Final Report Summary - LIFEVALVE (Living autologous heart valves for minimally invasive implantable procedures)
Executive Summary:
Background: Approximately 1% of all newborns suffer from congenital heart defects which cannot be treated efficiently due to the lack of growth of the clinically available "artificial" replacement materials, such as heart valve prostheses. Furthermore, current heart valves have significant limitations including limited durability and/or the need for lifelong medication. Today, heart valve prosthesis-associated problems occur in 30-35% of patients within 10 years, frequently necessitating high-risk, potentially life-threatening re-operations.
Objectives: The initiative “LifeValve” (Living autologous heart valves for minimally invasive implantable procedures), funded by the European Commission under the 7th Framework Programme, aimed at developing a tissue engineered heart valve which can be implanted by minimally invasive catheter technology. The valve was expected to have repair and regeneration capacity, thereby minimizing the need for future surgical intervention. LifeValve combined two novel life science technologies – tissue engineering and minimally invasive implantation technology for the benefit of children with congenital heart disease.
The technological objectives of this translational approach comprised (1) the in vitro creation of a valve with the capacity of growth by tissue engineering methods, (2) the development of intelligent stent systems, (3) a minimally invasive implantation technology (non-surgical procedure) of the LifeValve and (4) the preparation of a clinical study. This highly interdisciplinary collaborative project was implemented by eight renowned academic and industrial partners from Austria, Germany, Hungary, The Netherlands and Switzerland representing a comprehensive range of expertise in tissue engineering, biomedical and material research, medical technology and clinical practice.
Results: In the course of the 5.5 years initiative, a novel tissue engineered heart valve was developed. The multistep development included various optimization and adaptation steps. The valves of the first generation were made by seeding cells on a PGA/P4HB scaffold and subsequent cultivation in a bioreactor. Those living valves were limited by leaflet retraction phenomena leading to valvular incompetence in vivo as tested in a sheep model. In order to circumvent this problem, a new and alternative LifeValve biomaterial was investigated based on decellularized tissue-engineered extra cellular matrix. Further modifications steps were necessary focusing on optimization of the geometric design and the culture conditions. Finally, a valve was created showing good performance in vivo (sheep) with a follow-up time of 12 months. Regarding the stent, which is needed as a carrier for the tissue engineered valve and the fixation within the vessel, two systems were investigated: the already clinically established self-expandable stent material nitinol and a biodegradable stent which enables full incorporation of the minimally invasively implanted tissue engineered heart valve into the patients and subsequent growth. The design requirements of the nitinol stent were defined with the highest priority on radial force and secure anchoring. For the development of a bioresorbable stent, the nitinol stent served as a model for the computational based design. The chosen material is Poly L-lactic Acid (PLLA), known to be biocompatible and non-cytotoxic, that is subsequently modified with P4HB to improve the mechanical properties (flexibility). Stent development is in the final stage. Within LifeValve, minimally invasive implantation procedure was established. A specific delivery system for the LifeValve technology was developed, improved stepwise and adapted to the requirements in order to ensure a safe transcatheter based delivery. The device was routinely used with a high success rate in the pre-clinical study. In parallel, patients for a clinical study were identified and a first-in-man trial design was established.
Impact: Apart from the scientific and medical progress represented by this one-step next generation type of therapy for congenital heart defects, the socio-economic costs for European healthcare systems can be substantially reduced by the LifeValve technology, providing a growing heart valve replacement without the need for multiple re-operations and hospitalizations. The combination of heart valve tissue engineering and minimally invasive implantation technology by using self-expanding and biodegradable stents is a novel approach and can provide Europe with a unique opportunity to reinforce its competitiveness as to novel medical and life-science technologies and the related fast growing markets.
Project Context and Objectives:
Cardiovascular disease still represents the Killer Number One in all Western Societies accounting for the death of numerous patients and annual health care costs in the range of billion Euros. Valve replacement is the most common surgical therapy for valvular heart disease with almost 250.000 annual implantations worldwide. Valve replacement procedures are efficacious and substantially change the otherwise deadly natural history of valvular disease. However, although the overall performance of these devices is sufficient, prosthesis-associated problems occur within 10 years postoperatively in 30-35% of patients. Mechanical valves require lifelong anticoagulation therapy. Bioprosthetic valves have limited durability and may calcify prematurely particularly in young patients. More importantly, both mechanical and bioprosthetic valves are non-viable structures and do not have the ability to grow, repair or remodel, which is a specific problem in the pediatric patient population. Congenital heart disease (occurring in almost 1% of all newborns) represents a potentially life-threatening situation and treatment by highly invasive surgical interventions with substantial morbidity and mortality is often necessary. In many congenital heart malformations underdeveloped or malfunctioning heart valves are involved, which makes heart valve replacement necessary. A particular problem relates to the fact that all clinically available heart valve substitutes represent artificial or non-living structures which cannot grow with the paediatric patients. This normally necessitates several re-operations to adjust the heart valve prosthesis to the growing organism. Each of these re-interventions is associated with exponentially increasing morbidity and mortality.
The general objective of the LifeValve project was to develop a new therapeutic strategy to treat heart valve disease. In particular, the development of a tissue engineered living heart valve, with the capacity of growth in accordance with the growth of children, which can be implanted without an operation by minimally invasive catheter technology was to be addressed. The rationale for first focusing on paediatric applications was twofold: First, a novel, tissue engineered living heart valve with the potential for growth is addressing a clearly unmet medical need in congenital heart disease. All contemporary solutions with conventional heart valve prostheses are insufficient, do not provide a sustainable treatment and as a consequence create suffering and substantial socioeconomic costs. Second, successful use and increasing clinical experience within the paediatric patient population as entry indication will allow extending the clinical application to the much larger adult patient population with degenerative heart valve disease in a later stage.
To successfully realize this project, a highly interdisciplinary consortium consisting of academic research groups and clinicians as well as market oriented companies was formed having expertise in the fields of tissue engineering, biotechnology, cardiovascular surgery and medical devices. The integration of these partners represented the full range of life science technology yielding from basic science to clinical application and marketable products (“from bench to bedside”).
The LifeValve project aimed to combine two recent novel life science technologies, tissue engineering and minimally invasive (non-surgical) implantation technology. The technological objectives of the LifeValve project comprised (1) the in vitro creation of autologous cell based valve substitutes by tissue engineering methods (2) the development of an intelligent stent system, (3) a minimally invasive implantation technology (non-surgical procedure) of the LifeValve. The functionality of the valves should be tested in an animal model before initiating a clinical trial (4).
1. In vitro creation of valve substitutes (LifeValves) by tissue engineering methods
To address the substantial limitations of state of the art artificial heart valve prostheses, the ultimate goal of the tissue engineering approach was to construct tissues from their cellular components which combine most of the characteristics of the healthy native original. For a functional heart valve these include adequate mechanical characteristics, (mature extracellular matrix), durability, adequate haemodynamic performance, as well as the absence of immunogenic and/or inflammatory reactions. Moreover, the tissue engineered heart valve is supposed to offer the inherent capacity of growth, addressing a so far unmet medical need in children with congenital heart disease.
The success of the production of tissue engineered heart valves is dependent on four main issues: (a) the matrix (scaffold) which determines the three-dimensional shape and serves as an initial guiding structure for cell attachment and tissue development; (b) the cell source from which a living tissue is grown; and (c) the in vitro culture conditions of the construct before implantation and (d) the valve geometry.
Covering the whole production process, the project started by defining the optimal biodegradable biomaterial matrix requirements for the in vitro production of tissue engineered heart valves. Various synthetic biomaterials were tested in vitro as to adequate biocompatibility and biodegradation properties to eventually serve as a cell carrier for the tissue engineering process in the bioreactor to create a heart valve, which is finally able to accommodate growth in the patient. For the tissue engineering of the heart valves, different cell sources and culture conditions were tested for the suitability of tissue formation with regard to collagen formation (later replacing the initial scaffold) and mechanic stability. Later in the project the focus was put on the so-called decellularization technology, which included seeding cells on a scaffold and culturing in a bioreactor system to form tissue that finally takes over the function of the initial scaffold as a valve matrix with a specific design. All living cells were removed so that a matrix composed of extra-cellular components such as collagen is available that can be stored or directly used for implantation. The initial scaffold degrades stepwise.
Beside the matrix, the functionality of a tissue engineered heart valve strongly depends on the geometry of the valve. During the course of the project it turned out in the first pre-clinical trials that the initial geometric design of the valve let to functional insufficiencies in the animal, mainly because of retraction and thickening of the valve leaflets. These observations necessitated a substantial redesign of the valve using computational models and a subsequent adaption and optimization of the culture conditions.
The prerequisite for long term in vivo functionality of a tissue engineered valve is that the valve becomes part of the human body, which means that in vivo re-cellularization and remodelling occurs making the final valve similar to a native one in composition and function. Many cells are endowed with a reservoir of soluble factors that can exert paracrine effects on other cells ingrowth resulting in remodeling stimulation. It was thus hypothesized that recellularization directly prior implantation improves cell recruitment followed by generation of viable tissue.
Beside the establishment of such a recellularization procedure also cell tracking methods were investigated and utilized to follow the fate of the seeded cells during the in vivo phase of the experiments.
To assess tissue maturation and implantability, analysis of the stented tissue engineered heart valves comprised morphology (histology, immunohistology, electron microscopy), matrix protein quantification (collagen, GAG, DNA, elastin), biomechanical properties (tensile strength testing) and functional performance in a pulsatile flow simulator.
Furthermore, a special focus was on the establishment of stable (standard operation procedures) in vitro processes in accordance with GMP regulations in order to provide small scale production capacity for the consequent first-in-man clinical trials.
2. The stent system
A particular aim of the LifeValve project was the application of minimally invasive catheter techniques to avoid open heart surgery. The tissue engineered heart valve implanted by minimally invasive methods needed a carrier and an anchor to be fixed into the target position within the heart. For these purposes stents were used. It was a particular challenge to design a stent system that can be folded or crimped for delivery of the construct without being damaged, nor damaging the engineered tissue, and that, after expansion, can exert appropriate radial force and load on the conduit to resist the mechanical load exposed on the valve.
A clinically established stent based on nitinol allowing self-expansion (memory alloy) was used for the proof-of-concept of the tissue engineered heart valve within LifeValve.
However, in view of the applications in pediatric patients, nitinol is disadvantageous, as growth is restricted to the dimension of the used stent. An integral part of the LifeValve project was therefore the development of an intelligent stent system that accommodates the growth of children to avoid re-operations. Two routes regarding this issue were followed: the development of a biodegradable stent as well as the usage of an “oversized” nitinol stent.
Principally, the stent is part of the tissue engineered heart valve, and should not only fix the valve in place, but should also promote fusion of the engineered tissue with the native host tissue. Once this fusion is established, the stent loses its mechanical function. When following the biodegradable stent concept, the degradation of the stent allows growth of the tissue. Thus the developed stent was designed to have the following characteristics: (1) the used material should have a degradation time that is slow enough to allow in vitro production of the valve within the stent in a bioreactor system. On the other side degradation must be fast enough to allow growth of the surrounding native vessel; (2) the stent should be flexible so that it can be crimped down to small diameters (necessary for catheter based implantations) and reach the initial diameter after crimping and expansion; and (3) after implantation it should facilitate successful fixation and ingrowth of the valve. Therefore it must exhibit sufficient radial forces to hold the valve in the correct position. The biomaterial stent should be either self- or balloon-expandable. The stent system and the stent to biomaterial junction were of critical importance for the success of a functional and mechanically durable heart valve substitute.
In an alternative approach the usage of an oversized nitinol stent to accommodate growth of the implanted valve was investigated. The remaining extension capacity to the maximum diameter had to follow the growth of the native right ventricular outflow tract (RVOT) and pulmonary artery. For that purpose, oversized nitinol stents were implanted in juvenile sheep. Performance and tissue reactions (e.g. local vascular trauma, tissue remodeling) was assessed in order to prove the feasibility of that approach.
3. Minimally invasive implantation technology of the LifeValve
A further objective was the implantation of the LifeValve by transcatheter techniques to avoid open-heart surgery. An appropriate loading and delivery system was developed and continuously adapted to the ongoing modifications of the LifeValve. The loading system needed to be able to crimp the stented valve to an adequate small diameter and load it into the delivery system immediately before implantation under sterile conditions in the catheterization laboratory to ensure a minimum of stress to the valve. The delivery system had to be flexible and thin enough to be advanced by catheterization techniques through the narrow jugular vessels and the tortuous cardiac anatomy. Care had to be taken, that the tissue engineered valve remained unharmed during the procedure and that the deployment of the stent was exactly controllable and reproducible. Feasibility studies were performed in a large animal model. The most accepted animal model (FDA, EMA, Swissmedic) for the investigation of novel cardiovascular implants is the sheep. Thus, sheep were used to assess the functionality of the minimally invasively (transcatheter) implanted autologous living heart valves throughout the LifeValve project.
4. Clinical pilot trial
A specific focus was on the establishment of GMP (Good Manufacturing Practice) adequate logistics, which later can be utilized in the clinical part of a first-in-man project. Following proof of safety, long-term animal experiments were undertaken to demonstrate regeneration capacity and adaption of the valve to the biological environment. Finally, a clinical pilot study in paediatric patients with congenital heart defects was initiated.
Project Results:
1. Development of the tissue engineered heart valve (LifeValve)
1.1 In vitro creation of valve substitutes by tissue engineering methods
LifeValve aimed at developing a living tissue engineered heart valve with the ability of growth. Therefore, the choice of a biocompatible and biodegradable heart valve matrix (scaffold) allowing for growth of the implant was essential. Furthermore, as the LifeValve project aimed for development of a tissue engineered valve which can be implanted via the trans-catheter implantation technique, the final valve needed to be crimped to fit the very limited dimension of the trans-catheter implantation device. To reach this goal, the four main issues that define the success of the development of a functional tissue engineered heart valves were elaborated on: (1) the matrix (scaffold) which determines the three-dimensional shape and serves as an initial guiding structure for cell attachment and tissue development; (2) the cell source from which a living tissue is grown; (3) the in vitro culture conditions of the construct before implantation and (4) the valve geometry.
1.1.1 Scaffold material
In recent years, research groups have used a variety of different approaches and methods to develop tissue engineered heart valves (TEHVs) that are now at various stages of clinical development. The cyclic loading of the cardiovascular system (pulsative blood flow) requires an enormous strength, flexibility, and durability of the tissue-tissue engengineered cardiovascular replacements and consequently appropriate mechanical properties of the starting matrices to endure the cyclic stresses and strains exerted upon implantation. More specific, starter materials for TEHVs should contain stiffness around 0.5 MPa, while being elastic without any tendency to deform permanently under the influence of stresses. These requirements to the overall mechanical behavior of the material are determined by the intrinsic material properties (e.g. stiffness), the scaffold architecture (e.g. anisotropy) and the degradation rate. Moreover, due to the intense contact with the blood in cardiovascular applications, it is desirable to have a thromboresistant material e.g. by providing an endothelial surface layer. In general, cell infiltration into the scaffold is one of the prerequisites for successful tissue regeneration and is primarily determined by the scaffold microstructure, i.e. pore size and fiber thickness. Additionally, the microstructure of the scaffold can affect cell phenotype and influences the behavior of the infiltrating cells regarding cell adhesion, spreading, and proliferation. It is also used as an anchor to attach different bioactive molecules and signals that improve specific cell function, such as, pro-angiogenic signals. The mechanical properties of the scaffold also provide an important stimulus to the cells for extra cellular matrix (ECM) production and remodelling, as the cells experience different local stresses and strains depending on the scaffold stiffness. These mechanical cues, as result of the scaffold stiffness, can modulate the differentiation of cells into pathological phenotypes, e.g. osteoblastic or myofibroblastic differentiation. Additionally, bioactive factors can be incorporated into the scaffold that can direct local cellular function, or promote recruitment of specific cell types via chemotaxis. To conclude, the initial scaffold should meet the following requirements: (1) to be biocompatible and thromboresistant, (2) to be able to support cell infiltration, growth and cell-to-cell interaction, (3) to start with sufficient mechanical properties and degrade at a rate in relation to new tissue formation, (4) to have optimum architectural properties of pore size, porosity and permeability in order to allow diffusion of nutrients and metabolic waste products and, last but not least, (5) to be bioresorbable. [Source: Generali, Dijkman & Hoerstrup; Bioresorbable scaffolds for cardiovascular tissue engineering; EMJ 2015].
Different types of scaffold material have been used for heart valve tissue engineering, unfortunately none complying with all the aforementioned requirements so far. The matrices for valve replacement that comes closest to the geometry and structure of the native valve with native-like mechanical and physiological hemodynamics, is the xenograft (tissue from one species to an unlike species) or allograft (=homograft; tissue of the same species) valve. Obviously, the allograft valve is a more natural option compared to the xenograft valve, as the ideal valve replacement should show biomechanical characteristics similar to those of the native heart valve. Natural leaflet motion and flow patterns are supposed to avoid stress related calcification and early valve failure. The microstructure of the allograft valve is favourable with respect to proliferation, differentiation, and survival of reseeded cells. The availability of the allograft valve, however, is limited by donor shortage. Due to its anatomic similarity to human valves, the porcine heart valve represents an attractive alternative to allograft valves. Additionally, also xenogenic pericard is investigated as material for valve replacements. However, xenografts are associated with the risk of immunogenic reactions or disease transmission and the availability of homografts is limited. To eliminate immune reactions with xenogenic tissues, they are decellularized before implantation and are either reseeded prior to implantation or expected to become repopulated in vivo. Although the latter was partly demonstrated in animal models, clinical studies led to contrary reports about cellular infiltration in humans and proof of concept is still missing.
To overcome the limitations of xeno- and homografts, the applicability of various bioresorbable polymers, either synthetic (such as PGA, P4HB and PCL) or natural materials (such as fibrin and collagen) was investigated. Seeded with (autologous) cells and subsequently cultured in vitro, these materials have shown to be feasible for heart valve tissue engineering, with demonstrated functionality in vitro and in vivo. For example, autologous myofibroblast and endothelial cells (ECs) (or neonatal human dermal fibroblast) have been used to create fibrin-based TEHVs and have been evaluated in sheep up to three months. Although the leaflets remodelled in vivo, the retraction of the leaflets resulted in valvular insufficiency therewith hampering the functionality of the valve. Also the synthetic bioresorbable polymer PGA showed promising results when combined with P4HB or PLLA. When seeded with autologous myofibroblasts, mesenchymal stem cells (MSCs), and ECs valves can be engineered in vitro that undergo structural and functional remodelling in vivo without stenosis. However, also in these valves functionality was hampered by trivial to moderate regurgitation after six weeks due to thickening and retraction.[Source: Dijkman and Generali et al., Chapter Cardiovascular Tissue Engineering: Polymeric Starter Matrices for in Encyclopedia of Biomedical Polymers and Polymeric Biomaterials; 2015].
Classical tissue engineering approaches use patient specific cells to engineer tissue constructs in a so-called autologous approach. To investigate the potential of this approach for heart valve applications within the LifeValve project, cells were isolated from sheep that were used as the appropriate animal model. Sheep are prone to develop early heart valve calcification due to their high calcium metabolism and are therefore the so called “worst-case scenario” animal model. After isolating cells from a small piece of the vasculature, cell numbers were expanded in the laboratory to approximately 80 million, which is sufficient to culture one valve.
