The potential of physiological regeneration offered by stem-cell-mediated therapy has captured the imagination of both the public and the scientific community. This is especially true for the cardiovascular and nervous systems, which have long been considered terminally differentiated and post-mitotic with minimal capacity for regeneration and repair.
While the ability of embryonically derived stem cells to generate multiple cell types has long been appreciated, stem cell biology began in the 1950s with the discovery of bone marrow progenitors capable of reconstituting the hematopoietic system. Remarkably, the discovery of endothelial progenitor cells (EPCs)1 capable of homing to sites of ischemia was made only a decade ago, nearly 40 years after the identification of the hematopoietic stem cell. Despite this, the use of EPCs for the treatment of vascular disease has accelerated from pre-clinical studies to clinical application within a short period of time, with resulting debate about the need for further pre-clinical work2 as opposed to acceleration of clinical trials in this field.3
More recently, attention has also been focused on other adult stem cells, including mesenchymal stem cells derived from bone marrow stroma, myoblasts derived from autologous peripheral muscle, and cardiac stem cells grown from myocardial biopsies. Each of these sources lacks the ethical constraints that define the field of embryonic stem cell research, and represents a vigorous area of pre-clinical and clinical research.
Evidence for Progenitor-cell-mediated Repair
The presence of circulating progenitor cells1 capable of differentiating into EPCs and homing to sights of ischemia where they increase neovascularization led to the realization that the state of the vasculature reflects a delicate balance between damage and repair.4 Several creative avenues of research have indicated that stem-cell-mediated repair is an active and ongoing process in both the myocardium and the vasculature. Pre-clinical and human studies demonstrated that foreign bodies implanted in continuity with the vasculature become endothelialized with cells of bone marrow origin.5,6 The potency of progenitor-cell-mediated vascular repair has been demonstrated in both atherosclerotic mice7 and in human radiation injury.8 The presence of gender-mismatched organs after cardiac transplantation has offered investigators the opportunity to probe myocyte generation by cells of host origin.9,10
Although the frequency and extent of myocardial renewal varies among studies, the common observation that a small fraction of myocytes contain chromosomes of host origin points to the existence of stem-cell-mediated innate myocardial regenerative capacity in humans. Additionally, vascular regeneration as determined by similar identification of host cell differentiation into mature EPCs and smooth muscle cells occurs at a higher rate, suggesting that progenitor-cell-mediated vascular repair represents a more active and dynamic process.10,11 This concept has also been explored after bone marrow transplantation, in which terminally differentiated smooth muscle, epithelial, and hepatic cells— this time of donor origin—were demonstrated.12,13
Cellular repair mechanisms may be relatively quiescent until end organ injury occurs. For instance, myocardial infarction (MI) may stimulate expansion of a resident cardiac stem cell population10 or recruitment of circulating cells capable of myocardial regeneration.14 Similarly, EPCs, when infused into ischemic animals, home to the site of ongoing vascular injury.1
The paradigm that the state of the vasculature may represent a balance between injury and repair suggests that if reparative capacity remains adequate, clinically evident disease may remain quiescent. Clinically evident disease would then be heralded by exhaustion of reparative mechanisms. Considerable effort has been expended exploring the relationship of circulating EPCs to cardiovascular risk factors and the extent of coronary artery disease (CAD) noted at time of angiography, and as a marker for risk stratification of patients for future cardiovascular events.
The pioneering work of Vasa15 and Hill16 was the first to demonstrate an association of EPC numbers with the numbers of cardiovascular risk factors and with endothelial function as assessed using determinations of brachial reactivity. These findings were extended to studies assessing EPCs and the extent of CAD,17,18 with Kunz et al. demonstrating that only patient age was more predictive of advanced coronary disease. Ultimately, the usefulness of EPC as a prognostic tool is dependent on the strength of its association with future cardiovascular events; this is the focus of two studies that suggest that EPC measurement, should it become readily available, might serve such a purpose.19,20
Clinical Investigation of Cellular-mediated Repair
Attempts to augment cellular-mediated repair for the treatment of cardiovascular disease include an assortment of clinical experiments targeting a variety of conditions, strategies for cell delivery and/or mobilization, and cell types. Targeted conditions include advanced ischemic coronary disease, congestive heart failure due to systolic dysfunction, myocardial salvage after acute infarction, peripheral vascular disease, and conduction system disease, each of which has been treated with a variety of cell types and cell delivery strategies. Here we summarize the fields that have undergone the most study and which therapies are most advanced.
