilization of cells in patients after MI.47 In addition, other agents—such as the CXCR4 chemokine receptor antagonist plerixafor (AMD3100),48 statins,49–51 erythropoietin,52 vascular endothelial growth factor (VEGF),53–56 and angiopoietin-157—have all been shown to effectively promote the peripheral mobilization of CD34+ cells. Advanced age and chronic CVD tend to decrease both the functionality and the total count of CD34+ cells.58–64 In an effort to counter this decline, numerous studies have evaluated the use of a mobilizing agent alone or in various combinations to improve the cellular yield in patients with clinical conditions that cause either poor peripheral mobilization of CD34+ cells or poor yield (as reviewed by Cottler-Fox and colleagues65).
Complicating the problem is the fact that many environmental factors appear to influence the density of circulating CD34+ cells in human patients. Health factors such as smoking and alcohol abuse negatively affect CD34+ cell levels.61,66 Circulating levels of CD34+ cells appear to serve as an indicator of cardiovascular outcome in many clinical situations: the circulating numbers of CD34+ cells are commonly inversely proportional both to the severity of the disease51,63,67,68 and to the age of the patient. This last finding suggests the existence of a natural time-dependent decrease in the angiogenic potential of CD34+-mobilized progenitor cells. Conversely, exercise and the improvement of cardiovascular health tend to promote higher levels of circulating CD34+ cells.69,70
In addition to naturally occurring events and pharmacologic induction, certain pathophysiologic ischemic conditions also acutely mobilize CD34+ cells from the bone marrow. Both myocardial ischemia71,72 and peripheral ischemia18 are known to stimulate endogenous CD34+ cell mobilization. Upon mobilization, these cells tend to target zones of ischemia where they are thought to promote angiogenesis either through their direct incorporation into newly developing blood vessels or through their secretion of angiogenic growth factors that stimulate local peri-endothelial vascular development. Although the exact mechanisms by which this occurs are still being defined, abundant evidence supports the beneficial effect of these cells in the repair of ischemic tissue.
The use of autologous cells avoids the problems associated with transplantation of donor cells, as well as the potential toxicity of medications now in use for immunosuppression. Years of autotransplantation experience in hematologic and immune-disease settings have given us a defined safety profile for autologous CD34+ cells, and that—combined with their demonstrated efficacy in preclinical models of ischemic disease—makes the use of autologous stem cells an attractive option for tissue repair in ischemic diseases.
Although hematopoietic stem cells can be isolated on the basis of their CD34 antigen expression, it is important to clarify the fact that not all CD34-expressing cells are stem cells. Cells that harbor the CD34 antigen are in general considered to be relatively multipotent; of these, the hemangioblast is the most potent progenitor. As discussed earlier, endothelial cells also express CD34 and do not seem to possess inherent expansive potential, although it is possible that this viewpoint will change as we learn more about the plasticity of cells once thought to be terminally differentiated. It might well be discovered that the therapeutic benefit of using the heterogeneous population of isolated CD34+ cells arises from the synergistic effects of the many cell types that constitute such a population. As we improve our ability to effectively separate the various cell fractions and to identify the distinct functions within each fraction, we might become better able to target specific disorders with specific (and appropriate) cell types. The next 2 sections review both the preclinical data that suggested the use of human CD34+ cells to treat certain cardiovascular ischemic diseases and the human studies that have evaluated the usefulness of CD34+ cells in the treatment of ischemic conditions.
The multipotent nature of CD34+ cells was first established when it was shown that isolated CD34 expressing hematopoietic cells could fully reconstitute lethally irradiated baboons.7 This finding led to the cells' clinical use for hematopoietic reconstitution in patients undergoing radiation and chemotherapy treatment for various forms of cancer. The repertoire of the CD34+ cell expanded beyond oncology after the identification and characterization of circulating endothelial progenitor cells,4 a cell type that is often defined by its CD34 expression.
