Percutaneous interventional cardiovascular medicine

Introduction

Interventional cardiology is a key player in the field of regenerative medicine; cell delivery to the heart is more advanced than to any other organ, which is partly due to interventional cardiology’s success in taking on the challenges of translational medicine. So, what has driven this partnership? Although primary angioplasty has transformed the outcome of acute myocardial infarction, the number of patients suffering with heart failure has doubled globally over the past 30 years. In Europe, 15 million people are living with heart failure [1] [2], leading to 2 million admissions a year [3] and an expenditure of €15 billion (across 11 countries) [4].

The processes of myocardial damage and adaptation that lead to heart failure were once thought to be irreversible as the heart was considered a terminally differentiated post-mitotic organ. However, this dogma has been challenged, particularly with the demonstration of continued cell division within the adult heart following injury [5]. Researchers have also isolated and identified resident cardiac stem cells which can differentiate into multiple cardiac cell lineages such as cardiomyocytes and vascular smooth muscle cells [6]. However, the human heart’s own self-renewal capabilities cannot overcome the massive loss of cardiomyocytes (up to one billion cells) seen in acute myocardial infarction and heart failure (unlike non-mammalian vertebrates, such as the zebrafish, which have demonstrated the ability to regenerate up to 20% of the left ventricle following injury [7]). There is, therefore, a drive to determine whether the human heart can also be directed into eliciting a regenerative response following injury; either through the up-regulation of its own intrinsic repair mechanisms or through the addition of biological therapies such as cell-based treatments.

Trials have been performed in four distinct clinical settings: acute myocardial infarction, chronic ischaemic heart failure, dilated cardiomyopathy with a non-ischaemic aetiology and refractory angina. Although the therapeutic benefits seen in animal models have yet to be fully translated into man, these early Phase I/II trials provided important safety data and allowed interventional techniques to be refined. Cell therapy remains an exciting, novel treatment strategy and this chapter will provide an overview of the cellular mechanisms of cardiac regeneration and repair, a review of the latest clinical trial data and a summary of the field’s current challenges.

FOCUS BOX 1Rationale for cell-based therapy

The incidence of heart failure (either as a consequence of acute myocardial infarction or heart disease) continues to grow worldwide.
Despite modern pharmacological therapies targeting heart failure, related morbidity and associated health economic costs are increasing.
Cell-based therapies provide a method for delivering regenerative medicine that can potentially decrease the health and economic burden of heart failure.

What is the ideal cell type for cardiac repair ?

The ideal cell type for cardiac repair should possess the following characteristics:

Ability to repair/regenerate damaged myocardium
Easy to obtain
Safe (no tumour formation/toxicity)
No immunogenicity issues
Easy to store
Easy to deliver
No ongoing ethical issues
Cost-effective

Stem cells offer the most potential to fulfil the first prerequisite characteristic, namely the ability to contribute to cardiac repair. Although several different kinds of cell type have been shown to fulfil this criterion there is currently no cell type that meets all of the criteria above.

STEM CELL POTENTIAL – AN OVERVIEW

Stem cells are defined by two unique characteristics: they are unspecialised cells capable of unlimited self-renewal and they can differentiate into more specialised cells that form organs. The ultimate stem cell is the zygote which can develop into all cell types including the embryonic membranes - this potential is termed totipotent. The developing embryo contains embryonic stem cells (ESCs) which are pluripotent as they can develop into cells from all three germinal layers (the mesoderm, endoderm and ectoderm) and produce all the tissue types needed to form a functional organism.

Developed adult organs also contain undifferentiated stem cells - adult (somatic) stem cells (ASCs) – but, these occur in far fewer numbers. These cells were once thought to be limited in differentiation potential and only able to repair and replenish tissue in the organ in which they resided. However, ASCs have been shown to “transdifferentiate” into cell types from different germ layers (a property referred to as plasticity), which has been demonstrated in several scientific studies [8] [9]. Various cell types have been explored for cardiac repair and are classified as either allogeneic or autologous in origin ( Table 1).

ALLOGENEIC CELL TYPES

Embryonic stem cells (ESCs)

Given that multiple cell types are generated from a single ESC origin during embryonic development, this cell type’s potential for cardiac repair is self-evident. ESCs have diverse potential and their differentiation can be controlled to produce specific cell types useful for clinical application. However, their potential is counterbalanced by several significant drawbacks: the collection of cells from embryos raises ethical concerns, implanted cells may require immunosuppression in recipients to prevent rejection and, lastly, their potential for growth and differentiation can also result in tumour formation. Recently, the concept of using human-derived ESCs was tested in a small study which engrafted cells onto the heart using a fibrin patch during bypass surgery [10]. An improvement in cardiac function and symptoms was demonstrated, however larger controlled studies are needed to understand the true benefits of this approach. Therefore, although ESCs have significant capacity for cardiac repair, the problems inherent with their use mean that researchers have focussed on ASCs.

Foetal cardiomyocytes

Foetal cardiomyocytes were one of the first cell types to be investigated as a candidate for cardiac repair. Animal studies demonstrated that foetal cardiomyocyte transplantation improved the function of ischaemic and globally failing hearts [11] [12]. However, the use of foetal cardiomyocytes, like the use of ESCs, raises concerns regarding availability, immunogenicity and ethics. Therefore, other cell types have surpassed these as likely candidates for cardiac repair.

Human umbilical cord-derived stem cells (hUCSCs)

Umbilical cord blood (blood which remains in the placenta and the attached umbilical cord after childbirth) is a potent source of haematopoietic progenitor cells. Human umbilical cord blood mononuclear cells are currently used clinically to repopulate bone marrow in the treatment of bone marrow disorders such as acute leukaemia. Cord blood also contains a large number of non-haematopoietic progenitor cells which rarely express HLA class II antigens and appear to be immunologically naïve, thus reducing the risk of rejection. They are an accessible cell type (as they do not need to be harvested from a patient), they can be isolated and expanded in culture and they have a similar therapeutic potential to mesenchymal stem cells (they have been shown to be anti-fibrotic, pro-angiogenic and anti-inflammatory). Human umbilical cord-derived cells, therefore, provide an attractive off-the-shelf option for regenerative therapy [13] [14]. In animal models of acute myocardial infarction, the injection of human umbilical cord-derived cells has been associated with significant reductions in infarct size, particularly when given by the intramyocardial route [15]. Colicchia and colleagues have summarised the preclinical and clinical data for umbilical cord-derived cells in cardiovascular disease [16]. A growing number of clinical trials are adopting umbilical cord-derived cells and have shown promising results in cardiac diseases, including acute myocardial infarction [17] [18] and chronic heart failure [19] [20] [21] [22] [23] [24].

AUTOLOGOUS CELL TYPES

Unlike ESCs, autologous adult stem cells are virtually free from the risks of teratoma formation and immune rejection; however, their potential for differentiation is more limited. The types of adult cells that have been studied include induced pluripotent stem cells, adipose-derived stem cells, tissue-resident cardiac stem cells, skeletal myoblasts, mesenchymal stem cells, circulating endothelial progenitor cells and, most commonly, bone marrow-derived mononuclear progenitor cells.

Induced pluripotent stem cells (iPSCs)

An exciting alternative to ESCs is emerging in the form of induced pluripotent stem cells (iPSCs). These are adult stem cells (differentiated somatic cells) that have been successfully reprogrammed back to an undifferentiated pluripotent state through the insertion of regulatory genes (e.g. Oct3/4, Sox2, KL4 and c-Myc) [25] [26] [27]. These cells have the same morphological phenotype as ESSs and have been demonstrated in vivo and in vitro to have the same differentiation potential (i.e. the ability to form all three germ cell layers). Functioning cardiomyocytes have already been produced from iPSCs, demonstrating their potential for use in cardiovascular regenerative medicine [28]. Despite the potential to produce patient-individualised iPSCs, there are problems that need to be solved before they can be used in clinical trials. The main concern is tumour genesis as most iPSC lines have been derived by inserting putative oncogenes using integrating retroviruses, e.g. lentiviruses, into the host genome which can cause cancer. One mouse study demonstrated that up to 20% of offspring derived using iPSCs developed tumours [29].

One further practical barrier to the clinical use of iPSCs is the difficulty in producing enough cells for therapeutic purposes. Experiments to date have shown low conversion rates in the percentage of cells treated compared to those that are successfully induced into an iPSC phenotype - the conversion percentage varies from 0.0006-1% in the literature. To make matters more difficult, steps taken to combat tumour genesis, e.g. the use of adenoviruses, also significantly impair the conversion rate [30].

Recently, investigators have developed a better understanding of the mechanisms for reprogramming cells and the methods of preventing tumour genesis. This has led to the initiation of 4 clinical trials - the results of which are awaited [31]. Once the aforementioned issues have been resolved, there is a considerable expectation that iPSC production can be scaled up to provide a patient specific source of cardiovascular cells for use in regenerative medicine. Until these aims are accomplished, the main cell types used will remain adult-derived tissue-specific stem and progenitor cells.

Adipose-derived stem cells (ADSCs)

As adipose tissue contains a heterogeneous mixture of endothelial, haematopoietic and mesenchymal progenitor cells which can be harvested easily by liposuction [32], it has been investigated as a source of adult progenitor/stem cells for cardiac repair. Preclinical studies have shown that adipose-derived stem cells (ADSCs) are associated with improved ejection fraction in animal models of myocardial infarction, and neoangiogenesis via paracrine factors has been postulated as a potential mechanism of action [33].

Several clinical trials have explored the safety and feasibility of ADSCs, such as APOLLO [34] and ADVANCE [35] (for acute myocardial infarction), MyStromalCell [38] (for refractory angina) and CSCC_ASC [36] and PRECISE [37] (for chronic ischaemic heart failure). APOLLO demonstrated that liposuction in the acute phase of AMI and the intracoronary infusion of freshly isolated ADSCs is safe and feasible. At 6 months, cell treatment was associated with a trend towards improved cardiac function, a significant improvement in perfusion defect and a 50% reduction in myocardial scar formation [34]. MyStromalCell evaluated the intramyocardial injection of autologous VEGF-A₁₆₅ stimulated adipose-derived stromal cells in 60 patients with refractory angina. At 6 months, the results demonstrated a within group improvement in exercise time, but this was not significant when compared to placebo [38]. However, at 3 years follow up, the cell-treated patients had improved cardiac symptoms and unchanged exercise capacity, compared to a deterioration in the placebo group [39]. PRECISE and CSCC_ASC both demonstrated the safety of ADSCs delivered by the intramyocardial route in ischaemic heart failure, however neither established efficacy. In PRECISE, the injection of ADSCs in 21 patients did not increase left ventricular ejection fraction (LVEF) but did stabilise scar size [37]. CSCC_ASC showed a trend towards an improvement in cardiac function at 6-month follow-up: left ventricular end systolic volume decreased whilst LVEF and exercise capacity increased [36]. SCIENCE, a multicentre Phase II trial, built on these results, randomising 133 patients to an intramyocardial injection of allogeneic CSCC_ASCs or placebo in a 2:1 ratio [40] - the results are still awaited.

