Stem Cell Treatment of the Heart

Abstract

Stem cells are multipotent, undifferentiated cells capable of multiplication and differentiation. Preliminary experimental evidence suggests that stem cells derived from embryonic or adult tissues (especially bone marrow) may develop into myocardial cells. Some experts believe that this phenomenon occurs naturally in human beings, specifically during recovery from a myocardial infarction. Recently, stem cells have been used with the therapeutic intention of regenerating damaged tissues. Cardiac experiments, mainly with adult homologous stem cells, have proved that this therapy is safe and may improve myocardial vascularization and pump function.

We review current fundamental concepts regarding the normal development of embryonic stem cells into myocardial tissue and the heart as a whole. We describe the multiple conditions that naturally enable a stem cell to become a myocardial cell and a group of stem cells to become a heart. We also discuss the challenge of translating basic cellular and molecular mechanisms into effective, clinically relevant treatment options.

In presenting the Simon Dack Lecture at the 2004 meeting of the American College of Cardiology, Dr. Victor Dzau made a statement that is repeatedly heard by the cardiology community: “Cell and molecular therapies will soon enable the repair and regeneration of myocardium after myocardial infarction.”¹ This message suggests that cardiologists can now eliminate the frustrations of contemporary heart failure treatment by creating new contractile units in the damaged heart or even by creating a “cloned heart” from stem cells. Now that we are able to effectively treat arrhythmias with pacemakers and implantable defibrillators,² cardiac valve dysfunction with artificial valves,³ and coronary obstructions with surgical bypasses⁴ or drug-eluting stents,⁵ heart failure is the main unsolved problem in cardiology. Because homograft transplants are costly, dependent on a scarce donor-organ pool, and vulnerable to complications, they will never be adequate for widespread treatment of end-stage heart failure.⁶ Stem cell therapy is part of a new alternative called “regenerative medicine,”⁷ which is emerging as a multidisciplinary field involving molecular and cell biologists, embryologists, pathologists, clinicians, bioengineers, and ethicists.

Background

The use of stem cells in cardiology is frequently characterized as a matter of providing new myocytes, but it is much more complex than that. Whether global or segmental, heart failure is generally due to a specific cause, which must be removed as a precondition for the success of any reconstructive effort. For example, in a case of myocardial failure resulting from coronary occlusion (the most frequent cause of heart failure), cell treatment would be unlikely to yield a viable, functional myocardium if the normal blood supply were not also restored. Likewise, the mere generation of new vessels (by means of angiogenesis or vasculogenesis)^8,9 distal to the culprit coronary occlusion would hardly improve a restricted global blood supply (and would surely fail to compensate for cardiomyocyte loss). In other words, achieving an even distribution of an insufficient blood supply, as evidenced by increased capillary density, cannot be expected to be an adequate solution.

Even more important, unlike the progenitor cells used in bone marrow transplants, the elementary myocardial functional units are not lone cardiomyocytes but, rather, are myocardial cells that are integrated into a multicellular assembly of myofibers. These cells are oriented in specific directions (indeed, implanted cell therapy should avoid generating myofiber disarray, which is a disease state in itself). The elementary myocardial cells are connected with functional intercalated disks that integrate individual electrical activation and contraction into a pumping action. The myocardial cells are also 1) nourished by a capillary vascular network, which is connected with an arterial network that should provide variable nutrients and normal coronary blood flow; 2) activated by an integrated Purkinje fiber system that should produce rapid, orderly electrical activation while preventing reentry circuits or independent spontaneous pacemaker activity; 3) innervated by sympathetic neurofibers; and 4) located within a fibrous skeleton of functionally adequate geometric dimensions (an aneurysm, seeded with myofibers in the absence of proper remodeling, would yield little functional improvement). Therefore, the challenges of stem cell treatment for the heart are much more complex than those of blood transfusion for anemia and bone marrow transplantation for bone marrow failure, which are the only clinically successful cellular treatments thus far.

