Sonia Schikert

Closing Remarks and Future Perspectives

The design of biological bone substitute materials (BSMs) is highly complicated, as biological and mechanical requirements are often contradictory. In recent years, biological BSMs have been developed, which allow for a more efficient integration of the material with the surrounding osseous environment and hence, a higher mechanical stability of the treated defect. However, while these materials are promising, they are still far from ideal. Consequently, extensive pre-clinical experimentation is still required. Therefore, Chapter 2 aimed to provide a comprehensive overview of biomechanical considerations relevant for the design of biological BSMs. Further, the preclinical evaluation of biological BSMs intended for application in highly loaded skeletal sites was discussed. It was concluded that the selected animals and implantation sites should mimic the pathophysiology and biomechanical loading patterns of human bone as close as possible. In the literature, sheep are among the most frequently selected animal species for the evaluation of biomaterials intended for highly loaded skeletal sites. Regarding the anatomical sites, segmental bone defects created in the limbs and spinal column are suggested as most suitable. Furthermore, the outcome measurements used to assess biological BSMs for regeneration of defects in heavily loaded bone should be relevant and straightforward. Quantitative evaluation of bone defect healing through ex vivo biomechanical tests is a valuable addition to conventional in vivo tests (using histological 2D and/or imaging-based 3D techniques), as it determines the functional efficacy of BSM-induced bone healing. Finally, it was concluded that further standardization of pre-clinical studies is essential for reliable evaluation of biological BSMs in highly loaded skeletal sites.

Several types of fibers have previously been explored to reinforce CPCs and improve their mechanical properties to enable their application in load-bearing sites. Nevertheless, poly(vinyl alcohol) (PVA) fibers have not yet been investigated for this purpose, although this type of polymeric fiber has been extensively and successfully used to reinforce cementitious composites in civil engineering. Therefore, Chapter 3 aimed to firstly investigate the effect of PVA fibers on the mechanical properties of CPCs. Secondly, the in vitro cytocompatibility of these fibers was studied by using cell culture tests. Finally, an in vivo osteocompatibility study was carried out to assess tissue responses to PVA fiber-reinforced CPCs histologically after 6 and 12 weeks of implantation in the femoral condyle of rabbits. Results revealed that the incorporation of PVA fibers into CPCs was highly effective by strengthening and toughening CPCs, since the flexural strength and toughness of CPCs increased upon reinforcement with PVA fibers by more than 3-fold and 435-fold, respectively. In vitro cytocompatibility tests indicated that PVA fibers were cytocompatible, which was further confirmed by the in vivo results that showed that PVA fibers did not compromise the excellent osteocompatibility of CPCs.

Stabilization of dental implants by means of biomaterials such as bioceramic granules and cements is currently compromised by the poor mechanical properties of these bioceramics. In Chapter 3, a novel CPC reinforced with PVA fibers with improved flexural strength and toughness was developed and its biological safety was assessed. Chapter 4 evaluated the capacity of this PVA-reinforced CPC to stabilize dental implants in vitro and in vivo, using a range of mechanical and biological test methods. In vitro, filling of circumferential crestal peri-implant bone defects using synthetic bone analogues with fiber-reinforced CPC demonstrated superior implant stability compared to fiber-free CPC over a 12-week period. Similarly, filling of circumferential crestal peri-implant bone defects with fiber-reinforced CPC effectively stabilized dental implants installed in a rabbit femoral condyle defect, as assessed via both implant stability quotient (ISQ) and torque-out measurements. Moreover, histological and histomorphometric evaluation confirmed the osteocompatibility of fiber-reinforced CPC, as evidenced by the absence of soft tissue ingrowth, direct contact between the bone and cement, as well as gradual degradation of the CPC and replacement by newly-formed bone. The obtained data demonstrated that fiber-reinforced CPC successfully stabilized dental implants during osseointegration.