Resident cardiac stem cells

The heart was once considered to be a terminally differentiated organ which lacked self-renewal capabilities; however, this dogma has been challenged, particularly with the demonstration of continued cell division within the adult heart following injury such as myocardial infarction [5]. The discovery of myocardial chimeras further increased the suspicion that cardiac stem cells (CSCs) existed. Myocardial chimerism was demonstrated by testing for the presence of the Y chromosome in explanted hearts of male patients who had received female donor hearts. Male derived myocytes and endothelial cells were revealed within the female heart, which was interpreted as representing cardiac repair affected by circulating progenitor cells [41]. A second, similar study, in which the hearts of females who had received male bone marrow were examined for the presence of the Y chromosome, concluded that bone marrow progenitor cells were capable of transit to the heart and its subsequent repair [42].

The chimera studies suggest that a population of cardiac stem cells recruited from a non-cardiac source such as bone marrow exist. However, several independent investigators have also presented strong evidence for a resident population of cardiac progenitor cells within the heart. The isolated cardiac stem cells have standard stem cell defining characteristics and can differentiate into multiple cardiac cell lineages such as cardiomyocytes, endothelial and vascular cells [6]. These findings have been replicated in animal models and CSCs have been shown to repair and improve cardiac function following myocardial ischaemia [43]. CSCs are therefore an attractive option for clinical trials as they are intrinsically more likely to produce the cells needed to repair the damaged heart. However, as they are significantly more difficult to harvest and isolate compared to BMCs, the way ahead may be the activation and stimulation of a patient’s own CSCs rather than a transplantation.

A harvesting technique has been described that involved obtaining human myocardium from surgical or endomyocardial biopsies which was then partially enzymatically digested and expanded in culture for 14–24 days. Small round cells were seen to bud off from the primary ex-plant and divide in suspension. The isolated cells were then grown in a suspension culture containing a differentiation media which induced the formation of a ball of cells that has been termed a cardiosphere [6]. When cardiospheres were grown in co-culture with rat neonatal cardiomyocytes, approximately 20% of cardiospheres were seen to contract spontaneously. Moreover, spontaneous calcium transients were seen in a small percentage of these contracting cells.

Human cardiosphere-derived cells (CDCs), when injected into the border zone of a mouse model of myocardial infarction, engrafted and migrated into the infarct zone. After 20 days, the percentage of viable myocardium within the infarct zone was greater in the cell-treated group than in the fibroblast-treated or control group. Likewise, echocardiography revealed a higher LVEF in the cardiosphere-treated group (42.8±3.3%) than in the fibroblast-treated (25.0±2.0%, p<0.01) or control group (26.0±1.8%, p<0.01) [44].

Several clinical trials have demonstrated safety, feasibility and promising efficacy results using cardiac-derived cells. CADUCEUS was the first trial to use an intracoronary infusion of cardiosphere-derived stem cells. Autologous CDCs (12.5 to 25 x 10⁶) grown from endomyocardial biopsy specimens were infused via the intracoronary route in 17 patients with left ventricular dysfunction 1.5 to 3 months after myocardial infarction. At 1 year, preliminary indications of bioactivity included decreased scar size, increased viable myocardium and improved regional function of the infarcted myocardium [45].

CAREMI was the first randomised clinical trial to demonstrate the safety of allogeneic cardiac stem cells in the acute phase of STEMI; however, the trial was too small to report on efficacy endpoints [46]. More recently, ALLSTAR treated 134 patients (in a 2:1 ratio) with LVEF ≤45% and LV scar size ≥15% of LV mass (as assessed by MRI) with either an intracoronary infusion of allogeneic CDCs or placebo 4-12 weeks post-MI. The trial was stopped due to the low probability of meeting its primary efficacy endpoint (relative percentage change in infarct size at 12 months post-infusion) after an interim analysis at 6 months. The trial did however meet the primary safety endpoint (with no events at the time of CDC infusion or during the first month, including immunological surveillance) and showed a reduction in LV volumes and NT-proBNP - demonstrating the disease-modifying bioactivity of CDCs [47]. DYNAMIC tested the intracoronary infusion of allogeneic CDCs in 14 patients with dilated cardiomyopathy. The study demonstrated an improvement in ejection fraction, left ventricular end-systolic volume, Quality of Life scores and NYHA class at 6 months (the improvements in ejection fraction and Quality of Life scores remained significant at 12 months) [48].

A combination of mesenchymal and c-kit⁺ cardiac stem cells was tested in CONCERT-HF - a Phase II trial using a transendocardial injection in patients with ischaemic cardiomyopathy [49]. Although there was some evidence of efficacy, the trial was not definitive in establishing the role of this combination therapy and further exploration of this approach is unlikely [50].

In summary, although cardiac-derived cells have considerable theoretical advantages over other cell types, the clinical trials have not reflected this with a strong enough efficacy signal to counterbalance the practical issues. Furthermore, controversies surrounding work conducted using this cell type have damaged the field and make it difficult to pursue this line of research [51].

Epicardium-derived progenitor cells (EPDCs)

As mentioned previously, the zebrafish can fully regenerate its heart after injuring up to 20% of the ventricle [7]. Experimental evidence suggests that this regeneration may occur through the limited dedifferentiation of existing cardiomyocytes followed by proliferation [52], and through the activation and expansion of surrounding epicardial tissue which supports the regenerating myocardium [53]. These studies in zebrafish suggest that the epicardium may play an important role not only in adult heart repair, but also during the continuous growth of the adult heart [54]. A subset of epicardium-derived progenitor cells (EPDCs), expressing known markers of stem cells (i.e. c-kit and CD34), have been identified in the subepicardial space of foetal and adult human hearts [55]. Experimental studies have demonstrated that EPDCs have the potential to differentiate into cardiomyocytes [56], and the intramyocardial injection of human EPDCs in a mouse model of myocardial infarction has been shown to improve cardiac function [57] - supporting the hypothesis that EPDCs may play an important role in cardiac repair. The question arises as to why human EPDCs remain dormant following myocardial injury, and current work is focusing on whether paracrine factors may play a role in activating these cells. A potential stimulus has been identified in the form of thymosin β4 (Tβ4), an actin monomer-binding protein which has been shown to activate EPDCs to a pluripotent state, possibly by an epigenetic effect (a chemical change to DNA which alters gene expression) [58]. In a landmark experiment, the pre-treatment of mice with Tβ4 prior to inducing a myocardial infarct led to the activation of EPDCs which underwent cardiomyocyte differentiation, and infarct size and overall ejection fraction were significantly better in the treated mice compared to controls [59]. This study provides evidence that the activation and up-regulation of the heart’s own repair mechanism may be possible without the need for additional biological therapy. Despite the initial promising data, a Phase II clinical study (NCT01311518) evaluating the safety and efficacy of injectable Tβ4 in the treatment of acute myocardial infarction was suspended due to issues with GMP compliance in the drug manufacturing process. Unfortunately, despite the initial promise, the use of Tβ4 as a regenerative medical therapy for the treatment of cardiovascular disease is not part of ongoing trials.

Skeletal myoblasts

Skeletal myoblasts have been widely studied due to several favourable characteristics: their contractile phenotype, they are an autologous and easily harvested source of cells, they can be expanded with a minimal risk of tumour formation and they are resistant to ischaemia. Early preclinical animal studies demonstrated the ability of skeletal myoblasts to engraft, form myotubules and enhance cardiac function after transplantation into infarcted myocardium [60]. However, not all the preclinical studies produced positive results. Studies have consistently demonstrated significant cell loss (up to 84% loss in the first 24 hours) and the death of the transplanted myoblasts within a few days [61] [62]. Transplanted cells also down-regulate the major adhesion and gap junction proteins (N-cadherin and connexin43) of the intercalated disk [63], and are functionally isolated from the host myocardium with no evidence of electromechanical coupling [64]. This may account for the increased incidence of arrhythmias seen after myoblast transplantation [65].

Human studies have shown that the intramyocardial injection of these cells is feasible and has potential functional benefits [66] [67] [68]. The main limitations regarding skeletal myoblasts are that they remain committed to the skeletal muscle lineage and have been associated with arrhythmias [66] [68] [69] (the mechanism of which has yet to be fully defined, but may involve the processes described above). Currently, skeletal myoblasts are not being pursued in clinical studies; and are, instead, being investigated in conjunction with cell sheet technology [70] [71] [72] [73].

Mesenchymal stem cells (MSCs)

Mesenchymal stem cells (MSCs) can be found in bone marrow, muscle, skin and adipose tissue. They have a fibroblast-like morphology and can differentiate into bone, cartilage and fat cells [74]. Further to this well-described trilineage potential, MSCs have also been shown to transdifferentiate in vitro and in vivo towards cardiomyocyte and vascular cell phenotypes [75].

In vivo preclinical studies have demonstrated that MSCs have the capacity to stimulate both myocardial repair and neovascularisation in animal models of cardiac injury [76]. The differentiation of MSCs into cardiomyocytes and endothelial cells in vivo when transplanted into the heart in injury models has also been demonstrated. These transdifferentiated cells have been strictly characterised by immunohistochemistry and positively stain for cardiac and endothelial specific markers, as well as gap junction proteins [77]. Furthermore, MSC transplantation in acute myocardial infarction [76] and ischaemic cardiomyopathy [78] has been reported to induce functional benefits, including reduced scar formation and infarct size, improved regional and global ventricular function and increased vascular density and myocardial perfusion. There has also been evidence of benefit in studies using non-ischaemic models such as dilated cardiomyopathy [79]. The ability of MSCs to transdifferentiate into specialised cells that improve the function of a failing heart makes them a realistic option for cellular transplantation. This is further underlined by the relative ease by which they can be maintained and expanded in culture.

MSCs also appear to be relatively immunoprivileged as they lack various major histocompatibility complex and co-stimulatory cell-surface antigens and, therefore, may be used as an allogeneic graft [80]. This avoids the need for recipient harvest procedures such as bone marrow aspiration or liposuction, and MSCs can potentially be given intravenously as they have been shown to home to injured myocardium following acute myocardial infarction in animal models [81]. The safety of intravenous allogeneic MSCs (Prochymal®; Osiris Therapeutics, Inc., Columbia, USA) has been demonstrated in a Phase I randomised controlled trial of patients with acute myocardial infarction [82]. In this trial of 53 patients, adverse event rates were similar in the MSC and placebo treated groups and, interestingly, ejection fraction was significantly higher in the MSC treated group at 6 months. A larger Phase II study (NCT00877903) which aims to investigate these findings further has completed recruitment. Several trials have also reported positive outcomes using adipose tissue-derived stem cells (containing different proportions of MSCs) [36] [38].

There is a growing interest in guiding the transdifferentiation of MSCs into a cardiopoietic phenotype (i.e. lineage specified) prior to transplantation into the infarcted myocardium to improve the therapy’s efficacy. This has been successfully demonstrated in a murine model of myocardial infarction where the epicardial injection of lineage specified cardiopoietic MSCs (obtained from human bone marrow) achieved superior cardiac functional and structural improvements compared to the injection of unguided MSCs [83]. C-CURE was a Phase II trial that evaluated the endomyocardial injection of cardiopoietic stem cells (which were obtained from bone marrow-derived MSCs and treated with cardiogenic growth factor) in patients which heart failure. Safety, feasibility and efficacy were assessed at 6 months and 2 years; the results were encouraging and showed an increase in LVEF [84]. This led to CHART-1, a Phase III trial which analysed efficacy endpoints (all-cause mortality, worsening heart failure, Quality of Life scores, exercise capacity and cardiac function) at 39 weeks post‐injection in 271 patients. While the aim was to inject up to 600 million autologous cells, the bone marrow yield from each patient was not uniform. Consequently, the dosage (and therefore the number of injections) varied among patients. Although favourable trends were seen in functional parameters, CHART‐1 failed to meet its formal primary endpoints [85]. Despite some evidence of long-term benefits in a specific subgroup [86] this cell type is no longer being investigated in clinical trials.