Since the appearance of the original report describing the angiogenic potential of the endothelial progenitor cell,4 many groups have used CD34+ cells in animal models of peripheral, cerebral, and myocardial ischemia as a method to reestablish blood flow to underperfused tissues, presumably through regeneration of the blood-supplying vasculature. We will focus our preclinical and clinical reviews of the therapeutic usefulness of CD34+ cells on the 3 major forms of tissue ischemia listed above.
The first attempt to use CD34+ cells for the treatment of a peripheral vascular disorder was described in the seminal report that first characterized the endothelial progenitor cell. Asahara and colleagues4 used a murine model of hind-limb ischemia (in that instance, via surgical resection of a segment of the femoral artery) to show that chemically labeled human CD34+ cells integrated into new capillaries that selectively formed only in the ischemic leg and not in the contralateral uninjured limb. The fate of a portion of the systemically injected cells was shown to be differentiation into endothelial lineage cells, documented by the co-localization of the label used to mark the injected CD34+ cell with other endothelial cell markers such as CD31 and UAE-lectin, indicating that the CD34+ cells had become fully differentiated endothelial cells.4 Although the report did not evaluate the functional benefit of the injected cells in terms of blood flow within the ischemic limb, the article did establish the groundwork highlighting the angiogenic potential of CD34+ cells.
Schatteman and colleagues73 were the first to use the hind-limb ischemia model to evaluate whether CD34+ cells could restore blood flow in the ischemic limbs of diabetic mice and nondiabetic control mice. Interestingly, they found that direct intramuscular injections of CD34+ cells augmented the recovery of blood flow only in the ischemic limbs of diabetic mice—not in those of nondiabetic mice.73 Although the increased response in the diabetic mice was striking, the failure to observe a CD34+ cell-based therapeutic response in the nondiabetic mice might be the consequence of their naturally rapid recovery from the surgery. Nearly 60% of pre-ligation blood flow was restored within 10 days after surgery, versus 40% for diabetic mice, so the therapeutic window for comparison might have been too small to resolve differences between treatment groups.73 In support of this argument are subsequent reports that clearly show CD34+ cell-mediated augmentation of postnatal angiogenesis in nondiabetic animal models of peripheral ischemia.74,75 A separate study revealed a dose-dependent effect of injected CD34+ cells into ischemic limb muscle, which produced hemodynamic recovery (as measured by limb salvage) and both capillary and arteriole density in the affected limbs.76 Although the study used a subpopulation of CD34+ cells, selected on the basis of kinase insert domain-containing receptor (KDR) expression, it was shown that as few as 103 to 104 CD34+/KDR+ cells were capable of producing significant therapeutic benefit.76 Last, a direct comparison of the angiogenic qualities of peripheral blood mononuclear cells (PBMNCs) and PBMNCs depleted of CD34+ cells revealed a significant reduction in therapeutic neovascularization and blood-flow recovery within the ischemic limb in treatments devoid of CD34+ cells.77 This finding once again suggests the potential of CD34+ cells to directly and effectively combat the clinical problem of critical limb ischemia.
