Cell transplantation in heart failure: where do we stand in 2016?

Over the past 30 years, the clinical treatment options for ischaemic cardiomyopathy (both acute and chronic) have significantly evolved, from revascularization via surgical bypass or percutaneous interventions (and the hybrid technique of both approaches), to mechanical circulatory support therapy and cardiac transplantation. Although the basic tenets of these modalities, including restoring blood flow and preserving viable myocardium, have remained the same, the means to those ends have been expanded by the introduction of cell-based therapy. Cell transplantation in patients with ischaemic myocardium has been accelerated by exciting advances in the understanding of developmental biology as well as bolstered by new technology geared towards minimally invasive, precision delivery. Herein, we take a look at the history of cell therapy for heart failure, where we stand now, and what may lie ahead.

In 2001, Orlic et al . [ 1 ] published a groundbreaking manuscript concluding that local, epicardial delivery of lineage-depleted stem cells isolated from the bone marrow (Lin ⁻ ckit ⁺ ) could regenerate myocardium in a mouse model of ischaemic cardiomyopathy. Although other groups clearly showed that bone marrow cells do not transdifferentiate into myocardium [ 2–4 ], these findings, along with a multitude of publications showing significant improvements in myocardial perfusion and function in small and large animal models, effectively paved the way for numerous large-scale clinical trials of bone marrow mononuclear cells (BM-MCs) in humans. These first-generation trials were important in that they established a safety profile for intracoronary and epicardial delivery of BM-MCs, although uniformly, results revealed minimal meaningful clinical benefit. Despite the fact that the results of these trials did not live up to the high expectations, many researchers had for BM-MC therapy experiments pushed forward as new insight into the biology of bone marrow and cardiac stem cells, and their interaction with the host myocardium, unfolded. Investigators then began to ask what cell type would be most efficacious to deliver, and when and how to deliver these cells.

BONE MARROW MONONUCLEAR CELL THERAPY

In trials using bone marrow cells, patients undergo a bone marrow aspiration procedure and samples are subsequently processed to select for mononucleated cells. The draw of this method pertains to the history of well-defined protocols for bone marrow harvesting and delivery (bone marrow transplantation), its safety profile and rapid turnaround time. These features are likely why most clinical trials experimenting with cell transplantation have used BM-MCs. In October 2001, investigators began enrolment in TOPCARE-AMI, which was among the first large clinical trials to use BM-MCs [ 5 ]. In the initial pilot study, patients ( n  = 20) diagnosed with an acute myocardial infarction (MI) were randomized to receive intracoronary infusion of either BM-MC or CD34 ⁺ progenitor cells (isolated from the circulation) after revascularization with a coronary stent. Cells were infused through the stented vessel 3–4 days (mean: 4.3 ± 1.5 days) after MI, and patients were followed up with post procedural and 4-month interval cardiac catheterization with flow reserve, perfusion scan and echocardiography. The results showed significant improvements in left ventricular ejection fraction in both treatment groups compared with an internal reference group receiving standard-of-care treatment (51.6 ± 9.6% to 60.1 ± 8.6% compared with 51 ± 10% to 53.5 ± 7.9%). The promising results of the pilot study led to continued enrolment, where, in total, 59 patients were treated with cell therapy and followed for 1 year. The final results of this trial were published in 2004, and although the change in many of the measured parameters in the cell treatment group showed significant yet minimal improvements, the excellent safety profile led to larger, randomized, double-blind trials (REPAIR AMI, TIME, LATE-TIME and SWISS-AMI) [ 6–9 ]. During this time period, cardiac-gated MRI became available, which allowed investigators the ability to measure additional parameters pertaining to myocardial dynamics as well as scar characteristics. The results of increasingly powered trials with differing methodologies for LV functional assessment have shown that the delivery of intracoronary BM-MCs after AMI, either early or late, has very little additive benefit compared with standard-of-care treatment (percutaneous coronary intervention with reperfusion).

