Abstract
Background/Aim: The present study aimed to compare the tissue responses to biomaterials in the chick chorioallantoic membrane (CAM) model with those from the subcutaneous implantation model in rats at an early time point. It was especially investigated whether histopathological scoring according to DIN EN ISO 10993-6 is also possible after biomaterial implantation using the CAM model and to what extent the values differ from the data obtained from small animal experiments.
Materials and Methods: Implantation of a xenogeneic bone substitute using the CAM model for 24 h and subcutaneous implantation model in rats up to 10 days post implantation were conducted. Standardized histological and histopathological methods were used to apply for histopathological scoring according to DIN EN ISO 10993-6.
Results: The histological analysis as well as the histo-pathological scoring revealed that the tissue responses to the xenogeneic bone substitute were completely comparable in both organisms with no visible or statistical differences.
Conclusion: We suggest that bioincompatible biomaterials can already be sorted out in the context of THE preclinical in vivo test phase. Such pre-testing before the required small animal tests might clearly contribute to the 3R-concept to reduce the number of animals (REDUCE).
- Biocompatibility
- biomaterial
- medical device
- chicken chorioallantoic membrane (CAM)
- histopathology
- scoring
- DIN ISO 10993-6
- 3R-concept
Introduction
The development of a new medical device or biomaterial and its introduction into the market requires extensive testing procedures for safety and clinical performance. Various in vitro test methodologies are necessary based on the DIN EN ISO 10993 series including cytotoxicity assays (ISO 10993-5), genotoxicity testing (ISO 10993-3), hemocompatibility (ISO 10993-4) and irritation assays (ISO 10993-10) (1, 2). Although these tests are extensive and can already provide a lot of data on the cyto- or biocompatibility of a biomaterial, they have different limitations (3). The main limitations of in vitro assays are the lack of inactivation or removal of metabolic and toxic products (even in the case of static cell cultures), severely limited informative value regarding biofunctionality, such as osseointegration of bone replacement materials, the lack of both local and systemic factors as well as the difficulty of long-term studies (3, 4). Thus, in vivo trials are additionally necessary to safeguard the safety of biomaterials based on the DIN ISO 10993-6 norm (2). Thereby, in vivo results mostly confirm or disprove the in vitro results, but also increase the knowledge about the biomaterial to be tested for adverse events, such as systemic toxicity, as well as for the elucidation of local tissue reactions or (bone) tissue regeneration. Moreover, in vivo trials are needed to test their suitability for a specific indication.
In the context of in vivo studies, the first implantation model that needs to be conducted is the subcutaneous implantation in small animals, such as mice or rats, to determine the potential for unacceptable adverse biological responses resulting from contact of the (component) materials of the device with cells or tissues (5). Moreover, histo-pathological evaluation via a special scoring system is required based on the DIN EN ISO 10993-6 (6). The pathologists use histochemically-stained slices [i.e., slices stained with hematoxylin and eosin (HE)] to score the tissue reactions to a biomaterial, including the involvement of different cell types or tissue parameters, such as macrophages, lymphocytes, necrosis or fibrosis. Moreover, they have to calculate the irritancy value of this device in comparison to a selected control material to evaluate biocompatibility. To date, this examination step cannot be replaced by alternative “animal-free” methods.
A query of the German database “Animal test info” under the search terms “[bone substitute bone], [collagen], [dental], [animal experiments] and [animal numbers]” revealed the numbers of laboratory animals used for testing of medical devices in the last three years (7). It becomes clear that due to the stricter legislation for quality testing of biomaterials, the required number of experimental animals has increased 2.2-fold for rats and as much as 3.6-fold for mice. If all laboratory animals are considered, the number increased from 231,000 to 644,000, i.e., by 2.8 times in the period from 2020 to 2022. For this period, a total of approximately 1.5 million laboratory animals have been used. Accordingly, there is an indisputably urgent need for alternative test methods to demonstrate medical device compatibility.