Acute Myocardial Infarction
The cornerstone of modern therapy for acute MI is rapid and complete restoration of blood flow to the affected myocardium. Despite emphasis on educating the public about the signs and symptoms of acute infarction, delay in presentation to medical attention after symptom onset, the incomplete rate of restoration of normal coronary flow using thrombolytic therapy, and the delay inherent in mechanical reperfusion strategies, each leads to significant prolongation of ischemic time. Since myocardial cell death begins after only 40 minutes of ischemia, significant myocardial damage frequently cannot be avoided.
Human trials to date have largely focused on unselected bone marrow administration,21–29 with scattered reports exploring the use of selected bone marrow cells30,31 or mesenchymal stem cells.32 While some studies have focused on patients remote (>14 days) from MI,32–34 most have administered stem cells from three to six days after acute infarction. This time window has been targeted as a result of both practical and theoretical considerations, as acute cell infusion during infarction may lead to enhanced inflammation and myocardial injury. In addition, the human experience differs from animal models in that stem cell administration has been performed almost exclusively using intracoronary administration, while most pre-clinical models have employed intramyocardial injection. Collectively, these trials encompass cell administration to more than 1,000 patients. As the number of patients in randomized trials has accrued, preliminary conclusions are being compiled from assimilated data from these multiple, but relatively small, clinical trials.
Review of these trials highlights several concerns that cellular therapy trials must confront. For instance, demonstration of efficacy requires a control group in which both patients and their physicians are adequately blinded, especially given the power of placebo effects in patients with limited therapeutic options.35,36 As stem cells are most frequently obtained from autologous sources, placebo controls require invasive harvesting of cells. This procedure has significant potential complications, especially when performed on patients in the setting of an acute illness such as acute infarction. Complete blinding additionally mandates a ‘sham’ cell administration procedure. To date, at least four randomized clinical trials of bone marrow administration in the setting of acute infarction have met these criteria,32,37–39 setting a precedent for incorporating such a design into clinical trials in this field. Relatively small clinical studies have demonstrated both positive32,33,39–41 and negative28,37,42,43 results. Two systematic reviews of these studies have arrived at similar conclusions:44,45
- Myocardial administration of stem cells appears safe. Despite early studies suggesting that intracoronary delivery may accelerate restenosis,46 this has not been borne out in larger studies or in meta-analysis. Furthermore, while studies to date have not been powered to demonstrate an improvement in hard cardiovascular end-points, one trial has shown an improvement in major adverse cardiac events (MACE) after one year of follow-up, suggesting that this therapy may significantly improve clinical outcomes.47
- Myocardial delivery of bone marrow progenitor cells results in a measurable improvement in ejection fraction of several percentage points compared with no therapy or with the delivery of sham product. Cell delivery results in trends toward lower left ventricular volumes, possibly by preventing remodeling after infarctions.
- Although individual studies have suggested that bone marrow delivery may be particularly effective when administered late after infarction39 or when given in greater quantities after specific cell isolation,48 systematic overviews have failed to demonstrate consistent findings in this regard. Indeed, meta-analyses have failed to demonstrate any consistent interactions between cell dose, trial design (randomized controlled versus cohort studies), timing of cell administration after infarction, or the numbers or types of cells delivered. However, it should be noted that despite the substantial number of patients in these overviews, discrepancies between studies in terms of methods of cell preparation and delivery, overall trial design, entry criteria, target patient population, and cell dosing and timing of delivery make meaningful conclusions about interactions especially difficult to demonstrate.
These small studies have led to a call from national US authorities for large, multisite clinical trials addressing the efficacy of cell therapy for acute MI.49 Currently, over 10 clinical trials are listed with clinicaltrials.gov assessing bone marrow therapy in this setting, although the majority remain small with disparate strategies.
Cytokine-mediated Stem Cell Mobilization after Acute Myocardial Infarction
Cytokine stimulation results in the rapid and predictable marginalization of stem cells from the bone marrow compartment into the peripheral circulation. Use of cytokines such as granulocyte colony stimulating factor (G-CSF) for stem cell mobilization represents an alternative strategy obviating the need for cell harvest and local administration. In addition, this option appears attractive as the continuous movement of stem cells throughout the circulation system would theoretically result in the continuous exposure of ischemic tissue to a far greater total number of stem cells than would be predicted on the basis of local stem cell administration.