In conclusion, MSCs (whether autologous or allogeneic in origin) are an attractive option for regenerative therapy due to their ability to facilitate myocardial and vascular repair and the ease by which they can be collected, manipulated and expanded once in tissue culture. However, many aspects of their biology, especially with respect to their therapeutic use, still need to be understood in order to fully realise their potential. Unfortunately, despite the promising Phase II clinical trial data, very few Phase III clinical trials have been conducted using MSCs; and, therefore, the translation into clinical practice has not yet occurred. Further reviews of MSC application in heart failure and other heart diseases have been authored by Pandey et al. [87] and Majka et al. [88].

Endothelial progenitor cells

It is important to define a precise cellular phenotype when investigating a cell’s therapeutic potential - and the endothelial progenitor cell is an example where a rigorous definition is required [89]. Close consideration of its definition is needed to ensure agreement and consistency in preclinical research and avoid mistakes when this promising cell type is brought to the clinic.

Endothelial progenitor cells (EPCs) can be isolated from adult bone marrow or the peripheral circulation (these cells are termed circulating endothelial progenitor cells – CEPs). Adult-derived EPCs and CEPs can be distinguished from mature endothelial cells by a functional in vitro assay due to their high proliferation rate. Mature endothelial cells, EPCs and CEPs share several endothelial specific markers; however, only EPCs and CEPs express AC133 (CD133) [90]. Human CEPs have also shown the potential to differentiate into cardiomyocytes. When co-cultured with neonatal rat cardiomyocytes, human CEPs formed cells with a cardiomyocytic phenotype (as defined by positive staining for cardiac specific markers, such as troponin, atrial natriuretic peptide and MEF-2) [91].

EPCs and CEPs play an important role in neovascularisation in vivo [92]. Circulating EPCs are mobilised in response to organ ischaemia, trauma and acute myocardial infarction [91]. The increase in CEPs post-myocardial infarction is mirrored by a rise in levels of the migratory cytokine VEGF-A, which suggests a role for this factor in the mobilisation of progenitor cells [93]. Furthermore, HMG-CoA reductase inhibitors (statins) have been shown to augment the mobilisation of EPCs. This important observation may provide an alternative mechanism by which statins decrease morbidity and mortality in patients with ischaemic heart disease [94].

The transplantation of EPCs and CEPS has been shown to promote neovascularisation of the ischaemic heart and improve cardiac function. Transdifferentiation to endothelial cells, smooth muscle cells and cardiomyocytes has been characterised by immunohistochemistry [95] [96]. In animal models of myocardial infarction, the transplantation of EPCs or CEPs caused a significant improvement in cardiac function and an increase in capillary density, regional blood flow and collateral formation in the ischaemic heart [97] [98]. Encouraging results such as these led to human clinical studies of EPC transplantation for the treatment of acute myocardial infarction [99] [100] [101] [102] and heart failure [103] [104] [105] [106] [107] [108] [109] [110] [111] [112] which have used different delivery routes, including intracoronary, intramyocardial and adjunctive CABG. There have been no adverse outcomes as yet and most of the studies have shown improvements in cardiac function, myocardial perfusion and symptoms. Unfortunately, PERFECT, a Phase III study using EPCs in combination with bypass grafting, was terminated early due to recruitment issues [106].

Bone marrow-derived stem/progenitor cells (BMSCs/BMPCs)

The most widely studied cell type has been bone marrow-derived cells (BMCs). Strictly speaking, these are not true stem cells as they already express a phenotype and do not have the ability for infinite expansion and differentiation [113]. However, these cells have been shown in preclinical experiments to have both regenerative and reparative properties [114]. Bone marrow-derived cells consist of haematopoietic stem cells (HSCs), mesenchymal stem cells (MSCs) and endothelial progenitor cells (EPCs), all of which have been shown to have the potential to transdifferentiate to a cardiomyocyte phenotype [77] [91] [115]. BMCs can also be characterised by specific cell surface markers, for example CD34, which is expressed on primitive stem cells and differentiated progenitor cells but is absent on mature haematopoietic cells. BMCs can be further characterised by the presence or absence of a lineage marker (Lin - which, if present, represents cells committed to a particular lineage) and the stem cell factor receptor (c-kit).

The idea that the developmental potential of BMCs may expand beyond the haematopoietic lineages emerged after reports indicating that cells derived from bone marrow could give rise to multiple unexpected cell types, including neural cells [116], skeletal muscle cells [8] and hepatic cells [117]. The plasticity of BMCs has led to considerable excitement in using them in cell-based therapies and as vectors to deliver therapeutic agents. This is particularly attractive clinically because BMCs can be readily obtained from patients by bone marrow aspiration or by their mobilisation into the peripheral circulation with cytokines such as granulocyte colony stimulating factor (G-CSF). Moreover, if a patient’s own cells were taken for ex vivo expansion and directed to differentiate into cardiac cells, for example, no immune rejection problems would arise.

In an early animal study, Lin^- c-kit⁺ (undifferentiated) BMCs were injected directly into the contracting wall bordering a myocardial infarct which had been acutely induced by coronary artery ligation [118]. The transplanted cells appeared to undergo transdifferentiation into cardiomyocytes, with newly formed myocardium occupying a significant proportion of the infarcted area and a significant improvement in LVEF just 9 days after cell transplantation [119]. This observed transdifferentiation has not been universally reproduced and has, in fact, been challenged by different groups who have demonstrated that injected BMCs develop into haematopoietic cell types after transplantation rather than cardiomyocytes [120] [121]. It is therefore possible that adult stem cell plasticity (i.e. the ability to transdifferentiate into different cell types) has been overestimated, particularly regarding cardiomyogenic transdifferentiation in humans (as most of this work was based on animal experimentation). This has fuelled the ongoing debate regarding the mechanism of action by which cell therapy leads to cardiac repair. It is likely that the beneficial effects seen are multifactorial in origin; possible explanations include neovascularisation by differentiation into an endothelial phenotype, the paracrine effects of the cell infusate, cell fusion and the induction of myocardial regeneration [122].

Mixed cell types

Several studies have addressed the possibility that combinations of specific cell types might provoke a better therapeutic response in patients with cardiovascular disease. The largest of these trials, ixCELL-DCM, evaluated the efficacy of Ixmyelocel-T (a combination of autologous CD90⁺ MSCs and CD45⁺ CD14⁺ auto-fluorescent⁺ activated macrophages) in 126 heart failure patients. The trial demonstrated a significant reduction in cardiac events and an improvement in patient outcomes [123]. Unfortunately, attempts to perform larger clinical trials to confirm these findings have not reached completion. Furthermore, as the possible combinations of cell types are limitless; well-defined scientific rationale and rigorously designed clinical trials are required to effectively investigate the role of combined cell therapy.

Summary

To summarise, there a multiple cell types that offer therapeutic potential for the treatment of cardiovascular disease; however, none currently fulfil the full list of ideal criteria. ESCs continue to offer the greatest therapeutic potential whilst iPSCs may well provide the best compromise between these and the BMC population. To date, BMCs have had the most application in clinical studies due to preclinical science suggesting positive effects on cardiac function alongside the comparative ease with which they can be harvested using existing techniques. Work is still needed to identify a cell type that meets the ideal criteria and will move the field closer to delivering a viable biological therapy that can be used in routine clinical practice.

FOCUS BOX 2Candidate cells for cardiac repair

Embryonic stem cells have the broadest differentiation potential, but their use is limited by ethical and legal issues as well as concerns regarding malignant potential.
A variety of adult cells have been investigated as candidates for cardiac repair and have all shown therapeutic potential in clinical trials.
Due to promising preclinical science and practicality, adult bone marrow-derived cells have been most widely used in clinical studies.
There is now evidence of ongoing cell division within the adult heart. A resident cardiac stem cell has now been identified which theoretically has great therapeutic potential.

Mechanism of actions of cell based regenerative therapy

The mechanism by which adult and embryonic cellular therapy improves cardiac function and perfusion following pathological insults remains an area of ongoing investigation and debate. The initial theory and hope was that transplanted stem and progenitor cells formed new cardiomyocytes and necessary vascular conduits, directly leading to improved myocardial contraction and perfusion. This hypothesis was synthesised from a range of individual publications which, when taken together, supported the prospect of the direct “transdifferentiation” theory. This theory continues to be accepted as a mode of action for embryonic stem cells. However, for adult stem and progenitor cells, preclinical data has cast doubt over the contribution of transdifferentiation to functional benefit. Glimpses into many other possible mechanisms have since been reported ( Figure 1), including cell fusion, immunomodulation, paracrine effects leading to modulation of ischaemia-reperfusion, improved remodelling through the amelioration of scar tissue, cardiac stem cell stimulation and the direct promotion of angiogenesis.

The direct transdifferentiation theory was formed from several findings. Initially, the concept of exogenous cell delivery as a possible means of improving cardiac function was demonstrated. For example, when transplanted into the heart, embryonic cardiomyocytes formed stable grafts [124], while foetal cardiomyocytes improved cardiac function after cryo-injury [12]. Progress was also made with skeletal myoblasts, the native stem cell of striated muscle [125], with functional benefits also being shown when used to treat cryo-injured rabbit hearts [60].

Following these findings, adult stem and progenitor cells started to be seen as a possible donor cell source without the drawbacks of foetal or embryonic cell-based therapies. Several sources suggested that adult stem cells, particularly those found within the bone marrow, could contribute to de novo formation of cardiac and vascular cells. Animal experiments demonstrated that circulating endothelial progenitor cells of bone marrow origin could contribute to neovascularisation in adult tissues [126]. Evidence of haematopoietic stem cells contributing to cardiac muscle and vasculature was also seen [115]. Furthermore, human mesenchymal stem cells (MSCs) were shown to have multilineage potential [127] and were able to form cardiomyocytes in vitro when chemically treated [75]. These findings were supported by the demonstration of chimerism in human hearts, which reinforced the idea that circulating cells may be able to contribute to cardiac repair [128].

Multiple animal studies continued to support the direct transdifferentiation hypothesis, initially suggesting that all three cell lines from adult bone marrow were able to contribute to cardiac regeneration in models of myocardial ischaemia. MSCs were shown to engraft into ischaemic myocardium and express muscle proteins in models of myocardial infarction [77] [76]. Human EPCs were able to simulate neoangiogenesis within the infarcted vascular bed, reducing apoptosis and improving cardiac function in a mouse model [96]. In one study, bone marrow-derived cells were shown to regenerate substantial amounts of myocytes, endothelial cells and smooth muscle cells when injected into the hearts of mice following myocardial infarction. The authors felt that a subpopulation of primitive bone marrow cells regenerated the myocardium and therefore could be used to treat large myocardial infarcts [129].

Support for the direct transdifferentiation theory receded once muscle generation and transdifferentiation were not seen in two high-profile studies of bone marrow-derived cell therapy, although a small functional benefit was still seen [120] [121]. Furthermore, other studies observed that the rare transdifferentiation events which were thought to be de novo muscle or vessel formation could be attributed to donor and host cell fusion [130] [131].