Interestingly, the chicken chorioallantoic membrane (CAM) has been identified as an intermediate between in vitro and in vivo models (8, 9). The CAM model is performed on the chorioallantoic membrane of the chicken egg. The CAM is a vascularized membrane that is responsible for supplying the embryo with nutrients and for the regulation of the acid-base balance (10, 11). It forms around the fifth to sixth day of development and its composition including the cellular but also extracellular components are comparable to mammalian subcutaneous tissue (11). In this context, Valdes et al. have already described a new approach for testing biomaterials using the CAM model, as an alternative to mammalian implantation models (12). They revealed that the inflammatory response of the CAM tissue to biomaterials was comparable to that of the mammalian tissue and was also seen to vary according to different test materials. However, it has not yet been investigated whether histopathological scoring according to DIN EN ISO 10993-6 is also possible using this implantation model and to what extent the values differ from the data obtained from small animal experiments.
For this reason, the present study was designed to directly compare the tissue responses using the CAM model with those of the subcutaneous implantation model in rats at an early time point. For this purpose, previously established and standardized implantation models as well as histological and histopathological methods were used (13-16).
Materials and Methods
Biomaterial. For the present study, the bovine bone substitute (BBS) cerabone® (botiss biomaterials GmbH, Zossen, Germany) was used. This biomaterial is obtained from the femoral heads of cattle from registered slaughterhouses in New Zealand and Germany and purified from potentially immunogenic components via a multi-step-process (17). For the conduct of the present study cerabone® with a granule size of 0.5-1 mm was used.
CAM model. Fertilized specific pathogen-free (SPF) chicken eggs from Lohmann (Neu-Ulm, Germany) were incubated at 37°C and 70-80% humidity. Start of incubation was determined as embryonic development day (EDD) 1. At EDD8 the eggshell was opened by introducing a window followed by separation and removal of the shell membrane from the CAM using a small drop of Dulbecco’s phosphate-buffered saline (DPBS) solution. Immediately, the eggshell window was sealed with silk tape. At EDD 9, approximately 40 mg of the BBS granules were transplanted into the CAM (Figure 1), and the eggs were further incubated for 24 h until EDD10. At EDD10, the CAM tissue implanted with the BBS was harvested and fixed with 4% buffered formaldehyde until the histological workup (Figure 1). Directly after the harvest, the chick embryos were sacrificed via decapitation.
Practical and macroscopic representation of CAM implantation: (A) Scheme of the CAM implantation process and exemplary images of (B) the implantation of the bovine bone substitute (BBS) (black arrow) using the CAM model and (C) after explantation of the BBS at EDD 10.
In vivo study. The preclinical in vivo experiments were conducted at the Faculty of Medicine at the University of Niš (Serbia) as previously described (18-20). Initially, approval by the Local Ethical Committee based on the Veterinary Directorate of the Ministry of Agriculture, Forestry and Water Management of the Republic of Serbia (decision number 323-07-01762/2019-05/9, Date: 01 March 2019) was obtained. Afterwards, the Wistar rats were housed under standard conditions (water ad libitum, artificial light, and regular rat pellet, and standard pre- and postoperative care) until the implantation time point. Six females, 8-10-week-old Wistar rats (n=6) were used for implantation of the BBS for 10 days following an established and standardized subcutaneous implantation model based on the DIN ISO 10993-6:2016 guidelines (21). In brief, the rats underwent initial anesthesia via intraperitoneal injection of ketamine (50 mg/ml) in a dose 100 mg/kg of body weight and Xylazine (2%) in a dose of 5 mg/kg of body weight). Afterwards, the rostral subscapular were shaved and disinfected. The following implantation procedure included an incision followed by blunt subcutaneous pocket preparation and insertion of 0.2 g of the BBS granules and final stitching of the operation wounds. Ten days post implantation, euthanization of the rats was performed using an anesthesia medication overdose. The BBS and the adjacent tissue were explanted and subsequently fixed via a 4% buffered formaldehyde until the histological workup.
