Development of a Rat Model of Mandibular Irradiation Sequelae for Preclinical Studies of Bone Repair

Abstract

Repairing mandibular bone defects after radiotherapy of the upper aerodigestive tract is clinically challenging. Although bone tissue engineering has recently generated a number of innovative treatment approaches for osteoradionecrosis (ORN), these modalities must be evaluated preclinically in a relevant, reproducible, animal model. The objective of this study was to evaluate a novel rat model of mandibular irradiation sequelae, with a focus on the adverse effects of radiotherapy on bone structure, intraosseous vascularization, and bone regeneration. Rats were irradiated with a single 80 Gy dose to the jaws. Three weeks after irradiation, mandibular bone defects of different sizes (0, 1, 3, or 5 mm) were produced in each hemimandible. Five weeks after the surgical procedure, the animals were euthanized. Explanted mandibular samples were qualitatively and quantitatively assessed for bone formation, bone structure, and intraosseous vascular volume by using micro-computed tomography, scanning electron microscopy, and histology. Twenty irradiated hemimandibles and 20 nonirradiated hemimandibles were included in the study. The bone and vessel volumes were significantly lower in the irradiated group. The extent of bone remodeling was inversely related to the defect size. In the irradiated group, scanning electron microscopy revealed a large number of polycyclic gaps consistent with periosteocytic lysis (described as being pathognomonic for ORN). This feature was correlated with elevated osteoclastic activity in a histological assessment. In the irradiated areas, the critical-sized defect was 3 mm. Hence, our rat model of mandibular irradiation sequelae showed hypovascularization and osteopenia.

Impact statement

Repairing mandibular bone defects after radiotherapy of the upper aerodigestive tract is clinically challenging. Novel tissue engineering approaches for healing irradiated bone must first be assessed in animal models. The current rat model of mandibular irradiation sequelae is based on tooth extraction after radiotherapy. However, the mucosal sequelae of radiotherapy often prevent the retention of tissue-engineered biomaterials within the bone defect. We used a submandibular approach to create a new rat model of mandibular irradiation sequelae, which enables the stable retention of biomaterials within the bone defect and should thus facilitate the assessment of bone regeneration.

Introduction

With 600,000 new cases and more than 325,000 deaths per year worldwide, the treatment of carcinomas of the upper aerodigestive tract remains a major health care challenge.¹ This treatment is often based on a combination of broad surgical resection with postoperative, high-dose, external radiotherapy. Despite the implementation of preventive measures and the emergence of new techniques (such as intensity-modulated radiation therapy), surgery and radiotherapy can result in long-term aesthetic and functional disorders.² Mandibular osteoradionecrosis (ORN, defined as mucosal ulceration with exposed, necrotic bone that fails to heal within 3–6 months) is a severe side effect of radiotherapy. It may lead to mandibular fractures and swallowing and speech disorders.

The exact pathophysiological mechanism of ORN has not been determined, although several hypotheses have been put forward. These include the “3Hs”—hypoxia, hypovascularization, and hypocellularity³—and the “2Is”—infection and ischemia.⁴ In histological assessments, radiotherapy is seen to impair bone healing by damaging blood vessels, increasing intracellular damage and osteoclastic activity, and promoting bone marrow fibrosis.⁵

Between 5% and 15% of treated patients will display ORN. Localized ORN is treated with drugs,⁶ whereas the gold standard for extensive ORN is the microanastomosed free flap.⁷ However, the latter procedure induces locoregional complications in the recipient (irradiated) area and sequelae in the donor area.

In recent years, the development of calcium phosphate biomaterials has provided new perspectives for mandibular bone reconstruction. The rat hind limb model previously developed in our laboratory has been used to study sequelae of bone irradiation.⁸ However, this model is limited by its use of long bones and thus the difficulty of transposing the results to the jaw. Another model of mandibular irradiation sequelae is based on tooth extraction after radiotherapy.⁹ However, the mucosal sequelae induced by radiotherapy often prevent the biomaterial from being stably retained within the bone defect.