Despite the lacklustre data for BM-MCs in acute MI, investigators remained optimistic for a potential role in patients with chronic ischaemic heart disease with no revascularization options. The hypothesis supporting this optimism was that the delivery of BM-MCs could induce angiogenesis in the native myocardium, preventing infarct progression and limiting the ensuing adverse ventricular remodelling. What made this possible for application to clinical trials was the development of the NOGA endocardial mapping system (Biologics Delivery System, Cordis Corporation), where researchers could identify the infarct border zone and transendocardially deliver BM-MCs. The first large, randomized, double-blind trial to carry out the above hypothesis was the FOCUS-CCTRN trial, which was published in 2012 [ 10 ]. Patients with ischaemic heart failure with a perfusion defect on SPECT imaging without any revascularization options were eligible. Ninety-two patients were randomized in a 2 : 1 fashion to either endocardial BM-MCs or placebo, and outcomes were assessed using echocardiography, perfusion imaging and symptom improvement. Although this trial showed that transendocardial BM-MC delivery is feasible and safe, there was no significant difference in any of the primary or co-primary outcomes when compared with the control group. This is in contrast to what the ACT34-CMI investigators had published in 2011 [ 11 ]. In this Phase II trial, 168 patients with refractory angina and no revascularization options were randomized to receive intramyocardial CD34 ⁺ cells or control and outcomes were assessed based on anginal symptoms, quality of life and perfusion imaging. Importantly, this study did not use BM-MCs, but instead used CD34 ⁺ bone marrow progenitor cells that were mobilized using G-CSF and harvested via leucopheresis. These cells are thought to be effective due to their ability to participate in angiogenesis, inhibit apoptosis and promote myocyte retention. Using this cell type was not only safe, but patients had significantly less angina with improvements in exercise tolerance testing and overall quality of life. No change was seen in perfusion based on myocardial SPECT imaging between groups. Further support for the role of selecting a bone marrow progenitor cell from among BM-MCs before delivery comes from two separate studies with encouraging outcomes from patients undergoing coronary artery bypass grafting for ischaemic cardiomyopathy, receiving epicardial delivery of either CD34 ⁺ or CD133 ⁺ bone marrow progenitor cells [ 12 , 13 ]. These small, phase I pilot studies have shown the safety of directly injected bone marrow stem cells in patients receiving CABG with significant improvements in LV dynamics compared with patients receiving CABG alone. Owing to the success of bone marrow stem cells to date, larger randomized trials are under way (RENEW and IMPACT-CABG).

MESENCHYMAL STEM CELL THERAPY

Mesenchymal stem cells (MSCs) were first identified in 1970 as rare cells within the bone marrow that regulate and maintain the haematopoietic stem cell niche [ 14 ]. Since that time, MSCs have been isolated from numerous different tissue types (adipose, intestine, spleen, lung, umbilical cord blood and even peripheral blood), although it was not until the early 2000s that researchers applied the potential of these cells to ischaemic myocardium. In preclinical studies, these cells have been shown to stimulate and support angiogenesis as well as endogenous cardiac stem cells while also reducing apoptosis of cardiac myocytes in the border zone [ 15–17 ]. Another important feature of MSCs is their ability to suppress the immune response in many cell types within the innate and adaptive immune system, giving them a so-called immunoprivileged status and making them ideal for allogeneic therapy [ 18 , 19 ]. Several early-phase trials have utilized MSCs for ischaemic heart disease, and all have documented an excellent safety profile with very encouraging changes in LV structure and function along with improvements in exercise tolerance, heart failure symptoms and quality of life (C-CURE, PROMETHEUS, POSEIDON and TAC-HFT) [ 20–23 ]. MSCs have even been used in patients with end-stage cardiomyopathy requiring left ventricular assist device (LVAD) implantation. In an early-phase pilot study published in 2014, Ascheim et al . [ 24 ] recently reported very exciting results in the cell treatment group. In this multicentre, double-blind, placebo-controlled trial, 30 patients were randomized in a 2 : 1 fashion to receive LV injection of either 25 × 10 ⁶ allogeneic MSCs versus serum at the time of LVAD implantation. Owing to the concerns over immune sensitization, the number of infused MSCs was intentionally kept low. Despite this low cell dosage, the results seem to indicate that patients in the cell therapy group were more frequently successful at temporary weans from LVAD support and trended towards decreased 90-day mortality (Fig.  1 ). Importantly, there was no significant difference in major adverse cardiac events between groups, and at 1-year follow-up, there was no sensitization to donor-specific antibodies signifying the safety of allogeneic MSC injections—a finding similar to that from the POSEIDON study.

Figure 1:

Duration of left ventricular assist device wean. MPC: mesenchymal precursor cell. (Reproduced with permission from reference [ 24 ].)

CARDIAC PROGENITOR CELL THERAPY

Once believed to be functionally quiescent, data now exist showing rare cell types within the myocardium with regenerative capacity. Although these cells have no meaningful impact after an ischaemic insult under normal circumstances, new cell culture techniques have allowed for isolation and scaled production of these progenitor cells for clinical application. Two separate cell types with differing isolation protocols, namely cardiosphere-derived cells and lin ⁻ ckit ⁺ cardiac stem cells, have been developed and applied in early-phase clinical trials (CADUCEAUS and SCIPIO) [ 25 , 26 ]. Both of these exciting trials were powered as proof-of-concept studies, and each showed an excellent safety profile without evidence of clinically significant ventricular arrhythmias, tumour formation or major adverse cardiac events. Results show increased LV function, increased viable LV mass and decreased scar size. These are exceptionally meaningful results because they lead one to conclude that cardiac progenitor cell therapy may lead to de novo myocyte regeneration. This has led to a follow-up study to CADUCEUS named ALLSTAR, which aims to include allogeneic cardiosphere-derived cells in the treatment arm.