In considering these challenges, we will review 2 main relevant fields of basic investigation: 1) normal development of the heart in embryos and 2) evidence of the physiologic production of new myocardial tissue in adults. We will also re-evaluate the results of early experiments with adult and embryonic stem cells in animals and in adult human beings.

Normal Development of the Embryonic Heart

The normal embryonic heart is the product of at least 3 (perhaps 4) cellular populations:

1. The so-called cardiogenic or primordial cardiac mesoderm, which generates not only myocytes, but also fibroblasts, smooth muscle fibers, and some endothelial cells.

2. The pro-epicardial organ, which generates the epicardial coronary arteries and likely at least some of the endothelial and smooth muscle fibers.

3. The cardiac neural crest population of the primitive ectoderm, which is essential to the development of the media and to the septation of the great vessels.

4. Possibly blood-borne embryonic stem cells, which may contribute to cardiac development.

The development of the embryonic heart is a complex process, in which a crucial basic step is primordial cell determination. Some of the early embryo's undetermined (multipotent) cells, which are initially located within the lateral mesoderm (splanchnic plate) in paired symmetrical areas, migrate and fuse at the midline; this process is influenced by an instructive, inductive interaction with the anterior endoderm^10,11 and is likely related to the production of endogenous paracrine factors.^12–15 The existence of both proactive and inhibiting factors has been assumed or documented.^16,17 These factors include bone morphogenetic proteins (BMPs) and fibroblast growth factors (FGFs) on the one hand and the Wnts inhibiting factor on the other hand.¹⁸ Myocardial cell phenotypes are signified by gene expression of a number of specific, unique transcription factors (and protein molecules) that appear during embryonic development, in consistent temporal and spatial sequences, and that can be used as diagnostic markers.^17–20 The following cardiac transcriptional proteins seem to regulate the specialization of precontractile heart cells: Nkx-2.5; GATA-4, 5, and 6; Tbx5; serum response factor (SRF); myocardin; and the myocyte-specific enhancer binding factor (MEF 2c).^13,17

Experimental embryology data have recently shown that extracardiac mesoderm does not normally produce heart tissue but can do so when transplanted into a specific cardiac extracellular medium.^13,15 Likewise, extracardiac posterior mesoderm does not normally produce heart tissue but will express a cardiac phenotype when treated with extracellular regulators such as BMPs, FGFs, Wnt, and Dickkopf 1.^13,15,17 This phenomenon suggests that the embryonic cardiac jelly, or extracellular matrix, is a critical carrier of signaling instructions and enables the cellular migration necessary for the acquisition of proper intercellular spatial and functional organization.13,15,17,20,21 For the formation of definitive cardiac chambers, myocardial specialization (into atrial, ventricular, and conduction elements) is required.^16,17,20 This necessitates the influence of several genes, such as those that encode transcription factors Hand 1, Cited 1, and Irx 1/2/3; gap junction protein connexins 40 and 43; the secreted peptide atrial natriuretic factor; and the cytoskeletal protein Chisel.13,16,17,20 In addition, it has been proved that cells located in the cardiogenic areas during Hamburger-Hamilton stages 4 and 5 (which do not normally give rise to hemapoietic cells) produce differentiated blood cells when experimentally transplanted into a hematopoiesis-promoting environment, thereby suggesting that the precardiac mesoderm is still multipotential.¹³ In summary, primitive embryonic cells exhibit a wide spectrum of cell plasticity under experimental conditions.^14,22

Adult Stem Cells in the Normal and Abnormal Heart

Almost single-handedly, Anversa and colleagues^23–34 have raised new interest in the capacity of adult cardiac myocytes to replicate naturally. These investigators have proposed the innovative concept that, contrary to previous assumptions, the adult human heart is not terminally differentiated but has a significant population of stem cells capable of reproducing and differentiating into myocytes. Initially, by investigating pathologic anatomy, the Anversa group demonstrated that the adult human myocardium maintains some degree of reproductive-cell capacity: 0.015% of the myocytes found in histologic studies of failing hearts^25,27,29 and 0.08% of those found in infarcted hearts²⁶ were observed during mitosis.