As an appropriate scaffolding material, poly(glycolic acid) (PGA) was combined with P4HB in order to shape the scaffold material into a complex three-dimensional heart valve geometry, which was later on sutured into suitable stent devices. These stented heart valve scaffold constructs were then seeded with the isolated cells from the sheep and placed inside a sophisticated bioreactor system. This system was specifically engineered to mimic native pulsatile pulmonary pressure conditions. When the cell seeded constructs were subjected to these mechanical stimuli, cells showed an enhanced production of tissue, which gives the heart valve strength over time. After four weeks of culture, the cells have produced sufficient amounts of tissue, on the other hand the scaffold material has mostly degraded. The downside of using PGA based scaffolding material is that it loses mechanical integrity after already two weeks of culture time. As a result, the cells used during culture could deform the initially defined geometry of the valve by applying increasing rates of contractility.
In an attempt to overcome this problem of losing initial valve geometry, stiffer scaffold material such as poly(caprolactone) PCL, which has a much slower degradation rate, was investigated. Unfortunately this approach did not work out so well. Cells were still capable of deforming the valvular geometry, causing leaflet retraction. Even so, after culture the PCL is still abundantly present inside the tissue construct after culture and largely contributes to the mechanical integrity of the valve. PCL has a low fatigue resistance. Therefore the valve construct can quickly lose its mechanical support, while by that time the tissue is not strong enough to fully take over the load.
As a result, the PGA-P4HB combination was chosen as a final scaffold material in which the heart valves are from now on cultured in a so-called closed configuration, which means that the three leaflets are merged after culture, since they grew together during culture. In this way, the initial valvular geometry is lost during culture, but the leaflets are not retracted and the valves are still functional afterwards. Although scaffold remnants still reside in the valvular construct, they do not mechanically support the tissue anymore. In this case the engineered tissue is now carrying the full load. These so called autologous heart valves were then implanted into the same animals from which the cells were originating. However, over time it became apparent, that the leaflets partially retracted over time hampering their mobility and functionality.
In order to overcome this limitation, decellularized tissue engineered heart valves (dTEHVs) were investigated for their potential use as a suitable biomaterial for heart valve prostheses. The way these dTEHVs were produced is comparable to the living tissue engineered heart valve, but cells were not patient specific, but species specific and that after four weeks of culture, the cellular content was completely removed. In this way an entire non-living three-dimensional heart valve shaped tissue construct was produced. Prior to implantation, DNA content was removed to prevent the initiation of the donor’s immune system. Strikingly, these non-living constructs appeared to function perfectly fine up to eight weeks after implantation, not being stenotic and showing no signs of leaflet thickening. Furthermore, host recellularization was being observed as early as five hours after implantation, when cells started to migrate from the wall toward the leaflet tips. This observed amount of recellularization is unique compared to other valvular bio prosthesis, and is required to be able to provide future regeneration and growth. In addition, endothelial cell covering was shielding the entire heart valve construct which is essential to prevent thrombolytic events. Additionally, increasing amounts of tissue formation were observed after implantation, being a sign for remodelling potential. However, it was also observed that over time contractile cells were infiltrating while at the same time the heart valve became severely insufficient over time after four weeks of implantation. For this reason, adjustments to the valve design have been implemented to overcome this limitation as will be described below.
1.1.2 Cell sources
Despite being most important for the viable part of the engineered valve, the cell source represents the least controllable factor, in contrast to the controllable properties of the scaffold materials. Their regenerative capacity and quality to form tissue relies on the characteristics of the source and thus varies between individual donors and original organ tissue. A variety of cell types have been investigated for heart valve tissue engineering. For clinical application, the ideal cells should be non-immunogenic, functional, and easy to isolate and expand. The cell sources which have been mostly investigated are the non-immunogenic autologous endothelial cells (ECs), smooth muscle cells (SMCs) and (myo)fibroblasts, as they are normally found in the heart valve leaflets. The (myo)fibroblasts are well known for their suitability for tissue formation and specifically with regard to collagen formation, which provide mechanical stability of the valve. In order to obtain autologous cells, a piece of donor tissue is dissociated into individual cells, expanded in culture and finally attached to a scaffold and implanted. During the last years, also stem cells have attracted increasing attention and have become an important cell source for tissue engineering. Stem cells can be found in embryos, fetuses, and in adults and by definition they have the ability to reproduce themselves for a long period of time (self-renewal) and to give rise to different cells types (differentiation). Different stem cells exist, varying in their differentiation potential: the embryonic stem cells (ESCs) and the adult stem cells. The ESCs have the ability to give rise to all types of cells that form the three germ layers (mesoderm, endoderm, and ectoderm) from which all the cells of the body arise. ESCs are derived from the pre-implantation embryo, precisely from a group of pluripotent cells called inner cell mass (ICM). Once removed from the blastocyst, the cells of the ICM can be cultured under special conditions in vitro and are then called ESCs. Despite initial success, ethical issues, immunogenic and tumorigenic problems are the major drawbacks for ESCs. The adult stem cells have a more restricted (multipotent) differentiation potential compared to ESCs. They are found in adult organisms where they can give rise to specialized cells that will develop to the different tissues of the body. Adult stem cells serve for both homeostasis in healthy tissues and for regeneration of defective tissues. Most popular for cardiovascular tissue engineering are the bone marrow stromal cells (BMSCs) and adipose-derived stem cells (ADSCs). Recently, another interesting cell source for regenerative medicine, the induced pluripotent stem cell (iPSC), has become available. As part of this approach differentiated, adult cells have been genetically reprogrammed to an embryonic stem–like state with pluripotent differentiation capabilities. iPSCs are very similar in morphology, proliferation and gene expression to ESCs and represent a powerful tool in regenerative medicine. Unfortunately, iPSCs have been associated with the risk of teratoma formation. Therefore, more in depth research is required into the use of these cells for autologous cell based therapies such as cardiovascular tissue engineering. Moreover, it is still questionable whether an optimal approach would involve seeding with undifferentiated iPScs or rather using these cells to make a differentiated cell line of SMCs and/or ECs before seeding them onto scaffold matrices for cardiovascular tissue engineering. [Source: Dijkman and Generali et al.; Chapter Cardiovascular Tissue Engineering: Polymeric Starter Matrices for in Encyclopedia of Biomedical Polymers and Polymeric Biomaterials; 2015].
In the LifeValve project, with the focus on a paediatric tissue engineering concept, various cell sources were compared to define the most appropriate one for the production of tissue engineered heart valves (TEHVs). As a novel approach, the use of cell sources which can be obtained non-invasively was explored in comparison to the established vascular derived cells, including cells isolated from peripheral blood vessels, umbilical cord tissue, bone marrow and amniotic fluid. In accordance to the results in terms of proliferation capacity, expression profile, availability and the paediatric requirements, vascular and umbilical cord tissue derived cells have been turned out to be the most appropriate cell type.
As a novel and clinically interesting cell source in regenerative medicine, also adipose derived stem cells (ADSC) were included in the investigations. Fat, which is easy and repeatably accessible, is rich in pluripotent adipose tissue-derived stromal cells. In order to evaluate ADSC as an alternative cell source for the production of TEHV, adult ADSC were isolated from fat excisions from plastic surgery and characterized by immunohistochemistry, flow cytometry and differentiation assays. The cells exhibited characteristics of stem cells such as the differentiation capacity into the osteogenic, adipogenic, chondrogenic and endothelial-like linage. The stem cell (mesenchymal) specific combination of surface markers was detectable by immunohistochemical staining as well as by flow cytometry. To evaluate the cellular behaviour on 3D tissue formation, TEHVs based on PGA/P4HB were seeded with human adult ADSC and cultivated using strain bioreactors. The analysed ADSC-TEHVs showed a homogenous and vital cell distribution throughout the whole valve structure. Furthermore, a mechanically stable matrix with collagen production was demonstrated and the mechanical properties were comparable to native tissue. These preliminary results indicated that ADSC represent an interesting alternative autologous human cell source with clinical relevance due to their easy accessibility and excellent proliferation and tissue formation capacities. However, due to the existing experience and clinical safety of vascular cells, these cells were used for the frozen design TEHVs in the LifeValve project.
1.1.3 In vitro culture conditions
In order to evaluate the best in vitro conditioning protocol for the cultivation of TEHVs, two different bioreactor systems, the flow (full cardiac cycle) and the strain bioreactor (diastolic cardiac cycle), were investigated. The resulting TEHVs were histologically analysed concerning morphology, tissue formation and tissue organization. Additionally, cell phenotypes were characterized by immunohistochemistry, biochemical matrix proteins were determined and mechanical properties analysed by using a biaxial tensile tester. The results demonstrated the principal suitability of both bioreactor systems. The compositions of the extracellular matrix components as well as the biomechanical properties were rather comparable after cultivation in both systems.
1.1.4 Valve geometry
It is known that (engineered) tissues grow and remodel in response to the mechanical stimuli to which they are exposed. Therefore, to optimize the short-term remodelling process and prevent the development of leaflet retraction in tissue engineered heart valves (TEHVs), a computational model to evaluate the mechanical state of the valve and to investigate if this is similar to the mechanical state of native heart valves was developed in the LifeValve project. Native valves have a highly aligned collagen network, with most fibers running in the circumferential direction. Hence, when the valve is loaded during diastole, this results in small circumferential strains and large radial strains, which is beneficial for valve closure.
The analysis of the mechanical state in TEHVs with the initial valve design and material properties revealed that these valves have a tendency to become compressed in the radial direction during the diastolic phase, instead of being extended. This radial compression is detrimental for the functionality of the TEHVs, as it strongly contributes to the retraction of TEHVs. For this reason, computational modelling to investigate the potential influences of changes in the valve design (geometry and alignment of the collagen fibers) on the mechanical state of the valve was used. Based on these results, the design with the largest radial strain, as predicted by the computational model, was used to optimize the TEHVs in the in vivo studies.
Next to that, a tissue remodelling algorithm was added to the model to predict under which circumstances there is a risk for the development of leaflet retraction. The results indicated that leaflet retraction is not very likely to occur in TEHVs implanted in the aortic position. On the other hand, leaflet retraction in TEHVs implanted in the pulmonary position can only be prevented when the intrinsic contractility of the cells is very low.
By using the developed computational model, a failing mechanism for the dTEHVs was suggested. It appeared that the valves in the current geometry were subjected to leaflet tissue compression in radial direction under pulmonary pressure conditions. This in combination with the infiltration of host cells with contractile capacity, made it reasonable that these valves were prone to cell mediated leaflet shortening. In order to prevent tissue compression under hydrostatic loading, the valve geometry was adjusted by improving belly curvature and increasing the coaptation area. By comparing both designs, it appeared that this improved design is now subjected to leaflet tissue stretch in radial direction. Therefore the hypothesis was that by adjusting the final heart valve geometry, cell mediated leaflet retraction could be prevented by subjecting the heart valve to tissue stretch rather than compression. Unfortunately, controlling the tissue engineered heart valve geometry during culture has been limited so far since the cells deform the initial geometry by their contractile nature. As a solution to this problem, a rigid bioreactor insert has been developed that can impose the desired geometry during culture. In this way, the leaflets will compact around the rigid insert posts and will adopt the insert shape.
Two different insert iterations have been investigated. The first version was a solid structure with small holes to provide nutrient exchange to the adjacent tissue. However, it appeared that tissue formation became weaker in the vicinity of the coaptation area. In this specific region, the combination of dense scaffolding with the rigid insert was causing insufficient nutrient supply. To address the issue, large holes were created in the coaptation region of the bioreactor insert at these specific locations where constraining was not required. This resulted in enhanced tissue formation in those regions and solved the problem of insufficient nutrient exchange. To further improve the tissue formation in the belly region serum concentration in the medium was increased 10 fold. After four weeks of culture and subsequent decellularization, the inserts were removed from the valve constructs and resulted in functional, strong tissue constructs. In this way, any computational valve design can be translated into a compatible bioreactor insert to impose the desired geometry.
1.2 Evaluation of functionality of the tissue engineered valve/in vivo performance
After defining the first prototype (design 1.0) of the LifeValve of this project, and for the first time, we investigated the long-term in vivo functionality, host repopulation capacity, and matrix remodelling of the decellularized homologous tissue engineered heart valves (dTEHVs) as pulmonary valve replacement in sheep. Therefore, in a first series (published in JACC 2014) twelve dTEHV with the initial design were minimally invasively surgically implanted (via transapical access) into sheep and evaluated up to 24 weeks as pulmonary valve replacements. Further, 15 valves of the same design were implanted using the transjugular catheter-based access.
As a control, tissue-composition and -structure were analyzed in identical non implanted TEHVs (n=5). The transapical (n=12) and transjugular (n=15) implantation procedures were successful. No perioperative morbidity or mortality occurred and all valves could be delivered successfully at the target site. Early-postoperatively, two transapically implanted animals presented valve migration into the right-ventricular outflow-tract and died within 24 hours. The remaining animals (n=10) made a swift recovery and completed their respective follow-up without any complications. The appropriate position and functionality of the implanted valve was visualized by angiography. In vivo functionality (heart rate, mean and maximum transvalvular pressure gradient, and grade of insufficiency) was monitored using trans-thoracic echocardiography during the procedure, immediately after implantation and after 1, 4, 8, 16 and 24 weeks postoperatively.
In this study, initial valve functionality was excellent displaying sufficient leaflet motion and coaptation with only minor paravalvular leakage in some animals. Transapical and transcatheter jugular implantations yielded comparable results. While leaflet mobility was maintained on long-term follow-up, the coaptation slowly decreased over time, which was most likely due to radial compression during diastole and merging of the leaflets with the pulmonary artery wall at the level of the hinge area as well as the occurrence of a single leaflet prolapse in some animals. Consequently, mild to moderate central regurgitation could be observed at 16 weeks that further increased to moderate central regurgitation at 24 weeks. Functional measurements demonstrated stable mean and peak transvalvular pressure gradients over time (4.4±1.6 and 9.7±3.0 mm Hg respectively). After proving in vivo functionality, the animals were sacrificed and valves were explanted after 1 day, 8 weeks, 16 weeks, and 24 weeks. All explanted dTEHVs demonstrated shiny and pliable leaflets and dense whitish wall tissue, irrespective of the implantation period. In all explants, the wall tissue was integrated into the surrounding native pulmonary wall. Excellent coaptation of the explanted valves was evident in the implant and was maintained up to eight weeks of implantation. Thereafter, valve closure was incomplete in line with the observed central regurgitation. Apparently, the line of attachment of the leaflets to the wall shifted upwards in time indicating tissue merging process at the level of the hinge area, associated with a reduction in leaflet size with time.
The second series (design 2.0) of long-term animal experiments was started after design optimization of the LifeValve as described above (augmentation of the hyperbolic curvature (belly), prolongation of the coaptation zone, implementation of a bioreactor insert, gradual changes in nutritional medium concentrations and diastolic pressure augmentation). This second generation valve was implanted in 18 sheep by transjugular catheter technique. In the first six animals severe regurgitation after 8 to 24 weeks follow-up occurred. They met the endpoint criteria of regurgitation of more than 30% measured by magnetic resonance 2-dimensional quantitative flow analysis (flow MRI). Analyses of explanted valves revealed ruptured leaflets. Holes appeared mainly at the transition from curvature to coaptation zone near the attachment to the wall. It is suspected that this is the area of highest tension in the leaflets. As a result, the protocol of valve-manufacturing was stepwise adapted and the valves were subdivided in groups (“iteration steps”). For one group, the shape-giving insert was still added right from the beginning. However, a changed serum concentration was used when compared to the first six. In another group the insert was added after ten days of culturing in the bioreactor. For a last group of valves, the inserts were added after 13 days of culturing, giving the cells more time to generate extracellular matrix, thus creating a thicker and stronger tissue formation.
The valves, in which the insert was added later, showed very good coaptation after transjugular implantation in the animals. Coaptation area length was usually between 2 and 6 mm and stayed there throughout follow-up of up to one year. As a result of adding the inserts later (yielding stronger tissue formation), there were no signs of leaflet rupture during the crimping process and the course of implantation.
In the intracardiac echocardiographic assessment color flow mode, continuous wave Doppler and pulsed-wave Doppler were used in order to evaluate valve functionality.
By measuring forward flow, pressure gradients and flow accelerations were recorded. There were no signs of an elevated pressure gradient or an accelerated flux through the pulmonary valve, consequently showing no signs of stenosis.
Throughout the course of the study intracardiac echocardiography (ICE) assessment revealed slight regurgitation jets in some animals, none of which could be graded as more than mild, though. Color flow mode was used to detect regurgitation jets. Most of them were central, only few animals showed decentral insufficiency jets.
To grade and evaluate regurgitation more precisely, vena contracta (VC), proximal isovelocity surface area (PISA) and pressure half-time (PHT) was calculated from the recorded data. Vena contracta is defined as the point where the stream diameter of the regurgitation jet has its smallest width, yet fastest flux. PISA describes a three-dimensional hemispheric surface area close to the origin of a regurgitation jet, in which the flow converges. It gives a measure of the orifice area of non-coaptation. Depending on the width of this surface area combined with the vena contracta, insufficiencies can be assessed quantitatively (ml/s). To distinguish between moderate and severe regurgitation, pressure half-time can be useful according to the guidelines of the European Society of Cardiology for echocardiographic assessment of valvular regurgitation.
Considering all these measurements, most valves showed very good functionality, with a mild regurgitation if at all. Directly after the implantation procedure ten out of 15 animals showed no regurgitation at all. Flow was further assessed by cardiac magnetic resonance imaging (cMRI) which represents the gold standard for flow quantification. A regurgitation fraction was calculated with the obtained data, with one animal presenting with only 8% regurgitation after 52 weeks.
Endpoints: Animals were explanted either if the regurgitation fraction assessed by cMRI exceeded 30% or after 12 months (52 weeks). At the current time of follow-up, two animals have reached the landmark of 12 months. One of these animals showed only mild regurgitation in ICE and a regurgitation fraction of below 8% by flow MRI until the very last assessment before explantation. Up to date median values of the regurgitation fraction are approximately 10% in eight remaining animals at up to 48 weeks of follow-up.
In summary: One-year follow-up results of a competent, naturally remodelled heart valve, which has the tissue characteristics of a native valve, were already produced in two animals. Another eight animals are about to successfully fulfill the one-year follow-up, too.
The leaflets were as thin as in a native valve and free from thrombotic material. The shape of the leaflets preserved the hyperbolic curvature of manufacture. When loaded with hydraulic pressure (diastolic blood pressure) the leaflets closed competently.
The last three batches (in total ten valves) of these completely tissue-engineered heart valves deliver functional results which already fulfill the criteria of the ISO norm for valve certification: During opening the LiveValve shows no signs of stenosis (no antegrade flow gradient measured by intracardiac echocardiography). Its opening behavior is very similar to a native valve with the sole difference of a prolonged coaptation zone and slightly asynchronical opening in a few valves. The prolonged coaptation zone was introduced in design 2.0 to prevent leaflet retraction.