Vertebral compression fractures (VCFs) are a very common problem among the elderly, which ultimately result in severe pain and a drastically reduced quality of life. Conventional treatment of VCFs involves minimally invasive augmentation of the damaged vertebrae through vertebroplasty and/or kyphoplasty. These surgical procedures treat the affected vertebrae by injection of poly(methyl methacrylate) (PMMA) cement into the vertebral body. However, clinical use of PMMA cement is associated with major drawbacks related to its poor biological compatibility with bone tissue and a substantial mechanical mismatch. Previously, Chapters 3 and 4 focused on the design of moldable fiber-reinforced CPC, showing that both mechanical and biological properties were advantageous and potentially suitable for treating VCFs. However, in order to be delivered through a minimally invasive procedure (such as a vertebro- and/or kyphoplasty), the reinforced CPCs should be highly injectable. Therefore, in Chapter 5, the handling and mechanical properties of a CPC functionalized with both PVA fibers and carboxymethyl cellulose (CMC) were analysed and further fine-tuned to render this composite material suitable for vertebro- and kyphoplasty procedures. The obtained results demonstrated that the addition of CMC rendered PVA-reinforced CPCs injectable without negatively affecting their mechanical properties. Further, ex vivo mechanical analyses clearly showed that extravasation of CPC-PVA-CMC into trabecular bone was reduced compared to PMMA cement. Finally, it was observed that the ex vivo biomechanical performance of vertebrae treated with CMC-PVA-CMC was similar to PMMA-treated vertebrae. The obtained data suggest that CPCs functionalized with CMC and PVA fibers possess adequate handling, mechanical and structural characteristics for vertebro- and kyphoplasty procedures.

2. Closing Remarks and Future Perspectives

The present thesis explored the development and optimization of a CPC reinforced with PVA fibers, specifically intended for dental and orthopedic load-bearing indications. A wide range of in vitro, ex vivo and in vivo strategies has been employed to specifically characterize the physicochemical, handling, mechanical, and biological properties of this newly developed material. Ultimately, a tough and osteocompatible bioceramic has been obtained (Chapter 3). With respect to mechanical performance in terms of work-of-fracture and flexural strength, the reinforced CPC herein analyzed has surpassed a large number of fiber-reinforced CPCs previously reported in literature. [1-4] Further, fine-tuning of the injectability and cohesion of these toughened CPCs rendered them both applicable via minimally invasive injection and, once in situ, capable of remaining structurally intact. The joint action of enhanced handling and mechanical properties actively contributed to enhanced dental implant stability (Chapter 4) and reinforcement of ex vivo porcine vertebrae (Chapter 5). Importantly, the additions and modifications that were implemented to obtain this superior performance (i.e. addition of fibers, porogens and polymeric lubricants) did not alter the intrinsic osteocompatibility of CPCs, as demonstrated in Chapters 3 and 4.

While the previously mentioned accomplishments provide a step forward towards creation of load-bearing bioceramics, further research and development is still needed to enable clinical translation. In this context, future research directions are presented, not only to i) improve the biomechanical properties of the PVA-reinforced CPCs, but also to ii) expand the preclinical body-of-evidence for the use of BSMs as basis for iii) clinical assessment of the biocompatibility and biofunctionality of these materials.

2.1. Further improvement of the biomechanical properties of PVA-reinforced CPCs
Biodegradability
It is widely accepted that an ideal BSM should be fully degradable and ultimately lead to complete replacement of the biomaterial by native bone, shortly after implantation. In its unaltered form, apatitic CPCs have a very slow degradation profile, and extensive research efforts have been dedicated to improvement of their degradation behavior. The inclusion of poly(lactic-co-glycolic acid) (PLGA) microparticles within the matrix of CPCs has shown previously [5-7] and in this thesis (Chapter 3 and 4) to cause apatitic CPCs to degrade in vivo owing to the combination of surface increase (i.e. due to the porosity creation) and the acid production upon dissolution of PLGA microspheres and concomitant resorption of the CPC matrix. However, while PLGA microparticles enable successful degradation of the PVA-reinforced CPC matrix (Chapter 3 and 4), histology clearly demonstrated that the reinforcing PVA fibers did not alter their morphology, even after 12 weeks of implantation. Instead, after this implantation period, the intact fibers that once were embedded within the cement matrix were gradually surrounded by healthy and newly formed bone. In fact, although PVA is recognized as one of the very few vinyl polymers soluble in water and susceptible to ultimate biodegradation [8], the heavily cross-linked nature of these specific PVA fibers, while providing for excellent tensile strength, suppressed hydrolysis and rendered them non-degradable. Although no cytotoxicity (Chapter 3) or significant in vivo adverse reactions (Chapter 3 and 4) were observed from the direct contact between the newly formed bone and the fibers (i.e. suggesting that these fibers are inert), the long-term effect and fate of these unaltered fibers in the body remains unknown.