Unfortunately, several randomized placebo-controlled trials failed to demonstrate an improvement in ejection fraction using this strategy.50–52 Although one small study suggested that early (<2 hours) administration of G-CSF may prove effective, enthusiasm for this waned, partly due to an increase in inflammatory markers associated with G-CSF administration, which is perhaps responsible for the propensity of G-CSF to promote the acceleration of angina.53–55
Chronic Ischemic Heart Disease
Despite significant advances in revascularization technologies, a substantial population of patients experience significant lifestyle limitations due to angina from unrevascularizable coronary disease. Through no lack of effort, therapies for these patients remain limited, in part due to rather disappointing results with novel treatments such as transmyocardial revascularization, enhanced external counterpulsation, and angiogenic gene therapy.
The demonstration that EPCs home in on areas of ischemia and stimulate new blood vessel formation immediately identified EPC cell therapy as a promising treatment modality. Significant pre-clinical data have led to clinical studies of CD34+ cell administration for refractory angina. These trials are significant in that relief of angina has now been accepted by governing agencies as a primary efficacy end-point for therapy approval.
Recently reported phase I data have demonstrated that intramyocardial administration of EPCs (CD34+ cells) is safe, with no incidence of procedural or arrhythmic complications noted. While this trial was not powered to show statistically significant improvements in efficacy, angina frequency, nitroglycerin usage, exercise time, and Canadian Cardiovascular Society (CCS) angina class all showed trends that were favorable in patients receiving cell therapy compared with placebo patients.56 Currently, a phase II study is enrolling with plans for expanding to a larger pivotal study, should results remain favorable.
Chronic Heart Failure
Recent emphasis on rapid and complete restoration of blood flow after acute coronary occlusion has led to marked improvements in survival after acute infarction, but many of these patients go on to develop significant myocardial injury leading to progressive left ventricular (LV) dilatation and the development of congestive heart failure. Deaths from heart failure have increased by 148% over the last 20 years, with about 400,000 new cases diagnosed each year. Furthermore, the one-year survival rate for heart failure is worse than that for breast, prostate, and bladder cancer.
The search for cells capable of restoring myocardial function represents the apogee of cellular therapy, i.e. the regeneration of tissues that have limited conventional reparative capacity. Despite the discovery of ongoing stem-cell-mediated repair in the myocardium and the identification of resident cardiac stem cells capable of in vitro myocyte differentiation and expansion, the fact remains that myocardial regeneration after infarction is essentially clinically insignificant.
However, pre-clinical models in more primitive organisms suggest that stem-cell-mediated repair might be harnessed to effect significant myocardial regeneration. For instance, zebrafish withstand removal of >20% of myocardial tissue, and respond with a brisk and complete reparative response with complete regeneration of the lost myocardium.57 Of course, it remains to be determined whether such reparative capacity resides within more complex organisms such as humans.
To date, most clinical work in this area has focused on the use of skeletal myoblasts obtained from autologous peripheral muscle. Skeletal muscle myoblasts offer an autologous source of expandable cells, ease of harvest, in vitro expansion, and the ability to readily grow into functional contractile cells. Early uncontrolled studies utilized cell injection in the setting of incomplete revascularization during coronary artery bypass grafting, with cell injection into areas of unrevascularized scarred myocardium. Soon after cell administration, significant arrhythmias necessitating automatic implantable cardioverter–defibrillator (AICD) implantation were observed,58,59 thought to be due to the inability of myoblasts to be effectively incorporated into a cardiac syncytium with uniform electrical properties.60
Recently, larger controlled studies have been limited by slow enrollment, leading to premature termination of the Myoblast Autologous Grafting in Ischemic Cardiomyopathy (MAGIC) study. While failing to demonstrate an improvement in ejection fraction in treated patients, this study alleviated concerns about myoblast arrhythmogenicity while suggesting that myoblast administration may improve LV remodeling.61
This field will now be advanced by an international multisite trial of percutaneous myoblast administration to areas of non-contracting myocardium. The Multicenter Study to Assess the Safety and Cardiovascular Effects of Myocell Implantation (MARVEL) study will enroll 330 patients at 30 international sites with previous infarctions and an akinetic wall motion abnormality. The trial will assess the safety and efficacy of myoblast administration, with particular emphasis on the safety end-points of arrhythmias and all-cause cardiovascular death and rehospitalization. To ensure safety, all patients will have AICDs implanted prior to enrollment in the trial.
