- Email: [email protected]
Animal model of radiogenic bone damage to study mandibular osteoradionecrosis
Available online at www.sciencedirect.com
American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291 – 300 www.elsevier.com/locate/amjoto
Animal model of radiogenic bone damage to study mandibular osteoradionecrosis☆,☆☆ Marc Cohen, MDa , Ichiro Nishimura, DDS, DMSc, DMDb , Matthew Tamplen, BSa , Akishige Hokugo, DDS, PhDb , John Beumer, DDS, MSb , Michael L. Steinberg, MDc , Jeffrey D. Suh, MDa , Elliot Abemayor, MD, PhDa , Vishad Nabili, MDa,⁎ a
Division of Head and Neck Surgery, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA UCLA School of Dentistry, Weintraub Center for Reconstructive Biotechnology, Los Angeles, CA, USA c Department of Radiation Oncology, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA Received 27 April 2010 b
Abstract
Objective: The objective of the study was to create an animal model to study mandibular osteoradionecrosis (ORN) using high–dose rate (HDR) brachytherapy. Methods: Ten Sprague-Dawley male rats were used in this study. Six rats received a single dose of 30 Gy using an HDR remote afterloading machine via a brachytherapy catheter placed along the left hemimandible. The remaining 4 rats served as controls with catheter placement without radiation (sham). On the day following irradiation or sham, all 3 left mandibular molars were atraumatically extracted. Twenty-eight days after irradiation, mandibles were examined using nondecalcified histology with sequential fluorochrome labeling, decalcified histology, and micro–computed tomography scanning. Results: Irradiated rats demonstrated exposed bone at the extraction sockets, whereas the control animals had complete mucosalization. Alopecia was also seen in the irradiated group. Both histologic and radiologic analyses of the mandible specimens demonstrated a reduction in bone formation in the radiated mandibles as compared with controls. Conclusions: Our HDR brachytherapy model incorporating postradiation dental extractions has successfully demonstrated reproducible radiogenic mandibular bone damage analogous to the clinical ORN. Although clinical criteria continue to be used today in describing ORN, this model can serve as a platform for future studies to define ORN and delineate its pathogenesis. © 2011 Elsevier Inc. All rights reserved.
1. Introduction ☆
Presented at the American Academy of Otolaryngology Head and Neck Surgery Annual Meeting, October 6, 2009, San Diego, CA. ☆☆ Justification for additional authors: This project entailed expertise from multiple areas with each author playing a key role in providing input and resources including Head and Neck Surgical Oncology (Drs Cohen, Suh, and Abemayor), Microvascular Reconstruction (Dr Nabili), Dental biology (Drs Nishimura, Hokugo, and Beumer), and Radiation access and physicist input (Dr Steinberg). ⁎ Corresponding author. Division of Head and Neck Surgery, David Geffen School of Medicine at UCLA, 10833 LeConte Ave, CHS RM 62132, Los Angeles, CA 90095-1624, USA. Tel.: +1 310 709 7970; fax: +1 310 206 1393. E-mail address: [email protected] (V. Nabili). 0196-0709/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjoto.2010.06.001
Despite more target-specific advances in radiation technology for the treatment of head and neck cancer, including intensity-modulated radiation therapy and high– dose rate (HDR) brachytherapy, patients continue to suffer from adverse effects of this treatment modality. Significant morbidity can ensue from radiation therapy months to years after treatment. One specific adverse effect of radiation treatment for head and neck cancer patients is osteoradionecrosis (ORN) of the mandible. Being cured of their head and neck cancers, patients with advanced mandibular ORN may still succumb to extensive microvascular reconstructive surgery once reserved initially as an oncologic treatment
292
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
option. These heroic measures may still prove futile because ORN can recur and little is still known about the disease. ORN of the mandible is a pathologic condition that has been commonly described by clinical criteria, whereas a true definition has been elusive as the pathogenesis of this disease still remains undefined. Over the past 35 years, several authors have attempted to define ORN. Beumer et al [1] stated that ORN occurs “when bone in the radiation field [is] exposed for at least 2 months in the absence of local neoplastic disease.” Marx [2] defined ORN as “an area greater than one centimeter of exposed bone in a field of irradiation that had failed to show any evidence of healing for at least 6 months.” Hutchinson [3] defined ORN to only require 2 months of exposed bone, whereas Harris [4] used the 3-month marker. Despite the variability in these clinical descriptions, several commonalities are present in clinically defining ORN: (1) exposed bone for at least 2 to 6 months, (2) a history of radiation therapy to the region of exposed bone, (3) the presence of necrotic or devitalized bone, and (4) no evidence of tumor recurrence. Patients with