The search for a complete explanation behind the benefits gained with cell therapy led to the proposal that the local release of paracrine factors could be the significant key effect of transplanted cells rather than myogenesis. Supporting this, it is clear that adult-derived cells are capable of secreting a wide range of cytokines [132]. The released paracrine factors have been shown to produce a number of beneficial effects, including the induction of angiogenesis in host tissues [96], reduced apoptosis [132] and the immunomodulation of injury [133]. A further documented paracrine benefit is the proposed stimulation of resident cardiac stem cells [134].

Two notable studies support the hypothesis that adult stem cells stimulate an acute paracrine effect. Akt (protein kinase B) is a kinase which regulates several cellular processes, including cell cycle progression, transcription, glucose uptake and apoptosis. In a rat MI model, both MSCs transfected with Akt and Akt-MSC-conditioned media significantly limited infarct size when administered into the infarct border zone one hour after ischaemia was established [135]. The genes for VEGF, FGF-2, HGF, IGF-1 and Tβ4 were all found to be up-regulated within the Akt-MSCs, especially in the setting of hypoxia. This data strongly supports a paracrine-mediated mechanism for myocardial protection in acute myocardial infarction. The second supportive study used cardiac cryo-injured rats. Immediately following injury, MSCs were injected into the border zone [136]. Infarct size was assessed 10 weeks later and found to be significantly reduced, but no evidence of myogenesis or vasculogenesis was seen – providing indirect evidence for a paracrine effect rather than transdifferentiation.

Furthermore, bone marrow-derived cells have been demonstrated in vitro and in vivo to significantly reduce ischaemia-reperfusion injury. The mechanism of this benefit has been suggested both by our laboratory and by others to be due to a paracrine-related triggering of mediators of the reperfusion injury salvage kinase (RISK) pathway [137] [138] [139]. The RISK pathway refers to a group of protein kinases which, when specifically activated during myocardial reperfusion, invoke cardioprotection by preventing lethal reperfusion injury. The RISK pathway is thus a mediator of cell survival [140] [141].

Another component of the amelioration of reperfusion injury, aside from the activation of the RISK pathway, could be a direct effect on reactive oxygen species (ROS). MSCs, in particular, can reduce ROS generation as they express the enzymes required to manage ROS, in particular superoxide dismutase [143] and high levels of glutathione peroxidase [142].

It is possible, of course, that different mechanisms are important at different time points along the disease trajectory from acute myocardial infarction to chronic heart failure. Apart from very early reperfusion [144], the only other consistently observed mechanism for reducing infarct size, with a subsequent preservation of cardiac function, is the prevention of reperfusion injury [145]. The possibility of injected adult cells transdifferentiating into cardiomyocytes and endothelial cells within a matter of hours following myocardial infarction seems an unlikely prospect. From a logical perspective, the mechanism of acute repair is more likely to be related to the release of local factors which act on cells within the myocardium, reducing ischaemia-reperfusion injury [146].

In chronic cardiac dysfunction situations, such as chronic ischaemic heart failure, where the acute insult may have occurred weeks or months previously, there is no acute reperfusion injury; and thus no benefit would be gained from trying to reduce this. As predicted, there are no studies that demonstrate a benefit of reperfusion injury modulation in this chronic scenario. However, there are multiple sex-mismatch cardiac transplant studies in humans and one novel study of carbon-dated human cardiomyocyte DNA which demonstrate de novo generation of myocardium and endothelial cells in the chronic setting [147] [148] [149]. Studies have even suggested that the new cell formation is increased in pathological conditions such as heart failure, in a hitherto undocumented homoeostatic response [150]. Hence, in chronic cardiac conditions, it is likely that a different mechanism of action explains the positive observations seen in clinical trials - this may well include the direct or induced production of new myocytes and endothelial cells, as well as mechanisms of myocardial salvage (cf hibernating myocardium).

Thus, a growing body of evidence supports paracrine mechanisms as the main effectors of the functional benefits seen with adult cell use in acute myocardial ischaemia; while in chronic conditions, direct/indirect myogenesis, vasculogenesis and myocardial salvage are likely to be the predominant mechanisms. Understanding the complete pathways involved, the importance of these mechanisms and how they interact at specific time points in the progression towards chronic heart failure, is essential to the utilisation and optimisation of adult cell-based therapy.

It may well be through an understanding of these pathways that a panel of candidate proteins is identified which constitutes the repair “signature”. Manipulating these proteins using new technology (e.g. RNA silencing/promoting), may provide a more direct method of repairing the myocardium in the future. For the time being, however, the complex pathways involved in myocardial salvage remain hidden within the realms of cell therapy.

FOCUS BOX 3Proposed mechanisms by which cell-based therapy promotes cardiac repair

Neovascularisation
Cardiomyogenesis
Cell fusion leading to myocardial salvage
Production of paracrine factors promoting myocardial salvage/repair
Direct effects of myocardial salvage pathways

Cell therapy in acute myocardial infarction

Over the last 15 years, numerous clinical trials (from pilot through to Phase III - Table 2) have explored the efficacy of cell therapy in the setting of acute myocardial infarction (AMI). A major drive for this, is the human heart’s inadequate regenerative response to the myocardial necrosis sustained following AMI. Cell death from the ischaemic damage can lead to progressive ventricular dilation and dysfunction through adverse remodelling. This had been regarded as an inevitable process; however, the relatively recent demonstration of low frequency cell division within the adult heart [5] suggests a regenerative potential (albeit one that is unable to cope with the scale of cardiomyocyte loss following myocardial infarction). The existence of these endogenous repair mechanisms, as well as the concept of adult stem cell plasticity, suggests that cardiac repair may be achieved therapeutically in this setting. The demonstration of the ability of bone marrow-derived cells to transdifferentiate into cardiomyocytes and improve LV function following AMI in a mouse model [129] prompted the rapid initiation of human clinical trials. Although most of the trials have been conducted using bone marrow-derived progenitor cells, other cell types have also been tested and are discussed below.

Injection of cells with no sham procedure in controls

As is the case with the development of any new treatment, the initial clinical trials had no sham procedure in controls; instead, they compared cell infusion in addition to primary angioplasty to either primary angioplasty alone or historical controls.

In one of the first human trials, Strauer et al. evaluated the intracoronary infusion of BMCs in 20 patients 7 days after AMI [151]. This study appeared to confirm the findings of the earlier animal studies, with the cell-treated group showing a significant improvement in myocardial perfusion and a reduced infarct region. Conversely, ASTAMI, which only included patients with anterior AMI, showed that an intracoronary infusion of BMCs 4-8 days post-infarction did not have a beneficial effect on LVEF compared to PCI alone at 6 months [152]. This lack of beneficial effect may be explained by the different cell processing protocol used in this trial - a comparison of different isolation protocols revealed a vastly reduced recovery of mononuclear cells and a nullification of the neovascularisation capacity when the ASTAMI cell isolation and storage protocol was used [153]. BOOST, which was slightly larger, randomised 60 patients to either an intracoronary infusion of BMCs or standard therapy 4.8 days post PCI. At 6 months, there was a significant improvement in global LVEF in the cell-treated group, although there was no effect on LV remodelling. However, this improvement was not maintained at 18 months. The mean number of bone marrow cells that were infused contained 9.5 x 10⁶ CD34⁺ and 3.6 x 10⁶ haematopoietic colony-forming cells. The improvement in LVEF did not correlate to the number of CD34⁺cells or haematopoietic colony-forming cells [154]. The first long-term follow up in BALANCE (involving patients who underwent an intracoronary BMCs transplantation 7 days post-AMI) resulted not only in an early significant improvement in ejection fraction and infarct size, but significant improvements in LV performance as well as Quality of Life scores and mortality at 5 years [155].

However, the interpretation of these studies is questionable given the lack of an interventional placebo infusion in the control group. The solution is, therefore, trials that involve a placebo infusion arm; although the ethical considerations surrounding an invasive, interventional procedure with little therapeutic benefit in patients must be solved.

RANDOMISED CONTROLLED TRIALS INVOLVING CONTROLS WITH PLACEBO REINFUSION

In order to create a suitable comparator group, a number of clinical trials have been published where the control group had a bone marrow aspiration and a placebo infusion procedure. In TCT-STAMI, patients who received intracoronary BMCs had a significant improvement in LVEF at 6 months compared to the control group [156]. In REPAIR-AMI (a controlled trial which included 204 patients), the treatment group showed a significant improvement in global LVEF at 4 months as measured by quantitative left ventricular angiography [157]. Furthermore, the pre-specified cumulative endpoint of death, MI or revascularisation was significantly reduced at 4 months and maintained at 1-year follow-up [158]. The mean increase in LVEF in the BMSC group was 2.5%, and there was an inverse relationship between the baseline EF and the degree of improvement. For example, patients with a baseline EF below the median value (48.9%) had an absolute increase in global EF that was 3 times higher than that in the placebo group. By contrast, the improvement in LVEF in patients with a baseline EF that was above the median value was not significant (0.3%). The timing of cell infusion post-PCI also affected the primary endpoint; patients in whom the cells were infused ≥5 days post-PCI were the only ones who derived benefit.

By contrast, LEUVEN-AMI showed that an intracoronary infusion of BMCs within 24 hours of reperfusion, although associated with a greater reduction in infarct size and an improvement in regional systolic function, had no effect on the global left ventricular function compared to controls [159]. The early infusion of BMCs during the ischaemia reperfusion window (i.e. within 6 to 8 hours of reperfusion) was assessed in REGENERATE-AMI. In this study, 100 patients presenting with anterior MI (within 24 hours of primary PCI) were randomised to either the intracoronary delivery of autologous BMCs (n = 55) or placebo (n = 45). The primary outcome, change in LVEF at 1 year, increased in both the cell-treated and control groups but the difference was not significant between them. A secondary endpoint of myocardial salvage index, however, was significantly improved in the cell-treated patients [160]. Although this study failed to achieve its primary endpoint, it did suggest that early cell administration may lead to a mechanistic benefit which would need a larger trial to demonstrate a clinical difference.

However, the results of more contemporary studies, such as BOOST-2 (which reassessed the therapeutic potential of clonogenic vs nonclonogenic nucleated BMCs over a decade after the original BOOST trial), did not support the use of nucleated BMCs in patients with STEMI and moderately reduced LVEF treated according to the current standards of early PCI and drug therapy [161]. PreSERVE-AMI, the largest trial of autologous cell therapy conducted in the US, also failed to demonstrate a difference between cell-treated patients (CD34⁺) and a diluent control in the primary endpoint (change in myocardial perfusion). Secondary analysis did however confirm an improvement in ejection fraction in patients receiving cell therapy in addition to primary angioplasty [162].