Histological work up. Dehydration by series of increasing alcohol concentrations followed by xylol exposure was initially performed. Then, the tissue samples were stepwise immersed at 4°C in Technovit 9100 polymer embedding medium (Technovit 9100, Kulzer GmbH, Hanau, Germany) the resulting tissue blocks were trimmed into shape using a grinding machine (EcoMet 30, Bühler, Esslingen, Germany). Afterwards, sections with a thickness of 4 μm were prepared using a rotation microtome for hard tissue samples (CUT4060E, microTec GmbH, Walldorf, Germany). Finally, the slides were histochemically stained with hematoxylin and eosin (H&E).
Histopathological analyses. Histopathological evaluation of the tissue responses in both study groups was performed as already published (6, 19, 22, 23). The slices were evaluated to assess the local tissue response. Stained histological sections of the implant sites were evaluated by the Principal Investigator (MB) for a number of parameters evaluating safety and efficacy. The sections were analyzed and graded according to cell type and responses following the irritancy/reactivity grading scheme adapted from the ISO 10993-6 Annex E (Table I). Microsoft® Excel was utilized for data capture (scoring analysis).
Histologic evaluation system for irritancy/reactivity cell type/response.
The irritancy/reactivity scores were calculated as follows for each implantation site based on the parameters listed in Table I:
The irritancy scores for each study group were then calculated by averaging the irritancy scores of all implantation sites.
Statistical analyses. The scoring raw data were subjected to statistical analysis using the Mann-Whitney U-test to compare two independent groups via the GraphPad Prism 10.2 software (GraphPad Software Inc., La Jolla, CA, USA). Differences were designated as significant if the p-values were less than 0.05 (*p≤0.05), and highly significant if the p-values were less than 0.01 (**p≤0.01) or less than 0.001 (***p≤0.001). The data are presented as means±standard deviations.
Results
Histopathological comparison. The histological analysis revealed that the tissue reactions in both implantation models were comparable (Figure 2). The implanted bovine bone substitute (BBS) induced a mild tissue reaction involving mainly macrophages at the material surfaces and within the surrounding connective tissue in both organisms (Figure 2). Also, moderate numbers of multinucleated giant cells were found at the material surfaces in both implantation models (Figure 2). Additionally, low numbers of granulocytes and lymphocytes were detectable that were involved in the tissue reactions in both implantation models beside moderate numbers of blood vessels (Figure 2).
Exemplary histological images of the tissue reactions to the bovine bone substitute (BBS) after implantation using the (A) CAM model at EDD 10 and (B) the subcutaneous implantation model in Wistar rats at day 10 post implantation. Black arrows: macrophages; black arrowheads: multinucleated giant cells; red arrows: blood vessels; yellow arrows: granulocytes; and blue arrows: lymphocytes (H&E-staining, 200× magnification, scalebar=20 μm).
Comparison of the tissue reactions via histopathological scoring. The histopathological scoring revealed that the inflammatory tissue responses in both study groups were mainly composed of comparable extents of macrophages, lymphocytes, polymorphonuclear cells and multinucleated giant cells (Table II). Furthermore, the neovascularization values were comparable in both implantation models (Table II). In both models, comparable extents of a slight necrosis were additionally found. Finally, no plasma cells, fatty infiltrates or necroses were detectable in both models.
Results of the biomaterial scoring evaluations in both study groups.
Based on the scoring values, the irritancy scores were calculated. This analysis showed that the BBS within the CAM model had an average treatment irritancy score of 14.40, and the BSM within the subcutaneous connective tissue evoked an average treatment irritancy score of 14.17. Thus, the values from both implantation models were comparable.
The statistical analysis of the data showed no differences of the aforementioned cell and tissue parameters between both implantation models (Figure 3).
Graphical presentation of the results of the biomaterial scoring evaluations in both study groups. A) Polymorphonuclear cells. B) Lymphocytes. C) Macrophages. D) Giant cells. E) Necrosis. F) Neovascularization.