Valve closing is immediate and nearly complete. As in native pulmonary valves we see a tiny central regurgitation with a closing swash and a minimal leakage volume passing the valve during diastole. In ten animals this regurgitant volume adds to below 10% of the forward flow volume passing the valve during systolic ventricular contraction. The corresponding ISO norm demands less than 20% regurgitation for valve certification.
1.3 Remodelling
For all pre-clinical animal studies, the tissue-composition and -structure of the explanted dTEHV was analysed and compared to identical non implanted dTEHVs. A significant matrix remodelling was observed in the entire valve that corresponded with the rate of host cell repopulation. Prior to implantation, the dTEHVs revealed no cellular remnants and a well-developed extracellular matrix, mainly consisting of collagen, demonstrating an efficient decellularization procedure.
In the first pre-clinical animal studies with the first LifeValve prototype (design 1.0) it was observed that endogenous cellular repopulation occurred rapidly with first signs of cell infiltration as early as five hours after implantation of the dTEHV. Both leaflets and wall tissues were homogeneously repopulated over time with fastest repopulation at highest densities in the wall. After 24 weeks, cell repopulation density in the leaflets and hinge area approached that in the wall. Scaffold remnants remained longest visible in the leaflets with local increased cell densities. Minimal depositions, most likely blood platelets and fibrin, were present at the surface of the whole valve after five hours, but disappeared with time. The valvular tissues demonstrated abundant amounts of collagen that increased in density over time, in particular in the hinge area and wall and to a lesser extent in the leaflets. Elastic matrix formation was evident in the wall at eight weeks and later time points. In the hinge area the formation of elastic fibers was visible after 16 weeks and in the leaflets at 24 weeks. Calcification was not detected in any of the valves.
In order to analyse the repopulation of the dTEHV in more detail, cell phenotypes and distributions were analysed. Cells infiltrating the valve within five hours after implantation were all vimentin positive and alpha-SMA negative. After eight weeks, vimentin-positive cells were identified mainly in close vicinity of the polymeric scaffold remnants in the leaflet and hinge area, while more homogenously distributed and in higher amounts in the valvular wall. The cells in the leaflet and hinge area were all alpha-SMA negative and alpha-SMA positive in the valvular wall. After 16 weeks, homogenously distributed vimentin-positive cells were observed in all regions of the valve. These cells were alpha-SMA positive in the hinge area and wall and sparsely alpha-SMA positive in the leaflet. After 24 weeks, vimentin-positive cells were homogenously distributed over the valve. The level of alpha-SMA seemed lower in the hinge area and wall as compared to the 16 weeks explants. In the leaflets, more alpha-SMA positive cells were identified as compared to earlier time points.
As an endothelial layer on the valve leaflets is important to prevent blot clotting, the appearance of such a cell layer was investigated both by immune histological staining and SEM analyses. After eight weeks the dTEHV demonstrated partly confluent endothelial lining as observed by CD31 staining. Similar features were observed at the surface of the explants at 16 and 24 weeks. As visualized by SEM, the cell lining showed the typical cobblestone morphology, representative for endothelial cells, in all time points. The degree of endothelialization varied between locations and explants with decreasing endothelialization at the hinge area with implantation time and increasing endothelialization of the leaflet and valvular wall surface with implantation time.
Quantitative tissue analyses of the explanted dTEHV demonstrated that the DNA content (representative for the amount of cells) of the leaflets increased over the time in vivo. Although, after 16 weeks the DNA content was still lower than that in native leaflets, after 24 weeks the DNA content in the explanted dTEHVs was similar to that in native ovine valve leaflets. The amount of sulphated glycosaminoglycans (GAGs, that are important to keep the leaflets hydrated) was lower in freshly engineered dTEHVs as compared to that in native leaflets but approached native values after 24 weeks. The collagen content in the leaflets of the dTEHVs (being very important for the strength of the tissue) was equal to that in native ovine leaflets already before implantation and slightly increased after 16 and 24 weeks in-vivo.
1.4 Re-seeding
Host cell repopulation of the decellularized valve implant is the prerequisite for remodelling processes in vivo. In order to fasten and improve that process, the loading of the acellular valves with the patient’s own cells prior to implantation could be the method of choice. Especially, because endogenous repopulation is known to be enhanced by pre-seeding the graft prior to implantation with bone marrow mononuclear cells (BMMNCs), known for their paracrine effects. We hypothesized that a successful re-seeding method may facilitate enhanced remodelling of the TEHVs in vivo. Therefore, different tissue pre-treatment and seeding techniques were tested and analysed to find the optimal protocol for re-seeding of decellularized tissue engineered matrix. The method should guarantee the resistance of the re-seeded cells against the necessary crimping procedure and the flow/pressure stresses which the valves will experience directly after implantation. Finally, the most promising re-seeding procedure for decellularized tissue engineered cardiovascular constructs with BMMNCs prior to implantation was selected. This could potentially trigger a better host cells recellularization in vivo after implantation. Moreover, the process of reseeding was optimized towards an operator independent method that guarantees a standardized re-seeding quality. Due to time constraints within the project, no in vivo results of the performance of the re-cellularized TEHV could be generated.
1.5 In vivo tracking
In order to get insights about the behaviour, distribution and fate of the cells re-seeded on dTEHV, a PET (position emission tomography) reporter gene was used to track the cells in vivo. The PET-reporter gene (HSV1-tk) specifically binds a tracer called [18F]-FHBG that has suitable pharmacokinetics and dosimetry for clinical applications and imaging. The transfection method of mesenchymal stem cells isolated out of punctuated bone marrow was established.
The valves were implanted percutaneously into sheep in the pulmonary position under general anaesthesia (n=9), an additional sheep (n=1) did not undergo valve implantation to serve as control. Ten mCi [18F]-FHBG of PET tracer was produced for each procedure and serial PET-CT imaging of the sheep was performed 3h, 6h, 24h, 1 weeks, 3 weeks, 6 months and 7 months after the valve implantation. For the quantitative assessment of the number of cells survived in the 3D scaffold after in vivo implantation, vials containing 5x10*4, 2x10*5 and 4x10*5 transfected cells were mixed with the PET tracer for 1h, then the non-bound tracer was washed out and the vials were PET-CT imaged as in vitro control. To control the success of the transfection of the cells, the native (non-implanted) valves were investigated with PET-CT in vitro. In vitro PET-CT images of the valves showed the accumulation of the seeded cells mostly at the base of the leaflets. PET-CT images of the sheep 3h after implantation of the “living” valve showed a clear signal of the valves with the seeded transfected cells, with a mean estimated number of survived cells of 1.2x10*6. Extracardiac hot spots were seen in the mediastinal lymph nodes. No meaningful decrease of the cells living in the implanted scaffold occurred at 6h or 24h. Three weeks after valve implantation, living MSCs could be found in the valve location (estimated cell number 0.6x10*6) in one sheep, with some cardiac hot spots in the left ventricle. Immunofluorescense histology showed alpha-SMA positivity on the explanted valve surface one month after implantation. The explanted valve of one sheep after seven month FUP was investigated in vitro via PET-CT, which showed clear PET signal mostly at the valve base. Histology of the valves will be carried out. The remaining animals are still under investigation.
2. Stent systems
Endovascular stents were introduced in the early 1990s to restore the patency of occluded vessels for example of coronary, iliac or carotic arteries. In recent years, the stent concept has been extended to other indications and is now also being used as an anchor for artificial heart valves.
There are two general stent concepts: self-expandable and balloon-expandable stents. Self-expandable stents are made of nitinol. This is an alloy of nickel and titanium and has the two characteristics of superelasticity and thermal shape memory. Accordingly those stents are resistant to compression at body temperature. The stent appears compact in the compressed state, i.e. when integrated into the application system. At body temperature, the stent expands to its predetermined (unconstrained) diameter, thus restoring the patency of the lumen.
The nitinol stents are cut from a single nitinol tube and accordingly have no filament welding points or soldered joints. The stents feature a fine-meshed and flexible design. Therefore, they are ideal for insertion in an area of flexion and torsion. Most of the nitinol stents being implanted today have a mesh design.
Balloon-expandable stents can be made of various materials such as stainless steel, cobalt-chrome, platinum or out of various biodegradable materials, just to name a few. As all these types of material are not self-expandable, a balloon catheter is needed to deliver and dilatate them.
Minimally invasive implantation of heart valve replacement requires the use of stents for valve delivery and anchoring. However, the prosthetic valve becomes an integrated part of the heart within a few months and the stent itself is no longer needed. Furthermore, pediatric patients that receive a stented heart valve have to undergo multiple interventions as they outgrow the implant. Creating a minimally invasive implantable heart valve that can enable growth of the patient without the need of additional re-interventions requires not only a growing valve but also a growing stent. For this reason, the development of a biodegradable stent that vanishes after it has fulfilled its function was of paramount importance for the LifeValve project.
The stent has to meet very demanding conditions in order to enable heart valve implantation in pediatric patients. Some of them are pertinent to the implantation techniques and others are inherent to the material or the functionality of the implant. To begin with, the initial diameter of the stent should enable to host the heart valve, taking into account that the size of the valve depends on the size of the artery where it will be implanted which depends on the patient's body height and weight. Once the valve has been sutured to the stent, reducing the diameter of the prosthesis until sizes that enable minimally invasive insertion must be possible. For a transapical approach, the crimped diameter is about 10 mm. For a transcatheter insertion, the conditions are more challenging, requiring a crimping diameter of 8 or even 6 mm. These conditions impose an additional difficulty: the lower the crimping diameters, the higher the risk of permanent deformation. This means that after being pushed out of the catheter and self-expanded, the final diameter of the stent will be lower than the initial (plastic deformation). In other words, if this occurs the implanted valve could be smaller than the host tissue and the implant might migrate. To compensate for plastic deformation, the stent must be oversized so that after self-expansion, the resulting diameter is appropriate. The amount of oversizing was investigated within the project. To avoid migration, the stent must exert sufficient force against the artery to keep the implant in place (radial force). Little is known about the minimum force requirement; this topic was investigated as well during LifeValve. Since we aimed to create a stent that fully degrades and only the valve would remain, the stent material and its degradation products must be biocompatible. Metallic materials have excellent mechanical properties and provide the chance of very low or even no plastic deformation at all during crimping. On the other hand, most of them are not biodegradable and those that are, degrade fast. Polymers have the advantage of a good degradative potential but the downside of poor mechanical properties. For a proper functionality of the implant and due to the cyclic pulsatile loading of the implantation environment, the stent must be fatigue resistant for the first period of implantation. Finally, once the material has been selected and the conditions established, an appropriate manufacturing process has to be chosen. Such technique should not alter the mechanical properties of the material and should be reliable, accurate and reproducible.
In order to take into account all the previously mentioned aspects towards a biodegradable stent design that meets all the requirements, the following approach was followed within the LifeValve project:
- A non-biodegradable (nitinol) stent was taken as a reference to study and understand the implantation mechanisms and requirements.
- Computational simulations were created to resemble the nitinol stent geometry and material properties.
- Experimental data from ovine and pediatric arteries were obtained and implemented in the computational models to simulate a realistic host tissue.
- Forces during stent deployment and equilibrium with the artery were predicted computationally to set the requirements for the biodegradable stent.
- Different polymer alternatives and manufacturing procedures were evaluated.
- Based on the selected material, additional simulations were created to obtain a polymer stent geometry that adjusts to the requirements stated above.
- Polymer stent prototypes were manufactured and their forces were tested experimentally and compared with the computational results for validation of the methodology.
Experimental and computational methods were combined to study the forces and mechanisms during stent crimping of biodegradable and non-biodegradable materials. The nitinol system, used within LifeValve for ovine implantation of larger diameter valved stents, served as a reference to set an appropriate force requirement for the polymer stent for pediatric application. Pediatric and ovine tissue obtained after explantation helped to develop an appropriate host artery for the simulations and establish the equilibrium force and diameter after self-expansion. This scenario was crucial for the development of the polymer prototype.
Different biodegradable polymeric materials were evaluated for material selection. The different candidates were characterized for mechanical properties. The material that was closest to the set requirements was chosen as the material of choice for further testing and processing. The chosen material was a combination of two well-known biodegradable polymers in a 65-35 blend ratio.
After validation of the nitinol models and characterization of the biodegradable material of choice for computational implementation, an extensive study was performed assessing the influence of design parameters (such as the thickness, width and length of the struts) in the force of the stent. This method allowed us to optimize the geometry and obtain a theoretical design that meets the pursued force requirements, enabling crimpability in accordance with a transcatheter implantation approach and achieving a self-expanding diameter in accordance with the size of a juvenile ovine artery.
Various fabrication processes were tested during the Lifevalve project. These include melt extrusion, rapid prototyping, dip coating and laser cutting. The first demo prototypes were made in a two-step fabrication process. First, a tube of desired thickness from the chosen material is obtained by the process of dip coating. The process of dip coating was optimized for different diameters and thicknesses. After this, the tube is processed into a stent by laser cutting. A femto-second laser was used to avoid local heating effects on the stent. After laser cutting the dip coated tubes, manual crimping was evaluated and experimental radial force measurements were carried out to verify that the computational work provided accurate predictions. However, unexpected buckling of the stents was evidenced suggesting that the dip coating technique was not the optimal method for tube manufacturing. The resulting tubes were not homogeneous in thickness and appropriate crimping of the stents was not possible. Furthermore, validation of the computational models could not be achieved. This technique was further improved during the final stage of LifeValve. Nevertheless, these results could not be further materialized into a stent prototype before the end of the project.
To provide a reasonable alternative to validate the polymer computational models, a commercially available degradable filament was characterized and introduced in simulations until a suitable geometry was obtained and a 3D printed stent prototype was produced and tested experimentally. Buckling did not occur and experimental results were in good agreement with the computational predictions, validating the models. In addition, degradation of the material by hydrolysis and loss of mechanical properties was experimentally shown. Unfortunately, the composition of this material is not suitable for implantation due to the lack of biocompatibility.
In conclusion, the combination of computational and experimental results for the design and manufacture of a polymer stent with the pertinent mechanical properties to provide a stent alternative for heart valve replacement has been evaluated and successfully carried out. However, the simultaneous combination of a biocompatible material and a reliable manufacturing technique has not yet been achieved. To create functional stents based on laser cutting, it is of paramount importance to obtain tubes of good quality. If other rapid prototyping techniques such as 3D printing are to be considered, the selected material has to be supplied in a way that it is possible to be processed. Furthermore, the lack of precision of this technique should be accepted or accounted for, to have predictable computational tools. The most promising approach seems to be to focus future efforts on improving the tube production for laser cutting.
In an alternative approach allowing growth of the implanted valve, a study using oversized nitinol stents was performed. The hypothesis was that an oversized self-expanding nitinol stent will slowly expand over time and permit growth of the surrounding vessel. The radial force of the utilised nitinol stents could perhaps be used as driving force for native pulmonary vessel growth after valve-stent implantation: once implanted these metallic stents continuously tend to reach their maximum diameter. So they might support the native vessel to expand and grow with the valve-stent inside. Little was known about the radial force needed to keep the valve in place on one side and the effects of extensive radial pressure on the vessel structures on the other side. Twenty plus two (substitutes) animals were included into the oversized study: juvenile animals with a native pulmonary artery diameter between 16 and 18 mm were used to implant either (A) nitinol stents (length 30 mm) with a diameter of 20 mm which means ≈ 18 % oversized, (B) with a diameter of 25 mm which is oversizing of ≈ 47 % and (C) with a maximum diameter of 30 mm (≈ 76 %) oversized.
All stents were provided with a 20 mm custom made pericardial valve to avoid instantaneous free regurgitation into the right ventricle. Results indicated that using a self-expanding nitinol stent of the given design and radial force was not suitable as a growth permitting stent. It opened too fast and stretched the right ventricular outflow tract and the native pulmonary vessel.
3. Minimally invasive implantation technology (non-surgical procedure) of the LifeValve
Surgical heart valve implantations require the use of cardio-pulmonary bypass and open chest access. A hospital sojourn of several weeks including cardiopulmonary rehabilitation follows. Besides the undesired scar formation, an operation contains the risks of major complications like bleeding, infections, embolic complications, stroke and death.
As an alternative, minimally invasive procedures have been developed over the last decade. They erase the need for cardio-pulmonary bypass and open chest access. Up to date, heart valve replacements for minimally invasive implantation for the most part consist of a balloon expandable metallic stent covering a xenogenic valve. Use of balloon expansion requires over-pacing or short-term cardiac arrest (by use of adenosine e.g.) to reduce ventricular output.
In the LifeValve project a minimally invasive transcatheter implantation method was developed and applied in sheep. A heart valve was implanted through the jugular vein: an 11F (diameter of 3.7 mm) sheath was placed in the right jugular vein for the intracardiac echocardiography (ICE) probe. ICE was performed before and directly after implantation using an 8 or 10F ultrasound catheter and assessed 2D, color, pulsed wave and continuous wave Doppler to obtain dimensions and function of the native pulmonary valve before implantation. Fluoroscopy was used to perform angiographies and guide implantations. Pulmonary artery, right ventricular and right atrial angiographies were conducted and respective pressures were monitored and recorded. The access point at the right jugular vein was then dilated using commercially available dilators and a purse string suture created and tightened. A pigtail catheter bearing a pre-shaped ultra-stiff guide wire was advanced through the right heart and the pulmonary valve into either pulmonary artery.
In the catheter lab, the dTEHV was cut open and crimped to either 10 mm or – if feasible – to 8 mm. Crimping was done slowly over 20-30 min allowing liquids to be squeezed out of the leaflets without harming the tissue. In between the crimping device was reopened to investigate the leaflets for cracks or other damage. Finally, the dTEHV was loaded into the 10 mm or 8 mm capsule of the self-designed and tailor-made delivery system. The loaded system was then sheathlessly introduced over the ultra-stiff guide wire through the purse string suture into the jugular vein and forwarded into the right atrium. When using a 10 mm capsule in a smaller animal of below 40 kg it was sometimes a challenge to pass the tricuspid valve into the right ventricle and especially to take the bend from the right ventricle into the right ventricular outflow tract. In these small animals we experienced that the stiff length of the delivery system in relation to the diameter of the right ventricle is of utmost importance for the success of the procedure.
A delivery system, developed in the LifeValve project, was used for 55 successful implantations. These included tissue engineered (n=31) and pericardial valves (n=13) sewn into nitinol stents, an Andra-XXL balloon-expandable stent, one nitinol stent and several polymer stents (n=9). However, 78 attempts were necessary resulting in a success rate of 71%. Some of the failed implantation attempts can be attributed to the early stage of the development phase of the delivery system and have been solved in the course of the project (n=8). In two cases, the diameter of the delivery system was too large to be accommodated by the access vessel. The remaining problems were related to the length of the stent, the crimpability of the stented valve and the resulting catheter mechanics.
The delivery system is suited to be used for a wide range of implants. In its final version, outer diameter ranges from 13F (4.3 mm) to 32F (10.6 mm). Self-expanding as well as balloon-expandable stents, both with or without heart valves, can be implanted safely. Because of its great flexibility, almost every vessel can be accessed with the diameter of the delivery system and the stiff length of its tip in relation to the heart size being the limiting factor. For example, it could also be used for implantation of aortic valve prostheses.