TRIALS WHICH USED 2 DIFFERENT CELL POPULATIONS

Although the majority of trials have used unselected BMCs, a few have compared specific cell populations. The TOPCARE-AMI investigators randomised patients to an intracoronary infusion of either BMCs or ex vivo expanded circulating progenitor cells 4 days after AMI [163]. The mean number of cells infused was 245 x 10⁶ and contained haematopoietic progenitor cells, mesenchymal cells and stromal cells. At 4 months, both groups displayed a significant improvement in global and regional LV function, a beneficial effect on the post-infarction remodelling process (manifest by a profound improvement in wall motion abnormalities in the infarct area) and a significant reduction in end-systolic LV volume. LVEF further improved at 12 months, resulting in a total increase of 9.3% [164]. Interestingly, there was no difference between the two active treatment groups. The 5 year follow-up data also continued to demonstrate safety and favourable effects on LV function [165]. REGENT included patients with anterior MI who were randomised to receive an intracoronary infusion of unselected (n=80) or selected CD34⁺CXCR4⁺ (potentially a pure stem cell phenotype; n=80) BMCs or to standard therapy alone (n=40) [166]. Although patients in the treatment group had a 3% improvement in LVEF, it did not reach statistical significance. However, the primary endpoint analysis only included <60% of the total population of patients which decreased the statistical power of the trial to detect a difference between the groups. Subgroup analysis showed that baseline EF below the median value (37%) was an independent predictor of a significant (≥5%) increase in LVEF after treatment with BMCs. HEBE, one of the largest trials to date (n=200), compared the intracoronary infusion of BMCs or mononuclear peripheral blood cells (delivered 3-8 days post-AMI) to standard therapy. At 4 months, the three groups did not differ significantly in terms of changes to regional or global left ventricular function [167].

Mobilised progenitor cells in acute myocardial infarction

G-CSF is a haematopoietic cytokine [168] that has been shown to enhance the translocation of BMCs to the infarcted region post-MI [169] [129] [170]. G-CSF exerts its benefit on the myocardium through a number of mechanisms which include: cardiac regeneration [129], myocardial protection from apoptosis [171], the reduction of myocardial fibrosis [172] and an accelerated healing process [173]. Factors which influence the beneficial effect of G-CSF include: the timing of G-CSF administration, the patient’s age [319] and the presence of comorbidities such as diabetes, hypertension and dyslipidaemia [174]. Due to these multiple factors, there appears to be significant heterogeneity in the effect of G-CSF on LV function following AMI ( Table 3a). Although some trials demonstrated a significant improvement with G-CSF, meta-analyses have shown that G-CSF has no overall beneficial effect on LVEF [175] [176]. Importantly, the administration of G-CSF after AMI has not been shown to be as dangerous as some had predicted - a meta-analysis demonstrated that G-CSF therapy is safe and not associated with an increased incidence of in-stent restenosis [176]. A 10 year follow up of RIGENERA (the longest available follow up of G-CSF treatment in patients with AMI) also demonstrated no safety concerns, alongside an improvement in adverse LV remodelling and clinical status [177]. Unfortunately, STEM-AMI, the first Phase III trial, was terminated early due to slow recruitment [178]. The results of the main study are still awaited, although an imaging substudy has suggested a beneficial effect on imaging measurements of cardiac remodelling [179]. SITAGRAMI, the most recent and largest Phase III study, assessed the efficacy of G-CSF and Sitagliptin in 174 patients, but failed to show an improvement in LVEF at 6 months or a reduction in clinical events at 12 months [180] or long-term follow up [181].

Combination cytokine and cell therapy

In order to establish whether there is a synergistic effect between G-CSF and cell-based therapies, several trials have combined these treatments ( Table 3b). Whilst the earlier and smaller studies demonstrated an improvement on cardiac function [182] [183] [184], the largest trial [185] did not - again, demonstrating the mixed results of cell therapy in the setting of acute myocardial infarction.

Phase III clinical trials of cell-based therapy for acute myocardial infarction

Given the promising signals seen in the Phase II clinical trials, and as a result of the original ESC Task Force consensus concerning stem cells and cardiac repair [186], a pan-European academic Consortium was formed (comprising of some of the most experienced researchers in this field) to design and deliver BAMI - the first Phase III trial of cell therapy for acute myocardial infarction. The trial was rigorously designed (patient selection, cell processing and the timing of infusion) from existing Phase II trial data collected over the preceding decade and compared current best practice to an intracoronary infusion of BMMNCs 2-8 days after successful reperfusion for AMI [187].

BAMI, however, ran into several important difficulties. The trial highlighted the challenges regarding the regulatory processes surrounding the use of cell-based products in pan-European, multicentre trials (37 sites were initiated into the study, out of which 28 recruited and 23 delivered BMMNC therapy). Whilst the logistical issues were ultimately solved, the cost in time and money significantly impacted the trial’s success - only 375 patients were recruited out of the target of 3,000 [301].

The trial’s primary endpoint (all-cause mortality at 2 years) was powered around an estimation of 12% based on data at the trial’s design in 2012. However, as Bolli has indicated, all major cell therapy trials conducted in STEMI patients over the past decade have reported low rates of mortality (averaging 1.2% at 1 year - range 0–3%) and MACE (e.g. average rate of HF admissions at 1 year, ∼3%; range 0–7%) even in high-risk patients with moderate to severe LV dysfunction [188]. Thus, given the low number of patients developing post-infarction heart failure due to the success of primary angioplasty following AMI; the number of patients needed to demonstrate a significant treatment effect, alongside the logistics and costs involved, suggest that future trials in this area will be prohibitively expensive.

Therefore, BAMI has taught us several important lessons. Most importantly, if clinical endpoints are to be used in the conventional design of Phase III trials in AMI, then a very large number of patients will need to be recruited. Novel therapies that consume considerable logistic resource are therefore curbed, unless a more efficient approach to Phase III clinical trials is used (i.e. composite primary endpoints with high event rates).

Summary

The last 15 years of clinical research has produced over 50 trials of cell and cytokine therapy for acute myocardial infarction. Despite the large number of studies, meta-analyses show no consistent signal to support the role of BMCs in improving outcomes (mortality) after acute myocardial infarction ( Figure 2). The heterogenous results of these trials may be due numerous reasons which include differences in cells types, cell isolation and preparation protocols (cell numbers and composition), the timing of the infusion procedure, baseline ejection fraction and study design.

Despite being shown to be safe, the size of the effect suggested by the preclinical studies of cardiac regeneration appears to have been “lost in translation” [189]. It is important to note that human studies have differed from the animal experiments in several fundamental aspects listed below.

Myocardial infarction in animal models is induced by coronary artery ligation instead of plaque rupture and thrombotic occlusion. This may be important as the effects of an intracoronary cell infusion may be blunted by microvascular obstruction from thrombus. Furthermore, whilst experiments in animals occurred in otherwise healthy subjects, clinical trials target a population of patients with chronic disease in whom autologous cell functionality may be impaired [190]. Moreover, in the early animal models, the infarct-related artery was not reperfused and cell injection was directly into the border zone around the infarct.

Furthermore, cell delivery in the early animal models was via direct intramyocardial injection, whilst most of the human studies to date have used an intracoronary infusion. Only 2 human studies have used an intramyocardial injection with conflicting results [191] [192]; although, CellProthera’s EXCELLENT study (NCT02669810) is still in recruitment.

It would appear, therefore, that subject to the discovery of a more efficacious cell type that would lead to clear cut benefits in the acute myocardial infarction setting, there is currently no clear mandate supporting the role of cell therapy as an adjunct to primary angioplasty. However, whether cell therapy has a more efficacious role in other cardiac disease (e.g. heart failure and refractory angina) remains to be seen.

FOCUS BOX 4Cell therapy in acute myocardial infarction

There have been numerous randomised controlled trials published using BMCs in the setting of AMI which have varied in terms of design, timing of cell delivery and cell dose.
Intracoronary infusion has been the most utilised delivery method in the AMI setting.
Phase III clinical trials have faced recruitment difficulties and have therefore been unable to definitively define the role of cell therapy in AMI.

Cell therapy in chronic ischaemic heart failure

The prevalence of heart failure is ever-increasing with aging populations, and it is estimated that 15 million people in Europe, or around 2% of the population, are living with heart failure [193]. Regardless of aetiology, heart failure is associated with significant morbidity and mortality, reflected by survival rates of 80.8% at 1 year, 48.2% at 5 years and 26.2% at 10 years [194]. Although current optimal medical and device therapies have improved early mortality, long-term prognosis remains poor. Interestingly, the severity of heart failure symptoms predicts prognosis irrespective of the degree of systolic dysfunction; several studies have shown a correlation between New York Heart Association (NYHA) functional classification and mortality [195] [196] [197]. These realities have fuelled intensive efforts to develop novel therapeutic strategies for managing patients with heart failure and cell-based therapy has been recognised as one of these emerging treatments. In this section, we will review the clinical trials of cell therapy in patients with heart failure secondary to ischaemic heart disease ( Table 4).

Transepicardial delivery of cell therapy

Transepicardial injection has the advantage of the direct visualisation of injection sites and an accurate delivery of cells to the peri-infarct area; although some areas, such as the septum, may not be accessible. This method is highly invasive and probably only suitable for patients undergoing concomitant open-heart surgery. The feasibility and safety of this approach was confirmed in a Phase I, non-randomised, multicentre pilot study published in 2005 where 30 patients with ischaemic heart failure undergoing CABG or left ventricular assist device surgery had autologous skeletal myoblasts (obtained from culture of a prior muscle biopsy) injected into the epicardium during surgery. Myoblasts were successfully transplanted in all patients without any acute injection-related complications or significant long-term, unexpected adverse events. Follow-up positron emission tomography (PET) scans showed new areas of glucose uptake within the infarct scar suggestive of improved myocardial viability. Echocardiography measured an average improvement in LVEF from 28% at baseline to 35% at 1 year and 36% at 2 years [66]. MAGIC was the first randomised, placebo-controlled study of skeletal myoblast transplantation in patients with left ventricular systolic dysfunction secondary to previous myocardial infarction who required coronary artery bypass surgery [69]. Cells were injected into the epicardium within scarred areas during open heart surgery. This study was stopped prematurely for reasons of futility, as cell injection did not improve regional or global left ventricular function to the level needed to reach the endpoint.

Bone marrow-derived cells have also been transplanted via the transepicardial route. One of the initial surgical studies injected unmanipulated bone marrow, obtained from a sternal aspirate, directly into areas of scarred myocardium in 14 patients undergoing CABG. There appeared to be an improvement in regional contractility in the segments treated with revascularisation and cell injection [198]. However, a follow-up randomised controlled trial by the same group with 63 patients undergoing elective CABG, failed to show any benefit of BMCs therapy when given by epicardial injection or via the vein graft conduit [199]. In contrast, a non-randomised study with a similar study design (40 patients undergoing CABG) showed that the additional cell injection was associated with better improvements in global left ventricular function and myocardial perfusion at 6 months [103]. The study’s limitations included no sham injection of placebo in the control group and standard 2D echocardiography served as the only measurement of global LV contractility. Hence, PERFECT, a Phase III double blind, randomised, placebo-controlled multicentre trial was conducted in 6 centres across Germany. It aimed to recruit 142 patients to assess the efficacy of a CD133⁺ injection during CABG. The study was conducted from October 2009 to March 2016 but was stopped due to slow recruitment after a positive interim analysis in March 2015. Eighty-two patients were randomised to 2 groups: receiving an intramyocardial injection of 0.5-5×10⁶ CD133⁺ or placebo. The treatment was shown to be safe with no difference between groups in terms of serious adverse events. Efficacy was assessed in 58 patients, demonstrating a 9.6% improvement in LVEF at 180 days (p=0.001), but with no difference compared to the placebo group (ANCOVA: Placebo + 8.8% vs. CD133⁺ +10.4%, ∆CD133⁺ vs placebo +2.6%, p=0.4) [106].