Further, by using advanced histologic methods to identify the host Y-chromosome (from clinical biopsy or necropsy tissue) in sex-mismatched human cardiac transplants, researchers may gain a unique opportunity to observe whether any host male cells (carrying a Y-chromosome), especially those with myocytic features, can develop within the female donor heart.³⁵ Both early and late investigators of cardiac transplantation eventually acknowledged that the incidence of cardiac chimerism is highly variable (0% to 30% in different reports), and decreases with the age of the graft.^35–42 These findings in gender-mismatched transplanted hearts strongly suggest that the host's circulating stem cells can migrate and home into the transplanted heart, forming new myofibers.^35,43 This suggestion is further supported by evidence of a circulating pool of cells that express the cardiac lineage markers indicated above (for example, Nkx2.5, GATA-4, and MEF2C), which were derived from a nonadherent population of postnatal bone marrow cells that did not express hematopoietic markers.⁴⁴ Moreover, mobilization of these cells from the marrow to the myocardium was enhanced by infarction and chemoattractants, suggesting that this nonhematopoietic population of bone marrow cells represents cardiac progenitors that are potentially important in myocardial regeneration.⁴⁴

Alternatively, cell fusion may occur (with host nuclei penetrating the donor cells), or host inflammatory or stem cells may adhere closely to the donor myocytes, especially in the presence of allograft rejection.^45–50

These discordant data regarding the incidence and significance of chimerism after transplantation of sex-mismatched hearts have led to a pointed discussion about microscopic methods of identifying the Y-chromosome (conventional light microscopy versus fluorescent-labeling and confocal microscopy).^35,51,52 Using enhanced diagnostic methods, Hcht-Zeisberg and colleagues³⁸ recently re-evaluated such methods of identification. In autopsy samples from 5 men who had received female heterografts and died 1 to 28 days after an acute myocardial infarction, those authors identified Y-chromosome myocytes according to the following strict criteria: 1) fluorescence in situ hybridization (FISH) analysis, with simultaneous immunofluorescence microscopy detection of CD45 and CD68 (markers of inflammatory cells); 2) predefined morphologic criteria for cardiomyocytes, validated by immunofluorescence microscopy with cardiac-specific antibodies; and 3) 3-dimensional confocal microscopy to indisputably assign the nuclei to their cytoplasm. The authors were able to confidently determine that 80% of the cells initially deemed “host-derived cardiomyocytes” with fluorescence microscopy alone were eventually reassigned as “located outside the cardiomyocytes” (noninflammatory Y-chromosome progenitor cells adherent to the myocyte membrane). Only 0.21% of the thousands of examined cells were finally labeled as Y-chromosome myocytes in the postinfarction transplant patients; the control group of noninfarcted, sex-unmatched transplant patients had an even lower incidence (0.04%) of such myocytes. These observations suggest that only on exceptional occasions—especially after sustaining a myocardial infarction—can the donor heart produce recipient cardiomyocytes, probably from bone marrow–borne progenitor cells.

From their own research, Anversa and colleagues^28,29 concluded that, in addition to end-stage differentiated cardiomyocytes, other cells with multipotent capacity exist in the normal heart. These may be either resident cells, with the potential to replicate and mature into functional myocytes, or migrant bone marrow–derived multipotent cells, which continuously colonize the heart and can also produce mature cardiac cells.^28–30 The main basis for this conclusion seems to be an indirect one, from the same authors' investigations of apoptosis and necrosis in the normal adult heart: if there were no regenerative capacity, spontaneous cell death in the adult heart would have quickly resulted in atrophy of the organ.^29,30 From observations of rat hearts, those researchers calculated that approximately 94,200 myocytes could be lost during any 24-hour period.³⁰ Because the total number of myocytes in the rat heart is about 13 million, all of the cardiac cells would be lost within 5 months if regeneration did not occur.^28,30