4. Clinical pilot study
4.1 Identification of patients
For a later clinical study the database of patients of the department of congenital heart disease at the Deutsches Herzzentrum Berlin (German Heart Institute Berlin) was screened.
Inclusion criteria comprised:
• Adult patients (age >18 years)
• Primary diagnosis of tetralogy of Fallot (subpulmonary stenosis, ventricular septal defect, right ventricular hypertrophy, cyanosis)
• Surgical procedure: transannular patch reconstruction of the right ventricular outflow tract and closure of ventricular septal defect
• Magnetic resonance imaging (MRI) within the last three years
• Right ventricular dilation with an intraventricular diastolic volume of >140 ml/m2
Twelve patients (three female), with a median age of 25.8 years (range 17.0 to 31.8) were identified. They had MRI in median 2.1 years ago (range 0.6 to 3.2) with a median right ventricular volume of 154.5 ml/m2 body surface area (range 140 to 228 ml/m2). The normal mean value for the indexed enddiastolic right ventricular volume does not exceed 85 ml/m2 (standard deviation 60 to 100 ml/m2 maximum).
These patients are candidates for pulmonary valve implantation, e.g. LifeValve. We would appreciate to invite these patients to a first-in-human trial. However, the LifeValve pulmonary valve replacement has to prove well in the safety study and in further animal trials over a course of two years in accordance with regulary requirements.
The notified body for the clinical trial was identified and contacted. Necessary paperwork for ethical approval was prepared. Nevertheless the ethics application could not be handed in at the current state of the project as - according to the local authorities (LAGeSo) - two year animal studies regarding the special regulations are mandatory. Thus we applied for further funding to perform the necessary studies and to achieve the goal of a first-in-human study.
4.2 Translation of sheep protocols towards human tissue culture protocols.
For the translation of sheep to human tissue engineered heart valves, the cell source needs to be adjusted to make the dTEHVs compatible for human application. For the cell source, primary isolated human vena saphenous cells were expanded and used for culture. Human cells are known to exert less tension on the tissue, proliferate slower and require modifications in the culture medium. Based on the developed computational designs, the optimal valve geometry was predicted based on previously obtained tissue mechanical properties. Bioreactor inserts were produced accordingly and after 4 weeks of culture, human cell based tissue engineered heart valves were created and decellularized successfully. These valves have shown to maintain the imposed shape of the insert, including enlarged coaptation area and profound belly curvature. These valves were subjected to in vitro flow experiments, where they showed superior open and closing behavior, with leakage volumes even below 1%. Furthermore, these decellular constructs remained fatigue resistant up to 12 million cycles were tissue distribution was homogeneous throughout the thickness. Collagen anisotropy was observed near the coaptation region in circumferential orientation and became more isotopic towards the belly region. The tissue parameters of the eventually obtained design and mechanical properties were implemented in to the computational design, which revealed that these valves have a reduced amount of radial tissue compression compared to previously human tissue engineered heart valves. To conclude, human cell based dTEHVs can be successfully produced, which remain functional up to 12 million cycles. The geometrical improvements have shown that these valves are less prone to in vivo cell mediated leaflet retraction.
Potential Impact:
1. Benefits for patients
Valvular heart disease (VHD) represents a major global disease load with an increasing number of patients. For this reason VHD is a significant cause of morbidity and mortality worldwide. More than 250,000 heart valve replacements are performed worldwide per annum with a continuously increasing tendency that is expected to have tripled by the year 2050. Thus, the worldwide demand for heart valves is constantly rising due to age-related degeneration, improved survival of children with congenital heart defects and diseases like rheumatic fever. Surgical heart valve replacement using mechanical or bioprosthetic prosthesis represents the most common therapy for end-stage valvular disease and is a safe and efficient approach. Although currently available prostheses demonstrate excellent structural durability several limitations remain unsolved including the lack of growth capacity as well as in vivo repair and remodelling properties. Mechanical valves are known to have an increased risk of thromboembolic events. A lifelong anticoagulation therapy is required for these patients carrying a risk of spontaneous bleeding and/or embolism. Bioprotheses, either originating from animals (xenografts) or from human donors (homografts), do not require lifelong anticoagulation but are prone to dysfunctional, degenerative processes requiring high-risk redo operations. Therefore, these prostheses are less suitable for middle-aged and younger patients.
The development of a tissue engineered heart valve in the LifeValve project is a new approach to treat heart valve disease and a step towards a lifelong valve.
Tissue engineering is a still experimental approach to restore or replace damaged tissues in the human body. Instead of replacing organs or organ parts by foreign material (metals, polymers, foreign biological materials) it is attempted to make the organ parts from the patient´s own material. The “classical” heart valve tissue engineering concept comprising complex multistep procedures such as cell harvest, cell expansion, seeding on scaffolds, bioreactor in vitro culture, and time-critical implantation coordination of the delicate, living engineered autologous heart valves require high logistical and financial efforts. In this project we therefore investigated the concept of decellularized homologous tissue engineered heart valve, which is an innovative approach in order to create off-the-shelf availability of valve replacements without requiring a human or animal graft as starter material. The feasibility and long-term functionality of such homologous off-the-shelf tissue engineered heart valves was evaluated by implanting them via minimally-invasive transcatheter-based techniques in a relevant pre-clinical model (sheep). The LifeValve project has successfully demonstrated a well-functional tissue engineered valve for up to over one year. Although further research is necessary to refine the valve in order to improve the long-term stability, their remarkable rapid cellular repopulation and tissue remodelling, representing the self-repair capacity of these off-the-shelf valves demonstrates their potential as interesting alternative to current prostheses. The elimination of invasive cell harvesting from the patient, time consuming cell expansion and timeliness production of the patient specific graft will eliminate the waiting time for the patient and therewith increase the wellbeing of the patient and tremendously lower health costs.
Additionally, the decellularization methodology enables off-the-shelf and unlimited availability of the product in a large variety of sizes that will substantially contribute to the impact and success of the innovation and therefore may substantially simplify previous tissue engineering concepts towards clinical translation. The envisioned growth capacity of these valve replacements will eliminate the need for multiple reoperations, in particular relevant for pediatric patients, and therewith change the life of patients suffering from congenital cardiovascular disorders.
2. Socio-economic impact
The socio-economic costs of the European Healthcare System will be substantially reduced by the LifeValve technology, providing a growing heart valve replacement with repair and regeneration capacity and without the need for repeated reoperations and hospitalizations. Furthermore, faster recovery yielding shorter hospitalization will allow for faster turn-over in patient treatment. This will diminish treatment costs. Initially high costs for manufacturing will decrease over time as multiple companies will start to compete in tissue engineered valve production. Finally, pulmonary heart valve replacement will be a procedure of a three-day hospital sojourn, just like coronary artery stenting today.
Once tissue engineered valve replacement is established on the pulmonary side, this method will take over on the aortic side. This is when huge market numbers come into play. The catheter and medical device market is analysed to be the fastest growing sector in the medical industry within the next decade. Aortic valve replacement is estimated to hit the two billion US dollar benchmark in sales development within the coming five years.
The tissue engineered lifelong valve developed within the LifeValve consortium has the potential to foster the attractiveness of the European medical device market. Commercialisation is intended for both the delivery system and the dTEHV. The delivery system is designed and manufactured using materials already in use for comparable clinical products. This facilitates certification as a medical product. A patent was claimed and the patent application is published. The delivery system can be certified as a medical product on its own but would be preferably certified in combination with the LifeValve or its descendant (CE mark). Spin-off companies might attract venture capital leading to job creation in the field of medical and life-science technologies and the tissue engineering market.
3. Exploitable results
3.1 Decellularization technology
The decellularization methodology enables off-the-shelf and unlimited availability of the product. As this alternative approach for tissue engineering will also be applicable in other therapies for (congenital) cardiovascular diseases, it will not only extensively contribute to the knowledge in the field of heart valve disease, but more generally represent a tremendous step for clinical translation of cardiovascular tissue engineering worldwide.
3.2 Valve design and production
It is well known that (engineered) tissues grow and remodel in response to the mechanical stimuli to which they are exposed. Therefore, to optimize the short-term remodelling process and prevent the development of leaflet retraction in tissue engineered heart valves (TEHVs), we have developed a computational model to predict the tissue remodelling process in tissue engineered heart valves exposed to different conditions, i.e. different pressure differences over the valve to mimic the different hemodynamic conditions in valves implanted in the pulmonary or the aortic position, and different intrinsic contractilities of the cells inside the valves to mimic the different phenotypes. The results indicated that leaflet retraction is not very likely to occur in TEHVs implanted in the aortic position. On the other hand, leaflet retraction in TEHVs implanted in the pulmonary position can only be prevented when the intrinsic contractility of the cells is very low.
The computational model was used to redesign the LifeValve in order to improve the in vivo functionality. Controlling the geometry of tissue engineered heart valves during culture has been a major bottleneck for many years. With the development of constraining bioreactor inserts, heart valve geometry can be completely controlled and imposed. This resulted in the successful development of functional tissue engineered heart valves. The invention is not limited to pulmonary heart valve application, but can also be translated to aortic heart valve application. These bioreactors are therefore the key to produce functional tissue engineered heart valve, which otherwise would not be possible. A patent was filed about controlling tissue engineered heart valve geometry by using predefined inserts during culture.
All the above mentioned methods allow generating tissue engineered valves in a personalized design. Treatments nowadays address a broad market. Various products are mass produced, existing in different sizes but only available in limited varieties (e.g. Medtronic Melody® valve is offered in a narrow range of sizes). The tissue engineered valve is producible according to the patient’s needs from medical imaging data: according to size and form, fitting exactly into the patient. Moreover it can answer the need of a large market by being produced in an automated process.
3.3 Transcatheter Valves
The therapy for valvular heart disease is currently undergoing rapid changes. Besides conventional surgical valve replacement being the standard of care for several decades now, transcatheter techniques have been introduced for high-risk patients. Current results display promising outcomes and mid-term results. Transcatheter techniques can be performed in off-pump beating-heart fashion that can also be performed without general anesthesia in selected patients. As it is long-term safety proven via clinical, randomized trials, it can be assumed that the indications for this minimally invasive procedure will be continuously extended towards younger and less riskful candidates. Two specific routes for implantation have been established since then: the antegrade way via a surgical, direct transapical access, and the retrograde way using a transfemoral catheter-based concept. The transapical approach requires a left anterolateral mini-thoracotomy followed by a pledged purse-string suture to enter the apex, whereas the transfemoral method requires an adequate peripheral vascular access and can be performed fully percutaneously.
In the course of the LifeValve project, the minimally invasive transcatheter approach was followed in the animal experiments. For implantation by heart catheterization a proprietary delivery system was designed and manufactured.
Translating this approach to human patients would mean that the catheter carries the stented heart valve to its destination. It is possible to perform this operation under conscious sedation without the risks associated with general anesthesia. The physician accesses a vessel with a sheath and advances a deployment system into the beating heart. There is no major incision; by puncture of a vessel existing pathways leading straight to the heart are utilized. The catheter based intervention is a very gentle approach in comparison to open heart surgery. Usually the patient receives a compression bandage for 24 hours and can be released from hospital only a few days after the replacement valve has been implanted. Health care costs are extremely reduced using that approach. The delivery system is expected to work well in humans, especially as the crimping ratio is generally lower – i.e. less crimping is needed - compared to our sheep. Thus, less of the remaining problems should occur.
In this way, the LifeValve delivery system could be used to facilitate difficult implantations of commercially available prostheses such as the Medtronic Melody valve in which the Medtronic delivery system fails. This can be the case in Melody implantations which have to be conducted from the jugular access due to femoral thrombosis.
Currently, the innermost catheter is a 4F multipurpose catheter. For use with balloon expandable valved stents, this catheter must simply be exchanged for a balloon catheter. Corresponding technical drawings and calculations already exist to alter the existing delivery system.
The latest version of the delivery system has been designed toward ergonomics, safety and one-user operability. In addition, most parts can now be produced by injection molding. This is a major milestone toward serial production. For commercialization and certification, only minor changes remain necessary. The flexibility and the modular design of the delivery system enable versatile applications making it an ideal standardized platform for several applications.
3.4 Stents
The transcatheter approach neccessitates the usage of a stent system to deliver the valve, to fix the valve in place and to promote fusion of the engineered tissue with the native host tissue. Once this fusion is established, the stent not only loses its mechanical function, it also prevents the valve to grow, which is essential for the envisaged lifelong approach of the tissue engineered heart valve for pediatric patients. The approach used in LifeValve was the development of a biodegradable stent. Although the goal to develop a biodegradable stent could not be fully realized within the lifetime of the project, important cornerstones were laid, such as the design, the material and the production method.
There are numerous ways a suitable stent can be designed. However, designs have to fulfil certain criteria in order to be applicable. With the computational model developed within the project different stent designs can be investigated for their crimpability, self-expandability and the resulting reaction forces. These computational iterations can provide easy and fast evaluation and predict if a specific design is suitable for the application. Also, suggestions for further improvement can be provided based on this computational model. In this way, less labour intensive laboratory experiments are required before an appropriate stent design can be finalized. This will lead to significant R&D cost reductions (“time to market”).
4. Dissemination
The LifeValve dissemination aimed to promote knowledge sharing among the scientific community and to increase awareness of the project results on the part of the public. Various instruments were used to reach that goal.
The project website (http://www.remedi.uzh.ch/lifevalve) has been installed at the University of Zurich and was regularly updated. On the publicly accessible sites, the project homepage gives detailed information on the project. This included not only the scientific background and objectives of LifeValve, but also explanations on why we focus on children because of the unmet medical need of a living heart valve with the capacity of growth and which can be implanted minimally invasively. In order to highlight the collaborative nature of the project, the homepage also contains information on the work division between partners so that an interested audience could follow the plan of work within the project. In addition, there is a section giving detail information on all beneficiaries including a short introduction of their core research expertise, selected publications and contact information. The press section of the website contains links to several documents on LifeValve. For example, LifeValve has been chosen to be portraited as a StarProject by the EC Research & Innovation and was featured in a video by Euronews for Futuris, the European research programme. Furthermore, LifeValve has also been selected by the Communication and Support Action COMED as one cutting-edge health-research project funded within the Seventh Framework Programme. COMED aims communicating results and activities of EU-funded health research projects and improving the perception of sciences by the European public. The extensive science documentary about LifeValve with Berlin and Zurich as shooting locations used animations, background statistics, interviews and case studies to illustrate the objective and research approach of LifeValve. By linking the Project website to both the websites of StarProjects by the EC Research & Innovation and COMED, we aim to guide interested EU citizens to other cutting-edge projects within the EU-FP7.
The partners also addressed the LifeValve project on their institutional homepages.
The LifeValve project was presented at various international conferences to the scientific community. One of them, for example, was the annual meeting “Cardiovascular Cell Therapy and Innovative Strategies” in Madrid, Spain (29.5.2015) where the abstract entitled “Transcatheter delivery of homologous cell-derived off-the-shelf tissue engineered heart valves with self-repair capacity in a translational animal model” was appointed as the best one in the category translational/clinical studies.
Six peer reviewed publications arose from the project and more will come up in the near future. The first series of in vivo studies was published in the Journal of the American College of Cardiology (impact factor >15 in 2013). The study was highlighted by an editorial comment in the same issue of the journal.
Apart from scientific publications and dissemination activities directed at a general audience the consortium members also participated in events with key stakeholders in healthcare as the target audience. For example, the LifeValve project has been presented at the conference “Innovation in Healthcare. From Research to Market. SMEs in Focus” (organized jointly by the Research DG and Enterprise & Industry DG, Brussels, May 2010) and at the EU-China Science and Technology Week, organized as part of the European Union Pavilion at the World Expo 2010 Shanghai China (June 2010).
In 2012 e.g. the coordinator of LifeValve, Prof. Simon P. Hoerstrup, participated in the meeting “Industrial Technologies” in Aarhus, Denmark, where over 160 high profile international speakers from industry, government and research met to discuss visions for European research and industry in 2020, how Europe can succeed in the face of global competition, and the form and impact of Horizon 2020.
In the presentation “From cells to functional cardiovascular implants” he also pointed out those topics of importance that should be addressed in the upcoming Horizon2020 program in the field of regenerative medicine.
LifeValve related topics were also covered in the public press (e.g. in Switzerland). In the Horizon2020 brochure "Investing in success" (European Commission, Directorate-General for research and Innovation, 2012), LifeValve was highlighted as a story of success.
5.Outlook
For the first time ever, the feasibility and long-term functionality of transcatheter homologous off-the-shelf tissue engineered heart valves were demonstrated in a preclinical model. The concept of homologous off-the-shelf tissue engineered heart-valves will substantially simplify tissue engineering concepts towards clinical translation.
The combination of heart valve tissue engineering and minimally invasive implantation technology by using self-expanding and biodegradable stents is a novel approach and provides Europe with a unique opportunity to reinforce its competitiveness as to novel medical and life-science technologies.
This successful and highly innovative research shall be continued. The next step is preparation of a first-in-human trial. The ultimate goal is to bring the worldwide first tissue engineered heart valve on the market. This valve has the potential to solve the worldwide shortage of human heart valves as it can be produced on large scale - independent from human or animal valve donors. Therefore with members from the LifeValve project a new consortium was formed which has submitted a proposal in the frame of Horizon 2020.
List of Websites:
http://www.remedi.uzh.ch/lifevalve.html
Universität Zürich (Coordinator) (CH)
Prof. Dr. Dr. Simon P. Hoerstrup
Cardiovascular Surgery Research
Regenerative Medicine Program
Moussonstrasse 1
8091 Zurich, Switzerland
www.remedi.uzh.ch
Deutsches Herzzentrum Berlin (DE)
Prof. Dr. Felix Berger
Department of Congenital Heart Disease / Pediatric Cardiology
Augustenburger Platz 1
13353 Berlin, Germany
http://www.dhzb.de/de/kliniken/angeborene_herzfehler_und_kinderkardiologie/
Technische Universiteit Eindhoven (NL)
Prof. Dr. Frank Baaijens
Department for Biomedical Engineering
Den Dolech 2
5612 AZ Eindhoven, The Netherlands
http://www.tue.nl/en/university/departments/biomedical-engineering/
Medizinische Universität Wien (AT)
Prof. Dr. Mariann Gyöngyösi
Univ. Klinik für Innere Medizin II
Abteilung für Kardiologie
Waehringer Guertel 18-20, 1090 Wien, Austria
Pfm Produkte für die Medizin AG Köln (DE)
Dr. Frank Thiel
Wankelstr. 60, 50996 Cologne, Germany
http://www.pfmmedical.com/en/index.html
Xeltis AG (CH)
Laurent Grandidier
Muehlebachstrasse 26
8008 Zurich, Switzerland
http://www.xeltis.com/
Xeltis BV (NL)
Dr. Martijn Cox
De Lismortel R2 06 Cat Build 31
5612 AR, Eindhoven, The Netherlands
http://www.xeltis.com/
University of Debrecen (HU)
Department of Nuclear Medicine
Dr. Laszlo Balkay
Egyetem Ter 1, 4032 Debrecen, Hungary
www.nmc.dote.hu
Background: Approximately 1% of all newborns suffer from congenital heart defects which cannot be treated efficiently due to the lack of growth of the clinically available "artificial" replacement materials, such as heart valve prostheses. Furthermore, current heart valves have significant limitations including limited durability and/or the need for lifelong medication. Today, heart valve prosthesis-associated problems occur in 30-35% of patients within 10 years, frequently necessitating high-risk, potentially life-threatening re-operations.