The first evaluation of therapeutic cardiac neovascularization78 used athymic nude rats to show that a single systemic injection of human CD34+ cells after induction of an acute MI was sufficient to preserve cardiac function and reduce infarct size, collagen deposition, and apoptosis of cardiomyocytes. Presumably, these effects were the result of increased blood flow, a finding that agreed with the observation of CD34+ cell-induced angiogenesis and vasculogenesis in the infarct zones.78 The findings were specific to the CD34+ mononuclear cells (MNCs), because neither the CD34-negative cell fraction nor fully differentiated endothelial cells were capable of salvaging the infarcted tissue to the same extent.78
Although systemic injection is a relatively simple way to deliver the cells, a major drawback is that most of the injected cells fail to navigate to the infarcted tissue.79 Evidence from a study evaluating the terminal destination of 111-oxine labeled CD34+ cells in rats after intraventricular injection indicates that most of the cells localized to the liver, spleen, and kidney, although acute MI did stimulate a significant increase in localization to the heart.79 To circumvent this problem, our own laboratory evaluated direct injection of human CD34+ cells into the ischemic zone in a nude-rat model of acute MI and evaluated measurable factors associated with cardiac function.20 Histologically, the results indicated that animals in receipt of CD34+ cells showed a marked increase in capillary density, accompanied by a substantial decrease in the amount of fibrosis associated with the infarct. Histologic analysis revealed that CD34+ cells had integrated into foci of neovascularization located within the peri-infarct zone and that they expressed UAE-1 lectin, a marker of mature human endothelial cells—which is suggestive of differentiation into endothelial cells. This is in agreement with data from other groups.11 When evaluated functionally, local injection of CD34+ MNCs protected against the decreases in fractional shortening and regional wall motion seen after CD34-negative cell treatment and vehicle control.20 These results showed that CD34+ cells specifically diminished the global structural changes that occur after the induction of the infarct and showed, in addition, that local, site-specific injections could reduce the total number of cells needed to produce a beneficial outcome. This last point is extremely important, given that there is a finite limit to the number of CD34+ cells that can be collected from individual patients.
In addition, the therapeutic efficacy of mobilized, circulating human CD34+ cells has been compared to that of total MNCs (tMNCs) in a rat model of MI.19 Three treatment groups were compared: 1) a low-dose CD34+ cell group (5 × 105 cells/kg); 2) a low-dose tMNC cell group; and 3) a high-dose tMNC group, which contained the same absolute CD34+ cell dose as did group 1. Despite receiving the same absolute number of CD34+ cells, the high-tMNC treatment group produced increases in hemorrhagic MI as evaluated on postsurgical day 3. Tissue staining at that time point indicated an abundance of both hematopoietic and inflammatory cells derived from the xenotransplantation that were not found in the CD34+ cell group, which suggested that the tMNCs were responsible. The CD34+ cell group showed the greatest attenuation of structural changes attributable to the infarct, with the high-dose tMNC group showing an intermediate phenotype when compared with low tMNC or saline treatment.19 Overall, this report revealed a superior potency and therapeutic efficacy of CD34+ cells when compared with tMNC, and it further strengthened the case for the use of CD34+ cells in clinical settings.
The conclusions drawn from the studies outlined above agree with those from a nonhuman primate study that also evaluated the therapeutic efficacy of locally injected human CD34+ cells after acute MI.80 The authors showed that macaques that received intracardiac CD34+ cells showed improvements in regional blood flow and fractional shortening when compared with a saline-treated group. The mechanism for the improved functional response in these animals was suggested to be secretion of VEGF by CD34+ cells, which potentiated the observed angiogenic response.80
The exact nature of the beneficial course taken by CD34+ cells after acute MI and other ischemic conditions is still under debate. As mentioned, the 2 major hypotheses involve either the direct incorporation of injected cells into the newly developing vasculature or the production and secretion of angiogenic cytokines that support an ischemia-induced angiogenic response.81 In contrast to the studies discussed above, which indicate that CD34 cells can incorporate into newly formed vasculature, other studies have shown that, despite their migration to peri-endothelial regions, bone marrow MNCs fail to incorporate into the new vessels; rather, they promote vascular angiogenesis in a paracrine manner via secretion of well-established angiogenic factors.82–85 When one considers all the data in support of both hypotheses, it is probable that these mechanisms act synergistically to produce the observed outcomes. Yet further experimentation is clearly required to establish this idea as fact.
