Future potential of engineered heart tissue patches for repairing the damage caused by heart attacks

KEYWORDS:

• Cardiac regeneration
• engineered heart tissue
• heart failure
• ischaemic heart disease
• cellular therapy
• tissue engineering

Heart failure is a major cause of death in the western world, currently accounting for 1-2% of the entire UK NHS budget, and this is projected to increase in the near future as the population ages and more people are surviving post-myocardial infarction. Unlike the successes that have been achieved with treating myocardial infarction, the morbidity and mortality of heart failure have been stubbornly slow to improve with a mortality rate worse than several common cancers, with roughly 1 in 2 patients dying from the disease 5 years after diagnosis [1]. The heart, once thought of as a terminally differentiated organ, actually has a very restricted capacity to regenerate cardiomyocytes by itself (approximately 0.5–1% of cells) but significant loss from a myocardial insult/infarction can result in the heart failure syndrome [2].

Existing pharmacological and non-pharmacological therapies do improve survival but only retard the progression of disease. No current therapy replaces lost heart muscle, and few even increase the force of cardiac contraction. The single currently available curative option is heart transplantation, but this is of restricted supply. Therefore, there is an unmet need for novel therapies that can potentially reverse the disease process. This need brought about the cardiac regeneration field which aims to restore or repair the muscle that is lost after a heart attack.

There was a wealth of enthusiasm for the field in which clinical trials began at the start of this century but unfortunately the exciting results that were seen in early preclinical experiments were not replicated in many of the clinical trials to date. Part of the rush may have been premature without actually understanding the beneficial mechanism behind cellular therapy, and it has now become apparent that effects that were once believed to have explained observed improvements, for example transdifferentation of bone marrow cells to denovo cardiomyocytes, do not occur [3]. In addition, there are several caveats to existing delivery methods used in the current clinical trials. For example, with both intracoronary and intramyocardial injection very few cells remain after 24 h, with over a 90% washout almost immediately after intramyocardial injection. In addition, with intracoronary injection many cells end up in the liver or spleen just a few hours after injection. If they do survive, cells are subjected to a hostile environment that could prove to be deleterious/cause cell death.

Therefore, groups have been looking at ways to improve efficacy by increasing cell retention using different delivery methods or by using biomaterials which can help shield and protect cells during the early integration phase. Engineered heart tissue (EHT) was pioneered by Thomas Eschenhagen over 20 years ago, with the main long-term goal of replacing the muscle that is lost post-myocardial infarction. Engineered heart tissue involves the encapsulation of cells into a hydrogel matrix ex-vivo to create patches of heart like muscle ‘in a dish’. By encapsulating the cells into a hydrogel matrix, cells can be matured in vitro, and often beat in synchronization after a period of 1–2 weeks depending on size. The patches are applied to the epicardial surface of infarcted hearts with the aim of restoring the normal ventricular function. The field has steadily evolved over time and now clinically relevant patches (4 × 3 cm) are starting to appear and a few phase I studies have started with several planned [4].

The first EHT studies were created with rat neonatal cells but interest has now focussed on pluripotent stem cells which can either be from reprogramming of adult somatic cells (obtained from skin, urine or blood samples) or from embryonic stem cells. These have unique advantages since pluripotent cells are the only cell type from which bona fide cardiomyocytes can be created. Over time, there are now improved differentiation protocols that result in hundreds of millions/billions of cardiomyocytes (around a billion cardiomyocytes are lost in a myocardial infarction) being created in one batch and large animal preclinical data showing electromechanical coupling and sustained long-term retention of several months post grafting [5,6]. Embryonic cells have ethical issues surrounding their use. In contrast, induced pluripotent cells, for which Yamanaka won the Nobel prize, have fewer ethical issues and enable the potential for an autologous treatment option. For example, a patient could be seen in clinic; a somatic cell biopsy taken; cells differentiated into cardiomyocytes in-vitro and a patch made for transplantation at a later date. The cells transplanted would theoretically be immune-matched and able to restore the myocytes that were lost post-myocardial infarction.

