Evaluating the Risk of Calcification in Material-Driven in situ Tissue-Engineered Heart Valves: The Rocky Road to Clinical Translation

The technology of making Tissue-engineered Heart Valves (TEHVs) is designed to overcome the hurdles of current valve replacements. Current valve prostheses, used in conditions such as congenital or acquired heart valve disease, either necessitate reoperation due to valve deterioration (in the case of bioprosthetic valves) or lifelong anticoagulation therapy (in the case of mechanical valves). More importantly, they cannot accommodate the required change in size of a growing (congenital) patient. This can be partially overcome with the Ross procedure, though it still requires a pulmonary valve implant. One promising approach to make in situ TEHVs is the use of bioresorbable elastomers, electrospun into fibrous, valve-shaped scaffolds, that are available off-the-shelf and can be implanted directly at the functional site. Here, the synthetic valve evokes an immune response that subsequently remodels the scaffold into autologous tissue. Right after implantation, circulating plasma proteins adsorb to the scaffold, attracting neutrophils, which in turn attract macrophages. These inflammatory cells initiate the breakdown of scaffold fibers, while simultaneously attracting fibroblast-like cells that begin to produce a neo-tissue matrix. With time, the scaffold should be replaced entirely by new, autologous tissue, thus permitting growth and preventing immune rejection and coagulation issues. To test this innovative technology, many in vitro and preclinical studies have been conducted. These studies confirmed neo-tissue formation coinciding with scaffold breakdown, but also demonstrated potential hurdles to clinical translation. One of these hurdles is the formation of calcific nodules in the grafts. As calcification of bioprosthetic grafts and native heart valves is known to lead to structural valve problems that ultimately necessitate valve replacement, we wondered if and how calcification would affect the outcomes of in situ TEHVs. Therefore, the overall goal of this thesis was to evaluate the potential risk of calcification of current material scaffolds for in situ TEHVs. To that end, we analyzed existing preclinical study outcomes and developed in vitro calcification models to systematically probe the material-specific calcification potential under the influence of inflammation, representing a key aspect of the in situ tissue engineering process. To define the extent of the problem within the field, we first conducted a systematic review, examining the reporting of calcification in all preclinical in situ TEHV studies that replaced the pulmonary valve (Chapter 2). We found that calcification was strongly under-reported, with only half of all studies examining calcification. If reported, studies showed an overall calcification event rate of 35%, with calcification mainly consisting of micro-nodules, and to a much lesser extent than found in bioprosthetic controls (if available). The rate and extent of calcification appeared independent of the initial material, the method used to produce the valves, or of the animal models in which they were evaluated. Our analyses also highlighted the need for better reporting standards in pre-clinical models of TEHVs. To study calcification in the context of successful and unsuccessful tissue remodeling, we next performed a proteomics-based spatial analysis of in situ engineered pulmonary heart valves created from two different polycarbonate (PC)-based supramolecular polymers: PC-bisurea (PC-BU) and ureidopyrimidinone-extended-PC (PC-Upy). Based on this in-depth analysis, we confirmed in Chapter 3 that graft remodeling was primarily based on the innate inflammatory response and was spatio-temporally regulated, with scaffold-to-tissue transition progressing from the anastomotic site towards the mid, while the tip remained in an early stage of material-blood interaction even after 12 months. Examining calcification, we observed greater amounts and sizes of calcification in PC-Upy grafts than in PC-BU, albeit with considerable variation among the tested animals. We further found that macro-calcifications resembled those in diseased aortic valves, micro-calcifications were phenotypically, chemically, and biologically different. We then built in vitro models to systematically study and better understand the propensity of the used materials to calcify. In Chapter 4, we established a novel model for TEHV scaffold material calcification, based on known models of cell-induced valvular calcification. We used electrospun scaffolds seeded with fibroblast-like porcine valvular interstitial cells (pVICs) and cultured in a non-osteoinductive medium to avoid creating a disease environment. The model was applied to compare three TEHV scaffold materials (i.e., polycaprolactone (PCL), PC-BU, and PCL-BU) to glutaraldehyde-treated bovine pericardial tissue, a material commonly used in bioprostheses. We observed a significant increase in calcification in pericardial tissue compared to TEHV materials, consistent with the results presented in Chapter 2. The calcification of the biomaterials was cell-mediated, whereas pericardial tissue seemed to calcify independently of cell differentiation, via the precipitation of calcium phosphate in areas of DNA remnants in this decellularized material. Furthermore, the outcomes appeared unaffected by cyclic strain, suggesting that homeostatic strain does not influence material-driven TEHV calcification. Following these results, we further evaluated the potential correlation between the host inflammatory response to a biomaterial and graft calcification. To study the effect of the immune response, we used scaffold-associated macrophage-conditioned medium (MCM) to stimulate induced pluripotent aortic smooth muscle cells (iaSMCs) on the same-material scaffolds in Chapter 5. We used iaSMCs, since these cells resemble the endogenous cells that will infiltrate the tissue-engineered valve in situ after implantation. We found that scaffold-associated macrophages form an intermediate phenotype between pro-inflammatory and regenerative cells. The addition of MCM further significantly increased calcification and osteogenic cell differentiation in iaSMCs. This response not dependent of a subset of tested cytokines. Finally, in Chapter 6, we proposed a set of minimal requirements for testing TEHVs and made a plea for consensus on the minimum clinically necessary and scientifically significant reporting on TEHV material, scaffold, graft, implant, and explant evaluations at various stages of valve development in vitro and in vivo. The aim of with chapter was to aid the creation of trustworthy data that can be meaningfully interpreted by the multimodal expertise in the field, be benchmarked against suitable controls, combined to create high-quality large datasets, and be fitted into a safety framework. Eventually, this should all lead to better TEHV materials that can advance towards clinical translation. To conclude, in this thesis, we established that calcification, although understudied, may pose challenges for the clinical translation of material-driven TEHVs that need to be resolved or prevented. We further established novel ex vivo and in vitro techniques to study TEHV calcification. Using these methods, we found that the inflammatory response significantly influences both TEHV remodeling and graft calcification and warrants further study. Future studies should investigate how material properties and graft manufacturing can be used to change the inflammatory response towards a functional, but non-calcification-mitigating environment. The models and techniques described in this thesis could pave the way for the further development of calcification-free materials, thereby aiding the clinical translation of TEHV grafts.