Ischemic cardiomyopathy has also been targeted with bone marrow cell therapy. In the only randomized trial, 102 patients well removed (>3 months) from MI were randomized to receive either an intracoronary infusion of autologous bone marrow cells (n=51), an intracoronary infusion of cultured peripheral blood cells (n=35), or no cell infusion (n=16). Ejection fraction was marginally improved in the patients receiving bone marrow cells (increase of 3%), while no change was seen in the other groups. Atrial natriuretic peptide (ANP) serum levels decreased in both of the cell therapy groups, but not in the control groups.62
Other Stem Cell Types
Other multiple stem cells remain in pre-clinical development or in early clinical trials. Mesenchymal stem cells (MSCs) are multipotent stem cells obtained from bone marrow stroma. MSCs offer many properties that make them particularly attractive—including immuno-privilege— bypassing the requirement for autologous cell harvest and allowing administration of pre-grown cells from a healthy source.63 This may be vital, as bone marrow cells in patients with extensive vascular disease may be functionally impaired, limiting the effectiveness of autologous sources in those patients most in need of effective cellular therapy.
MSCs are easily isolated, cultured, and expanded, and can be used to generate a large supply of a homogeneous cellular product for future use.64 The vast differentiation capacity of these cells may broaden their use to a variety of conditions, and their intravenous administration allows for a minimally invasive approach to cellular delivery, obviating some of the ethical concerns previously discussed. A phase I double-blind, randomized, placebo-controlled, dose-ranging study exploring MSC administration as adjunctive therapy for MI in 53 patients (34 of whom received stem cells) demonstrated a trend toward improved ejection fraction in the treated patients, with lower rates of MACE.65
Cardiac stem cells have been identified and expanded from myocardial biopsy specimens and may generate mature myocytes with a greater propensity to properly electrically integrate into native myocardial tissue, not only minimizing electrical instability but also offering the possibility of generating a functional contracting tissue as well as an electrical syncytium with native myocardium.12,66,67 Their utility may be limited by their small numbers. Whether these cells are functionally or numerically depleted in patients with heart failure or chronic ischemic disease also remains to be determined.
As in any new therapeutic quest, the initial iterations of proposed cell therapies are likely to be the most primitive and least efficacious. The primitive nature of this field is attested to by the almost complete lack of any studies comparing treatment strategies. Until such studies are conducted, we will understand little about the best cell types, delivery methods, and timing of cell therapy for each treatment target. The adequacy of autologous sources is especially concerning, given that the development of ischemic heart disease may reflect impairments in cellular repair mechanisms and bone marrow progenitor cell exhaustion.7 Indeed, bone marrow cells obtained from patients with ischemic cardiomyopathies have been shown to lack migratory and proliferative capacities found in those of normal patients, raising concerns that autologous cell sources may lack adequate reparative capacity unless the etiology of these deficits can be identified and overcome.68
Cellular therapy offers the promise to harness nature’s fountain of youth, namely the regenerative capacity of intrinsic repair systems. Early results for the treatment of MI appear promising; however, attempts at cellular myocardial regeneration are less advanced. While early clinical reports have been encouraging, it should be recognized that other lines of investigation were accompanied by similar euphoria before more carefully controlled trials were undertaken. Requirements needed for advances to be made are not only carefully designed randomized, blinded trials with attention to safety, but also significant effort to better understand the mechanistic underpinnings of these therapies. In this regard the number of questions facing the field is daunting.
To overcome these substantial hurdles, the field will be best served by carefully designed studies with attention to randomized, blinded design, close monitoring for arrhythmias, precise non-invasive measurements, perhaps using cardiac magnetic resonance imaging (MRI) to assess the effects of therapy on extent of myocardial damage, recovery of LV function and LV dimensions, and ischemic burden, with long-term follow-up to both determine efficacy and exclude the possibility of a higher future risk of malignancy or long-term safety concerns with cellular administration.