Objectives: The initiative “LifeValve” (Living autologous heart valves for minimally invasive implantable procedures), funded by the European Commission under the 7th Framework Programme, aimed at developing a tissue engineered heart valve which can be implanted by minimally invasive catheter technology. The valve was expected to have repair and regeneration capacity, thereby minimizing the need for future surgical intervention. LifeValve combined two novel life science technologies – tissue engineering and minimally invasive implantation technology for the benefit of children with congenital heart disease.
The technological objectives of this translational approach comprised (1) the in vitro creation of a valve with the capacity of growth by tissue engineering methods, (2) the development of intelligent stent systems, (3) a minimally invasive implantation technology (non-surgical procedure) of the LifeValve and (4) the preparation of a clinical study. This highly interdisciplinary collaborative project was implemented by eight renowned academic and industrial partners from Austria, Germany, Hungary, The Netherlands and Switzerland representing a comprehensive range of expertise in tissue engineering, biomedical and material research, medical technology and clinical practice.
Results: In the course of the 5.5 years initiative, a novel tissue engineered heart valve was developed. The multistep development included various optimization and adaptation steps. The valves of the first generation were made by seeding cells on a PGA/P4HB scaffold and subsequent cultivation in a bioreactor. Those living valves were limited by leaflet retraction phenomena leading to valvular incompetence in vivo as tested in a sheep model. In order to circumvent this problem, a new and alternative LifeValve biomaterial was investigated based on decellularized tissue-engineered extra cellular matrix. Further modifications steps were necessary focusing on optimization of the geometric design and the culture conditions. Finally, a valve was created showing good performance in vivo (sheep) with a follow-up time of 12 months. Regarding the stent, which is needed as a carrier for the tissue engineered valve and the fixation within the vessel, two systems were investigated: the already clinically established self-expandable stent material nitinol and a biodegradable stent which enables full incorporation of the minimally invasively implanted tissue engineered heart valve into the patients and subsequent growth. The design requirements of the nitinol stent were defined with the highest priority on radial force and secure anchoring. For the development of a bioresorbable stent, the nitinol stent served as a model for the computational based design. The chosen material is Poly L-lactic Acid (PLLA), known to be biocompatible and non-cytotoxic, that is subsequently modified with P4HB to improve the mechanical properties (flexibility). Stent development is in the final stage. Within LifeValve, minimally invasive implantation procedure was established. A specific delivery system for the LifeValve technology was developed, improved stepwise and adapted to the requirements in order to ensure a safe transcatheter based delivery. The device was routinely used with a high success rate in the pre-clinical study. In parallel, patients for a clinical study were identified and a first-in-man trial design was established.
Impact: Apart from the scientific and medical progress represented by this one-step next generation type of therapy for congenital heart defects, the socio-economic costs for European healthcare systems can be substantially reduced by the LifeValve technology, providing a growing heart valve replacement without the need for multiple re-operations and hospitalizations. The combination of heart valve tissue engineering and minimally invasive implantation technology by using self-expanding and biodegradable stents is a novel approach and can provide Europe with a unique opportunity to reinforce its competitiveness as to novel medical and life-science technologies and the related fast growing markets.
Project Context and Objectives:
Cardiovascular disease still represents the Killer Number One in all Western Societies accounting for the death of numerous patients and annual health care costs in the range of billion Euros. Valve replacement is the most common surgical therapy for valvular heart disease with almost 250.000 annual implantations worldwide. Valve replacement procedures are efficacious and substantially change the otherwise deadly natural history of valvular disease. However, although the overall performance of these devices is sufficient, prosthesis-associated problems occur within 10 years postoperatively in 30-35% of patients. Mechanical valves require lifelong anticoagulation therapy. Bioprosthetic valves have limited durability and may calcify prematurely particularly in young patients. More importantly, both mechanical and bioprosthetic valves are non-viable structures and do not have the ability to grow, repair or remodel, which is a specific problem in the pediatric patient population. Congenital heart disease (occurring in almost 1% of all newborns) represents a potentially life-threatening situation and treatment by highly invasive surgical interventions with substantial morbidity and mortality is often necessary. In many congenital heart malformations underdeveloped or malfunctioning heart valves are involved, which makes heart valve replacement necessary. A particular problem relates to the fact that all clinically available heart valve substitutes represent artificial or non-living structures which cannot grow with the paediatric patients. This normally necessitates several re-operations to adjust the heart valve prosthesis to the growing organism. Each of these re-interventions is associated with exponentially increasing morbidity and mortality.
The general objective of the LifeValve project was to develop a new therapeutic strategy to treat heart valve disease. In particular, the development of a tissue engineered living heart valve, with the capacity of growth in accordance with the growth of children, which can be implanted without an operation by minimally invasive catheter technology was to be addressed. The rationale for first focusing on paediatric applications was twofold: First, a novel, tissue engineered living heart valve with the potential for growth is addressing a clearly unmet medical need in congenital heart disease. All contemporary solutions with conventional heart valve prostheses are insufficient, do not provide a sustainable treatment and as a consequence create suffering and substantial socioeconomic costs. Second, successful use and increasing clinical experience within the paediatric patient population as entry indication will allow extending the clinical application to the much larger adult patient population with degenerative heart valve disease in a later stage.
To successfully realize this project, a highly interdisciplinary consortium consisting of academic research groups and clinicians as well as market oriented companies was formed having expertise in the fields of tissue engineering, biotechnology, cardiovascular surgery and medical devices. The integration of these partners represented the full range of life science technology yielding from basic science to clinical application and marketable products (“from bench to bedside”).
The LifeValve project aimed to combine two recent novel life science technologies, tissue engineering and minimally invasive (non-surgical) implantation technology. The technological objectives of the LifeValve project comprised (1) the in vitro creation of autologous cell based valve substitutes by tissue engineering methods (2) the development of an intelligent stent system, (3) a minimally invasive implantation technology (non-surgical procedure) of the LifeValve. The functionality of the valves should be tested in an animal model before initiating a clinical trial (4).
1. In vitro creation of valve substitutes (LifeValves) by tissue engineering methods
To address the substantial limitations of state of the art artificial heart valve prostheses, the ultimate goal of the tissue engineering approach was to construct tissues from their cellular components which combine most of the characteristics of the healthy native original. For a functional heart valve these include adequate mechanical characteristics, (mature extracellular matrix), durability, adequate haemodynamic performance, as well as the absence of immunogenic and/or inflammatory reactions. Moreover, the tissue engineered heart valve is supposed to offer the inherent capacity of growth, addressing a so far unmet medical need in children with congenital heart disease.
The success of the production of tissue engineered heart valves is dependent on four main issues: (a) the matrix (scaffold) which determines the three-dimensional shape and serves as an initial guiding structure for cell attachment and tissue development; (b) the cell source from which a living tissue is grown; and (c) the in vitro culture conditions of the construct before implantation and (d) the valve geometry.
Covering the whole production process, the project started by defining the optimal biodegradable biomaterial matrix requirements for the in vitro production of tissue engineered heart valves. Various synthetic biomaterials were tested in vitro as to adequate biocompatibility and biodegradation properties to eventually serve as a cell carrier for the tissue engineering process in the bioreactor to create a heart valve, which is finally able to accommodate growth in the patient. For the tissue engineering of the heart valves, different cell sources and culture conditions were tested for the suitability of tissue formation with regard to collagen formation (later replacing the initial scaffold) and mechanic stability. Later in the project the focus was put on the so-called decellularization technology, which included seeding cells on a scaffold and culturing in a bioreactor system to form tissue that finally takes over the function of the initial scaffold as a valve matrix with a specific design. All living cells were removed so that a matrix composed of extra-cellular components such as collagen is available that can be stored or directly used for implantation. The initial scaffold degrades stepwise.
Beside the matrix, the functionality of a tissue engineered heart valve strongly depends on the geometry of the valve. During the course of the project it turned out in the first pre-clinical trials that the initial geometric design of the valve let to functional insufficiencies in the animal, mainly because of retraction and thickening of the valve leaflets. These observations necessitated a substantial redesign of the valve using computational models and a subsequent adaption and optimization of the culture conditions.
The prerequisite for long term in vivo functionality of a tissue engineered valve is that the valve becomes part of the human body, which means that in vivo re-cellularization and remodelling occurs making the final valve similar to a native one in composition and function. Many cells are endowed with a reservoir of soluble factors that can exert paracrine effects on other cells ingrowth resulting in remodeling stimulation. It was thus hypothesized that recellularization directly prior implantation improves cell recruitment followed by generation of viable tissue.
Beside the establishment of such a recellularization procedure also cell tracking methods were investigated and utilized to follow the fate of the seeded cells during the in vivo phase of the experiments.
To assess tissue maturation and implantability, analysis of the stented tissue engineered heart valves comprised morphology (histology, immunohistology, electron microscopy), matrix protein quantification (collagen, GAG, DNA, elastin), biomechanical properties (tensile strength testing) and functional performance in a pulsatile flow simulator.
Furthermore, a special focus was on the establishment of stable (standard operation procedures) in vitro processes in accordance with GMP regulations in order to provide small scale production capacity for the consequent first-in-man clinical trials.
2. The stent system
A particular aim of the LifeValve project was the application of minimally invasive catheter techniques to avoid open heart surgery. The tissue engineered heart valve implanted by minimally invasive methods needed a carrier and an anchor to be fixed into the target position within the heart. For these purposes stents were used. It was a particular challenge to design a stent system that can be folded or crimped for delivery of the construct without being damaged, nor damaging the engineered tissue, and that, after expansion, can exert appropriate radial force and load on the conduit to resist the mechanical load exposed on the valve.
A clinically established stent based on nitinol allowing self-expansion (memory alloy) was used for the proof-of-concept of the tissue engineered heart valve within LifeValve.
However, in view of the applications in pediatric patients, nitinol is disadvantageous, as growth is restricted to the dimension of the used stent. An integral part of the LifeValve project was therefore the development of an intelligent stent system that accommodates the growth of children to avoid re-operations. Two routes regarding this issue were followed: the development of a biodegradable stent as well as the usage of an “oversized” nitinol stent.
Principally, the stent is part of the tissue engineered heart valve, and should not only fix the valve in place, but should also promote fusion of the engineered tissue with the native host tissue. Once this fusion is established, the stent loses its mechanical function. When following the biodegradable stent concept, the degradation of the stent allows growth of the tissue. Thus the developed stent was designed to have the following characteristics: (1) the used material should have a degradation time that is slow enough to allow in vitro production of the valve within the stent in a bioreactor system. On the other side degradation must be fast enough to allow growth of the surrounding native vessel; (2) the stent should be flexible so that it can be crimped down to small diameters (necessary for catheter based implantations) and reach the initial diameter after crimping and expansion; and (3) after implantation it should facilitate successful fixation and ingrowth of the valve. Therefore it must exhibit sufficient radial forces to hold the valve in the correct position. The biomaterial stent should be either self- or balloon-expandable. The stent system and the stent to biomaterial junction were of critical importance for the success of a functional and mechanically durable heart valve substitute.
In an alternative approach the usage of an oversized nitinol stent to accommodate growth of the implanted valve was investigated. The remaining extension capacity to the maximum diameter had to follow the growth of the native right ventricular outflow tract (RVOT) and pulmonary artery. For that purpose, oversized nitinol stents were implanted in juvenile sheep. Performance and tissue reactions (e.g. local vascular trauma, tissue remodeling) was assessed in order to prove the feasibility of that approach.
3. Minimally invasive implantation technology of the LifeValve
A further objective was the implantation of the LifeValve by transcatheter techniques to avoid open-heart surgery. An appropriate loading and delivery system was developed and continuously adapted to the ongoing modifications of the LifeValve. The loading system needed to be able to crimp the stented valve to an adequate small diameter and load it into the delivery system immediately before implantation under sterile conditions in the catheterization laboratory to ensure a minimum of stress to the valve. The delivery system had to be flexible and thin enough to be advanced by catheterization techniques through the narrow jugular vessels and the tortuous cardiac anatomy. Care had to be taken, that the tissue engineered valve remained unharmed during the procedure and that the deployment of the stent was exactly controllable and reproducible. Feasibility studies were performed in a large animal model. The most accepted animal model (FDA, EMA, Swissmedic) for the investigation of novel cardiovascular implants is the sheep. Thus, sheep were used to assess the functionality of the minimally invasively (transcatheter) implanted autologous living heart valves throughout the LifeValve project.
4. Clinical pilot trial
A specific focus was on the establishment of GMP (Good Manufacturing Practice) adequate logistics, which later can be utilized in the clinical part of a first-in-man project. Following proof of safety, long-term animal experiments were undertaken to demonstrate regeneration capacity and adaption of the valve to the biological environment. Finally, a clinical pilot study in paediatric patients with congenital heart defects was initiated.
Project Results:
1. Development of the tissue engineered heart valve (LifeValve)
1.1 In vitro creation of valve substitutes by tissue engineering methods
LifeValve aimed at developing a living tissue engineered heart valve with the ability of growth. Therefore, the choice of a biocompatible and biodegradable heart valve matrix (scaffold) allowing for growth of the implant was essential. Furthermore, as the LifeValve project aimed for development of a tissue engineered valve which can be implanted via the trans-catheter implantation technique, the final valve needed to be crimped to fit the very limited dimension of the trans-catheter implantation device. To reach this goal, the four main issues that define the success of the development of a functional tissue engineered heart valves were elaborated on: (1) the matrix (scaffold) which determines the three-dimensional shape and serves as an initial guiding structure for cell attachment and tissue development; (2) the cell source from which a living tissue is grown; (3) the in vitro culture conditions of the construct before implantation and (4) the valve geometry.
1.1.1 Scaffold material
In recent years, research groups have used a variety of different approaches and methods to develop tissue engineered heart valves (TEHVs) that are now at various stages of clinical development. The cyclic loading of the cardiovascular system (pulsative blood flow) requires an enormous strength, flexibility, and durability of the tissue-tissue engengineered cardiovascular replacements and consequently appropriate mechanical properties of the starting matrices to endure the cyclic stresses and strains exerted upon implantation. More specific, starter materials for TEHVs should contain stiffness around 0.5 MPa, while being elastic without any tendency to deform permanently under the influence of stresses. These requirements to the overall mechanical behavior of the material are determined by the intrinsic material properties (e.g. stiffness), the scaffold architecture (e.g. anisotropy) and the degradation rate. Moreover, due to the intense contact with the blood in cardiovascular applications, it is desirable to have a thromboresistant material e.g. by providing an endothelial surface layer. In general, cell infiltration into the scaffold is one of the prerequisites for successful tissue regeneration and is primarily determined by the scaffold microstructure, i.e. pore size and fiber thickness. Additionally, the microstructure of the scaffold can affect cell phenotype and influences the behavior of the infiltrating cells regarding cell adhesion, spreading, and proliferation. It is also used as an anchor to attach different bioactive molecules and signals that improve specific cell function, such as, pro-angiogenic signals. The mechanical properties of the scaffold also provide an important stimulus to the cells for extra cellular matrix (ECM) production and remodelling, as the cells experience different local stresses and strains depending on the scaffold stiffness. These mechanical cues, as result of the scaffold stiffness, can modulate the differentiation of cells into pathological phenotypes, e.g. osteoblastic or myofibroblastic differentiation. Additionally, bioactive factors can be incorporated into the scaffold that can direct local cellular function, or promote recruitment of specific cell types via chemotaxis. To conclude, the initial scaffold should meet the following requirements: (1) to be biocompatible and thromboresistant, (2) to be able to support cell infiltration, growth and cell-to-cell interaction, (3) to start with sufficient mechanical properties and degrade at a rate in relation to new tissue formation, (4) to have optimum architectural properties of pore size, porosity and permeability in order to allow diffusion of nutrients and metabolic waste products and, last but not least, (5) to be bioresorbable. [Source: Generali, Dijkman & Hoerstrup; Bioresorbable scaffolds for cardiovascular tissue engineering; EMJ 2015].
Different types of scaffold material have been used for heart valve tissue engineering, unfortunately none complying with all the aforementioned requirements so far. The matrices for valve replacement that comes closest to the geometry and structure of the native valve with native-like mechanical and physiological hemodynamics, is the xenograft (tissue from one species to an unlike species) or allograft (=homograft; tissue of the same species) valve. Obviously, the allograft valve is a more natural option compared to the xenograft valve, as the ideal valve replacement should show biomechanical characteristics similar to those of the native heart valve. Natural leaflet motion and flow patterns are supposed to avoid stress related calcification and early valve failure. The microstructure of the allograft valve is favourable with respect to proliferation, differentiation, and survival of reseeded cells. The availability of the allograft valve, however, is limited by donor shortage. Due to its anatomic similarity to human valves, the porcine heart valve represents an attractive alternative to allograft valves. Additionally, also xenogenic pericard is investigated as material for valve replacements. However, xenografts are associated with the risk of immunogenic reactions or disease transmission and the availability of homografts is limited. To eliminate immune reactions with xenogenic tissues, they are decellularized before implantation and are either reseeded prior to implantation or expected to become repopulated in vivo. Although the latter was partly demonstrated in animal models, clinical studies led to contrary reports about cellular infiltration in humans and proof of concept is still missing.
To overcome the limitations of xeno- and homografts, the applicability of various bioresorbable polymers, either synthetic (such as PGA, P4HB and PCL) or natural materials (such as fibrin and collagen) was investigated. Seeded with (autologous) cells and subsequently cultured in vitro, these materials have shown to be feasible for heart valve tissue engineering, with demonstrated functionality in vitro and in vivo. For example, autologous myofibroblast and endothelial cells (ECs) (or neonatal human dermal fibroblast) have been used to create fibrin-based TEHVs and have been evaluated in sheep up to three months. Although the leaflets remodelled in vivo, the retraction of the leaflets resulted in valvular insufficiency therewith hampering the functionality of the valve. Also the synthetic bioresorbable polymer PGA showed promising results when combined with P4HB or PLLA. When seeded with autologous myofibroblasts, mesenchymal stem cells (MSCs), and ECs valves can be engineered in vitro that undergo structural and functional remodelling in vivo without stenosis. However, also in these valves functionality was hampered by trivial to moderate regurgitation after six weeks due to thickening and retraction.[Source: Dijkman and Generali et al., Chapter Cardiovascular Tissue Engineering: Polymeric Starter Matrices for in Encyclopedia of Biomedical Polymers and Polymeric Biomaterials; 2015].
Classical tissue engineering approaches use patient specific cells to engineer tissue constructs in a so-called autologous approach. To investigate the potential of this approach for heart valve applications within the LifeValve project, cells were isolated from sheep that were used as the appropriate animal model. Sheep are prone to develop early heart valve calcification due to their high calcium metabolism and are therefore the so called “worst-case scenario” animal model. After isolating cells from a small piece of the vasculature, cell numbers were expanded in the laboratory to approximately 80 million, which is sufficient to culture one valve.