mandibular ORN often have radiologic assessment in the form of computed tomography (CT) and/or panorex. Limitations in radiologic analysis of ORN are not due to technology but again more so due to lack of defined radiologic criteria for mandibular ORN. Common radiographic findings for mandibular ORN include but are not limited to radiolucencies, bony sequestration, and/or pathologic fracture(s). Posttreatment histopathologic analysis often shows chronic inflammation and necrotic bone, yet the pathogenesis is unclear. To investigate mandibular ORN further, we set out to first create an animal model as a platform to demonstrate the aforementioned criteria used today. Previous attempts at creating a model of ORN in a rat have been successful to some degree. However, these models were lacking in certain respects. Most of the models involved whole head irradiation, and this is not analogous to current radiation treatment for head and neck cancer. Recently, Niehoff et al [5] described using HDR brachytherapy to create an experimental model of radiogenic bone damage in a rat. The HDR brachytherapy is a technique that uses a relatively intense source of radiation, typically iridium-192, to deliver a therapeutic dose of radiation through temporarily placed catheters at a specified location such as a unilateral rat mandible. The authors successfully demonstrated mandibular bone damage after HDR radiation. Mandibular ORN, however, was not described or defined. It is known that dental trauma, especially tooth extractions in a radiated field, can increase the incidence of ORN [6]. Several previous animal models involved maxillary tooth extraction, rather than mandibular. This is not as clinically relevant because it is well established that ORN predominately occurs in the mandible and not in the maxilla [7]. Most studies used histologic examination of the specimens as the only evaluation tool. In reality, ORN is
characterized by clinical examination with the aid of histologic and radiologic assessments. The goal of our study was to improve and update the previous models of ORN not only using the most current technology, but incorporating the current knowledge of the disease process in our model. The rat model that has been designed involves irradiating rat mandibles with an extremely targeted device, extracting mandibular molars after radiation, and assessing outcomes using high-speed digital photography, micro-CT, and fluorescent and basic microscopy for histologic analysis. A thoroughly effective treatment plan with minimal morbidity does not exist for ORN and most likely stems from a lack of understanding of the pathophysiology of the disease process. It is for this reason that we have attempted to create an updated rat model of ORN. Our hope is to establish a reproducible and clinically relevant animal model of post–dental extraction radiogenic bone damage from which to define and study the pathogenesis of mandibular ORN.
2. Materials and methods 2.1. Experimental design Approval for the research protocol was obtained from the University of California Los Angeles Chancellor's Animal Research Committee. Ten male Sprague-Dawley rats of 7 weeks of age (192–224 g) were used in this study. Rats were obtained from Charles River Laboratories International, Inc (Wilmington, MA). The rats were kept in pairs and given a standard pelleted rodent diet and water ad libitum in accordance with the requirements of the United States of America Animal Welfare Act and the Public Health Service Policy on Humane Care and Use of Laboratory Animals. The 10 rats were divided into 2 groups: group 1—left mandibular irradiation followed by left molar extractions (n = 6) and group 2—left mandibular sham irradiation (placement of HDR catheter only) followed by left molar extractions (n = 4). 2.2. Irradiation Under inhalational isoflurane anesthesia, the left cheek skin of the rat was shaved and prepared with betadine. A sterile plastic HDR catheter (Alpha Omega Services, Long Beach, CA) was implanted along the lateral aspect of the left mandibular body. The catheter was inserted via a stab incision made just lateral to the left central incisor in the gingivolabial sulcus. The catheter was advanced submucosally along the inferior edge of the mandible. The catheter exited the oral cavity via a second stab incision at the posterior border of the mandibular ramus and advanced just inferior to the external auditory meatus. This advancement of the external portion of the catheter was consistently 10 mm and served as a reproducible length along with the above-
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
293
microscopy using UV light (Olympus BX51 Research Microscope, Tokyo, Japan) was used to analyze bone apposition at mandibular tooth sockets. 2.5. Killing of animals The animals were killed 28 days post–tooth extraction. Mandibles were extracted and photographed using a Canon Rebel Ti DSLR camera (Canon, Tokyo, Japan). Mandibles were then placed in 70% isopropyl alcohol. 2.6. Radiologic analysis Fig. 1. The HDR brachytherapy catheter is inserted along the lateral border of the left mandibular body.