In summary, the theory that the adult heart has an intrinsic regenerative capacity has not yet been fully proved (with regard to its extent and relevance), but it has stimulated the imagination of many investigators.⁵¹ Moreover, a recent molecular discovery appears to have authenticated the pioneering work of Anversa's group. Using Cre/lox technology, Laugwitz and coworkers⁴⁴ were able to specifically mark a resident population of cardiogenic precursor cells in rats, mice, and human beings that express islet-1: a gene initially iterated in early embryonic mesodermal cells that were clearly committed to a myocardial lineage.⁴⁴ Culture data indicated that islet-1⁺ cells isolated from postnatal hearts can proliferate and differentiate into cells that express myocardial markers and can generate an action potential without evidence of fusion. Whether the islet-1⁺ cells also express hematopoietic or nonhematopoietic markers of bone marrow–derived stem cells remains to be determined, but the presence of a resident population of cells that express a very early embryonic marker associated with cardiogenesis holds added potential for regenerative applications.

Evidence of Experimental Stem Cell Plasticity, with Cardiogenic Potential, in Adult Animals and Human Beings

Recent observations in the adult heart have suggested that adult cardiac and noncardiac stem cells, such as those obtained from the bone marrow, brain, skeletal muscle, adipose tissue, liver, or peripheral blood, may become cardiomyocytes after undergoing natural migration or experimental transplantation into the heart.^53–60 This evidence indicates that the presence of such cells in the adult extracellular cardiac environment (which may need to be ischemic or damaged for this phenomenon to occur) induces the maturation of cardiac phenotypes; thus, stem cell therapy might be effective for regenerating infarcted myocardium.25–43,45–61

Adult stem cell implantation for myocardial regeneration was originally performed in animals^54,55 but was soon extended to human beings. In one of the more optimistic animal studies, Orlic and associates⁵⁴ used a transgenic mouse model to identify the destiny of acutely implanted bone marrow cells that were injected directly into the border region of an experimental myocardial infarction. At 9 days, newly formed myocardium was reported as occupying 68% of the treated portion of the ventricle.

More recently, Murry and coworkers⁶¹ used a similar transgenic mouse line into which a nuclear-localized β-glycosidase reporter was inserted to monitor the differentiation of transplanted hematopoietic stem cells in models of myocardial infarction induced by liga-tion or cauterization. In these settings, the transplanted Lin⁻, c-Kit⁺ hematopoietic stem cells did not show evidence of transdifferentiation into cardiomyocytes.⁶¹ In another experiment with bone marrow transplantation in lethally irradiated mice, the same authors found only very infrequent new myocytes of donor origin in the peri-infarct area.⁶¹ This finding is consistent with that of a previous report by Jackson and colleagues,⁵⁵ who observed donor cardiomyocytes in 0.02% of the myofibers in the peri-infarct zone following similar bone marrow transplantation (involving side-population, hematopoietic stem cells) that was performed 60 minutes after coronary ligation.

Balsam and co-authors⁶² recently reported that 30% of the hematopoietic stem cells implanted in a mouse model of acute myocardial infarction were alive at 10 days; only 0.2% remained alive at 30 days, when the stem cells had developed mostly into mature myeloid or lymphoid cells but not myocardial cells.

Using an immunodeficient mouse model and molecular markers, Yeh and associates⁵⁹ reported that circulating CD34⁺ peripheral blood cells were able to engraft into the heart and blood vessels and transdifferentiate into cardiomyocytes, vascular smooth muscle, and other cardiovascular cell types. However, they found that engraftment was largely dependent upon ischemic injury. Importantly, on the basis of chromosomal markers, they found that CD34⁺ cells isolated from the peripheral blood were able to transdifferentiate directly into cardiac cell types. Although cell fusion was observed, it was not the basis for their observations of stem cell plasticity.

While importance of cell fusion,⁶⁰ the mechanisms of plasticity, and the requirement of injury, importantly, continue to be studied in animal models, several pilot investigations have been initiated in human beings to test the potential of stem cells to regenerate myocytes and functional cardiac tissue (including its vascular supply).^63–86

By early 2004, more than 150 persons had undergone some kind of “cellular cardiomyoplasty” with use of autologous adult stem cells for myocardial regeneration.⁶³ In Germany, Strauer and associates⁶⁴ tested the use of autologous intracoronary mononuclear bone marrow cells, administered 5 to 9 days after coronary angioplasty, which was performed within 12 hours after an acute myocardial infarction. Compared with a 10-member control group treated only by coronary angioplasty, the 10 stem cell recipients seemed to show increased segmental wall motion at 3 months.