Although many studies have evaluated the short-term efficacy of CD34+ cell therapy, little is known about the long-term consequences of CD34+ cell treatment for myocardial ischemia. A recent paper86 has provided further insight by evaluating the timeline of the CD34+ cell existence once injected into the heart, in an effort to determine whether the well-characterized short-term functional benefits lead to sustained long-term functional improvements. Wang and associates86 retrovirally transduced CD34+ cells with a luciferase reporter and then locally injected the cells into the peri-infarct region of a severe-combined-immunodeficient mouse that had undergone coronary artery ligation. Repeated evaluation using bioluminescence tracking determined that the CD34+ cells remained localized within the heart for up to 52 weeks after injection. Micro-computed tomographic and micro-positron emission tomographic scanning further localized the bioluminescent signal to the left anterior ventricular wall, which showed that the cells failed to migrate significantly from the injection site. Last, magnetic resonance imaging was used to determine that an improvement in left ventricular ejection fraction (LVEF) related to the CD34+ cell therapy also persisted for up to 52 weeks.86 These findings highlight the potency of CD34+ cells as a therapeutic treatment: a single dose of cells was sufficient to promote sustained functional improvement for a whole year after administration. Although increased LVEF does not necessarily imply improved survival, these results do support the idea that these cells can become a valuable tool in treating ischemic disorders in human patients.
Recently, preclinical studies have been focused on developing techniques to increase the therapeutic efficacy of CD34+ cells in models of myocardial ischemia. These have included attempts to deliver angiogenic genes to CD34+ cells via transfection or viral transduction,87 the co-delivery of neovascularization-promoting gene therapy along with CD34+ cells,22,88 the isolation, in vitro expansion, and delivery of subsets of the CD34+ cell population in search of an optimized cell,89–91 and, last, the co-administration of binary RGDS-presenting nanofibers along with CD34+ cells in the hope that cells will more effectively adhere to a scaffold within the ischemic region of interest.92,93 As we come to learn more about CD34+ cells, their complex actions, and the various manipulations they can tolerate, we begin to realize that the most effective cell therapy for treating ischemic or hypoxic conditions associated with acute MI will probably use various combinations of all these new and exciting strategies. Most importantly, the information gathered from these preclinical studies has led to the development and implementation of human trials that test the usefulness of the CD34+ cell for treating problems associated with myocardial dysfunction, some of which are discussed in the last section.
In comparison with myocardial or peripheral ischemia, far fewer preclinical reports have evaluated the use of CD34+ cells in the treatment of cerebral ischemic conditions. A study by Taguchi and co-investigators in 2004 showed a negative correlation between the number of circulating CD34+ cells and the number of cerebral infarctions experienced by patients, which suggested that increased numbers of circulating CD34+ cells might provide protection against the occurrence of ischemic cerebral events.67 From this original observation, the hypothesis that CD34+ cells could be used as a therapeutic tool to treat acute cerebral infarction was then tested in animal models. The first study of this kind used a murine model of middle cerebral artery ligation to evaluate the therapeutic efficacy of systemically injected human CD34+ cells.94 The authors discovered that CD34+ cells accelerated the neovascularization of infarcted neuronal tissue and reduced the cognitive and behavioral deficits associated with the infarct, a finding not seen in animals treated with the CD34-negative fraction of cells or saline. After CD34+ cell treatment, the investigators also saw greater recovery of motor deficits and increased neurogenesis, in comparison with saline-treated control mice. The observed increase in vascular density and cerebral blood flow in CD34+ cell-treated animals suggested that accelerated neovascularization within the infarcted tissue was the mechanism behind the improved outcomes.94
After the systemic-injection study, a separate group undertook the task of locally injecting CD34+ cells directly into the infarct zone after the occlusion of the middle cerebral artery in rats.95 In agreement with Taguchi and co-investigators,94 Shyu and colleagues95 also determined that CD34+ cells promoted neovascularization and increased cortical blood flow within the infarct zone. Functionally, the superior blood flow in CD34+-treated rats was associated with a vast improvement of both motor and behavioral endpoints, when compared with control rats treated with saline. The injected CD34+ cells were found to differentiate into endothelial and glial cells, as well as neurons, which again emphasized the inherent stem-cell capacity of CD34+ cells.95 Presumably, the neuronal and glial fates of the injected cells were dictated by the strong neuronal cytokine and growth factor gradients found within the brain. Unfortunately, a more appropriate control treatment would have been CD34+ cell-depleted MNCs, especially given that negative outcomes associated with local injections of MNCs have been described in the MI model.19 Nonetheless, the strikingly beneficial effects of the injected CD34+ cells (in comparison with saline) further established their clinical usefulness for the treatment of cerebral ischemia.