To tackle this issue, two different approaches could be followed. Firstly, a long-term assessment of the biosafety of these fibers should be performed, with the purpose of analyzing whether they remain inert on long-term, or still alter morphologically and chemically after prolonged times. In case these fibers do degrade, it is thereafter of utmost importance to study if these degradation byproducts do not trigger an adverse inflammatory reaction or immune response. According to the International Organization for Standardization (ISO), the long-term evaluation of local effects after implantation in bone should be performed in animals with a relatively long life-expectancy for implantation periods of up to 104 weeks. [9] Secondly, the chemical composition and fabrication method of the PVA fibers could be analyzed in depth in order to improve the potential biodegradation profile of these fibers. Ideally, the cross-linking performed during the production of these fibers should be adapted to combine the mechanical strength of these fibers with slow and steady hydrolysis-mediated fiber degradation. [10] Alternatively, previous studies have also been carried out, in which the biodegradability of cross-linked PVA was improved by using gamma irradiation as a pre-treatment strategy. [11] Although this strategy might render crosslinked PVA degradable, their mechanical strength and affinity to the cement matrix might have been negatively affected. In this case, the extent to which gamma irradiation affects the mechanical properties of PVA fibers and their interaction with the CPC matrix should be investigated in detail, for instance by resorting to tensile [12] and pull-out tests [13], respectively. Regardless of the selected approach, reinforcement of CPCs using biodegradable PVA fibers would enable to combine mechanical support at an initial stage of implantation, followed by the subsequent degradation of the cement matrix and the fibers on long-term.

Biofunctionalization
As described in Chapter 2, the biological enhancement of BSMs - including CPCs - has been widely explored over the recent years. Although PVA-reinforced CPCs combine improved mechanical performance with favorable biological properties, enhancement of their regenerative capacity would allow for application of CPCs in skeletal sites with a compromised healing capacity caused by e.g. diseases such as infection or osteoporosis.

A popular biofunctionalization approach of CPCs includes mixing living cells, mostly from an osteogenic precursor line, into the CPC paste. In principle, employing a CPC in combination with living cells into a bone defect should allow for faster and more efficient bone regeneration than filling a defect without cells. [14] Previous studies have explored the inclusion of stem cells into injectable CPCs and concluded that stem cell inclusion significantly enhances osteogenic cell attachment and proliferation. [15, 16] Pre-clinical studies further confirmed the efficacy of this strategy, by demonstrating that placing CPCs enriched with bone marrow mesenchymal stem cells (MSCs) in large bone defects in mini-pigs, resulted in considerably enhanced bone regeneration and blood vessel density as compared to the cell-free CPC control. [17] Another study also employed CPCs enriched with MSCs in cranial defects of nude rats and demonstrated to have improved new bone formation by approximately 45%, when compared to the non-cell CPCs. [18] Therefore, while the addition of cells into the material is not directly responsible for forming new bone, they play an important immunomodulatory role, from which the regeneration of the bone defect will ultimately benefit from. It should however be stressed that the inclusion of living cells into CPCs is a complicated procedure, as mixing forces, ionic exchange and pH fluctuations during CPC setting are detrimental to cell viability [14], for which cell encapsulation techniques should be used to protect cells. CPCs have also been functionalized to improve their bone regenerative capacity by incorporation of various growth factors such as bone morphogenic proteins (BMPs) [19, 20], transforming growth factors (TGFs) [21], platelet-derived growth factor and vascular endothelial growth factor. [22] This approach confers CPCs with bone-inductive properties, thereby enabling stimulation of bone cell differentiation both in vitro [21] and in vivo. [19, 20] Recently, BMP-2 was added to a PLGA fiber-reinforced CPCs to treat bone defects created in a sheep lumbar osteopenia model. [20] Results demonstrated that, even at a low dose (i.e. ≤ 100 μg), the addition of BMP-2 to these fiber-reinforced CPCs resulted in a significant increase in bone formation.

Generally, it should be emphasized that biofunctionalization approaches unavoidably come along with numerous complications in terms of product design, upscaling and regulation. Finally, it is questionable if the preparation of such biofunctionalzied CPCs matches with the current procedures in the surgical theatre and can be performed by the surgical team. As such, the clinical benefit of these biofunctionalized medical devices should largely outweigh the increased costs.

2.2. Expanding the preclinical body-of-evidence for fiber-reinforced CPCs
The current thesis stressed the importance of performing pre-clinical studies to accurately assess the biomechanical performance of PVA-reinforced CPCs. Nonetheless, the preclinical studies carried out in this thesis (Chapter 3 and 4) focused primarily on confirmation of the biocompatibility of PVA-reinforced cements due to the initial stage of this research development. Chapter 5 further included ex-vivo evidence that paves the way for future preclinical studies on the feasibility of treating vertebral compression fractures using injectable fiber-reinforced CPCs. Therefore, future research should continue studying the feasibility of fiber-reinforced CPCs by means of clinically relevant in vivo studies.