However, this route is time-consuming, expensive and more complex than previously thought with cells taken from diseased donors often expressing phenotypes of the disease. These reasons are why in part the first clinical studies have utilized an allogenic approach. Menasché et al. reported a phase I feasibility in which cardiovascular progenitors were seeded onto a fibrin-based patch in patient undergoing coronary artery bypass grafting [7]. Cells were from an allogenic cell source and all patents received immunosuppression for a period of 1 month. Feasibility was shown in this 6-patient trial. A similar trial has recently been granted approval in Japan using cell sheet technology and plans to use 100 million cells again from an allogenic cell source. These trials are primarily studying safety including the incidence of arrhythmias. There may be several reasons for an allogenic approach over autologous, including the possibility of extensive preclinical testing of a specific cell line to ensure viability and effectiveness, and the potential for an off the shelf product and more viable business model. For example, autologous approaches of other organ systems are limited but autologous human corneal epithelial cells (Holoclar) cost £80,000 per eye (NHS list price) when used to treat advanced limbal stem cell deficiency after eye burns.

The allogenic approach, if a muscle replacement strategy is to be pursued, will require long-term immunosuppression. In the first clinical trials, immunosuppression was only given for 1–2 months, and therefore there will be no cells remaining after this point. There are risks regarding long-term immunosuppression including, renal toxicity, allograft vasculopathy and the risk-benefit ratio favors long-term immunosuppression for organ transplantation including renal, liver and heart. At present, the data regarding cellular therapy means that the risk of long-term immunosuppression may outweigh the benefit of cellular therapy and therefore seeking ethical approval for long-term immunosuppression may prove more difficult. Regulatory approval for induced pluripotent stem cell products varies between countries, and Japan has recently adopted a fast track approval system allowing clinical use only if products have been deemed safe there are hints of efficacy from feasibility trials.

Work is ongoing regarding a hypoimmunogenic cell line, using CRISPR-Cas9 technology to inactivate certain immune cell markers and promising data has been shown in small animal models [8]. However, preclinically more work will be needed before translation to ensure cell line stability and efficacy. Although engineered heart patches of clinically relevant dimensions have been reported, several caveats regarding the technology exist. These include lack of extensive preformed vascular network, graft thickness, and electromechanical coupling. Data so far have reported that the grafts are able to couple but not in all cases. This may be in part due to a fibrous layer that exists between the graft and the host. Preformed vascular networks have been shown but vessel sizes are small and flow is slow. Most grafts are approximately 1–1.5 mm in thickness and if a muscle replacement strategy is needed this is insufficient. Finally, measures such as conduction velocity and force measurement are quantitatively still some way behind human myocardium. But nevertheless, the modality is exciting and promising preclinical data regarding improvements in ventricular function, together with a good safety profile, have been reported from several groups. Another caveat is that most patches in preclinical use today require a thoracotomy to place them. Therefore, the patch cannot be implanted in a minimally invasive way. The patient most likely to receive the patches will be those undergoing coronary artery bypass grafting or left ventricular assist device insertion so the patch can be concurrently sutured to the epicardium. Work is ongoing regarding memory EHTs that can be inserted via a small needle upon which they regain original shape but this is currently at a small animal model preclinical stage [9].

If these roadblocks can be overcome and the human clinical trials are successful, the hoped reality could be that patches of engineered heart muscle could become a viable treatment option, implanted in a minimally invasive way as a day case procedure (similar to permanent pacemaker insertion) without the need for long-term immunosuppression. Further preclinical work is needed to improve the patches and understand the long-term data regarding effects on cellular retention, electromechanical coupling and arrhythmia generation.

Declaration of interest

The authors have no relevant affiliations or financial involvement with any organization or entity with a financial interest in or financial conflict with the subject matter or materials discussed in the manuscript. This includes employment, consultancies, honoraria, stock ownership or options, expert testimony, grants or patents received or pending, or royalties.

Reviewer disclosures

Peer reviewers on this manuscript have no relevant financial or other relationships to disclose.

Additional information

Funding

This paper was funded by BHF Centre of Regenerative Medicine, Imperial College London [BHF RM/17/1/33377]