Although CD34+ cells have been used for nearly 20 years as a means to reconstitute the hematopoietic systems of cancer patients undergoing radiation therapy, their use for the treatment of ischemic CVD is relatively novel. Arising from the preclinical studies described above, a substantial number of nonrandomized clinical studies96–101 have evaluated the therapeutic efficacy of unselected bone marrow cells in human patients who recently experienced MI. As a whole, these studies reported improvements in both left ventricular function and myocardial salvage in patients treated with unselected mononuclear bone marrow cells via intracoronary infusion. For example, intracoronary infusion of either tMNCs or circulating progenitor cells in the TOPCARE-AMI trial (which evaluated cell-based treatment in patients who recently experienced an ST-elevation MI and were acutely treated with revascularization via stent placement) produced statistically significant improvements in LVEF and left ventricular wall motion within the infarct region and also decreased end-systolic left ventricular volume at 4 months after infusion. They also reported a reduction in infarct size at the 1-year follow-up.96,98 Similar functional improvements have been seen in patients undergoing infusion of autologous bone marrow MNCs during coronary artery bypass grafting.100 Although these studies have shown promising results in regard to functional improvement, the use of heterogeneous MNC populations as the therapeutic agent renders it nearly impossible to discern which specific cell type mediates the effect. Despite this caveat, these studies used bone marrow cell fractions that definitely contained CD34+ cells (the CD34+ cell content in the infusions ranged from ~0.5% to 2.5%), so it is likely that at least a portion of the realized therapeutic benefit can be attributed to their function. Another published report102 found no functional benefit of intracoronary tMNC infusion in 5 patients with large anterior MIs, but the total cell number delivered was only one quarter of that used in the TOPCARE-AMI trial. Although this finding disagrees with those of the other studies, the small sample size and the possibly subtherapeutic cell dose could explain the negative findings. There also remains the possibility that co-injection of multiple cell types within the mononuclear fraction could result in a situation wherein the positive therapeutic effects of some cell types are diminished or eliminated by the negative therapeutic effects of others. These important issues will be clarified only when larger, randomized, double-blinded, controlled clinical studies are performed in which selected CD34+ cells are directly compared with subsets of unselected bone marrow MNCs in terms of their therapeutic efficacy in ischemic patients.
Although the use of heterogeneous bone marrow MNCs and some distinct subsets of bone marrow MNCs may explain why all reports have not been in complete agreement, randomized clinical trials21,103–107 have generally shown positive findings. In the REPAIR-AMI trial,103 204 patients underwent intracoronary infusion of bone marrow-derived progenitor cells or placebo after an acute ST-elevation MI that had successfully been reperfused with stent implantation. Overall, the study determined that infusion of the cells resulted in improved recovery of left ventricular contractile function as compared with placebo treatment.103 A separate randomized trial by the same group determined that cells derived from bone marrow aspirate produced significant improvement in LVEF after intracoronary delivery versus control treatment.107 Once again, the CD34+ cell content of these infusions (approximately 1% of the total cell dose) could theoretically be expected to play a positive role in the improved functionality seen in both studies, given their profound proliferative and therapeutic potential. Yet this does not rule out the possibility that non-CD34+ cells are also therapeutic.