As an appropriate scaffolding material, poly(glycolic acid) (PGA) was combined with P4HB in order to shape the scaffold material into a complex three-dimensional heart valve geometry, which was later on sutured into suitable stent devices. These stented heart valve scaffold constructs were then seeded with the isolated cells from the sheep and placed inside a sophisticated bioreactor system. This system was specifically engineered to mimic native pulsatile pulmonary pressure conditions. When the cell seeded constructs were subjected to these mechanical stimuli, cells showed an enhanced production of tissue, which gives the heart valve strength over time. After four weeks of culture, the cells have produced sufficient amounts of tissue, on the other hand the scaffold material has mostly degraded. The downside of using PGA based scaffolding material is that it loses mechanical integrity after already two weeks of culture time. As a result, the cells used during culture could deform the initially defined geometry of the valve by applying increasing rates of contractility.
In an attempt to overcome this problem of losing initial valve geometry, stiffer scaffold material such as poly(caprolactone) PCL, which has a much slower degradation rate, was investigated. Unfortunately this approach did not work out so well. Cells were still capable of deforming the valvular geometry, causing leaflet retraction. Even so, after culture the PCL is still abundantly present inside the tissue construct after culture and largely contributes to the mechanical integrity of the valve. PCL has a low fatigue resistance. Therefore the valve construct can quickly lose its mechanical support, while by that time the tissue is not strong enough to fully take over the load.
As a result, the PGA-P4HB combination was chosen as a final scaffold material in which the heart valves are from now on cultured in a so-called closed configuration, which means that the three leaflets are merged after culture, since they grew together during culture. In this way, the initial valvular geometry is lost during culture, but the leaflets are not retracted and the valves are still functional afterwards. Although scaffold remnants still reside in the valvular construct, they do not mechanically support the tissue anymore. In this case the engineered tissue is now carrying the full load. These so called autologous heart valves were then implanted into the same animals from which the cells were originating. However, over time it became apparent, that the leaflets partially retracted over time hampering their mobility and functionality.
In order to overcome this limitation, decellularized tissue engineered heart valves (dTEHVs) were investigated for their potential use as a suitable biomaterial for heart valve prostheses. The way these dTEHVs were produced is comparable to the living tissue engineered heart valve, but cells were not patient specific, but species specific and that after four weeks of culture, the cellular content was completely removed. In this way an entire non-living three-dimensional heart valve shaped tissue construct was produced. Prior to implantation, DNA content was removed to prevent the initiation of the donor’s immune system. Strikingly, these non-living constructs appeared to function perfectly fine up to eight weeks after implantation, not being stenotic and showing no signs of leaflet thickening. Furthermore, host recellularization was being observed as early as five hours after implantation, when cells started to migrate from the wall toward the leaflet tips. This observed amount of recellularization is unique compared to other valvular bio prosthesis, and is required to be able to provide future regeneration and growth. In addition, endothelial cell covering was shielding the entire heart valve construct which is essential to prevent thrombolytic events. Additionally, increasing amounts of tissue formation were observed after implantation, being a sign for remodelling potential. However, it was also observed that over time contractile cells were infiltrating while at the same time the heart valve became severely insufficient over time after four weeks of implantation. For this reason, adjustments to the valve design have been implemented to overcome this limitation as will be described below.
1.1.2 Cell sources
Despite being most important for the viable part of the engineered valve, the cell source represents the least controllable factor, in contrast to the controllable properties of the scaffold materials. Their regenerative capacity and quality to form tissue relies on the characteristics of the source and thus varies between individual donors and original organ tissue. A variety of cell types have been investigated for heart valve tissue engineering. For clinical application, the ideal cells should be non-immunogenic, functional, and easy to isolate and expand. The cell sources which have been mostly investigated are the non-immunogenic autologous endothelial cells (ECs), smooth muscle cells (SMCs) and (myo)fibroblasts, as they are normally found in the heart valve leaflets. The (myo)fibroblasts are well known for their suitability for tissue formation and specifically with regard to collagen formation, which provide mechanical stability of the valve. In order to obtain autologous cells, a piece of donor tissue is dissociated into individual cells, expanded in culture and finally attached to a scaffold and implanted. During the last years, also stem cells have attracted increasing attention and have become an important cell source for tissue engineering. Stem cells can be found in embryos, fetuses, and in adults and by definition they have the ability to reproduce themselves for a long period of time (self-renewal) and to give rise to different cells types (differentiation). Different stem cells exist, varying in their differentiation potential: the embryonic stem cells (ESCs) and the adult stem cells. The ESCs have the ability to give rise to all types of cells that form the three germ layers (mesoderm, endoderm, and ectoderm) from which all the cells of the body arise. ESCs are derived from the pre-implantation embryo, precisely from a group of pluripotent cells called inner cell mass (ICM). Once removed from the blastocyst, the cells of the ICM can be cultured under special conditions in vitro and are then called ESCs. Despite initial success, ethical issues, immunogenic and tumorigenic problems are the major drawbacks for ESCs. The adult stem cells have a more restricted (multipotent) differentiation potential compared to ESCs. They are found in adult organisms where they can give rise to specialized cells that will develop to the different tissues of the body. Adult stem cells serve for both homeostasis in healthy tissues and for regeneration of defective tissues. Most popular for cardiovascular tissue engineering are the bone marrow stromal cells (BMSCs) and adipose-derived stem cells (ADSCs). Recently, another interesting cell source for regenerative medicine, the induced pluripotent stem cell (iPSC), has become available. As part of this approach differentiated, adult cells have been genetically reprogrammed to an embryonic stem–like state with pluripotent differentiation capabilities. iPSCs are very similar in morphology, proliferation and gene expression to ESCs and represent a powerful tool in regenerative medicine. Unfortunately, iPSCs have been associated with the risk of teratoma formation. Therefore, more in depth research is required into the use of these cells for autologous cell based therapies such as cardiovascular tissue engineering. Moreover, it is still questionable whether an optimal approach would involve seeding with undifferentiated iPScs or rather using these cells to make a differentiated cell line of SMCs and/or ECs before seeding them onto scaffold matrices for cardiovascular tissue engineering. [Source: Dijkman and Generali et al.; Chapter Cardiovascular Tissue Engineering: Polymeric Starter Matrices for in Encyclopedia of Biomedical Polymers and Polymeric Biomaterials; 2015].
In the LifeValve project, with the focus on a paediatric tissue engineering concept, various cell sources were compared to define the most appropriate one for the production of tissue engineered heart valves (TEHVs). As a novel approach, the use of cell sources which can be obtained non-invasively was explored in comparison to the established vascular derived cells, including cells isolated from peripheral blood vessels, umbilical cord tissue, bone marrow and amniotic fluid. In accordance to the results in terms of proliferation capacity, expression profile, availability and the paediatric requirements, vascular and umbilical cord tissue derived cells have been turned out to be the most appropriate cell type.
As a novel and clinically interesting cell source in regenerative medicine, also adipose derived stem cells (ADSC) were included in the investigations. Fat, which is easy and repeatably accessible, is rich in pluripotent adipose tissue-derived stromal cells. In order to evaluate ADSC as an alternative cell source for the production of TEHV, adult ADSC were isolated from fat excisions from plastic surgery and characterized by immunohistochemistry, flow cytometry and differentiation assays. The cells exhibited characteristics of stem cells such as the differentiation capacity into the osteogenic, adipogenic, chondrogenic and endothelial-like linage. The stem cell (mesenchymal) specific combination of surface markers was detectable by immunohistochemical staining as well as by flow cytometry. To evaluate the cellular behaviour on 3D tissue formation, TEHVs based on PGA/P4HB were seeded with human adult ADSC and cultivated using strain bioreactors. The analysed ADSC-TEHVs showed a homogenous and vital cell distribution throughout the whole valve structure. Furthermore, a mechanically stable matrix with collagen production was demonstrated and the mechanical properties were comparable to native tissue. These preliminary results indicated that ADSC represent an interesting alternative autologous human cell source with clinical relevance due to their easy accessibility and excellent proliferation and tissue formation capacities. However, due to the existing experience and clinical safety of vascular cells, these cells were used for the frozen design TEHVs in the LifeValve project.
1.1.3 In vitro culture conditions
In order to evaluate the best in vitro conditioning protocol for the cultivation of TEHVs, two different bioreactor systems, the flow (full cardiac cycle) and the strain bioreactor (diastolic cardiac cycle), were investigated. The resulting TEHVs were histologically analysed concerning morphology, tissue formation and tissue organization. Additionally, cell phenotypes were characterized by immunohistochemistry, biochemical matrix proteins were determined and mechanical properties analysed by using a biaxial tensile tester. The results demonstrated the principal suitability of both bioreactor systems. The compositions of the extracellular matrix components as well as the biomechanical properties were rather comparable after cultivation in both systems.
1.1.4 Valve geometry
It is known that (engineered) tissues grow and remodel in response to the mechanical stimuli to which they are exposed. Therefore, to optimize the short-term remodelling process and prevent the development of leaflet retraction in tissue engineered heart valves (TEHVs), a computational model to evaluate the mechanical state of the valve and to investigate if this is similar to the mechanical state of native heart valves was developed in the LifeValve project. Native valves have a highly aligned collagen network, with most fibers running in the circumferential direction. Hence, when the valve is loaded during diastole, this results in small circumferential strains and large radial strains, which is beneficial for valve closure.
The analysis of the mechanical state in TEHVs with the initial valve design and material properties revealed that these valves have a tendency to become compressed in the radial direction during the diastolic phase, instead of being extended. This radial compression is detrimental for the functionality of the TEHVs, as it strongly contributes to the retraction of TEHVs. For this reason, computational modelling to investigate the potential influences of changes in the valve design (geometry and alignment of the collagen fibers) on the mechanical state of the valve was used. Based on these results, the design with the largest radial strain, as predicted by the computational model, was used to optimize the TEHVs in the in vivo studies.
Next to that, a tissue remodelling algorithm was added to the model to predict under which circumstances there is a risk for the development of leaflet retraction. The results indicated that leaflet retraction is not very likely to occur in TEHVs implanted in the aortic position. On the other hand, leaflet retraction in TEHVs implanted in the pulmonary position can only be prevented when the intrinsic contractility of the cells is very low.
By using the developed computational model, a failing mechanism for the dTEHVs was suggested. It appeared that the valves in the current geometry were subjected to leaflet tissue compression in radial direction under pulmonary pressure conditions. This in combination with the infiltration of host cells with contractile capacity, made it reasonable that these valves were prone to cell mediated leaflet shortening. In order to prevent tissue compression under hydrostatic loading, the valve geometry was adjusted by improving belly curvature and increasing the coaptation area. By comparing both designs, it appeared that this improved design is now subjected to leaflet tissue stretch in radial direction. Therefore the hypothesis was that by adjusting the final heart valve geometry, cell mediated leaflet retraction could be prevented by subjecting the heart valve to tissue stretch rather than compression. Unfortunately, controlling the tissue engineered heart valve geometry during culture has been limited so far since the cells deform the initial geometry by their contractile nature. As a solution to this problem, a rigid bioreactor insert has been developed that can impose the desired geometry during culture. In this way, the leaflets will compact around the rigid insert posts and will adopt the insert shape.
Two different insert iterations have been investigated. The first version was a solid structure with small holes to provide nutrient exchange to the adjacent tissue. However, it appeared that tissue formation became weaker in the vicinity of the coaptation area. In this specific region, the combination of dense scaffolding with the rigid insert was causing insufficient nutrient supply. To address the issue, large holes were created in the coaptation region of the bioreactor insert at these specific locations where constraining was not required. This resulted in enhanced tissue formation in those regions and solved the problem of insufficient nutrient exchange. To further improve the tissue formation in the belly region serum concentration in the medium was increased 10 fold. After four weeks of culture and subsequent decellularization, the inserts were removed from the valve constructs and resulted in functional, strong tissue constructs. In this way, any computational valve design can be translated into a compatible bioreactor insert to impose the desired geometry.
1.2 Evaluation of functionality of the tissue engineered valve/in vivo performance
After defining the first prototype (design 1.0) of the LifeValve of this project, and for the first time, we investigated the long-term in vivo functionality, host repopulation capacity, and matrix remodelling of the decellularized homologous tissue engineered heart valves (dTEHVs) as pulmonary valve replacement in sheep. Therefore, in a first series (published in JACC 2014) twelve dTEHV with the initial design were minimally invasively surgically implanted (via transapical access) into sheep and evaluated up to 24 weeks as pulmonary valve replacements. Further, 15 valves of the same design were implanted using the transjugular catheter-based access.
As a control, tissue-composition and -structure were analyzed in identical non implanted TEHVs (n=5). The transapical (n=12) and transjugular (n=15) implantation procedures were successful. No perioperative morbidity or mortality occurred and all valves could be delivered successfully at the target site. Early-postoperatively, two transapically implanted animals presented valve migration into the right-ventricular outflow-tract and died within 24 hours. The remaining animals (n=10) made a swift recovery and completed their respective follow-up without any complications. The appropriate position and functionality of the implanted valve was visualized by angiography. In vivo functionality (heart rate, mean and maximum transvalvular pressure gradient, and grade of insufficiency) was monitored using trans-thoracic echocardiography during the procedure, immediately after implantation and after 1, 4, 8, 16 and 24 weeks postoperatively.
In this study, initial valve functionality was excellent displaying sufficient leaflet motion and coaptation with only minor paravalvular leakage in some animals. Transapical and transcatheter jugular implantations yielded comparable results. While leaflet mobility was maintained on long-term follow-up, the coaptation slowly decreased over time, which was most likely due to radial compression during diastole and merging of the leaflets with the pulmonary artery wall at the level of the hinge area as well as the occurrence of a single leaflet prolapse in some animals. Consequently, mild to moderate central regurgitation could be observed at 16 weeks that further increased to moderate central regurgitation at 24 weeks. Functional measurements demonstrated stable mean and peak transvalvular pressure gradients over time (4.4±1.6 and 9.7±3.0 mm Hg respectively). After proving in vivo functionality, the animals were sacrificed and valves were explanted after 1 day, 8 weeks, 16 weeks, and 24 weeks. All explanted dTEHVs demonstrated shiny and pliable leaflets and dense whitish wall tissue, irrespective of the implantation period. In all explants, the wall tissue was integrated into the surrounding native pulmonary wall. Excellent coaptation of the explanted valves was evident in the implant and was maintained up to eight weeks of implantation. Thereafter, valve closure was incomplete in line with the observed central regurgitation. Apparently, the line of attachment of the leaflets to the wall shifted upwards in time indicating tissue merging process at the level of the hinge area, associated with a reduction in leaflet size with time.
The second series (design 2.0) of long-term animal experiments was started after design optimization of the LifeValve as described above (augmentation of the hyperbolic curvature (belly), prolongation of the coaptation zone, implementation of a bioreactor insert, gradual changes in nutritional medium concentrations and diastolic pressure augmentation). This second generation valve was implanted in 18 sheep by transjugular catheter technique. In the first six animals severe regurgitation after 8 to 24 weeks follow-up occurred. They met the endpoint criteria of regurgitation of more than 30% measured by magnetic resonance 2-dimensional quantitative flow analysis (flow MRI). Analyses of explanted valves revealed ruptured leaflets. Holes appeared mainly at the transition from curvature to coaptation zone near the attachment to the wall. It is suspected that this is the area of highest tension in the leaflets. As a result, the protocol of valve-manufacturing was stepwise adapted and the valves were subdivided in groups (“iteration steps”). For one group, the shape-giving insert was still added right from the beginning. However, a changed serum concentration was used when compared to the first six. In another group the insert was added after ten days of culturing in the bioreactor. For a last group of valves, the inserts were added after 13 days of culturing, giving the cells more time to generate extracellular matrix, thus creating a thicker and stronger tissue formation.
The valves, in which the insert was added later, showed very good coaptation after transjugular implantation in the animals. Coaptation area length was usually between 2 and 6 mm and stayed there throughout follow-up of up to one year. As a result of adding the inserts later (yielding stronger tissue formation), there were no signs of leaflet rupture during the crimping process and the course of implantation.
In the intracardiac echocardiographic assessment color flow mode, continuous wave Doppler and pulsed-wave Doppler were used in order to evaluate valve functionality.
By measuring forward flow, pressure gradients and flow accelerations were recorded. There were no signs of an elevated pressure gradient or an accelerated flux through the pulmonary valve, consequently showing no signs of stenosis.
Throughout the course of the study intracardiac echocardiography (ICE) assessment revealed slight regurgitation jets in some animals, none of which could be graded as more than mild, though. Color flow mode was used to detect regurgitation jets. Most of them were central, only few animals showed decentral insufficiency jets.
To grade and evaluate regurgitation more precisely, vena contracta (VC), proximal isovelocity surface area (PISA) and pressure half-time (PHT) was calculated from the recorded data. Vena contracta is defined as the point where the stream diameter of the regurgitation jet has its smallest width, yet fastest flux. PISA describes a three-dimensional hemispheric surface area close to the origin of a regurgitation jet, in which the flow converges. It gives a measure of the orifice area of non-coaptation. Depending on the width of this surface area combined with the vena contracta, insufficiencies can be assessed quantitatively (ml/s). To distinguish between moderate and severe regurgitation, pressure half-time can be useful according to the guidelines of the European Society of Cardiology for echocardiographic assessment of valvular regurgitation.
Considering all these measurements, most valves showed very good functionality, with a mild regurgitation if at all. Directly after the implantation procedure ten out of 15 animals showed no regurgitation at all. Flow was further assessed by cardiac magnetic resonance imaging (cMRI) which represents the gold standard for flow quantification. A regurgitation fraction was calculated with the obtained data, with one animal presenting with only 8% regurgitation after 52 weeks.
Endpoints: Animals were explanted either if the regurgitation fraction assessed by cMRI exceeded 30% or after 12 months (52 weeks). At the current time of follow-up, two animals have reached the landmark of 12 months. One of these animals showed only mild regurgitation in ICE and a regurgitation fraction of below 8% by flow MRI until the very last assessment before explantation. Up to date median values of the regurgitation fraction are approximately 10% in eight remaining animals at up to 48 weeks of follow-up.
In summary: One-year follow-up results of a competent, naturally remodelled heart valve, which has the tissue characteristics of a native valve, were already produced in two animals. Another eight animals are about to successfully fulfill the one-year follow-up, too.
The leaflets were as thin as in a native valve and free from thrombotic material. The shape of the leaflets preserved the hyperbolic curvature of manufacture. When loaded with hydraulic pressure (diastolic blood pressure) the leaflets closed competently.
The last three batches (in total ten valves) of these completely tissue-engineered heart valves deliver functional results which already fulfill the criteria of the ISO norm for valve certification: During opening the LiveValve shows no signs of stenosis (no antegrade flow gradient measured by intracardiac echocardiography). Its opening behavior is very similar to a native valve with the sole difference of a prolonged coaptation zone and slightly asynchronical opening in a few valves. The prolonged coaptation zone was introduced in design 2.0 to prevent leaflet retraction.