Hence, the primary objective of this study was to develop a novel rat model of mandibular irradiation sequelae, to facilitate the assessment of radiotherapy's harmful effects on bone structure, intraosseous vascularization, and bone regeneration. We specifically chose a submandibular approach so that biomaterials could be stably retained in the resulting bone defect. The study's secondary objective was to determine the critical-sized defect in the irradiated jaw bone, that is, the smallest defect that would not heal spontaneously.

Materials and Methods

After 10 days of acclimatization, an 8-week-old inbred male Lewis 1A-haplotype RT1a rat was irradiated with a single 80 Gy dose to the jaws, using an XRAD225Cx irradiator (Precision X-Ray, Inc., North Bandfort, CT). General anesthesia was induced by inhalation of isoflurane (Isoflo^®; Abbott France, Rungis, France) at a concentration of 4% for induction and 2% for maintenance. Next, the animal was placed in a feet-first prone position in the irradiator (Fig. 1). The target volume was the mandibular angle, and the field diameter was 10 mm at the isocenter. An irradiation time of 561 s (voltage: 225kVp; intensity: 13 mA) was required for an 80 Gy dose to the bone.

FIG. 1.

An animal in the feet-first prone position in the irradiator. Color images are available online.

Three weeks after irradiation with a single 80 Gy dose, a 0, 1, 3 or 5 mm diameter circular defect was created on each hemimandible to test tissue engineering strategies for mandibular regeneration after irradiation. Mandibular defects were created under general anesthesia, according to the protocol mentioned earlier. Under aseptic conditions, the masseter muscle was incised and retracted to expose the mandibular ramus. A full-thickness 1, 3, or 5 mm diameter circular defect was created in the mandibular angle on both sides of the body, using a high-speed drill and cooling with physiological saline solution. The muscle and skin lesions were closed with resorbable suture (Vicryl^® 4.0; Ethicon, France).

Postoperative analgesia was provided by (1) buprenorphine (0.05 mg/kg, Buprecare^®; Animalcare, York, UK) as a subcutaneous injection every 12 h for 4 days, and (2) meloxicam (1 mg/kg, Metacam^®; Boehringer, Ingelheim Am Rhein, Germany) as a subcutaneous injection every 24 h for 5 days and then gavaged in sterile water on days 6 and 7. A 5-day course of antibiotic (marbofloxacin 3 mg/kg, Marbocyl^®; Vetoquinol, Magny-Vernois, France) was initiated on the day of the surgery.

Experiment

Experimental design

Animals

Twenty 8-week-old inbred male Lewis 1A-haplotype RT1a rats were provided by a certified breeding center (Janvier Labs, Le Genest-Saint-Isle, France). The study protocol was approved by the regional Animal Care and Use Committee (reference: CEEA.201610272139518/AFAPIS 7387) and performed in accordance with European Directive 2010/63/EU. During a 10-day acclimatization period, the animals' health status, behavior, and food intake were examined daily. The rats were weighed twice a week throughout the experiment.

Experimental design

This study is summarized graphically in Figure 2.

FIG. 2.

Study design.

On day 0, 10 animals were irradiated with a single 80 Gy dose to the jaw and 10 animals were used as controls. Three weeks after irradiation, mandibular bone defects were created on each hemimandible. For each of the 20 rats, two defects of different sizes (0, 1, 3, or 5 mm) were created in each mandibular angle. In each group of 10 rats (irradiated and nonirradiated), five animals had a 5 mm defect on one side and no defect (i.e., a “0 mm defect” of) on the other, and the other five had a 3 mm defect on one side and a 1 mm defect on the other. Hence, there were 10 defects of each size and 40 defects in total. Five weeks after the surgical procedure, the animals were euthanized.

Two rats from each of the four subgroups received an intravenous injection of contrast agent, so that vascularization could be studied. The mandibles were harvested for qualitative and quantitative analyses.

The irradiation procedure

The dose of radiation described in the Materials and Methods section was delivered at the Institut de Cancérologie de l'Ouest (Centre de Lutte Contre le Cancer René Gauducheau, Saint Herblain, France).

The surgical procedure

Mandibular defects were created 3 weeks after irradiation, as described in the Materials and Methods section. A full-thickness circular defect (diameter: 0, 1, 3, or 5 mm) was created in the bilateral mandibular angle.