FUTURE DIRECTIONS

As we look towards the future to anticipate what lies ahead, it is helpful to review developments in the basic science arena and preclinical assessments of new and innovative technology. Murry's group out of Washington has recently published a non-human primate study where human embryonic stem cell-derived cardiomyocytes (hESC-CMs) were used to regenerate myocardium [ 27 ]. In this fascinating work using a large animal macaque model of ischaemic cardiomyopathy, 10 ⁹ hESC-CMs were directly injected into the infarct and borderzone 14 days following temporary balloon occlusion of the mid left anterior descending coronary artery. This group convincingly showed excellent cell retention with injected hESC-CMs occupying nearly 40% of the infarct (Fig.  2 ). Additionally, these cells functionally incorporated into the host myocardium, exhibiting evidence for electromechanical coupling during rapid atrial pacing. There was, however, evidence of ventricular arrhythmias in all cell-treated animals, although none were sustained or fatal. Even though the mechanism for these arrhythmias needs to be explored, it seems likely from the overall results that it will not be long before human pluripotent embryonic stem cells enter into the equation of which cell type is optimal in the treatment of cardiomyopathy.

Figure 2:

( A – I ) Confocal immunofluorescence of macaque hearts subjected to myocardial infarction and transplantation of hESC-CMs. Grafts were studied at Day 14 ( A – G ) and Day 84 ( H and I ) post-engraftment. ( A ) Remuscularization of a substantial portion of the infarct region (dashed line) with hESC-CMs co-expressing GFP. The contractile protein α-actinin (red) is expressed by both monkey and human cardiomyocytes. Scale bar: 2000 μm. ( B – F ) Images from the peri-infarct region of the same heart shown in ( A ), demonstrating extensive hESC-CM engraftment. Scale bars: 1000 μm ( B – E ) and 200 μm ( F ). ( G ) Graft–host interface (arrows) at Day 14 with interconnected α-actinin (red)-expressing cardiomyocytes (arrows). Note that host sarcomeric cross-striations (asterisks) show greater alignment than hESC-CM graft. Scale bar: 25 μm. ( H – I ) Day 84 hESC-CM grafts contain host-derived blood vessels lined by CD31 ⁺ endothelial cells. Scale bar: 20 μm. Inset scale bar: 10 μm. hESC-CMs: human embryonic stem cell-derived cardiomyocytes. (Reproduced with permission from reference [ 27 ].)

Another research topic that merits attention is the application of tissue engineering to cell delivery. One of the major drawbacks of cell treatment is the poor retention rate, mainly from the in-hospitable environment of post-infarct myocardium. Sawa's group from Osaka University has pioneered a cell culturing technique allowing for the production of cell sheets, which addresses the issue of cell retention. These cell sheets can be made up of MSCs, cardiac stem cells and even hm-ESC-derived cardiomyocytes and are directly applied to the infarct territory. They have published very favourable results in large animal models of ischaemic cardiomyopathy [ 28 , 29 ].

In summary, a tremendous amount of work and effort has gone into the development of cell therapy in cardiomyopathy; however, this field remains in its infancy with many unanswered questions. What we have learned from the body of evidence in the literature is that unfractionated BM-MCs likely have a negligible effect on LV structure and function after an acute MI, which is supported by a recent meta-analysis (ACCRUE). It is also apparent that delivery of these varying cell types, regardless of whether they are intracoronary, transendocardial or epicardial, is feasible and safe warranting larger, phase III trials to adequately assess effectiveness. It will also be imperative to better delineate the mechanism of action of each individual cell therapy. Currently, there are many hypotheses as to why treatment groups derive benefit, whether that is realized as improved LV structure and function, decreased infarct size, increased myocardial mass or relief of symptoms. Are these outcomes due to a paracrine effect from cell delivery on native cells, are the cells incorporating into the host myocardium (and if so, how long do they remain) and is myocardium actually being regenerated? These are difficult questions to answer, but they are questions that must be addressed as the field moves forward.

Funding

This work was supported in part by a grant from the National Institutes of Health (1R01 HL089315-01, Y.J.W.)

Conflict of interest: none declared.