Valve closing is immediate and nearly complete. As in native pulmonary valves we see a tiny central regurgitation with a closing swash and a minimal leakage volume passing the valve during diastole. In ten animals this regurgitant volume adds to below 10% of the forward flow volume passing the valve during systolic ventricular contraction. The corresponding ISO norm demands less than 20% regurgitation for valve certification.
1.3 Remodelling
For all pre-clinical animal studies, the tissue-composition and -structure of the explanted dTEHV was analysed and compared to identical non implanted dTEHVs. A significant matrix remodelling was observed in the entire valve that corresponded with the rate of host cell repopulation. Prior to implantation, the dTEHVs revealed no cellular remnants and a well-developed extracellular matrix, mainly consisting of collagen, demonstrating an efficient decellularization procedure.
In the first pre-clinical animal studies with the first LifeValve prototype (design 1.0) it was observed that endogenous cellular repopulation occurred rapidly with first signs of cell infiltration as early as five hours after implantation of the dTEHV. Both leaflets and wall tissues were homogeneously repopulated over time with fastest repopulation at highest densities in the wall. After 24 weeks, cell repopulation density in the leaflets and hinge area approached that in the wall. Scaffold remnants remained longest visible in the leaflets with local increased cell densities. Minimal depositions, most likely blood platelets and fibrin, were present at the surface of the whole valve after five hours, but disappeared with time. The valvular tissues demonstrated abundant amounts of collagen that increased in density over time, in particular in the hinge area and wall and to a lesser extent in the leaflets. Elastic matrix formation was evident in the wall at eight weeks and later time points. In the hinge area the formation of elastic fibers was visible after 16 weeks and in the leaflets at 24 weeks. Calcification was not detected in any of the valves.
In order to analyse the repopulation of the dTEHV in more detail, cell phenotypes and distributions were analysed. Cells infiltrating the valve within five hours after implantation were all vimentin positive and alpha-SMA negative. After eight weeks, vimentin-positive cells were identified mainly in close vicinity of the polymeric scaffold remnants in the leaflet and hinge area, while more homogenously distributed and in higher amounts in the valvular wall. The cells in the leaflet and hinge area were all alpha-SMA negative and alpha-SMA positive in the valvular wall. After 16 weeks, homogenously distributed vimentin-positive cells were observed in all regions of the valve. These cells were alpha-SMA positive in the hinge area and wall and sparsely alpha-SMA positive in the leaflet. After 24 weeks, vimentin-positive cells were homogenously distributed over the valve. The level of alpha-SMA seemed lower in the hinge area and wall as compared to the 16 weeks explants. In the leaflets, more alpha-SMA positive cells were identified as compared to earlier time points.
As an endothelial layer on the valve leaflets is important to prevent blot clotting, the appearance of such a cell layer was investigated both by immune histological staining and SEM analyses. After eight weeks the dTEHV demonstrated partly confluent endothelial lining as observed by CD31 staining. Similar features were observed at the surface of the explants at 16 and 24 weeks. As visualized by SEM, the cell lining showed the typical cobblestone morphology, representative for endothelial cells, in all time points. The degree of endothelialization varied between locations and explants with decreasing endothelialization at the hinge area with implantation time and increasing endothelialization of the leaflet and valvular wall surface with implantation time.
Quantitative tissue analyses of the explanted dTEHV demonstrated that the DNA content (representative for the amount of cells) of the leaflets increased over the time in vivo. Although, after 16 weeks the DNA content was still lower than that in native leaflets, after 24 weeks the DNA content in the explanted dTEHVs was similar to that in native ovine valve leaflets. The amount of sulphated glycosaminoglycans (GAGs, that are important to keep the leaflets hydrated) was lower in freshly engineered dTEHVs as compared to that in native leaflets but approached native values after 24 weeks. The collagen content in the leaflets of the dTEHVs (being very important for the strength of the tissue) was equal to that in native ovine leaflets already before implantation and slightly increased after 16 and 24 weeks in-vivo.
1.4 Re-seeding
Host cell repopulation of the decellularized valve implant is the prerequisite for remodelling processes in vivo. In order to fasten and improve that process, the loading of the acellular valves with the patient’s own cells prior to implantation could be the method of choice. Especially, because endogenous repopulation is known to be enhanced by pre-seeding the graft prior to implantation with bone marrow mononuclear cells (BMMNCs), known for their paracrine effects. We hypothesized that a successful re-seeding method may facilitate enhanced remodelling of the TEHVs in vivo. Therefore, different tissue pre-treatment and seeding techniques were tested and analysed to find the optimal protocol for re-seeding of decellularized tissue engineered matrix. The method should guarantee the resistance of the re-seeded cells against the necessary crimping procedure and the flow/pressure stresses which the valves will experience directly after implantation. Finally, the most promising re-seeding procedure for decellularized tissue engineered cardiovascular constructs with BMMNCs prior to implantation was selected. This could potentially trigger a better host cells recellularization in vivo after implantation. Moreover, the process of reseeding was optimized towards an operator independent method that guarantees a standardized re-seeding quality. Due to time constraints within the project, no in vivo results of the performance of the re-cellularized TEHV could be generated.
1.5 In vivo tracking
In order to get insights about the behaviour, distribution and fate of the cells re-seeded on dTEHV, a PET (position emission tomography) reporter gene was used to track the cells in vivo. The PET-reporter gene (HSV1-tk) specifically binds a tracer called [18F]-FHBG that has suitable pharmacokinetics and dosimetry for clinical applications and imaging. The transfection method of mesenchymal stem cells isolated out of punctuated bone marrow was established.
The valves were implanted percutaneously into sheep in the pulmonary position under general anaesthesia (n=9), an additional sheep (n=1) did not undergo valve implantation to serve as control. Ten mCi [18F]-FHBG of PET tracer was produced for each procedure and serial PET-CT imaging of the sheep was performed 3h, 6h, 24h, 1 weeks, 3 weeks, 6 months and 7 months after the valve implantation. For the quantitative assessment of the number of cells survived in the 3D scaffold after in vivo implantation, vials containing 5x10*4, 2x10*5 and 4x10*5 transfected cells were mixed with the PET tracer for 1h, then the non-bound tracer was washed out and the vials were PET-CT imaged as in vitro control. To control the success of the transfection of the cells, the native (non-implanted) valves were investigated with PET-CT in vitro. In vitro PET-CT images of the valves showed the accumulation of the seeded cells mostly at the base of the leaflets. PET-CT images of the sheep 3h after implantation of the “living” valve showed a clear signal of the valves with the seeded transfected cells, with a mean estimated number of survived cells of 1.2x10*6. Extracardiac hot spots were seen in the mediastinal lymph nodes. No meaningful decrease of the cells living in the implanted scaffold occurred at 6h or 24h. Three weeks after valve implantation, living MSCs could be found in the valve location (estimated cell number 0.6x10*6) in one sheep, with some cardiac hot spots in the left ventricle. Immunofluorescense histology showed alpha-SMA positivity on the explanted valve surface one month after implantation. The explanted valve of one sheep after seven month FUP was investigated in vitro via PET-CT, which showed clear PET signal mostly at the valve base. Histology of the valves will be carried out. The remaining animals are still under investigation.
2. Stent systems
Endovascular stents were introduced in the early 1990s to restore the patency of occluded vessels for example of coronary, iliac or carotic arteries. In recent years, the stent concept has been extended to other indications and is now also being used as an anchor for artificial heart valves.
There are two general stent concepts: self-expandable and balloon-expandable stents. Self-expandable stents are made of nitinol. This is an alloy of nickel and titanium and has the two characteristics of superelasticity and thermal shape memory. Accordingly those stents are resistant to compression at body temperature. The stent appears compact in the compressed state, i.e. when integrated into the application system. At body temperature, the stent expands to its predetermined (unconstrained) diameter, thus restoring the patency of the lumen.
The nitinol stents are cut from a single nitinol tube and accordingly have no filament welding points or soldered joints. The stents feature a fine-meshed and flexible design. Therefore, they are ideal for insertion in an area of flexion and torsion. Most of the nitinol stents being implanted today have a mesh design.
Balloon-expandable stents can be made of various materials such as stainless steel, cobalt-chrome, platinum or out of various biodegradable materials, just to name a few. As all these types of material are not self-expandable, a balloon catheter is needed to deliver and dilatate them.
Minimally invasive implantation of heart valve replacement requires the use of stents for valve delivery and anchoring. However, the prosthetic valve becomes an integrated part of the heart within a few months and the stent itself is no longer needed. Furthermore, pediatric patients that receive a stented heart valve have to undergo multiple interventions as they outgrow the implant. Creating a minimally invasive implantable heart valve that can enable growth of the patient without the need of additional re-interventions requires not only a growing valve but also a growing stent. For this reason, the development of a biodegradable stent that vanishes after it has fulfilled its function was of paramount importance for the LifeValve project.
The stent has to meet very demanding conditions in order to enable heart valve implantation in pediatric patients. Some of them are pertinent to the implantation techniques and others are inherent to the material or the functionality of the implant. To begin with, the initial diameter of the stent should enable to host the heart valve, taking into account that the size of the valve depends on the size of the artery where it will be implanted which depends on the patient's body height and weight. Once the valve has been sutured to the stent, reducing the diameter of the prosthesis until sizes that enable minimally invasive insertion must be possible. For a transapical approach, the crimped diameter is about 10 mm. For a transcatheter insertion, the conditions are more challenging, requiring a crimping diameter of 8 or even 6 mm. These conditions impose an additional difficulty: the lower the crimping diameters, the higher the risk of permanent deformation. This means that after being pushed out of the catheter and self-expanded, the final diameter of the stent will be lower than the initial (plastic deformation). In other words, if this occurs the implanted valve could be smaller than the host tissue and the implant might migrate. To compensate for plastic deformation, the stent must be oversized so that after self-expansion, the resulting diameter is appropriate. The amount of oversizing was investigated within the project. To avoid migration, the stent must exert sufficient force against the artery to keep the implant in place (radial force). Little is known about the minimum force requirement; this topic was investigated as well during LifeValve. Since we aimed to create a stent that fully degrades and only the valve would remain, the stent material and its degradation products must be biocompatible. Metallic materials have excellent mechanical properties and provide the chance of very low or even no plastic deformation at all during crimping. On the other hand, most of them are not biodegradable and those that are, degrade fast. Polymers have the advantage of a good degradative potential but the downside of poor mechanical properties. For a proper functionality of the implant and due to the cyclic pulsatile loading of the implantation environment, the stent must be fatigue resistant for the first period of implantation. Finally, once the material has been selected and the conditions established, an appropriate manufacturing process has to be chosen. Such technique should not alter the mechanical properties of the material and should be reliable, accurate and reproducible.
In order to take into account all the previously mentioned aspects towards a biodegradable stent design that meets all the requirements, the following approach was followed within the LifeValve project:
- A non-biodegradable (nitinol) stent was taken as a reference to study and understand the implantation mechanisms and requirements.
- Computational simulations were created to resemble the nitinol stent geometry and material properties.
- Experimental data from ovine and pediatric arteries were obtained and implemented in the computational models to simulate a realistic host tissue.
- Forces during stent deployment and equilibrium with the artery were predicted computationally to set the requirements for the biodegradable stent.
- Different polymer alternatives and manufacturing procedures were evaluated.
- Based on the selected material, additional simulations were created to obtain a polymer stent geometry that adjusts to the requirements stated above.
- Polymer stent prototypes were manufactured and their forces were tested experimentally and compared with the computational results for validation of the methodology.
Experimental and computational methods were combined to study the forces and mechanisms during stent crimping of biodegradable and non-biodegradable materials. The nitinol system, used within LifeValve for ovine implantation of larger diameter valved stents, served as a reference to set an appropriate force requirement for the polymer stent for pediatric application. Pediatric and ovine tissue obtained after explantation helped to develop an appropriate host artery for the simulations and establish the equilibrium force and diameter after self-expansion. This scenario was crucial for the development of the polymer prototype.
Different biodegradable polymeric materials were evaluated for material selection. The different candidates were characterized for mechanical properties. The material that was closest to the set requirements was chosen as the material of choice for further testing and processing. The chosen material was a combination of two well-known biodegradable polymers in a 65-35 blend ratio.
After validation of the nitinol models and characterization of the biodegradable material of choice for computational implementation, an extensive study was performed assessing the influence of design parameters (such as the thickness, width and length of the struts) in the force of the stent. This method allowed us to optimize the geometry and obtain a theoretical design that meets the pursued force requirements, enabling crimpability in accordance with a transcatheter implantation approach and achieving a self-expanding diameter in accordance with the size of a juvenile ovine artery.
Various fabrication processes were tested during the Lifevalve project. These include melt extrusion, rapid prototyping, dip coating and laser cutting. The first demo prototypes were made in a two-step fabrication process. First, a tube of desired thickness from the chosen material is obtained by the process of dip coating. The process of dip coating was optimized for different diameters and thicknesses. After this, the tube is processed into a stent by laser cutting. A femto-second laser was used to avoid local heating effects on the stent. After laser cutting the dip coated tubes, manual crimping was evaluated and experimental radial force measurements were carried out to verify that the computational work provided accurate predictions. However, unexpected buckling of the stents was evidenced suggesting that the dip coating technique was not the optimal method for tube manufacturing. The resulting tubes were not homogeneous in thickness and appropriate crimping of the stents was not possible. Furthermore, validation of the computational models could not be achieved. This technique was further improved during the final stage of LifeValve. Nevertheless, these results could not be further materialized into a stent prototype before the end of the project.
To provide a reasonable alternative to validate the polymer computational models, a commercially available degradable filament was characterized and introduced in simulations until a suitable geometry was obtained and a 3D printed stent prototype was produced and tested experimentally. Buckling did not occur and experimental results were in good agreement with the computational predictions, validating the models. In addition, degradation of the material by hydrolysis and loss of mechanical properties was experimentally shown. Unfortunately, the composition of this material is not suitable for implantation due to the lack of biocompatibility.
In conclusion, the combination of computational and experimental results for the design and manufacture of a polymer stent with the pertinent mechanical properties to provide a stent alternative for heart valve replacement has been evaluated and successfully carried out. However, the simultaneous combination of a biocompatible material and a reliable manufacturing technique has not yet been achieved. To create functional stents based on laser cutting, it is of paramount importance to obtain tubes of good quality. If other rapid prototyping techniques such as 3D printing are to be considered, the selected material has to be supplied in a way that it is possible to be processed. Furthermore, the lack of precision of this technique should be accepted or accounted for, to have predictable computational tools. The most promising approach seems to be to focus future efforts on improving the tube production for laser cutting.
In an alternative approach allowing growth of the implanted valve, a study using oversized nitinol stents was performed. The hypothesis was that an oversized self-expanding nitinol stent will slowly expand over time and permit growth of the surrounding vessel. The radial force of the utilised nitinol stents could perhaps be used as driving force for native pulmonary vessel growth after valve-stent implantation: once implanted these metallic stents continuously tend to reach their maximum diameter. So they might support the native vessel to expand and grow with the valve-stent inside. Little was known about the radial force needed to keep the valve in place on one side and the effects of extensive radial pressure on the vessel structures on the other side. Twenty plus two (substitutes) animals were included into the oversized study: juvenile animals with a native pulmonary artery diameter between 16 and 18 mm were used to implant either (A) nitinol stents (length 30 mm) with a diameter of 20 mm which means ≈ 18 % oversized, (B) with a diameter of 25 mm which is oversizing of ≈ 47 % and (C) with a maximum diameter of 30 mm (≈ 76 %) oversized.
All stents were provided with a 20 mm custom made pericardial valve to avoid instantaneous free regurgitation into the right ventricle. Results indicated that using a self-expanding nitinol stent of the given design and radial force was not suitable as a growth permitting stent. It opened too fast and stretched the right ventricular outflow tract and the native pulmonary vessel.
3. Minimally invasive implantation technology (non-surgical procedure) of the LifeValve
Surgical heart valve implantations require the use of cardio-pulmonary bypass and open chest access. A hospital sojourn of several weeks including cardiopulmonary rehabilitation follows. Besides the undesired scar formation, an operation contains the risks of major complications like bleeding, infections, embolic complications, stroke and death.
As an alternative, minimally invasive procedures have been developed over the last decade. They erase the need for cardio-pulmonary bypass and open chest access. Up to date, heart valve replacements for minimally invasive implantation for the most part consist of a balloon expandable metallic stent covering a xenogenic valve. Use of balloon expansion requires over-pacing or short-term cardiac arrest (by use of adenosine e.g.) to reduce ventricular output.
In the LifeValve project a minimally invasive transcatheter implantation method was developed and applied in sheep. A heart valve was implanted through the jugular vein: an 11F (diameter of 3.7 mm) sheath was placed in the right jugular vein for the intracardiac echocardiography (ICE) probe. ICE was performed before and directly after implantation using an 8 or 10F ultrasound catheter and assessed 2D, color, pulsed wave and continuous wave Doppler to obtain dimensions and function of the native pulmonary valve before implantation. Fluoroscopy was used to perform angiographies and guide implantations. Pulmonary artery, right ventricular and right atrial angiographies were conducted and respective pressures were monitored and recorded. The access point at the right jugular vein was then dilated using commercially available dilators and a purse string suture created and tightened. A pigtail catheter bearing a pre-shaped ultra-stiff guide wire was advanced through the right heart and the pulmonary valve into either pulmonary artery.
In the catheter lab, the dTEHV was cut open and crimped to either 10 mm or – if feasible – to 8 mm. Crimping was done slowly over 20-30 min allowing liquids to be squeezed out of the leaflets without harming the tissue. In between the crimping device was reopened to investigate the leaflets for cracks or other damage. Finally, the dTEHV was loaded into the 10 mm or 8 mm capsule of the self-designed and tailor-made delivery system. The loaded system was then sheathlessly introduced over the ultra-stiff guide wire through the purse string suture into the jugular vein and forwarded into the right atrium. When using a 10 mm capsule in a smaller animal of below 40 kg it was sometimes a challenge to pass the tricuspid valve into the right ventricle and especially to take the bend from the right ventricle into the right ventricular outflow tract. In these small animals we experienced that the stiff length of the delivery system in relation to the diameter of the right ventricle is of utmost importance for the success of the procedure.
A delivery system, developed in the LifeValve project, was used for 55 successful implantations. These included tissue engineered (n=31) and pericardial valves (n=13) sewn into nitinol stents, an Andra-XXL balloon-expandable stent, one nitinol stent and several polymer stents (n=9). However, 78 attempts were necessary resulting in a success rate of 71%. Some of the failed implantation attempts can be attributed to the early stage of the development phase of the delivery system and have been solved in the course of the project (n=8). In two cases, the diameter of the delivery system was too large to be accommodated by the access vessel. The remaining problems were related to the length of the stent, the crimpability of the stented valve and the resulting catheter mechanics.
The delivery system is suited to be used for a wide range of implants. In its final version, outer diameter ranges from 13F (4.3 mm) to 32F (10.6 mm). Self-expanding as well as balloon-expandable stents, both with or without heart valves, can be implanted safely. Because of its great flexibility, almost every vessel can be accessed with the diameter of the delivery system and the stiff length of its tip in relation to the heart size being the limiting factor. For example, it could also be used for implantation of aortic valve prostheses.