mentioned landmarks. The external distal portion of the catheter was secured to the non–hair-bearing neck and cheek skin using Steri-Strip adhesive skin closure tape (3M, St Paul, MN) (Fig. 1). A 3-dimensional treatment plan for irradiation was devised that encompassed a depth of 5 mm from a point 1 cm from the distal aspect of the catheter tip. The center of the radiation field was placed lateral to the midline of the left mandibular body. A single dose of radiation was applied with an HDR afterloading remote machine (GammaMed 12it; Varian Medical Systems, Charlottesville, Inc, VA) providing 30 Gy with predefined isodose lines (Fig. 2). The catheter was removed after irradiation was complete in group 1 and immediately removed after placement in group 2 without radiation. 2.3. Tooth extraction Under inhalational isoflurane anesthesia, all rats underwent atraumatic extraction of all 3 of the left mandibular molars 1 day following irradiation (or sham catheter placement). Extreme care was taken to avoid breaking the tooth roots from the crown. The teeth were examined after extraction with 3.5× surgical magnifying loupes to evaluate for completeness of extraction. Postoperative pain management was achieved with buprenorphine (Buprenex; Reckitt Benckiser Healthcare Ltd, Hull, England) at a dose of 0.03 mg/kg given subcutaneously. The diets were supplemented with sliced apples for the first 3 days after molar extractions.
CT of the extracted mandibles was performed using a desktop cone-beam micro-CT scanner (μCT 40 Scanco Medical, Brüttisellen, Switzerland). Three-dimensional reconstructions and volume analysis of the microradiographs were accomplished using μCT Evaluation Software v6.0 (Scanco Medical). The mandibular volume that was analyzed encompassed the region from the most superior aspect of the tooth socket where both medial and lateral cortices were seen (in an axial view) down to the nadir of the incisor root. Because of the relatively large volume that the incisor root socket normally encompasses in a rat mandible, these sockets were subtracted from volume calculations to limit the focus to the molar teeth regions. Bone volume (BV) and total volumes (TVs) were measured, and the ratio of these numbers was analyzed. 2.7. Histologic analysis After micro-CT analysis was performed, both the radiated and control specimens were randomized into 2 groups to allow for non–decalcified-based fluorochrome analysis and decalcified-based paraffin embedded analysis, respectively. The first group consisted of 4 nondecalcified specimens (2 irradiated and 2 controls). These samples were taken from
2.4. Fluorochrome labeling Sequential different-colored fluorochrome labeling of mineralizing bone was performed [8]. Two fluorochromes were used in this experiment: calcein (green fluorescence) (1% in 2% NaHCO3 solution, 20 mg/kg body weight; Sigma Aldrich, St Louis, MO) and demeclocycline (orange fluorescence) (1% in 2% NaHCO3 solution, 20 mg/kg body weight; Sigma Aldrich). All rats underwent intraperitoneal injection of calcein 2 days before irradiation and demeclocycline 2 days before being killed. Fluorescent
Fig. 2. The isodose lines of the 30-Gy radiation field as calculated by the GammaMed HDR brachytherapy device.
294
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
Fig. 3. Clinical manifestations of irradiation. Alopecia of the facial skin in the irradiated rat (A) and complete hair regrowth in the nonirradiated rat (B).
70% isopropyl alcohol and immediately embedded in methyl methacrylate. Five-micrometer–thick sagittal sections of left hemimandibles were cut on a Jung polycot microtome (Reichert-Jung, Heidelberg, Germany) and either stained with toluidine blue or left unstained for fluorochrome analysis. Bone formation and resorption were analyzed at 2 standardized sites using a Pyser-SGI 5-mm stage micrometer scale (Pyser-SGI, Kent, United Kingdom) and an infinity eyepiece with dimensions of 10 mm by 0.1 mm divisions (Fisher Scientific, Pittsburgh, PA) at 20× magnification. Bone mineralization was detected via fluorescent microscopy in the center of the tooth extraction socket immediately above inferior alveolar nerve at a standardized location. Bone mineral apposition rate (MAR) was calculated along the lower cortical rim of the mandible directly below the posterior molar extraction socket and the inferior alveolar nerve at a second standardized location. All
fluorochrome analysis was performed using an Olympus BX51 Research Microscope. The 5 remaining specimens (3 irradiated and 2 controls) were decalcified in 10% EDTA (pH 7.4) for 10 days, fixed in formalin overnight, and paraffin embedded by the UCLA Translational Pathology Core Laboratory. Four-micrometer sagittal sections of the left hemimandibles were stained with hematoxylin and eosin (H&E) and viewed for qualitative histologic analysis using camera-assisted light microscopy (Nikon Labophot-2; Nikon, Tokyo, Japan). 3. Results 3.1. Clinical Nine rats survived the study period. One rat from the irradiated group died 12 days after tooth extraction and was
Fig. 4. Clinical manifestations of irradiation. Retardation of central incisor growth in the irradiated rat (A) and uninhibited growth in the nonirradiated rat (B).