Assmus and colleagues⁷⁹ confirmed this early clinical finding in a similar pilot series of 20 patients who underwent stem cell therapy 4 days after emergency stent-angioplasty for acute myocardial infarction. Either bone marrow cells (n = 9) or circulating blood-derived progenitor cells (n = 11) were injected into the culprit vessel, with functional re-evaluation at 4 months as an endpoint.

In the United States, Perin and collaborators⁸¹ were the 1st investigators to receive Federal Drug Administration authorization to perform a clinical pilot study designed to ascertain the safety of using bone marrow–derived stem cells for treating heart failure at hibernating myocardial territories. In their pilot human experiments, Perin's group⁸¹ considered the possibility that bone marrow stem cell therapy could promote angiogenesis, thereby accounting for a marked improvement in global ejection fraction (from 0.20 to 0.29) and a 75% decrease in areas of reversible ischemia. Those researchers used a catheter with a short needle to inject autologous bone marrow stem cells subendocardially with the aid of electromechanical mapping. The series included 14 study patients and 7 control patients, all of whom had a chronic myocardial infarction, an ejection fraction of less than 0.40, inoperable coronary obstructive disease, and nuclear scintigraphic evidence of reversible perfusion deficits. Other investigators58,63,69,82,83,85,86 have successfully injected an alternative type of autologous stem cell—skeletal myoblast grafts carrying cells⁸⁷—into myocardial scar tissue during coronary bypass surgery or catheter-based procedures.

The hypothesis that supplementation with cytokines can increase both the bone marrow release and the homing and differentiation efficiency of adult stem cells^88–95 has been tested in experimental models of infarction. A pilot study⁹³ using granulocyte colony stimulating factor and stem cell factor suggested that neovascularization, but not myocyte regeneration, could be enhanced by these methods.

Embryonic stem cells may also have a potential role in cell therapy.^14,96–101 Currently, however, these cells are available only in limited numbers, and their therapeutic use would likely introduce ethical and regulatory dilemmas,^102,103 as well as the risk of allograft immunologic reactions. Until now, it has not been possible to promote, in an in vitro culture, the development of a layer of pure cardiomyocyte lineages derived from embryonic cells. Nevertheless, early attempts using cardiomyocytes generated from marrow stromal cells have been successful in tissue cultures.¹⁰³ Nuclear transfer technology (therapeutic cloning)¹⁰⁴ might solve the problem of generating an adequate cell pool and eliminating the potential for allogenic rejection,¹⁰⁵ but this technology is still experimental,¹⁰⁶ unreliable, and expensive, and it may not solve the ethical objections to the use of embryonic cells.¹⁰⁷

Recent technical innovations introduced by Hwang and associates in South Korea¹⁰⁸ may dramatically improve the chances of realizing the embryonic cell option. These researchers' effective in vitro production of patient-specific, immunologically compatible, human embryonic (totipotent) cells might possibly lead to the production of unlimited numbers of ideal cells. Nevertheless, further animal experiments are necessary in order to establish reliable in vitro directed differentiation before clinical testing can be warranted. Moreover, the likelihood is quite low that local tissue conditions (at the border zone of an acute myocardial infarction or in an area of hibernating myocardium) could automatically induce (by means of “paracrine effects”¹⁰⁹) proper differentiation of multipotent or totipotent cells.

The additional option exists, at least potentially, of using not cells but only signaling cytokines that are yet to be fully identified. This approach could improve adult stem cell efficiency¹¹⁰ or, alternatively, it could promote dedifferentiation of adult cells back into a pluri- or multipotent state.¹¹¹ The attempt to reinitiate cell cycling simply through the overexpression of cell cycle–promoting factors has led to immediate cardiomyocyte apoptosis in early experiments.⁷⁰ However, current evidence favors the conclusion that embryonic stem cells have multiple possible pathways of differentiation and that the embryonic interstitial matrix normally carries the messengers and inductors of final choice (which are not yet well known). It is generally agreed that, once a stem cell has differentiated, it loses its capacity for further change (most likely, adult myocardial cells do not reproduce or dedifferentiate under normal conditions).