4. Clinical pilot study
4.1 Identification of patients
For a later clinical study the database of patients of the department of congenital heart disease at the Deutsches Herzzentrum Berlin (German Heart Institute Berlin) was screened.
Inclusion criteria comprised:
• Adult patients (age >18 years)
• Primary diagnosis of tetralogy of Fallot (subpulmonary stenosis, ventricular septal defect, right ventricular hypertrophy, cyanosis)
• Surgical procedure: transannular patch reconstruction of the right ventricular outflow tract and closure of ventricular septal defect
• Magnetic resonance imaging (MRI) within the last three years
• Right ventricular dilation with an intraventricular diastolic volume of >140 ml/m2
Twelve patients (three female), with a median age of 25.8 years (range 17.0 to 31.8) were identified. They had MRI in median 2.1 years ago (range 0.6 to 3.2) with a median right ventricular volume of 154.5 ml/m2 body surface area (range 140 to 228 ml/m2). The normal mean value for the indexed enddiastolic right ventricular volume does not exceed 85 ml/m2 (standard deviation 60 to 100 ml/m2 maximum).
These patients are candidates for pulmonary valve implantation, e.g. LifeValve. We would appreciate to invite these patients to a first-in-human trial. However, the LifeValve pulmonary valve replacement has to prove well in the safety study and in further animal trials over a course of two years in accordance with regulary requirements.
The notified body for the clinical trial was identified and contacted. Necessary paperwork for ethical approval was prepared. Nevertheless the ethics application could not be handed in at the current state of the project as - according to the local authorities (LAGeSo) - two year animal studies regarding the special regulations are mandatory. Thus we applied for further funding to perform the necessary studies and to achieve the goal of a first-in-human study.
4.2 Translation of sheep protocols towards human tissue culture protocols.
For the translation of sheep to human tissue engineered heart valves, the cell source needs to be adjusted to make the dTEHVs compatible for human application. For the cell source, primary isolated human vena saphenous cells were expanded and used for culture. Human cells are known to exert less tension on the tissue, proliferate slower and require modifications in the culture medium. Based on the developed computational designs, the optimal valve geometry was predicted based on previously obtained tissue mechanical properties. Bioreactor inserts were produced accordingly and after 4 weeks of culture, human cell based tissue engineered heart valves were created and decellularized successfully. These valves have shown to maintain the imposed shape of the insert, including enlarged coaptation area and profound belly curvature. These valves were subjected to in vitro flow experiments, where they showed superior open and closing behavior, with leakage volumes even below 1%. Furthermore, these decellular constructs remained fatigue resistant up to 12 million cycles were tissue distribution was homogeneous throughout the thickness. Collagen anisotropy was observed near the coaptation region in circumferential orientation and became more isotopic towards the belly region. The tissue parameters of the eventually obtained design and mechanical properties were implemented in to the computational design, which revealed that these valves have a reduced amount of radial tissue compression compared to previously human tissue engineered heart valves. To conclude, human cell based dTEHVs can be successfully produced, which remain functional up to 12 million cycles. The geometrical improvements have shown that these valves are less prone to in vivo cell mediated leaflet retraction.
Potential Impact:
1. Benefits for patients
Valvular heart disease (VHD) represents a major global disease load with an increasing number of patients. For this reason VHD is a significant cause of morbidity and mortality worldwide. More than 250,000 heart valve replacements are performed worldwide per annum with a continuously increasing tendency that is expected to have tripled by the year 2050. Thus, the worldwide demand for heart valves is constantly rising due to age-related degeneration, improved survival of children with congenital heart defects and diseases like rheumatic fever. Surgical heart valve replacement using mechanical or bioprosthetic prosthesis represents the most common therapy for end-stage valvular disease and is a safe and efficient approach. Although currently available prostheses demonstrate excellent structural durability several limitations remain unsolved including the lack of growth capacity as well as in vivo repair and remodelling properties. Mechanical valves are known to have an increased risk of thromboembolic events. A lifelong anticoagulation therapy is required for these patients carrying a risk of spontaneous bleeding and/or embolism. Bioprotheses, either originating from animals (xenografts) or from human donors (homografts), do not require lifelong anticoagulation but are prone to dysfunctional, degenerative processes requiring high-risk redo operations. Therefore, these prostheses are less suitable for middle-aged and younger patients.
The development of a tissue engineered heart valve in the LifeValve project is a new approach to treat heart valve disease and a step towards a lifelong valve.
Tissue engineering is a still experimental approach to restore or replace damaged tissues in the human body. Instead of replacing organs or organ parts by foreign material (metals, polymers, foreign biological materials) it is attempted to make the organ parts from the patient´s own material. The “classical” heart valve tissue engineering concept comprising complex multistep procedures such as cell harvest, cell expansion, seeding on scaffolds, bioreactor in vitro culture, and time-critical implantation coordination of the delicate, living engineered autologous heart valves require high logistical and financial efforts. In this project we therefore investigated the concept of decellularized homologous tissue engineered heart valve, which is an innovative approach in order to create off-the-shelf availability of valve replacements without requiring a human or animal graft as starter material. The feasibility and long-term functionality of such homologous off-the-shelf tissue engineered heart valves was evaluated by implanting them via minimally-invasive transcatheter-based techniques in a relevant pre-clinical model (sheep). The LifeValve project has successfully demonstrated a well-functional tissue engineered valve for up to over one year. Although further research is necessary to refine the valve in order to improve the long-term stability, their remarkable rapid cellular repopulation and tissue remodelling, representing the self-repair capacity of these off-the-shelf valves demonstrates their potential as interesting alternative to current prostheses. The elimination of invasive cell harvesting from the patient, time consuming cell expansion and timeliness production of the patient specific graft will eliminate the waiting time for the patient and therewith increase the wellbeing of the patient and tremendously lower health costs.
Additionally, the decellularization methodology enables off-the-shelf and unlimited availability of the product in a large variety of sizes that will substantially contribute to the impact and success of the innovation and therefore may substantially simplify previous tissue engineering concepts towards clinical translation. The envisioned growth capacity of these valve replacements will eliminate the need for multiple reoperations, in particular relevant for pediatric patients, and therewith change the life of patients suffering from congenital cardiovascular disorders.
2. Socio-economic impact
The socio-economic costs of the European Healthcare System will be substantially reduced by the LifeValve technology, providing a growing heart valve replacement with repair and regeneration capacity and without the need for repeated reoperations and hospitalizations. Furthermore, faster recovery yielding shorter hospitalization will allow for faster turn-over in patient treatment. This will diminish treatment costs. Initially high costs for manufacturing will decrease over time as multiple companies will start to compete in tissue engineered valve production. Finally, pulmonary heart valve replacement will be a procedure of a three-day hospital sojourn, just like coronary artery stenting today.
Once tissue engineered valve replacement is established on the pulmonary side, this method will take over on the aortic side. This is when huge market numbers come into play. The catheter and medical device market is analysed to be the fastest growing sector in the medical industry within the next decade. Aortic valve replacement is estimated to hit the two billion US dollar benchmark in sales development within the coming five years.
The tissue engineered lifelong valve developed within the LifeValve consortium has the potential to foster the attractiveness of the European medical device market. Commercialisation is intended for both the delivery system and the dTEHV. The delivery system is designed and manufactured using materials already in use for comparable clinical products. This facilitates certification as a medical product. A patent was claimed and the patent application is published. The delivery system can be certified as a medical product on its own but would be preferably certified in combination with the LifeValve or its descendant (CE mark). Spin-off companies might attract venture capital leading to job creation in the field of medical and life-science technologies and the tissue engineering market.
3. Exploitable results
3.1 Decellularization technology
The decellularization methodology enables off-the-shelf and unlimited availability of the product. As this alternative approach for tissue engineering will also be applicable in other therapies for (congenital) cardiovascular diseases, it will not only extensively contribute to the knowledge in the field of heart valve disease, but more generally represent a tremendous step for clinical translation of cardiovascular tissue engineering worldwide.
3.2 Valve design and production
It is well known that (engineered) tissues grow and remodel in response to the mechanical stimuli to which they are exposed. Therefore, to optimize the short-term remodelling process and prevent the development of leaflet retraction in tissue engineered heart valves (TEHVs), we have developed a computational model to predict the tissue remodelling process in tissue engineered heart valves exposed to different conditions, i.e. different pressure differences over the valve to mimic the different hemodynamic conditions in valves implanted in the pulmonary or the aortic position, and different intrinsic contractilities of the cells inside the valves to mimic the different phenotypes. The results indicated that leaflet retraction is not very likely to occur in TEHVs implanted in the aortic position. On the other hand, leaflet retraction in TEHVs implanted in the pulmonary position can only be prevented when the intrinsic contractility of the cells is very low.
The computational model was used to redesign the LifeValve in order to improve the in vivo functionality. Controlling the geometry of tissue engineered heart valves during culture has been a major bottleneck for many years. With the development of constraining bioreactor inserts, heart valve geometry can be completely controlled and imposed. This resulted in the successful development of functional tissue engineered heart valves. The invention is not limited to pulmonary heart valve application, but can also be translated to aortic heart valve application. These bioreactors are therefore the key to produce functional tissue engineered heart valve, which otherwise would not be possible. A patent was filed about controlling tissue engineered heart valve geometry by using predefined inserts during culture.
All the above mentioned methods allow generating tissue engineered valves in a personalized design. Treatments nowadays address a broad market. Various products are mass produced, existing in different sizes but only available in limited varieties (e.g. Medtronic Melody® valve is offered in a narrow range of sizes). The tissue engineered valve is producible according to the patient’s needs from medical imaging data: according to size and form, fitting exactly into the patient. Moreover it can answer the need of a large market by being produced in an automated process.
3.3 Transcatheter Valves
The therapy for valvular heart disease is currently undergoing rapid changes. Besides conventional surgical valve replacement being the standard of care for several decades now, transcatheter techniques have been introduced for high-risk patients. Current results display promising outcomes and mid-term results. Transcatheter techniques can be performed in off-pump beating-heart fashion that can also be performed without general anesthesia in selected patients. As it is long-term safety proven via clinical, randomized trials, it can be assumed that the indications for this minimally invasive procedure will be continuously extended towards younger and less riskful candidates. Two specific routes for implantation have been established since then: the antegrade way via a surgical, direct transapical access, and the retrograde way using a transfemoral catheter-based concept. The transapical approach requires a left anterolateral mini-thoracotomy followed by a pledged purse-string suture to enter the apex, whereas the transfemoral method requires an adequate peripheral vascular access and can be performed fully percutaneously.
In the course of the LifeValve project, the minimally invasive transcatheter approach was followed in the animal experiments. For implantation by heart catheterization a proprietary delivery system was designed and manufactured.
Translating this approach to human patients would mean that the catheter carries the stented heart valve to its destination. It is possible to perform this operation under conscious sedation without the risks associated with general anesthesia. The physician accesses a vessel with a sheath and advances a deployment system into the beating heart. There is no major incision; by puncture of a vessel existing pathways leading straight to the heart are utilized. The catheter based intervention is a very gentle approach in comparison to open heart surgery. Usually the patient receives a compression bandage for 24 hours and can be released from hospital only a few days after the replacement valve has been implanted. Health care costs are extremely reduced using that approach. The delivery system is expected to work well in humans, especially as the crimping ratio is generally lower – i.e. less crimping is needed - compared to our sheep. Thus, less of the remaining problems should occur.
In this way, the LifeValve delivery system could be used to facilitate difficult implantations of commercially available prostheses such as the Medtronic Melody valve in which the Medtronic delivery system fails. This can be the case in Melody implantations which have to be conducted from the jugular access due to femoral thrombosis.
Currently, the innermost catheter is a 4F multipurpose catheter. For use with balloon expandable valved stents, this catheter must simply be exchanged for a balloon catheter. Corresponding technical drawings and calculations already exist to alter the existing delivery system.
The latest version of the delivery system has been designed toward ergonomics, safety and one-user operability. In addition, most parts can now be produced by injection molding. This is a major milestone toward serial production. For commercialization and certification, only minor changes remain necessary. The flexibility and the modular design of the delivery system enable versatile applications making it an ideal standardized platform for several applications.
3.4 Stents
The transcatheter approach neccessitates the usage of a stent system to deliver the valve, to fix the valve in place and to promote fusion of the engineered tissue with the native host tissue. Once this fusion is established, the stent not only loses its mechanical function, it also prevents the valve to grow, which is essential for the envisaged lifelong approach of the tissue engineered heart valve for pediatric patients. The approach used in LifeValve was the development of a biodegradable stent. Although the goal to develop a biodegradable stent could not be fully realized within the lifetime of the project, important cornerstones were laid, such as the design, the material and the production method.
There are numerous ways a suitable stent can be designed. However, designs have to fulfil certain criteria in order to be applicable. With the computational model developed within the project different stent designs can be investigated for their crimpability, self-expandability and the resulting reaction forces. These computational iterations can provide easy and fast evaluation and predict if a specific design is suitable for the application. Also, suggestions for further improvement can be provided based on this computational model. In this way, less labour intensive laboratory experiments are required before an appropriate stent design can be finalized. This will lead to significant R&D cost reductions (“time to market”).
4. Dissemination
The LifeValve dissemination aimed to promote knowledge sharing among the scientific community and to increase awareness of the project results on the part of the public. Various instruments were used to reach that goal.
The project website (http://www.remedi.uzh.ch/lifevalve) has been installed at the University of Zurich and was regularly updated. On the publicly accessible sites, the project homepage gives detailed information on the project. This included not only the scientific background and objectives of LifeValve, but also explanations on why we focus on children because of the unmet medical need of a living heart valve with the capacity of growth and which can be implanted minimally invasively. In order to highlight the collaborative nature of the project, the homepage also contains information on the work division between partners so that an interested audience could follow the plan of work within the project. In addition, there is a section giving detail information on all beneficiaries including a short introduction of their core research expertise, selected publications and contact information. The press section of the website contains links to several documents on LifeValve. For example, LifeValve has been chosen to be portraited as a StarProject by the EC Research & Innovation and was featured in a video by Euronews for Futuris, the European research programme. Furthermore, LifeValve has also been selected by the Communication and Support Action COMED as one cutting-edge health-research project funded within the Seventh Framework Programme. COMED aims communicating results and activities of EU-funded health research projects and improving the perception of sciences by the European public. The extensive science documentary about LifeValve with Berlin and Zurich as shooting locations used animations, background statistics, interviews and case studies to illustrate the objective and research approach of LifeValve. By linking the Project website to both the websites of StarProjects by the EC Research & Innovation and COMED, we aim to guide interested EU citizens to other cutting-edge projects within the EU-FP7.
The partners also addressed the LifeValve project on their institutional homepages.
The LifeValve project was presented at various international conferences to the scientific community. One of them, for example, was the annual meeting “Cardiovascular Cell Therapy and Innovative Strategies” in Madrid, Spain (29.5.2015) where the abstract entitled “Transcatheter delivery of homologous cell-derived off-the-shelf tissue engineered heart valves with self-repair capacity in a translational animal model” was appointed as the best one in the category translational/clinical studies.
Six peer reviewed publications arose from the project and more will come up in the near future. The first series of in vivo studies was published in the Journal of the American College of Cardiology (impact factor >15 in 2013). The study was highlighted by an editorial comment in the same issue of the journal.
Apart from scientific publications and dissemination activities directed at a general audience the consortium members also participated in events with key stakeholders in healthcare as the target audience. For example, the LifeValve project has been presented at the conference “Innovation in Healthcare. From Research to Market. SMEs in Focus” (organized jointly by the Research DG and Enterprise & Industry DG, Brussels, May 2010) and at the EU-China Science and Technology Week, organized as part of the European Union Pavilion at the World Expo 2010 Shanghai China (June 2010).
In 2012 e.g. the coordinator of LifeValve, Prof. Simon P. Hoerstrup, participated in the meeting “Industrial Technologies” in Aarhus, Denmark, where over 160 high profile international speakers from industry, government and research met to discuss visions for European research and industry in 2020, how Europe can succeed in the face of global competition, and the form and impact of Horizon 2020.
In the presentation “From cells to functional cardiovascular implants” he also pointed out those topics of importance that should be addressed in the upcoming Horizon2020 program in the field of regenerative medicine.
LifeValve related topics were also covered in the public press (e.g. in Switzerland). In the Horizon2020 brochure "Investing in success" (European Commission, Directorate-General for research and Innovation, 2012), LifeValve was highlighted as a story of success.
5.Outlook
For the first time ever, the feasibility and long-term functionality of transcatheter homologous off-the-shelf tissue engineered heart valves were demonstrated in a preclinical model. The concept of homologous off-the-shelf tissue engineered heart-valves will substantially simplify tissue engineering concepts towards clinical translation.
The combination of heart valve tissue engineering and minimally invasive implantation technology by using self-expanding and biodegradable stents is a novel approach and provides Europe with a unique opportunity to reinforce its competitiveness as to novel medical and life-science technologies.
This successful and highly innovative research shall be continued. The next step is preparation of a first-in-human trial. The ultimate goal is to bring the worldwide first tissue engineered heart valve on the market. This valve has the potential to solve the worldwide shortage of human heart valves as it can be produced on large scale - independent from human or animal valve donors. Therefore with members from the LifeValve project a new consortium was formed which has submitted a proposal in the frame of Horizon 2020.
List of Websites:
http://www.remedi.uzh.ch/lifevalve.html
Universität Zürich (Coordinator) (CH)
Prof. Dr. Dr. Simon P. Hoerstrup
Cardiovascular Surgery Research
Regenerative Medicine Program
Moussonstrasse 1
8091 Zurich, Switzerland
www.remedi.uzh.ch
Deutsches Herzzentrum Berlin (DE)
Prof. Dr. Felix Berger
Department of Congenital Heart Disease / Pediatric Cardiology
Augustenburger Platz 1
13353 Berlin, Germany
http://www.dhzb.de/de/kliniken/angeborene_herzfehler_und_kinderkardiologie/
Technische Universiteit Eindhoven (NL)
Prof. Dr. Frank Baaijens
Department for Biomedical Engineering
Den Dolech 2
5612 AZ Eindhoven, The Netherlands
http://www.tue.nl/en/university/departments/biomedical-engineering/
Medizinische Universität Wien (AT)
Prof. Dr. Mariann Gyöngyösi
Univ. Klinik für Innere Medizin II
Abteilung für Kardiologie
Waehringer Guertel 18-20, 1090 Wien, Austria
Pfm Produkte für die Medizin AG Köln (DE)
Dr. Frank Thiel
Wankelstr. 60, 50996 Cologne, Germany
http://www.pfmmedical.com/en/index.html
Xeltis AG (CH)
Laurent Grandidier
Muehlebachstrasse 26
8008 Zurich, Switzerland
http://www.xeltis.com/
Xeltis BV (NL)
Dr. Martijn Cox
De Lismortel R2 06 Cat Build 31
5612 AR, Eindhoven, The Netherlands
http://www.xeltis.com/
University of Debrecen (HU)
Department of Nuclear Medicine
Dr. Laszlo Balkay
Egyetem Ter 1, 4032 Debrecen, Hungary
www.nmc.dote.hu