Transendocardial delivery of cell therapy

Percutaneous transcatheter intramyocardial injection of cell therapy has shown promise in early Phase I/II clinical studies. The main additional risks of this approach are ventricular perforation with pericardial effusion/tamponade and procedure-related ventricular arrhythmias - although the reported incidences of these have been low in the published clinical trials.

CAuSMIC used a NOGA-guided intramyocardial injection of skeletal myoblasts into areas of viable myocardium in 12 patients with severe ischaemic heart failure. At 1 year follow up, there was an improvement in NYHA functional class, Quality of Life scores and evidence of reverse ventricular remodelling in the cell-treated group [68]. In a similar study design, SEISMIC confirmed the safety and feasibility of a catheter-based intramyocardial injection of skeletal myoblasts (n=26). There appeared to be some improvement in patient symptoms, but the study failed to show any significant improvement in LVEF [67]. The transendocardial injection of BMCs has also been shown to be safe and have potentially beneficial effects. An early non-randomised study of 20 patients showed that the NOGA-guided intramyocardial injection of BMCs was safe and associated with improvements in exercise tolerance and myocardial perfusion [200]. Further small studies showed improvements in patient symptoms and LV function following a NOGA-guided intramyocardial injection of BMCs [201], and improvements in regional contractility and evidence of reverse remodelling in using the BioCardia Helix catheter system in chronically infarcted myocardium [202].

Larger phase II studies have built on these early results, demonstrating the safety and efficacy of cellular therapy. REGENERATE-IHD was a randomised, placebo‐controlled, single‐centre trial involving 90 patients with symptomatic ischaemic cardiomyopathy and no further treatment options. Randomisation was to one of three arms: peripheral, intracoronary (IC) or intramyocardial (IM). Patients were further randomised to active treatment or placebo in each arm. All patients, apart from the peripheral placebo group (saline only), received G-CSF for 5 days. At 1 year, only the IM BMC group showed a significant improvement in LVEF (4.99%; 95% confidence interval 0.33–9.6%; p = 0.038). This group also showed a reduction in NYHA class at 1 year and NT‐proBNP at 6 months. This finding was supported by post‐hoc between‐group comparisons [203]. TAC-HFT, a Phase I/II randomised, blinded, placebo-controlled study involving 65 patients with ischaemic cardiomyopathy (LVEF <50%) compared the injection of MSCs (n=19) with placebo (n = 11), and BMCs (n = 19) with placebo (n = 10). In these patients, the transendocardial injection of MSCs and BMCs appeared to be safe, with no patients having treatment-emergent serious adverse events at day 30. At 1 year of follow-up, the incidence of serious adverse events was 31.6% (95% CI, 12.6% to 56.6%) for MSCs, 31.6% (95% CI, 12.6%-56.6%) for BMCs and 38.1% (95% CI, 18.1%-61.6%) for placebo [204]. Despite promising results for efficacy, the sample size and multiple comparisons in both TAC-HFT and REGENERATE-IHD preclude a definitive statement about clinical effects. Additionally, not all studies in this patient group have shown positive results. FOCUS-CCTRN randomised 92 subjects with chronic ischemic heart failure to autologous BMCs (n = 61) or placebo (n=31). In this study, the transendocardial injection of BMCs compared with placebo did not improve LVESV (-0.9 mL/m(2) [95% CI, -6.1 to 4.3]; p = 0.73), maximal oxygen consumption (1.0 [95% CI, -0.42 to 2.34]; p = 0.17) or reversibility on SPECT (-1.2 [95% CI, -12.50 to 10.12]; p = 0.84) [205].

Two later trials also compared the efficacy of different cell types. POSEIDON was an early, Phase I/II randomised, non-controlled pilot study (n=30) that evaluated the transendocardial delivery of autologous versus allogeneic MSCs. Three different cell doses (20, 100, and 200 million cells) were tested in both treatment groups. Surprisingly, the lowest dose yielded the best outcomes in terms of LV volumes and LVEF. Moreover, despite its small size, POSEIDON provided preliminary evidence of the comparable safety and efficacy between autologous and allogeneic MSCs [206] [207]. CONCERT-HF was a Phase II trial of autologous MSCs and c-kit⁺ cardiac cells (alone or in combination) in patients with ischaemic heart failure. One hundred and twenty-five patients were randomised (1:1:1:1) to a transendocardial injection of MSCs and CPCs, MSCs alone, CPCs alone or placebo. CPCs and MSCs were associated with improved clinical outcomes (MACE and Quality of Life scores) without affecting left ventricular function or structure - suggesting possible systemic or paracrine cellular mechanisms. The combination of CPCs and MSCs appeared to be complimentary. CPCs alone were associated with reduced HF-MACE and MSCs alone were associated with increased Quality of Life scores; the combined cell group, however, had both improved Quality of Life scores and reduced HF-MACE [49]. Taken together, these trials demonstrated the technique’s safety and supported a rationale for larger studies to provide definitive efficacy evidence.

However, despite the promising results seen in Phase II, very few Phase III trials have been conducted. DREAM-HF, a Phase III clinical trial of mesenchymal precursor cells in chronic heart failure, has now been completed [208]. The final results have not been published but preliminary reports suggest that this trial has demonstrated a significant decrease in clinical cardiovascular events. CardiAMP is an ongoing Phase III clinical trial evaluating point of care BMMNC therapy in ischemic heart failure and so far has released favourable data (functional, echo and quality of life trends) from the roll-in cohort (n=10) [209].

Intracoronary delivery of cell therapy

There have been relatively few studies assessing the intracoronary delivery of BMCs in chronic ischaemic heart failure. In a randomised, crossover design TOPCARE-CHD evaluated the intracoronary delivery of CPCs versus BMCs. The BMC group showed an improvement in LVEF of 2.9% with no major adverse cardiac events [210]. By contrast, DanCell-CHF, in which patients received 2 treatments of intracoronary BMCs 4 months apart, showed no improvement in LV function at 1-year follow-up, although there was significant improvement in patient symptoms [211]. This study adds to the growing opinion that using ejection fraction as a marker of the beneficial effects of cell therapy may be inadequate, as patient symptoms seem to improve without any significant objective change in LVEF. STAR-Heart is the largest study to date of BMC therapy in chronic ischaemic heart failure. Of the eligible patients screened, 191 patients underwent intracoronary BMC therapy and the control group consisted of 200 patients who declined to have the active intervention. Over a 5-year follow-up period, intracoronary BMC therapy was associated with a significant improvement in LVEF as well as exercise capacity. It was also the first trial with long-term follow up to show a survival benefit in patients receiving cell therapy [212]. A criticism of this study has been the use of patients who declined the intervention in the control arm as this introduces an element of selection bias since they may be less likely to be compliant with medications and follow-up.

Cytokine and cell therapy for ischaemic heart failure

Treatment with G-CSF is an attractive option as it may provide a non-invasive method for promoting cardiac repair. There have only been a few clinical trials in the post-MI chronic heart failure setting. Two small, non-randomised studies demonstrated possible beneficial effects on symptoms and LV function; but, they also raised concerns regarding the worsening of angina during the treatment phase [213] [214].

The combination of G-CSF followed by progenitor cell harvest and direct delivery to the heart has the theoretical advantage of promoting cell proliferation, maturation and functional activation prior to injection; and there is some preliminary clinical data to support this approach [170] [215] [216]. There have only been a handful of trials assessing the safety and efficacy of combined G-CSF and cell therapy in patients with chronic ischaemic heart failure which have generally shown improvements in cardiac function, Quality of Life scores and NYHA class [203] [217] [218] [219] [220] .

A few trials have assessed the efficacy of G-CSF alone and in conjunction with cell therapy. TOPCARE-G-CSF randomised 32 patients to G-CSF alone or G-CSF stimulated CPCs and demonstrated safety and small improvements in cardiac function in both groups [217]. REGENERATE-IHD was a randomised, double-blind, placebo-controlled trial (n=30) which demonstrated that G-CSF alone lead to no clinical improvement in symptoms, LVEF or outcome compared to placebo. However, the combination of G-CSF and cell therapy injected via the intramyocardial route did lead to an improvement in cardiac function - suggesting a synergy between cytokine and cell therapy [203].

Summary

In summary, numerous studies using different cell types and different delivery methods have assessed whether cell therapy is beneficial in patients with established heart failure secondary to ischaemic heart disease. The intramyocardial route has been the most utilised cell delivery method in the chronic ischaemic heart failure setting, but the intracoronary route is more practical. As in the acute myocardial infarction setting, cell therapy appears to be safe, with potentially beneficial effects in terms of patient symptoms alongside modest improvements in LV function. However, in contrast to the acute myocardial infarction setting, meta-analysis also suggests cell therapy improves mortality ( Figure 3) and a recent review of cell therapy for chronic ischaemic heart disease and congestive heart failure concluded that patients with heart failure are more likely to derive benefit from cell-based therapies than those with acute myocardial infarction [221]. Further large, multicentre, randomised controlled trials are required (and are in process) to define the ideal cell type, delivery method and patient population that will derive the most benefit from this biological therapy.

FOCUS BOX 5Cell therapy in chronic ischaemic heart failure

Skeletal myoblasts, bone marrow derived cells and mesenchymal stem cells have been the most commonly investigated cell types.
The intramyocardial delivery of cells may be the preferred delivery method, but the intracoronary delivery is more practical.
Phase I/II trials have provided evidence of safety and symptomatic improvement. Phase III trials are needed.
Chronic ischaemic heart failure appears to be a more promising target for cell-based therapies than acute myocardial infarction.

Cell therapy in dilated cardiomyopathy

Dilated cardiomyopathy (DCM) is a leading cause of heart failure and remains the most common indication for cardiac transplantation worldwide [222] [223]. The prevalence of DCM is estimated at 40 cases per 100,000 individuals with an annual incidence 7 cases per 100,000 individuals [224]. The prognosis of DCM is highly variable; earlier studies reported 5 year mortality rates of 50% which have declined to 20% in more recent reports. This improvement in survival reflects both early disease detection and advances in heart failure therapy. However, the prognosis and quality of life in symptomatic heart failure patients remain worse than many malignancies and serious chronic conditions such as arthritis and chronic lung disease [225].

Dilated cardiomyopathy represents the final common morphologic outcome of various biologic insults resulting in myocardial necrosis and chronic fibrosis. The proposed mechanisms involved in the pathogenesis of DCM include genetic predisposition, viral infection, myocardial ischaemia, antibody-mediated cytotoxicity and the apoptosis of cardiomyocytes. Prognostic markers of increased mortality in DCM include the degree of LV dysfunction, a higher NYHA functional class and high brain natriuretic peptide (BNP) levels. BNP is a powerful predictor of all-cause mortality in heart failure [226] and may also predict the response to therapy [227].

Current conventional therapies for DCM do not correct underlying defects in cardiac muscle. The only therapeutic option which currently addresses fibrosis and cardiomyocyte loss is heart transplantation. This has a significant impact on mortality with a survival rate following heart transplantation of 83% at 1 year and 72% at 5 years [228]. However, many patients do not receive a transplant due to the shortage of donors and the stringent selection criteria. Therefore, the emergence of cell therapy as a novel therapeutic option to address the progression of myocardial dysfunction has generated a lot of enthusiasm. This section reviews cell and G-CSF therapy in DCM, highlighting some of the major animal and clinical trials and discussing some of the challenges which lie ahead.