Intracardiac injections of contrast agent and euthanasia

Five weeks after surgery, two rats from each of the four subgroups received an injection of contrast agent. Under general anesthesia with isoflurane, the abdominal cavity and the thorax were opened. A catheter was placed at the apex of the left ventricle and connected via a peristaltic pump to a container filled with warmed (40°C) contrast agent (50% barium sulfate (Micropaque^®; Guerbet, Roissy Charles de Gaulle, France) in 50% phosphate buffered saline (PBS) with 1.5% gelatin (Sigma-Aldrich, St. Louis, MO)). After resection of the right atrium, ∼200 mL of contrast agent was injected into the bloodstream at a constant rate until all the animal extremities turned white. The animals died (under anesthesia) a few minutes after the start of the injection.

Animals that did not receive contrast agent were euthanized with an intracardiac injection of sodium thiopental (Dolethal^®; Vetoquinol). Lastly, the mandibles were harvested and fixed in a solution of paraformaldehyde in 4% PBS for 6 days.

Micro-computed tomography image acquisition and analysis

All samples were imaged with a micro-computed tomography (Micro-CT) system (SkyScan 1272^®; Bruker, Kontich, Belgium), with the following acquisition parameters: 0.5 mm aluminum +0.038 mm copper filter; resolution: 18 μm; voltage: 90 kV; amperage: 111 μA; rotation step: 0.8°. The images were processed offline with NRecon^® (version 1.6.9.8), DataViewer^® (version 1.5.1.2 64-bit) and CTAnalyser^® (version 1.14.4.1) software (all from Bruker micro-CT), to create three-dimensional reconstructions and thus study new bone formation and intraosseous vascularization.

The percentage of new bone formation was defined as the bone volume/tissue volume (BV/TV) ratio in the volume of interest (i.e., the defect's initial volume). The vascular volume (in mm³) was quantified in a volume of interest of 5.4 mm in length (composed of 300 slices) starting immediately distal to the third molar.

Histological examinations

After the samples had been fixed, the two hemimandibles were separated for easier inclusion. Nondecalcified, explanted bone specimens were dehydrated with a graded series of ethanol-xylene solutions and then embedded in a methylmethacrylate resin (Technovit 9100; Kulzer, Weirheim, Germany) that hardens at low temperatures. After various pre-infiltration and infiltration procedures, the samples were embedded in the polymerization mixture at −20°C for 1 week. Serial 6 μm coronal sections were then prepared with a microtome for nondecalcified tissues (Polycut MEICA SM2500; Leica Microsystems, Wetzlar, Germany).

The bone sections were stained with Movat pentachrome. The tartrate-resistant acid phosphatase (TRAP) activity was also measured. Lastly, bone sections were observed with a digital slide scanner at a magnification of × 20 (NanoZoomer 2.0 HT; Hamamatsu, Hamamatsu, Japan).

Scanning electron microscopy and image analysis

Each bone sample was sanded, then carbon-coated on a Desk III (Denton Vacuum, Moorestown, NJ), and finally studied in backscattered electron mode in a scanning electron microscope (Leo 1450 VP; Zeiss, Oberkochen, Germany). Bone defects, bone contours, and bone formation were observed in 2D at the same cutting levels as the specimens for histological staining at magnification 50. The area of newly formed bone was determined by using image analysis software (ImageJ; U.S. National Institutes of Health, Bethesda, MD). The percentage of newly formed bone was the ratio between the bone area and the total area corresponding to the initial defect area.

Statistical analysis

Groups of animals were compared with regard to the bone and vascular volumes obtained from micro-CT and scanning electron microscopy (SEM) data, using the Mann-Whitney test. Statistical and graphical analyses were performed with GraphPad software (version 5.00; GraphPad Software, Inc., San Diego, CA). The threshold for statistical significance was set to p < 0.05.

Results

Animals and clinical observations

None of the animals died before the scheduled euthanasia. Both the irradiation procedure and the surgical procedure were well tolerated. A breakpoint in the weight curve was detected after irradiation and lasted for 1 week in the irradiated group. A similar breakpoint was observed after surgery in all groups (data not shown).