In the BOOST trial, it was determined that intracoronary bone marrow MNC infusions (which contained at least 3 million CD34+ cells) accelerated the recovery of LVEF at 6 months after treatment. However, the single infusion of cells in the ST-elevation MI patients failed to maintain this functional enhancement at both the 18-month104 and 5-year108 follow-up time points, in comparison with the placebo control group. These findings suggest that additional infusions at defined intervals might be needed if the intracoronary delivery method is to be used. In addition, a separate study failed to show any left ventricular functional benefit of bone marrow-MNC infusion in ST-elevation MI patients, although the reasons for this discrepancy are not absolutely clear.105
Our own group undertook a randomized, double-blinded, placebo-controlled phase I/IIa clinical trial of autologous CD34+ cell treatment in patients with severe, advanced, inoperable coronary heart disease that was refractory to the best medical therapy.21 After mobilization of circulating CD34+ cells with a 5-day treatment of G-CSF, mobilized blood MNCs were collected with an apheresis system. A magnetic microbead selection process (Isolex 300i, Baxter Healthcare; Deerfield, Ill) enabled the isolation of pure CD34+ cells from the MNC fraction and, with the aid of a NOGA electromechanical mapping system, the CD34+ cells were then locally injected intramuscularly into the ischemic myocardium. The trial (unique identifier NCT00081913) tested whether direct cardiac injections of CD34+ cells could improve symptoms associated with intractable angina in 24 human patients and determined that patients receiving autologous CD34+ cells fared better than those receiving placebo treatment in the efficacy measurements tested. These values included the patient's frequency of angina, use of nitroglycerin, exercise tolerance, ranking on the Canadian Cardiovascular Society angina classification scale, and results of quality-of-life testing. Of importance was the study's provision of safety and feasibility data, because no new incidents of MI, congestive heart failure, or other major cardiovascular events were observed. This study established the foundation for the initiation of a phase IIb trial, which has just recently concluded a 2-year follow-up.109
The recently completed REGENT clinical trial110 compared intracoronary infusion of bone marrow-derived unselected MNCs with selected CD34+/CXCR4+ cells in patients with both acute MI and reduced LVEF (< 0.40). The study evaluated the primary endpoints (LVEF and LV volumes) before treatment and 6 months after treatment. Overall, the study did not detect endpoint differences between the unselected or selected cell-treatment groups, and both cell treatments showed an improvement over the no-cell-treatment control group. However, in a closer inspection of patients with severely reduced LVEF (< 0.37 before treatment), treatment with CD34+/CXCR4+ cells produced a level of improvement similar to that produced by unselected BMCs, albeit at a 100-fold lower cell dose. It should also be noted that follow-up evaluations of patients in both the unselected and selected cell-treatment groups were lower than expected (46 and 51 of 80 patients, respectively), and this might have limited the ability to resolve endpoint differences independent of the initial severity of LVEF.110
A recent meta-analysis of 18 randomized controlled trials attempted to define the long-term impact of progenitor cell therapy in the treatment of MI.111 Despite the inconsistencies among the individual studies (as described above), the analysis indicated that bone marrow cell therapy after MI results in more rapid improvements in systolic cardiac function, including LVEF and LV end-systolic and end-diastolic volumes, than in control groups; further, these improvements were sustained for 6 months after therapy.111 The authors also observed statistically and clinically significant benefits in the regional cardiac anatomy after cell treatment and showed that, in the baseline-impaired LVEF subgroup, LVEF was improved after bone marrow cell therapy when compared with the control treatment. Of importance is the fact that, as a whole, the cell therapy was safe, although there was no observation of a reduction in cardiovascular events. Last, subgroup analyses suggested that cell infusion after acute MI had a positive effect on LVEF.111 Together, these findings indicate that the delivery of CD34+ cell-containing treatments to infarcted and ischemic myocardial tissue is capable of promoting functional recovery of damaged myocardium.