Techniques for Administering Stem Cell Therapy

Administration of different types of progenitor cells (Table I) by means of endovascular (intravenous or intracoronary) injection has been attempted in pilot studies, with promising early results.64,79,80,85,112 Readily available angioplasty catheters can be used for the intracoronary route but may entail a risk of microvascular obliteration and of poor therapeutic efficiency if the stem cells are to cross the coronary wall and migrate extravascularly—especially when targeting a territory supplied by occluded coronary arteries. Bone marrow–derived (stem) cells appear to migrate through the arterial or capillary wall better than do skeletal myoblasts.^113,114 The vascular approach seems generally less promising than direct intramural injection into the target myocardium, either surgically (via the epicardial approach)^63,82,83 or by catheter (via the endocardial approach.73,77,81,85,115 Still, even a single, direct, intramural administration of stem cells might be inefficient in an ischemic territory (revascularization may be a precondition for myocardial cell differentiation and growth).

TABLE I. Clinical Sources of Stem Cells

Specific Clinical Issues Regarding Cardiac Cellular Therapy: The Biology of Senescence

Myocardial infarction is primarily a disease of older persons, and the senescent myocardium may differ biologically from the myocardium of young persons and from that of the small adult animals typically used in initial stem cell experiments.¹¹⁶ In particular, the cells (myocytes) themselves and the intercellular messaging milieu in the interstitial space may be profoundly different under each clinical condition. Nadal-Ginard and associates²⁹ recently characterized the senescent myocardium by the predominance of large myofibers (volume >90,000 mm³) expressing p16INK4, a marker of cellular aging and increased apoptosis. Most likely, the molecular signals produced by such cells and their extracellular environment are not as favorable for stem cell differentiation, migration, and integration as are the signals present in younger hearts. These conditions need to be further characterized, because they might allow physicians to modify the environment, making it more conducive to successful stem cell treatment. Potentially significant experiments in this regard were recently performed in cultured endothelial progenitor cells.¹¹⁷ When more than 15 mg/mL of high-sensitivity C-reactive protein (CRP) was added to the medium, it significantly inhibited endothelial differentiation, increased apoptosis, and impaired angiogenesis.¹¹⁷ Such increases in CRP levels are frequently observed in older patients with myocardial infarction. The discussion on donor age also applies to the nuclear transfer clonic techniques, as shown recently by Hwang and colleagues.¹⁰⁸

Biologic Characterization of Morbid Conditions to Be Treated with Stem Cell Therapy

From the perspective of stem cell therapy, acute versus chronic myocardial infarction states differ profound-ly in terms of anatomic structure and biologic behavior. The fundamental functional event that causes an acute myocardial infarction is a sudden decrease in coronary blood flow so that it is below the levels required for contractile function and anatomic integrity.¹¹⁸ Development of functional myocardium by means of stem cell therapy probably cannot occur in a critically ischemic environment.^78,118 Revascularization of the occluded culprit epicardial coronary artery could be a precondition for success, and the degenerated, dependent microvascular bed^116,119 could also present a formidable challenge. Early experiments with stem cell therapy in such a setting suggested that neovascularization frequently occurs, but only a dis-tal capillary network develops.^32,70,77 No evidence is yet available concerning stem cell generation of actual new proximal arteries (provided with a medial layer), although the transformation of capillaries (bare endothelial channels) into arteries is commonly observed in the long term, in both embryos and adult humans.

The recently infarcted myocardial territory undergoes both myofiber cell degeneration (within the 1st few hours after infarction) and tissue reabsorption (soon after cell degeneration), mediated by inflammatory-cell invasion of the tissues.¹¹⁸ Most likely, this environment establishes an interstitial milieu that expresses signals conducive to stem cell differentiation into inflammatory or myeloid cells, which, if confirmed, would suggest that stem cell injections should be accompanied by aggressive, artificial signal enhancement designed to promote the development of myofibers or arteriogenesis, and not scar tissue.¹²⁰ This issue needs to be explored prospectively in animal models, although healing of the acute injury is still essential.