Five or 6 weeks after irradiation, the irradiated animals showed an area of depilation on the cheek.

Micro-computed tomography study

New bone formation

The micro-CT assessment evidenced centripetal bone regrowth in the 1 and 3 mm nonirradiated defects (Fig. 3a, d), whereas no bone regrowth was observed in the 5 mm defects (Fig. 3g). No mandible fractures were observed for 1 mm defects, and only one fracture was observed for 3 mm defects. In contrast, two fractures (40%) were observed for mandibles with a 5 mm defect. In the irradiated group, little bone formation was observed within the bone defects (Fig. 3b, e); jaw fractures were observed in all five animals (100%) with a 5 mm defect (Fig. 3h) and in three of the animals (60%) with a 3 mm defect.

FIG. 3.

Micro-CT 3D reconstructions of mandible defects in irradiated animals (b, e, h) and in nonirradiated animals (a, d, g), and quantification of newly formed bone (c, f). micro-CT, micro-computed tomography. Color images are available online.

We were able to analyze bony formation in 10 hemimandibles with a 1 mm defect, six with a 3 mm defect, and three with a 5 mm defect. The other samples were fractured, which compromised the observation of bone formation. The extent of new bone formation (expressed as BV/TV) tended to decrease as the size of the defect increased. The bone formation was significantly greater in nonirradiated 1 mm bone defects than in irradiated bone defects of the same size (p = 0.0317) (Fig. 3c). The difference was not significant for 3 mm defects (Fig. 3f).

Intraosseous vascularization and bone volume

Before euthanasia, four irradiated animals and four nonirradiated animals were injected with contrast agent. Macroscopically, major destruction of intraosseous vessels was not observed. To limit variations in BV, only hemimandibles with a 1 mm defect or without a defect were included in the qualitative micro-CT analysis. One rat from the nonirradiated group was excluded from this analysis because of poor injection quality. Hence, three mandibles from the nonirradiated group and four from the irradiated group were assessed. Bone and vessel volumes were significantly lower in the irradiated group than in the nonirradiated group (p = 0.0286) (Fig. 4).

FIG. 4.

Quantification of bone and vessel volumes in micro-CT reconstructions of the jaw.

SEM analysis

Qualitative SEM analysis

In the irradiated group, the SEM study revealed the presence of large, multilobed gaps (Fig. 5a) that were suggestive of Dambrain periosteocytic lysis.⁴ There was no new bone formation in the defects, regardless of the size. The defects' edges were flat, with no convexity toward the center of the defect (Fig. 5c). In contrast, the bone microarchitecture was maintained in the nonirradiated samples (Fig. 5b). Likewise, the cortical bone in the nonirradiated samples was smooth, and all the defects contained newly formed bone with a medullary aspect (Fig. 5d, in yellow).

FIG. 5.

(a, b) SEM images of bone architecture in irradiated and nonirradiated animals. A large number of polycyclic gaps can be seen (white stars); these features are suggestive of periosteocytic lysis, described by Dambrain as being pathognomonic for ORN.⁴ (c, d) SEM images of bone defects in irradiated and nonirradiated animals, and quantification of new bone formation (in yellow) (e). ORN, osteoradionecrosis; SEM, scanning electron microscopy. Color images are available online.

Quantitative SEM analysis

In the 1 mm defect group, two hemimandibles were excluded from the quantitative analysis because the section did not pass through the defect. Therefore, eight samples were included. In the 3 mm defect group, three irradiated mandibles were fractured and one section did not pass through the defect. In the nonirradiated group, one explant could not be used and so five were included in the analysis. All the 5 mm defects in the irradiated group were fractured, preventing an analysis of this size of defect. For 1 mm defects, the bone ingrowth ratio was significantly higher in nonirradiated bone than in irradiated bone (p = 0.0286) (Fig. 5e). The same type of difference was found for 3 mm bone defects but was not statistically significant.