Dental application of PVA-reinforced CPCs
In Chapter 4, an unsupported dental implant was analyzed with respect to its primary and secondary stability, upon placement of a PVA-reinforced CPC into its surrounding distance. This study was performed in a femoral condyle of a rabbit. This model is not only reasonable from a practical, economical and ethical point of view, but also provides a trabecular bone fraction, covered by a thin cortical bone layer that is relatively similar to the alveolar condition. [23, 24] Considering the promising results obtained herein, a future study should therefore study unsupported dental implants placed in a clinically more relevant context. Firstly, the selected animal model should have maxillofacial dimensions and oral conditions relatively similar to those of humans. Dogs, for example, have a dentition consisting of 3 incisors, 1 canine, 4 pre-molars and 2-3 molars, and a natural susceptibility to periodontal disease. [25] Furthermore, the maxillofacial dimensions of dogs allow for placement of clinically available dental implants. In vivo studies analyzing the effect of Bio Oss® (Geistlich Pharma AG, Wolhusen, Germany) on the stability and/or histological bone formation/bone-to-implant contact of immediately placed dental implants surrounded by bone defects in dogs have been successfully carried out in the past. [26-28] These studies included both the extraction of a tooth followed by placement of a dental implant (i.e. a surgical protocol nearly identical to the clinical one) and artificial creation of a bone defect surrounding the dental implant. Interestingly, all aforementioned studies did not load the implant immediately (i.e. placement of a functional crown subjected to occlusal and masticatory loads after the surgical procedure). This omission might be related to that fact that Bio Oss® consists of discrete brittle granules with limited reinforcing efficacy. [29] The use of PVA-reinforced CPCs, on the other hand, could potentially lead to optimized load transfer and more efficient osseointegration of the implant, resulting in implant stability shortly after their installation.

In addition to dogs, minipigs have also been vastly selected for dental implant research, as they are considered to be a suitable animal for the evaluation of periodontal diseases and osseointegrated dental implants. [30] Minipigs are considered to be a close representative of the humans, as they demonstrate similarities to human bone regarding anatomy, morphology, healing, remodeling and bone mineral concentration. [31] Intra-orally, the dentition of minipigs is similar to that of humans, even though their teeth are somewhat larger, longer and with a higher number of roots. [30, 31] Minipigs have been previously selected for immediately loaded implant studies [32, 33], as well as studies analyzing regeneration strategies for the peri-implant crestal bone. [34-36]

Orthopedic application of PVA-reinforced CPCs
Vertebral compression fractures treated through vertebra- and kyphoplasty have been previously simulated through in vivo surrogates, by means of an ovine model. [37-40] As mentioned before (i.e. Chapter 2), sheep are popular for orthopedic and musculoskeletal research, not only due to their pathophysiological similarity with humans, but also because these animals have shown striking biomechanical similarities with humans in certain analogous anatomical regions (e.g. vertebral column). Therefore, considering the potential suitability of PVA-reinforced CPCs for application in vertebral augmentation procedures (Chapter 5), the next logical step would be to further confirm this suitability by means of in vivo experimentation. In this context, an ovine model of high translational value could compare the biological and mechanical performance of reinforced CPCs in a straightforward manner, by using both unreinforced CPC and commercially available PMMA cement as controls upon implantation in a standardized bone defect created in vertebrae of the animal. The underlying hypothesis of this future study would be that, as result of their superior biological performance and lower stiffness than PMMA cement, injectable and toughened PVA-reinforced CPC would directly connect with the surrounding bone and not lead to fracture of the subsequent vertebrae, while remaining stable enough to provide adequate support to the treated vertebra. This hypothesis should be further validated by resorting to outcome measurements such as ex-vivo mechanical evaluations (i.e. compressive analysis of the excised vertebrae), micro-computed tomography (μ-CT) and quantitative histomorphometry.

In addition to spinal applications, the mechanically improved CPC explored throughout this thesis also shows promising characteristics for repairing complex bone fractures. In this sense, the biomechanical performance of a reinforced CPC could be further explored to treat e.g. tibial fractures by using both ovine [41, 42] and canine [43] preclinical models.

2.3. Clinical assessment of the efficacy PVA-reinforced CPCs.
Currently, there are no commercially available CPCs with improved mechanical properties (i.e. through fiber-reinforcement or other strategy) available in the market for human use. For this reason, while extensive research has been conducted regarding the broadening of clinical applications of CPCs, their clinical application remains limited to mechanically unchallenged skeletal sites. Therefore, to bring CPCs reinforced with PVA fibers from the bench to the bedside, clinical studies should be performed after having obtained a solid preclinical body of evidence. These clinical studies should confirm the safety and superior efficacy of these cements as compared to commercially available predicate devices such as unreinforced CPCs as well as conventional granular bone substitutes (e.g. Bio Oss®).