Normal embryologic development of the myocardium (not to mention the whole heart, with its valves and complex architecture) depends on a series of coordinated, sequential, irreversible events (cellular differentiation, genetic expression, migration, and autocrine secretion)²⁰ that are hard to imagine as being reproducible in the adult heart under normal conditions. Although transplanting stem cells into the borders of an experimentally created infarction may be a means of establishing that some stem cells will indeed develop into myocardial cells, this approach can hardly solve the problem of effective myocardial regrowth throughout the entire infarcted area. Stem cell migration into an environment so different from the embryonic cardiac jelly seems quite unlikely; at the same time, satisfactory spatial and functional integration of the new myocytes into the remnant of the viable myocardium would likely constitute a formidable challenge.

The protocol of Perin and co-authors⁸¹ suggests that the postinfarction state known as hibernating myocardium should become the target of cell therapy; however, this condition involves a chronically deficient blood supply and a myocardium that is at least partially recoverable, even if the function and metabolism remain diminished. Without a substantial solution to the problem of an adequate blood supply in similar “inoperable” patients, it is unlikely that new myofibers, if they developed, would improve the condition; rather, they would add to the metabolic deficit. Therefore, in hibernating myocardium, the primary need may be not as much for new myofibers as for the recovery of existing ones if blood supply can be increased. Such a recovery of blood supply may have caused the reported improvements, as Perin and his group suggested.⁷⁷

As opposed to the problems encountered in the setting of acute myocardial infarction, the challenges of cell therapy in the setting of an established chronic infarction in the absence of revascularization include the following: the culprit epicardial coronary artery has usually undergone diffuse sclerosis and relative atrophy; the microvascular network has already shrunk substantially; the myofibers have been digested and reabsorbed; and the interstitial connective tissue is greatly thickened and stiffened because of abundant collagen deposition around isolated mummified myocytes. The local cardiac skeleton and architecture have also undergone some degree of remodeling, which may involve substantial, irreversible volume expansion.¹²⁰ In this setting, the injection of stem cells can hardly enable the return of new, functional myocardium, even when surgical coronary revascularization is feasible.

Endpoints

The endpoints to be used for evaluating the results of clinical stem cell therapy are a matter of great interest and of some confusion in the literature. To date, the initial pilot studies have used imprecise markers, because no technology is yet available for labeling the implanted cells and tracking their destiny in the clinical setting. The early small studies have relied on segmental and global contractile function, exercise per-formance, and nuclear perfusion imaging. Despite the importance of clinically relevant endpoints (especially improved contractile function), nonspecific surrogate markers of cell-based therapeutic effects are inadequate for firmly establishing the mechanisms of clinical improvement. Therefore, alternative interpretations of the results should remain open to discussion.

Conclusions

This review has used the evidence supplied both by basic science and by clinical arguments to encourage prudence and patience despite the current explosive interest in stem cell therapy for the heart. Although such therapy may indeed become an ideal substitute for transplantation or for the artificial heart, many observers consider its scientific basis too primitive to justify clinical studies at present.21,67,121–125 Significant functional myocardial regeneration in animals must first be documented. Then researchers should rule out eventual degeneration of the grafted cells, arrhyth-mogenesis, and local or remote teratogenic consequences. Although treatments that use a patient's own stem cells might dramatically solve the crucial problems of donor-organ scarcity, high cost, and allograft rejection, these treatments are still highly experimental. We hope that the apparent lack of immediate commercial or industrial interest will not discourage the scientific community from adopting a disciplined strategy in pursuing this field. For now, the main challenge is to improve the translation of cellular and molecular concepts into clinically relevant endpoints. Once it has been proved that stem cells from differ-ent sources can develop into myocytes, biologists will need to teach clinicians how to optimize this process in order to achieve significant myocardial regeneration.