Discussion
Preclinical testing of biomaterials for biocompatibility is crucial in ensuring their safety and efficacy before clinical use. Traditionally, animal trials, including the subcutaneous implantation model, have been the standard approach for evaluating biocompatibility (24-26). However, high costs and regulatory pressures have fueled the exploration of alternative methods. Furthermore, the society has become increasingly sensitive to ethical considerations associated with animal experimentation (27, 28). Thus, the exploration of alternative methods for preclinical biocompatibility testing of biomaterials becomes imperative.
Advancements in cell culture techniques have led to the development of sophisticated in vitro models that mimic the in vivo environment to assess biocompatibility. These models include: i) Cell monolayers: Culturing cells on biomaterial surfaces to evaluate cell adhesion, proliferation, and viability (29, 30). ii) Three-dimensional (3D) cultures: Mimicking tissue-like structures using scaffolds to study cell-material interactions more realistically (31, 32). iii) Organ-on-a-Chip: Microfluidic devices that replicate organ-level physiology, allowing for complex in vitro testing (33, 34).
Additionally, computational modeling has been shown to offer a cost-effective and ethical alternative by simulating interactions between biomaterials and biological systems. Techniques include: i) Finite element analysis (FEA): Predicting mechanical responses and stress distribution within tissues in response to biomaterial implantation (35, 36). ii) Molecular dynamics (MD) simulation: Modeling atomic-level interactions between biomaterials and biological molecules to predict biocompatibility (37, 38).
Finally, ex vivo tissue models utilizing tissue explants or organ slices from human or animal sources provide a more physiologically relevant environment for evaluating biocompatibility (39, 40). These models mimic tissue architecture and cellular interactions, allowing for rather complex testing (40, 41).
However, all of the aforementioned methodologies include the eminent disadvantage that they can not mimic the complex in vivo situation starting with the initial protein agglomeration at the biomaterial surface followed by its interactions with cells embedded in the multifaceted network of cell-cell-interactions and mediators such as cytokines (26, 42-44). This means that to obtain a complete picture of the processes during cell or tissue reactions to a biomaterial, it is necessary to better understand the underlying complex cellular, biochemical and molecular biological processes. Current in vitro cell culture models can only mimic the initial aspects of sub-steps of the complex cascade (45, 46). This discussion highlights the diverse array of alternative methods available for preclinical biocompatibility testing, emphasizing their potential to mitigate the reliance on animal trials and specifically addressing the need for alternatives to the subcutaneous implantation model.
Depending on the question to be answered, the chicken egg CAM model represents a very convincing alternative to implantation models in mice or rats and can be used in this context as a powerful in vivo replacement method to test the early biocompatibility of a biomaterial. It is worth mentioning that the CAM per se is not innervated and biocompatibility experiments can be performed from day 9 to day 13, thus in the first 2/3 period of embryo development (11, 47). Methods such as the tumour xenograft CAM model and different angiogenesis CAM assays have already been established (48-50). Thereby, this methodology usually requires less bureaucracy regarding animal testing applications. Owing to the faster implementation, they also allow a higher experimental frequency thus representing a promising alternative to animal experiments, especially from an ethical point of view.
Moreover, the CAM model has already been shown to bridge the gap between in vitro and in vivo studies in the translational biocompatibility analysis of new biomaterials (51, 52). The main difference between the evaluation using the subcutaneous implantion model in mammalians and the CAM model is the length of the incubation period, which is limited to 10 days in the CAM model (52). However, owing to its simplicity, high throughput and low costs, the CAM model could represent a first indicator of the in vivo reactivity of a biomaterial in order to sort out bioincompatible material and ultimately significantly minimize the number of test animals in subsequent preclinical in vivo tests. Although, different studies have evaluated the angiogenic potential of biomaterials via the CAM model, even in comparison to subcutaneous implantation in rodents, the usage of this model to examine the tissue responses to biomaterials in accordance to the histopathological analysis scheme of the DIN EN ISO 10993-6 has still not been evaluated. Thus, the present study was conducted to distinguish the tissue responses using the CAM model with those using the subcutaneous implantation model in rats at an early time point, i.e., 10 days post implantation. For this purpose, previously established and standardized implantation models as well as histological and histopathological methods were used (53-55).