Preclinical and clinical trials in dilated cardiomyopathy

Several animal studies assessing the safety and efficacy of cell therapy for DCM have published encouraging results. In one rat study, the intramyocardial injection of mesenchymal stem cells was associated with enhanced LV remodelling, an improvement in survival, the induction of angiogenesis and the inhibition of myocardial fibrosis [79]. The promising results from these animal studies led to the publication of case reports in man which showed that cell transplantation led to an improvement in cardiac function, symptoms and ventricular remodelling [229] [231] [233]. An early case report detailed a clear improvement in LVEF in a 58 year old man with end-stage heart failure secondary to DCM following an intramyocardial injection of bone marrow-derived CD133⁺ cells through a minimally invasive surgical approach [229]. Another case report demonstrated an improvement in LVEF from 20% to 45% in a critically ill 2 year old child with DCM following an intracoronary injection of autologous bone marrow-derived progenitor cells [233]. These positive results led to the quick initiation of clinical trials ( Table 5).

One of the first clinical trials, ABCD, assessed the intracoronary delivery of autologous BMMNCs in 44 patients. At 6 months, the cell-treated group showed a significant improvement in LVEF by 5.4% and an improvement in NYHA functional class [234]. In TOPCARE-DCM, the intracoronary administration of BMCs was associated with a regional and global improvement in LVEF which was linked to an improvement in microvascular function [235]. This may be important as microvascular dysfunction in DCM is associated with an increase in mortality and the progression of heart failure [236]. However, the most recent trials to use BMMNCs have failed to show an improvement in cardiac function - suggesting the need for a different cell type or combination therapy [237] [238].

Two clinical trials have explored mesenchymal stem cells. Butler and colleagues evaluated the safety and efficacy of intravenously applied allogeneic MSCs (1.5 × 10⁶/kg) in 22 patients with DCM in a placebo-controlled crossover trial. Although at 90 days no improvement in LVEF was observed, the data suggested patients had a better exercise capacity and quality of life after cell treatment [239]. The larger POSEIDON-DCM trial compared the safety and efficacy of a transendocardial injection of either autologous or allogeneic MSCs in 37 patients with non-ischaemic DCM. The results demonstrated a significantly better response to allogeneic MSCs compared to autologous MSCs; the former was associated with a greater improvement in LVEF (8% versus 4%), a decrease in myocardial inflammation and an increase in patient exercise capacity and quality of life [240]. Several factors may have accounted for the greater efficacy of allogeneic MSCs, including MSC donor age (the mean age in the allogeneic MSC group was about 50% of the autologous MSC group) and the possible adverse impact of the systemic pro-inflammatory milieu of heart failure on autologous MSCs. Interestingly, the only study that has compared MSCs to BMMNCs showed similar efficacy between the 2 cell types; however, it should be noted that the number of patients in each treatment arm was small and therefore the study may have been underpowered to show meaningful differences [241].

The potential benefit of cell therapy in DCM is still being evaluated; and although there have been some positive signals using cell therapy alone, the effectiveness has been mixed. In order to address this, several studies have combined cell and cytokine therapy which are discussed below.

Cytokine and cell therapy in dilated cardiomyopathy

Cytokine therapy with G-CSF in animal models of DCM is associated with reduced cardiomyocyte apoptosis as well as an improvement in cardiac function [242] [243] [244]. There have been single case reports [245] and small studies [214] suggesting possible beneficial effects of G-CSF in patients with DCM; however, these preliminary findings have not been pursued in larger studies.

Numerous trials have evaluated the efficacy of combined cytokine and cell therapy. REGENERATE-DCM assessed whether G-CSF administration with or without adjunctive cell therapy improved global LV function. At 12 months, patients who received both the intracoronary infusion of BMMNCs and G-CSF had significantly improved LVEF (7.04%), NYHA class and Quality of Life scores compared to placebo. However, no benefit was seen with G-CSF therapy alone, highlighting the need for combined cell and cytokine therapy for clinical effect [246]. Multiple other Phase II trials (all but one have used CD34⁺ cells mobilised by G-CSF) have consistently shown improvements in cardiac function and symptoms [247] [248] [249] [250] [251] [252] - demonstrating the safety and efficacy of this therapy in the treatment of DCM.

Summary

To date, the results of the clinical trials targeting patients with DCM (particularly in combination with cytokine therapy) have demonstrated consistently positive results. Although the exact mechanism by which combination therapy has beneficial effects remains unclear, these trials do present some of the most convincing data for the use of cell and cytokine therapy in the treatment of non-ischaemic heart failure.

FOCUS BOX 6Cell therapy in dilated cardiomyopathy

Dilated cardiomyopathy is a leading cause of heart failure and remains the most common indication for cardiac transplantation worldwide.
Cytokine therapy with G-CSF in animal models of dilated cardiomyopathy is associated with reduced cardiomyocyte apoptosis as well as an improvement in cardiac function.
Randomised clinical trials have demonstrated the benefit of cell-based therapy in improving cardiac function and reducing heart failure symptoms - particularly in conjunction with cytokine therapy (G-CSF).

Cell therapy in refractory angina

Refractory angina is an ever-growing problem as populations age and CAD-related survival rates increase. In the multinational, observational CLARIFY study, 20% of patients with stable coronary artery disease had anginal symptoms; and these were associated with worse clinical outcomes over the 2 years of follow up [253]. The management of angina remains challenging using conventional treatment strategies; a significant number of patients remain symptomatic despite PCI, CABG and anti-anginal drugs. Based on the COURAGE trial (which assessed the clinical outcomes of patients with stable coronary disease on optimal medical therapy with or without PCI), 42% of patients in the optimal medical therapy arm and 24% of patients in the PCI arm still had angina at 1 year [254].

Pre-clinical and early clinical studies have indicated that cell therapy represents a potential treatment modality for these patients. The Phase I-II trials, which have predominantly utilised the intramyocardial delivery method, have reproducibly demonstrated the therapy’s safety and established a beneficial effect ( Table 6). As with acute myocardial infarction, ischaemic heart failure and dilated cardiomyopathy, most studies so far have used autologous bone marrow-derived cells. The first non-randomised study in 2003 by Tse et al. (n=8) showed that the transendocardial injection of autologous BMMNCs increased myocardial perfusion, wall thickening and motion [255]. Unfractionated BMCs have since continued to demonstrate benefits to clinical outcomes and physiological parameters [201] [256] [257] [258] [259] [260] [261] [262] [263] [264] [265]. Numerous studies that have used endothelial progenitor cells (CD133⁺) have also shown similar improvements in symptoms and perfusion [107] [108] [109] [110] [112] [266] [267]. Two trials have examined adipose-derived cells. In ATHENA, 31 patients underwent an intramyocardial delivery of autologous adipose-derived cells or placebo. Changes in NYHA, CCS and V02 max favoured the cell-treated group, but there was no difference in left ventricular function or volumes [268]. MyStromalCell randomised 60 patients to an intramyocardial injection of VEGF-A165 culture-stimulated adipose-derived stromal cells or placebo. Although exercise capacity improved at 6 months, there was no significant difference when compared to placebo [38]. At 3 years, the cell-treated patients had improved cardiac symptoms (CCS) and unchanged exercise capacity, compared to a deterioration in the placebo group [39].

The intracoronary delivery method has been used in fewer studies; but has demonstrated similarly promising improvements in clinical and physiological markers [109] [266] [269] [270] [271]. To date, the largest trial to examine this simpler method of cell administration was a randomised, placebo-controlled trial by Wang et al. published in 2010. In this study, 112 patients with intractable angina, no further revascularisation options and evidence of perfusion defects on SPECT imaging, were randomised to an intracoronary injection of autologous bone marrow-derived CD34⁺ cells or placebo. There were no serious adverse events in either group and the cell-treated group showed a reduction in angina episodes and improved myocardial perfusion [269].

Combined cell and cytokine therapy for refractory angina

Phase I/II clinical trials have also assessed the combination of G-CSF and cell therapy (so far either using CD34⁺ or CD133⁺ cells), and have demonstrated symptomatic improvements and functional benefit [109] [111] [270] [272] [273]. The first study to examine combined cell and cytokine therapy was published by Losordo et al. in 2007. The trial involved 24 patients and demonstrated that an intramyocardial injection of autologous, peripherally harvested CD34⁺ cells following a course of G-CSF was safe and associated with a symptomatic improvement in angina class; although there was no clear improvement in myocardial perfusion [272]. ACT34-CMI built on these results, randomising 167 ‘no-option’ patients with CCS class III-IV angina to low-dose (1x10⁵ CD34/kg body weight), high-dose (5 x 10⁵ CD34/kg) or placebo. At 6 and 12 months, the low-dose group had a significant reduction in angina frequency (p = 0.02, 0.035) and improvements in exercise tolerance testing (p = 0.014, 0.017) compared to the placebo group [273]. At 24 months, both the low and high dose cell-treated groups had a significant reduction in angina frequency (p = 0.03). Autologous CD34⁺ cell therapy was also associated with an improvement in MACE (33.9% in control, 21.8% in low dose and 16.2% in high dose group) and a trend towards a reduction in mortality (7 deaths, 12.5% in the control group; 1, 1.8% in the low dose group; 2, 3.6% in the high dose group) [274].

Late Phase clinical trials for refractory angina

In light of all these positive studies, larger studies have been initiated; although none have reached completion (RENEW was terminated early by the sponsor for logistical reasons, REGENT-VSEL due to a slow recruitment rate and the ATHENA programme due to procedure-related issues). RENEW, the only Phase III trial, was designed to definitively address the role of G-CSF and CD34⁺ cell therapy in the treatment of refractory angina [275]; however, it only recruited 112 out of the planned 444 patients. The results were consistent with observations from the earlier Phase studies, demonstrating a positive effect on exercise capacity and angina frequency, but the study was not adequately powered to conclusively define the therapy’s efficacy [276].

The combination of the positive signals seen in the Phase I/II trials and the failure to conduct a definitive Phase III trial has left important unanswered questions. In the absence of data from late Phase clinical trials, pooled and meta-analyses have been performed to look for clinical signals in larger groups of patients. A pooled analysis evaluating the intramyocardial injection of BMCs in refractory angina patients with chronic myocardial ischemia (n=100) demonstrated significant improvements in clinical outcomes and Quality of Life scores at 5 years, alongside safety at 10 years [277]. Another patient-level pooled analysis, which revisited 3 of the CD34⁺ trials (including the aborted RENEW trial, n = 304), compared the intramyocardial injection of autologous CD34⁺ cells with placebo. CD34⁺ cell therapy resulted in clinically meaningful improvements in total exercise time and angina frequency at 3, 6 and 12 months. Importantly, an impressive reduction in 24 month mortality was also reported (12.1% vs. 2.5%; p = 0.0025), alongside numerically reduced MACE (38.9% vs. 30.0%; p = 0.14) [278]. Caution is needed, however, as these analyses were performed retrospectively from three different trials which used comparable but not identical designs. Nevertheless, the most recent metanalysis by Jones et al. which looked at 8 randomised controlled trials (n = 526), also identified that, alongside improvements in indices of angina (episodes and use of antianginal medications), cell therapy also improved cardiovascular outcomes (mortality and MACE - Figure 4) [279]. These results are exciting; displaying safety and efficacy with respect to pain relief and potential improvements in harder endpoints (i.e. mortality) and underlines the need for further definitive trials.