Histological examinations

In the nonirradiated samples, newly formed bone (surrounded by a large number of medullar cells) was observed on the edges of the defect (Fig. 6a, b). The bone marrow showed high cellularity and was highly vascularized (Fig. 6c). The edges of the defects were smooth, and the TRAP activity was low (Fig. 6h). In the irradiated group, no bone regrowth was observed (Fig. 6d, e). The defects were filled with acellular, possibly fibrotic tissue. The bone marrow showed low cellularity. Multinucleated cells (possibly osteoclasts) were observed on the edges of the defects (Fig. 6f). Moreover, the level of TRAP activity was higher in the irradiated bone than in the nonirradiated bone. The level was particularly high in the multilobed gaps observed in the SEM study (Fig. 6g).

FIG. 6.

Histological assessments with Movat pentachrome staining in irradiated animals (e, f) and nonirradiated animals (b, c), and the correlation with the qualitative SEM study (a, and d). In the nonirradiated group, new bone is formed around a cell-rich bone marrow (black stars) and on the edges of the defects (black arrows). TRAP activity (g) is more intense on the edges of the irradiated jaws (red arrows) than on the edges of the nonirradiated jaws (h); likewise, Movat staining revealed the presence of multinucleated cells, which might be osteoclasts (white arrows). TRAP, tartrate-resistant acid phosphatase. Color images are available online.

Discussion

The histological and biomechanical effects of bone irradiation appear at doses above 60 Gy. As the last stage in bone tissue degradation, ORN is characterized by the occurrence of gaps, fibrosis, and necrotic areas. In our present laboratory, the sequelae of irradiation have typically been studied in the rat posterior limb model. However, this model is limited by the use of long bones (whose embryological origin differs from that of the jaw) and by the presence of a mixed ossification process.¹⁰

Although several research groups^11–13 have studied animal models of mandibular ORN induced by low-energy X-rays, the irradiation was usually diffuse; in this study, however, we used targeted irradiation. Other researchers have used the high-energy X-rays techniques⁹ commonly used to treat patients in the clinic and that are believed to harm the bone less than low-energy X-rays do at an equivalent dose level. Likewise, a number of models have been based on the fractioned radiotherapy techniques used in the clinic,^11–13 although implementation with animals is complicated by the need for general anesthesia and repeated transport—both of which can induce stress. Therefore, the results will vary according to the irradiation dose and protocol.

In an earlier model,¹⁴ rats received a single 80 Gy dose to their hind limbs, and pathognomonic ORN lesions were observed. However, specific lesions (such as ORN) were not observed at a lower (50 Gy) dose—even though fractures were just as frequent. Hence, we chose to keep a single 80 Gy dose in this study.

The critical-sized circular defect in the mandibular angle is often used in experiments that do not involve irradiation.^15–20 Another model involves segmental mandibulotomy stabilized by an internal fixator. A critical-sized defect will not heal spontaneously^21,22 and so mandibular regeneration requires a therapeutic intervention. According to the literature, the critical size of the bone defect in a healthy jaw is around 4 or 5 mm. This model offers an ideal platform for testing the osteoconductive properties of bone substitutes²³ and membranes in the presence^17–19 or absence of growth factors.²⁴

Lastly, it is known that trauma (especially tooth extraction) in an irradiated area can induce ORN.²⁵ Therefore, some models are based on molar extraction a few days or a few weeks after irradiation.^9,26,27 However, the mucosal sequelae of irradiation prevent the defect from being filled with biomaterial in a lasting manner.

This study is the first to have assessed a rat model of mandibular irradiation sequelae with a circular submandibular defect in an irradiated area. This model avoids mucosal lesions and thus enables the evaluation of bone remodeling with tissue engineering techniques. We sought to model the adverse events associated with bone irradiation and so we observed the clinical and histological characteristics of ORN. Seven days after irradiation, we observed a break in the weight curve in the irradiated group—suggesting oral mucositis. Five weeks after irradiation, the irradiated area had shed its hairs.