The results showed that the implanted BBS induced in both organisms a mild tissue reaction involving mainly macrophages at the material surfaces and within the surrounding connective tissue. Also, moderate numbers of multinucleated giant cells and low numbers of granulocytes and lymphocytes were found at the material surfaces in both implantation models beside moderate numbers of blood vessels. Thus, there was no visible histological difference in the tissue reactions to the BBS in both organisms. This result is not surprising, since the stroma of the CAM contains the same cell types also found within the connective tissue of mammals and it is known that the mammalian tissue responses (including acute and chronic inflammation, granulation tissue formation, and fibrosis) and those of the CAM tissue are similar (16). In this context, it has already described by Valdes and colleagues that the inflammatory response of the CAM to biomaterials was fully similar to that of mammalian tissue and was also varying in accordance to test materials with different physical material characteristics, i.e., structure and geometry of the test materials (12). Thus, they also suggested that this animal model is attractive for the rapid in vivo screening of the biocompatibility of biomaterials.
Additionally, the histopathological scoring of the tissue reactions as a novelty in this study revealed that in both models comparable scoring results were obtained. Furthermore, the irritancy scores that were calculated on the basis of the scoring values were completely comparable. Finally, the statistical comparison of the scoring data also showed no differences in the cell and tissue parameters in both implantation models.
Altogether, these data further demonstrate the clear correspondence between the two implantation models in the two organisms at this early post-implant study time point. In this context, Russell and Burch called for the implementation of the so-called “3R-concept” (56). This concept is also suggested in the Directive of the European Union Directive of 2010 (57). “Replace” refers to the use of animal replacement methods on non-sentient matter. In the context of “reduction”, the number of animals used in animal experiments should be reduced to the lowest possible level. “Refinement” includes both experiments that reduce suffering and pain due to improved technology, as well as improved husbandry conditions. In addition, the European Union Directive explicitly points out that an implementation of the 3Rs to reduce animal experiments is required and that especially alternative models should be supported (56, 58). The similarity of the response of the CAM with that of the mammalian model proven in the present study make this test model attractive for the screening of biomaterials as a first indicator of the in vivo reactivity and can contribute thereby clearly to the 3R-concept in view of an animal number reduction (REDUCE). In the future, it might allow to “sort out” bioincompatible materials as an integral part of the preclinical investigation cascade and ultimately significantly minimize the number of test animals in subsequent preclinical in vivo tests using the same histopathological methodology to evaluate the tissue responses to biomaterials in accordance to the DIN EN ISO 10993-6.
Acknowledgements
The Authors would like to thank Ms Mirijam Schäfer, Ms Aenne Foth and Ms Denise Pankalla for their technical support.
Footnotes
Authors’ Contributions
Conceptualization: M.B. and R.S.S.; methodology: M.B., S.S., S.N., K.H. and R.S.S.; CAM model and imaging: K.H.; software: M.B.; validation, M.B.; formal analysis: M.B. and K.B; investigation: M.B.; resources: M.B., S.N., O.J. and R.S.S.; data curation: M.B. and K.B.; writing – original draft preparation: M.B. and K.B.; writing – review and editing: M.B., O.J., T.A. and R.K.; visualization: M.B.; supervision, M.B., S.N., R.S. and R.S.S.; project administration: M.B.; funding acquisition: M.B., S.N., O.J. and R.S.S. All Authors have read and agreed to the published version of the manuscript.
Funding
The Authors gratefully acknowledge the funding by the German Research Foundation (Deutsche Forschungsgemeinschaft, DFG) for the subproject 2 and 6 within the Research Unit 5250 “Permanent and bioresorbable implants with tailored functionality” (No. 449916462).
Conflicts of Interest
The Authors declare no conflicts of interest in relation to this study.
- Received May 10, 2024.
- Revision received June 17, 2024.
- Accepted July 1, 2024.
- Copyright © 2025 The Author(s). Published by the International Institute of Anticancer Research.
This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).