In addition to the treatment of myocardial ischemic conditions, several other clinical trials have evaluated, or are in the process of evaluating, the use of CD34+ cell-containing therapies for peripheral, critical limb, and cerebral ischemia in human patients. One randomized controlled trial112 was designed to evaluate whether injection of bone marrow MNCs into the gastrocnemius of a critically ischemic leg was safe and feasible, and whether the procedure improved the ankle-brachial index and rest pain. Strikingly, in limbs injected with the cells, ankle-brachial index, transcutaneous oxygen tension, collateral vessel formation, blood flow, and pain-free walking time were all significantly improved in comparison with the saline-treated limb.112 This finding agrees with other reports of small trials of bone-marrow-MNC treatment of ischemia, including critical limb ischemia113–115 and hand ischemia.116
The first completed trial evaluating the implantation of purified CD34+ cells for the treatment of critical limb ischemia came from Asahara and colleagues in 2009.117 Although the trial was not randomized, 3 different CD34+ cell doses (6 patients received 105 cells/kg, 8 patients received 5 × 105 cells/kg, and 3 patients received 106 cells/kg) were evaluated in no-option patients with atherosclerotic peripheral arterial disease or Buerger disease with critical limb ischemia. The G-CSF-mobilized cells were injected intramuscularly into the gastrocnemius muscle in the leg of each patient who exhibited highly severe ischemia. The primary analysis involved evaluating the total walking distance on a standardized treadmill test, the toe-brachial index in the limb receiving the treatment, and the Wong-Baker FACES pain rating scale, all of which were evaluated at 4 and 12 weeks after treatment. Secondary endpoints included the Rutherford score, skin ulcer size, pain-free walking distance, ankle-brachial pressure index, and transcutaneous oxygen pressure. The 3 primary endpoints were converted to a treatment efficacy score, with positive numbers indicating improvement and negative numbers indicating impairment. All 3 treatment groups showed positive efficacy scores, which indicated that the treatments had reversed some of the ischemia, although no differences were seen between the treatment doses. Since no dose response was seen in the efficacy scores, all other endpoints were considered at 4 and 12 weeks after treatment, regardless of the dose regimen received. At 4 weeks after treatment, only transcutaneous partial oxygen pressure, total walking distance, and pain-free walking distance were significantly improved; but at 12 weeks after treatment, all endpoints except ankle-brachial pressure index were significantly improved. Some patients encountered difficulty in mobilizing sufficient numbers of CD34+ cells to qualify for the high-dose treatment arm, but the results of this study117 and at least 1 other21 indicate that higher numbers of cells do not necessarily result in better therapeutic outcomes.
The next step in the evaluation of CD34+ cells in the treatment of critical limb ischemia comes in the form of a larger, controlled, randomized and double-blinded phase IIb trial: the ACT34-CLI (critical limb ischemia) trial, run in conjunction with Baxter Healthcare, has just recently concluded its 1-year follow-up study, and the data should be forthcoming pending a comprehensive analysis.
Last, there are also some currently active and recently completed phase I/II trials (NCT00950521 and NCT00761982 at ClinicalTrials.gov) that are evaluating the use of isolated CD34+ cells as a means to ameliorate—via infusion of the cells into the middle cerebral artery—ischemia caused by acute or chronic stroke. A separate, recently concluded trial (NCT01019733) has evaluated the intrathecal delivery of CD34+ cells as a means to treat acute hypoxic or ischemic brain injury or cerebral palsy in pediatric patients. One other trial (NCT00535197) aims to determine whether intracerebral arterial infusion of CD34+ cells is effective in treating patients who have acute total anterior circulation syndrome. As the data from these trials become available, it will be interesting to see if the multifunctional nature of CD34+ cells can also be applied to treating neurologic disorders in human patients.
Currently, a phase III clinical trial of autologous CD34+ cell therapy for refractory angina is being planned. If successful, this approach would augment the current therapeutic armamentarium available for patients with advanced myocardial ischemia and would be used in standard clinical practice, alongside pharmaceutical treatment and mechanical revascularization. In a manner quite different from the development and testing of pharmaceutical agents or devices, autologous cell therapy presents unusual challenges in defining dose, potency, and endpoints for therapies that largely target quality of life as their initial indication. However, if the bioactivity of autologous cell therapies is verified in clinical trials, additional uses will undoubtedly be explored, and these could change the prospects of patients, affording hope not only for improved quality of life but perhaps for altering the natural history of the disease.
Address for reprints: Douglas W. Losordo, MD, Program in Cardiovascular Regenerative Medicine, Northwestern Memorial Hospital & Northwestern University, Galter 11–240, 201 E. Chicago Ave., Chicago, IL 60611