A micro-CT analysis showed a loss of BV after 80 Gy irradiation and thus confirmed the direct impact on bone microarchitecture. Further, a large number of fractures were observed during the explantation and the micro-CT acquisitions. Fractures of the inferior mandibular border in nonirradiated areas have already been reported but were infrequent¹⁷; this suggests that a critical defect in an irradiated area is smaller than a critical defect in a nonirradiated area. Likewise, Tchanque-Fossuo et al.²⁸ described a dose–response effect for the deterioration of the murine mandible's biomechanical properties after irradiation.

In this study, we found that bone regrowth decreased as the size of the defect increased. However, there were no significant differences between the defect conditions. In nonirradiated areas, almost no bone formed in 5 mm defects—confirming the critical-sized defect reported in the literature. Five weeks after surgery, partial bone regrowth was observed in nonirradiated 1 and 3 mm bone defects. Although this timeframe was short to observe complete healing, it was enough to observe bone regeneration. Our histological data showed increased osteoclastic activity in the irradiated area—suggesting that the various types of bone cell differ in their radiosensitivity. Osteoblasts appear to be the most radiosensitive, followed by osteocytes and then osteoclasts.²⁹

We showed a decreased intraosseous vascular network in the irradiated animals, in accordance with previous studies also using micro-CT^8,30 or Doppler.³¹ In vitro, Dudziak et al. showed that vascular endothelial growth factor expression, essential for neoangiogenesis, was decreased beyond an 8-Gy irradiation.³² The pathologic effects of radiation are mediated in part through endothelial lesions of capillaries and distal vessels. It results in progressive obliterative endarteritis, leading to decreased blood flow and fibrosis. Mandibular bone is particularly prone to radiation damage due to its tenuous vascular supply from the inferior alveolar artery, limited collateral circulation, and compact bony structure.^33,34 This set of anatomical features could explain the pathological fractures recorded in the study.

According to our micro-CT and SEM studies, a 1 mm defect appeared to be too small for accurate quantification of bone regrowth: It closed quickly after being filled with biomaterials. Therefore, a 3 mm defect appears to be a good compromise for a critical-sized defect: Bone regrowth was noted in a healthy jaw but not at all in an irradiated jaw.

This study had some limitations. First, the number of animals in each group was low; however, this was in accordance with current guidelines on reducing the use of laboratory animals. Second, we created a circular defect that does not resemble the naturally occurring bone loss induced by irradiation. Third, our follow-up period was limited to 5 weeks; clinically, the sequelae of bone irradiation develop over several months. However, rodent bone regenerates more quickly than human bone. Despite these differences, our model reproduced alterations in bone healing. From a cost perspective, a small animal model is an advantage. In the future, we, nevertheless, intend to develop a large animal model.

Our study also had several strengths. For example, we used several imaging techniques to assess bone remodeling. Scanning electron microscopy is the gold standard analytical technique in our lab.^14,35,36 However, SEM measurements are limited to a single bone section, whereas new bone grows in all three dimensions. Micro-CT has a lower spatial resolution than SEM but seems to be of value in the assessment of bone healing in three dimensions.

Conclusion

This study highlighted the impact of radiation on the intraosseous vasculature and bone microarchitecture of the rat mandible. With a view to studying bone remodeling, the critical-sized defect in the irradiated jaw appears to be 3 mm. Despite practical and time constraints, we demonstrated that our model effectively reproduces the sequelae of irradiation of the jaw and may be a useful research tool. Further studies of larger groups of rats will be necessary to obtain consistent quantitative data in this new model and to test tissue-engineering strategies for mandibular regeneration after irradiation.

Footnotes

Acknowledgment

The authors thank Sophie Chiavassa from the Institut de Cancérologie de l'Ouest (Saint Herblain, France) for her assistance with the irradiation procedures.

Ethical Approval

All applicable international, national, and/or institutional guidelines for the care and use of animals were followed. The protocols were approved by the Pays-de-Loire Animal Care and Use Committee (reference: CEEA.201610272139518/AFAPIS 7387), in accordance with the European directive 2010/63/EU.

Disclosure Statement

No competing financial interests exist.

Funding Information

This study was funded by grants from the Côtes d'Armor and Loire-Atlantique branches of the La Ligue Contre Le Cancer foundation, from the Les Gueules Cassées foundation, and from IXBONE (ANR-18-CE18-0014).