Summary

Overall, cell therapy for refractory angina appears to be a very promising treatment option. As with the other cardiac conditions, refinement regarding cell type, targeting and patient selection (e.g. the use/development of better biomarkers to identify subpopulations that will respond the best) may well lead to even better results.

FOCUS BOX 7Cell therapy in refractory angina

Phase I/II clinical trials have reproducibly demonstrated the benefit of cell therapy for refractory angina.
Intramyocardial and intracoronary delivery methods have both shown beneficial effects in terms of improvement in angina symptoms, exercise capacity and myocardial perfusion.
Phase III clinical trials of cell therapy for refractory angina are needed.

Current limitations and future challenges

Following two decades of clinical trials, there is still no consensus regarding which patient groups to target - although the current beneficial effects are mainly seen in heart failure patients, particularly those with dilated cardiomyopathy and refractory angina. Moreover, although a few consensus documents have been produced [186] [280] [281], there is no universally-accepted, standardised method for cell choice, cell processing technique or administration route (and regulatory approvals are increasingly challenging).

One of the main sticking points has been the failure of Phase III trials to definitively resolve these questions. BAMI, the only Phase III trial of cell therapy in acute myocardial infarction, encountered complex logistical issues (relating to the processing of cells and differing local regulations across Europe) and an unexpectedly low event rate, meaning that being able to demonstrate a significant outcome would require vast numbers of patients and be extremely costly. In the heart failure setting, although more Phase III trials have commenced; unfortunately, many have been terminated prematurely - mainly due to logistical issues. Therefore, cell therapy continues to be a complex and costly procedure in its research phase, making Phase III trials untenable. If research is to continue in the field, the logistical issues of cell processing (e.g. point of care delivery) as well as regulatory issues surrounding cell products (and the associated expense) must be overcome.

Another limitation has been the debate around the use of surrogate endpoints in clinical trials due to their lack of acceptance for regulatory approval. The complexities of translational research mean that surrogate endpoints can be used to reduce the size and duration of these expensive trials of new, innovative therapies. Without reducing the cost of these studies, they will remain out of reach of academia and perhaps represent too much risk for industry. Industry partners are also reluctant to invest in this field due to the difficulties in securing IP around cell-based products, meaning that the large sums of money needed to take a product to market are out of reach.

A more recent obstacle has arisen with the withdrawal of the NOGA intramyocardial injection system from the market. Although other intramyocardial injection systems are available, they have been used far less in clinical trials and do not have the same targeting capability. Therefore, from a practical perspective, further exploration of cell delivery via the intracoronary route as an alternative method is important. This may well prove similarly efficacious and, on a positive note, it is a far more accessible approach (it is cheaper and no specialist training is required).

In summary, we still need to translate the exciting preclinical and early Phase data into cost effective clinical trials which, if positive, can be adopted by the medical community to change clinical practice.

Conclusion

Interventional cardiology plays an important role in translating the preclinical promise of regenerative medicine into the reality of clinical trials. Given the need for new therapeutic solutions in the treatment of heart failure and acute myocardial infarction, the translation from bench to bedside has occurred rapidly. Whilst embryonic stem cells theoretically remain the best candidate for myocardial regeneration, trials have focussed on adult-derived progenitor cells (whether autologous or allogeneic in origin), as they provide a far more pragmatic approach to cell therapy. The positive early Phase I/II studies (which tested different cell types and delivery methods) established biologics as a new therapeutic reality; and, based on these results, Phase III trials with standardised methodology and hard clinical endpoints were designed. However, the trials themselves have proved difficult (both to initiate and to complete) due to funding, logistic and regulatory issues. The challenge therefore remains to successfully translate these positive preclinical and early Phase I/II studies into meaningful Phase III trials and establish this therapy in routine clinical practice.

Personal perspective - Anthony Mathur

Regenerative medicine still has the potential to change the practice of medicine. Although the field has not lived up to initial expectations, the concept of cell therapy, particularly if autologous, can still be seen as an elegant therapeutic solution to a catalogue of complex diseases. Interventional cardiology has played an important role in translating the exciting preclinical data into the reality of clinical trials. Whilst a convincing efficacy signal from a Phase III trial is still awaited, we have made several important steps in understanding how to deliver biologics to patients. We know that we can deliver cell therapy to the heart in a number of different ways, and in so doing we have addressed many of the issues relating to safety. The significance of this pioneering step should not be underestimated. Several different cell types have been tested and, undoubtedly, more will enter into clinical trials as basic research learns from the successes and failings. The field has extended beyond the relatively straightforward approach of using BMCs to the complexities of engineered cells. The embryonic stem cell will continue to represent the ultimate cell for regeneration of the heart. However, overcoming the ethical and technical issues to do this safely will not be easy and may yet take several decades. It is likely that a cell type will be identified which represents a compromise between the simplicity of autologous BMCs and the potential of ESCs. It is also likely that the current interest in the field will lead to a better understanding of the proteome/secretome which controls myocardial regeneration, and that this will open the way for targeted therapies using modified cells or genes. Bioscaffolds, cell sheet technology and the bioengineering of the decellularised heart will all offer alternative solutions to direct cell therapy.

There can be no illusion that changing the practice of medicine will be quick and the last 15 years of research in this field has demonstrated the pitfalls of trying to translate successful basic science into meaningful clinical results. The ultimate “holy grail” of the field - to grow new heart muscle which completely restores cardiac function - remains some way off, but along the way new discoveries (both biological and technical) are changing patients’ lives. It is important that interventional cardiology continues to support this area of research to ensure that the translation into man can progress. Clearly there is still much to be done.

We wish to acknowledge the help of Dr Martina Colicchia, Dr Abdul Mozid and Dr Daniel A. Jones in the preparation of this chapter.

Recommended bibliography

The following list contains meta-analyses for specific disease states and key publications (both basic science and clinical trials) that have been pivotal in developing this field.

Meta-analyses

Jones DA, Weeraman D, Colicchia M, Hussain MA, Veerapen D, Andiapen M, Rathod KS, Baumbach A, Mathur A. The Impact of Cell Therapy on Cardiovascular Outcomes in Patients With Refractory Angina. Circ Res. 2019;124:1786-1795.
Fisher SA, Doree C, Mathur A, Taggart DP, Martin-Rendon E. Cochrane Corner: stem cell therapy for chronic ischaemic heart disease and congestive heart failure. Heart. 2018;104:8-10.
Fisher SA, Zhang H, Doree C, Mathur A, Martin-Rendon E. Stem cell treatment for acute myocardial infarction. Cochrane Database Syst Rev. 2015;30:CD006536.

Basic science

Orlic D, Kajstura J, Chimenti S, Jakoniuk I, Anderson SM, Li B, Pickel J, McKay R, Nadal-Ginard B, Bodine DM, Leri A, Anversa P. Bone marrow cells regenerate infarcted myocardium. Nature. 2001;410:701-5.
Lovell MJ, Mathur A. The role of stem cells for treatment of cardiovascular disease. Cell Prolif. 2004;37:67-87.
Bergmann O, Zdunek S, Felker A, Salehpour M, Alkass K, Bernard S, Sjostrom SL, Szewczykowska M, Jackowska T, Dos Remedios C, Malm T, Andrä M, Jashari R, Nyengaard JR, Possnert G, Jovinge S, Druid H, Frisén J. Dynamics of Cell Generation and Turnover in the Human Heart. Cell. 2015;161:1566-75.

Clinical trials

Malliaras K, Makkar RR, Smith RR, Cheng K, Wu E, Bonow RO, Marbán L, Mendizabal A, Cingolani E, Johnston PV, Gerstenblith G, Schuleri KH, Lardo AC, Marbán E. Intracoronary cardiosphere-derived cells after myocardial infarction: evidence of therapeutic regeneration in the final 1-year results of the CADUCEUS trial (CArdiosphere-Derived aUtologous stem CElls to reverse ventricUlar dySfunction). J Am Coll Cardiol. 2014;63:110-22.
Patel AN, Henry TD, Quyyumi AA, Schaer GL, Anderson RD, Toma C, East C, Remmers AE, Goodrich J, Desai AS, Recker D, DeMaria A; ixCELL-DCM Investigators. Ixmyelocel-T for patients with ischaemic heart failure: a prospective randomised double-blind trial. Lancet. 2016;387:2412-21.
Menasché P, Vanneaux V, Hagège A, Bel A, Cholley B, Parouchev A, Cacciapuoti I, Al-Daccak R, Benhamouda N, Blons H, Agbulut O, Tosca L, Trouvin JH, Fabreguettes JR, Bellamy V, Charron D, Tartour E, Tachdjian G, Desnos M, Larghero J. Transplantation of Human Embryonic Stem Cell-Derived Cardiovascular Progenitors for Severe Ischemic Left Ventricular Dysfunction. J Am Coll Cardiol. 2018;71:429-438.
Mathur A, Fernández-Avilés F, Bartunek J, Belmans A, Crea F, Dowlut S, Galiñanes M, Good MC, Hartikainen J, Hauskeller C, Janssens S, Kala P, Kastrup J, Martin J, Menasché P, Sanz-Ruiz R, Ylä-Herttuala S, Zeiher A; BAMI Group. The effect of intracoronary infusion of bone marrow-derived mononuclear cells on all-cause mortality in acute myocardial infarction: the BAMI trial. Eur Heart J. 2020;41:3702-3710.

Cell-based regenerative therapy

Introduction

What is the ideal cell type for cardiac repair ?

STEM CELL POTENTIAL – AN OVERVIEW

ALLOGENEIC CELL TYPES

Embryonic stem cells (ESCs)

Foetal cardiomyocytes

Human umbilical cord-derived stem cells (hUCSCs)

AUTOLOGOUS CELL TYPES

Induced pluripotent stem cells (iPSCs)

Adipose-derived stem cells (ADSCs)

Resident cardiac stem cells

Epicardium-derived progenitor cells (EPDCs)

Skeletal myoblasts

Mesenchymal stem cells (MSCs)

Endothelial progenitor cells

Bone marrow-derived stem/progenitor cells (BMSCs/BMPCs)

Mixed cell types

Summary

Mechanism of actions of cell based regenerative therapy

Cell therapy in acute myocardial infarction

Injection of cells with no sham procedure in controls

RANDOMISED CONTROLLED TRIALS INVOLVING CONTROLS WITH PLACEBO REINFUSION

TRIALS WHICH USED 2 DIFFERENT CELL POPULATIONS

Mobilised progenitor cells in acute myocardial infarction

Combination cytokine and cell therapy

Phase III clinical trials of cell-based therapy for acute myocardial infarction

Summary

Cell therapy in chronic ischaemic heart failure

Transepicardial delivery of cell therapy

Transendocardial delivery of cell therapy

Intracoronary delivery of cell therapy

Cytokine and cell therapy for ischaemic heart failure

Summary

Cell therapy in dilated cardiomyopathy

Preclinical and clinical trials in dilated cardiomyopathy

Cytokine and cell therapy in dilated cardiomyopathy

Summary

Cell therapy in refractory angina

Combined cell and cytokine therapy for refractory angina

Late Phase clinical trials for refractory angina

Summary

Current limitations and future challenges

Conclusion

Personal perspective - Anthony Mathur

Recommended bibliography

Meta-analyses

Basic science

Clinical trials

RELATED MEDIA

REFERENCES

SHARE YOUR COMMENT