References

Ferlay

, Soerjomataram

, Dikshit

, et al. Cancer incidence and mortality worldwide: sources, methods and major patterns in GLOBOCAN 2012. Int J Cancer, 136, E359, 2015.

O'Sullivan

, Rumble

R.B.

, Warde

.; and Members of the IMRT Indications Expert

Panel.

Intensity-modulated radiotherapy in the treatment of head and neck cancer. Clin Oncol (R Coll Radiol), 24, 474, 2012.

Marx

R.E.

Osteoradionecrosis: a new concept of its pathophysiology. J Oral Maxillofac Surg, 41, 283, 1983.

Dambrain

[The pathogenesis of osteoradionecrosis]. Rev Stomatol Chir Maxillofac, 94, 140, 1993.

Fenner

, Park

, Schulz

, et al. Validation of histologic changes induced by external irradiation in mandibular bone. An experimental animal model. J Craniomaxillofac Surg, 38, 47, 2010.

Delanian

, Chatel

, Porcher

, Depondt

, and Lefaix

J.L.

Complete restoration of refractory mandibular osteoradionecrosis by prolonged treatment with a pentoxifylline-tocopherol-clodronate combination (PENTOCLO): a phase II trial. Int J Radiat Oncol Biol Phys, 80, 832, 2011.

Rice

, Polyzois

, Ekanayake

, Omer

, and Stassen

L.F.

The management of osteoradionecrosis of the jaws—a review. Surgeon, 13, 101, 2015.

Michel

, Blery

, Pilet

, et al. Micro-CT analysis of radiation-induced osteopenia and bone hypovascularization in rat. Calcif Tissue Int, 97, 62, 2015.

Bléry

, Espitalier

, Hays

, et al. Development of mandibular osteoradionecrosis in rats: importance of dental extraction. J Craniomaxillofac Surg, 43, 1829, 2015.

10.

Parada

, and Chai

Mandible and Tongue Development. Curr Top Dev Biol, 115, 31, 2015.

11.

Williamson

R.A.

An experimental study of the use of hyperbaric oxygen to reduce the side effects of radiation treatment for malignant disease. Int J Oral Maxillofac Surg, 36, 533, 2007.

12.

Tchanque-Fossuo

C.N.

, Monson

L.A.

, Farberg

A.S.

, et al. Dose-response effect of human equivalent radiation in the murine mandible: part I. A histomorphometric assessment. Plast Reconstr Surg, 128, 114, 2011.

13.

Donneys

, Nelson

N.S.

, Perosky

J.E.

, et al. Prevention of radiation-induced bone pathology through combined pharmacologic cytoprotection and angiogenic stimulation. Bone, 84, 245, 2016.

14.

Michel

, Blery

, Henoux

, et al. Bone marrow cell extract promotes the regeneration of irradiated bone. PLoS One, 13, e0196145, 2017.

15.

Schortinghuis

, Ruben

J.L.

, Meijer

H.J.

, Bronckers

A.L.

, Raghoebar

G.M.

, and Stegenga

Microradiography to evaluate bone growth into a rat mandibular defect. Arch Oral Biol, 48, 155, 2003.

16.

Van Leeuwen

A.C.

, Van Kooten

T.G.

, Grijpma

D.W.

, and Bos

R.R.

In vivo behaviour of a biodegradable poly(trimethylene carbonate) barrier membrane: a histological study in rats. J Mater Sci Mater Med, 23, 1951, 2012.

17.

Gielkens

P.F.

, Schortinghuis

, de Jong

J.R.

, Raghoebar

G.M.

, Stegenga

, and Bos

R.R.

Vivosorb, Bio-Gide, and Gore-Tex as barrier membranes in rat mandibular defects: an evaluation by microradiography and micro-CT. Clin Oral Implants Res, 19, 516, 2008.

18.

Kowalczewski

C.J.

, Tombyln

, Wasnick

D.C.

, et al. Reduction of ectopic bone growth in critically-sized rat mandible defects by delivery of rhBMP-2 from kerateine biomaterials. Biomaterials, 35, 3220, 2014.

19.

Jiang

, Zhao

, Wang

, et al. Mandibular repair in rats with premineralized silk scaffolds and BMP-2-modified bMSCs. Biomaterials, 30, 4522, 2009.

20.

Lattanzi

, Parrilla

, Fetoni

, et al. Ex-vivo transduced autologous skin fibroblasts expressing human Lim Mineralization Protein-3 efficiently form new bone in animal models. Gene Ther, 15, 1330, 2008.

21.

Kaban

L.B.

, Glowacki

, and Murray

J.E.

Repair of experimental mandibular bony defects in rats. Surg Forum, 30, 519, 1979.

22.

Hollinger

J.O.

, and Kleinschmidt

J.C.

The critical size defect as an experimental model to test bone repair materials. J Craniofac Surg, 1, 60, 1990.

23.

Kaban

L.B.

, and Glowacki

Induced osteogenesis in the repair of experimental mandibular defects in rats. J Dent Res, 60, 1356, 1981.

24.

Saadeh

P.B.

, Khosla

R.K.

, Mehrara

B.J.

, et al. Repair of a critical size defect in the rat mandible using allogenic type I collagen. J Craniofac Surg, 12, 573, 2001.

25.

Chrcanovic

B.R.

, Reher

, Sousa

A.A.

, and Harris

Osteoradionecrosis of the jaws—a current overview—part 2: dental management and therapeutic options for treatment. Oral Maxillofac Surg, 14, 81, 2010.

26.

Tamplen

, Trapp

, Nishimura

, et al. Standardized analysis of mandibular osteoradionecrosis in a rat model. Otolaryngol Head Neck Surg, 145, 404, 2011.

27.

Cohen

, Nishimura

, Tamplen

, et al. Animal model of radiogenic bone damage to study mandibular osteoradionecrosis. Am J Otolaryngol, 32, 291, 2011.

28.

Tchanque-Fossuo

C.N.

, Monson

L.A.

, Farberg

A.S.

, et al. Dose-response effect of human equivalent radiation in the murine mandible: Part II. A biomechanical assessment. Plast Reconstr Surg, 128, 480e, 2011.

29.

Jegoux

, Malard

, Goyenvalle

, Aguado

, and Daculsi

Radiation effects on bone healing and reconstruction: interpretation of the literature. Oral Surg Oral Med Oral Pathol Oral Radiol Endod, 109, 173, 2010.

30.

Deshpande

S.S.

, Donneys

, Farberg

A.S.

, Tchanque-Fossuo

C.N.

, Felice

P.A.

, and Buchman

S.R.

Quantification and characterization of radiation-induced changes to mandibular vascularity using micro-computed tomography. Ann Plast Surg, 72, 100, 2014.

31.

Okunieff

, Wang

, Rubin

, Finkelstein

J.N.

, Constine

L.S.

, and Ding

Radiation-induced changes in bone perfusion and angiogenesis. Int J Radiat Oncol Biol Phys, 42, 885, 1998.

32.

Dudziak

M.E.

, Saadeh

P.B.

, Mehrara

B.J.

, et al. The effects of ionizing radiation on osteoblast-like cells in vitro. Plast Reconstr Surg, 106, 1049, 2000.

33.

Farberg

A.S.

, Jing

X.L.

, Monson

L.A.

, et al. Deferoxamine reverses radiation induced hypovascularity during bone regeneration and repair in the murine mandible. Bone, 50, 1184, 2012.

34.

Tchanque-Fossuo

C.N.

, Donneys

, Sarhaddi

, et al. The effect of Amifostine prophylaxis on bone densitometry, biomechanical strength and union in mandibular pathologic fracture repair. Bone, 57, 56, 2013.

35.

Lerouxel

, Weiss

, Giumelli

, et al. Injectable calcium phosphate scaffold and bone marrow graft for bone reconstruction in irradiated areas: an experimental study in rats. Biomaterials, 27, 4566, 2006.

36.

Espitalier

, Vinatier

, Lerouxel

, et al. A comparison between bone reconstruction following the use of mesenchymal stem cells and total bone marrow in association with calcium phosphate scaffold in irradiated bone. Biomaterials, 30, 763, 2009.