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
295
Fig. 5. Gross pathologic evaluation of rat mandibles. Exposed bone at the site of extraction in the irradiated rat (A) and complete wound mucosalization in the nonirradiated rat (B).
excluded from further evaluation; the cause of death was attributed to malnutrition. The remainder of the irradiated rats survived the study time of 4 weeks and continued to gain weight. However, the irradiated group average weight gain was significantly less than the control group (124.8 vs 196.8 g, P = .008). The rats from the irradiated group demonstrated adverse effects of target-specific HDR radiation as evidenced by unilateral left cheek skin alopecia in 100% of the group, whereas all of the animals in the control group had complete hair regrowth (Fig. 3). Likewise, ipsilateral incisor growth was also retarded in the radiated animals as compared with nonirradiated animals (Fig. 4). Evaluation of the intraoral inferior alveolar ridges using high-resolution digital photography demonstrated exposed bone at the site of dental extraction in 100% of the irradiated rats. Complete wound mucosalization was seen in all of the animals in the control group (Fig. 5). 3.2. Micro-CT evaluation Before using the analysis software for the microradiographs, the raw micro-CT data were visually analyzed by 2 independent observers to assess for efficacy of tooth extractions. Two rats (one from each group) had evidence of more than 2 retained tooth roots. Because of the variability in analysis caused by retained tooth elements, these 2 rats were excluded from microradiograph volumetric analysis. As described in the methods section, BV was measured in relation to the TV of analyzed mandible and expressed as a Table 1 Micro-CT data of hemimandible volume analysis
Rats 4 and 9 (shaded) were excluded from statistical analysis because of retained tooth roots.
ratio (BV/TV) (Table 1). The average BV/TV in the irradiated group was 0.559 as compared with the control group ratio of 0.745 (P = .009). Evaluation of the 3-dimensional CT reconstructions of the rat mandibles did allow for ultrastructural viewing of bone regrowth in the tooth extraction socket sites and qualitative comparisons between radiated and nonradiated samples (Fig. 6). Lack of bone in the exposed sockets in the irradiated mandibles is seen in these CT reconstructions. 3.3. Fluorescent microscopic analysis Bone MAR was measured at the second standardized location site along the inferior border of the mandible below the central incisor and inferior alveolar nerve for the 4 nondecalcified specimens (2 irradiated and 2 controls). This area demonstrated lamellar nontraumatized bone with presence of both calcein green and demeclocycline orange labeling in all samples. MAR for each sample were calculated by dividing the distance of bone deposition between the 2 labels by a total of 28 days between fluorochrome injections. Both irradiated samples showed far less bone deposition, averaging an MAR of 6.23 µm/d compared with a 12.5-µm/d average for the control samples (Fig. 7 and Table 2). The tooth extraction sockets in all of the animals demonstrated a disarray of fluorochrome labeling that prevented any quantitative analysis. Qualitatively, it was noted that the second label, demeclocycline (orange), was not present in any of the radiated extraction sites while still being present along the inferior cortical border of all mandibles as mentioned above. The control group showed clear evidence of double labeling within the extraction sockets indicative of new bone formation postextraction. Qualitative analysis of toluidine-stained slides from the nondecalcified specimens further demonstrated a lack of bone present in the extraction sites of irradiated samples when compared with controls (Fig. 7).
296
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
Fig. 6. Three-dimensional micro-CT reconstructions of rat mandibles. Irradiated mandible (A) with a lower bone density at extraction sites vs nonirradiated mandible (B).
3.4. Decalcified histologic analysis The radiated specimens showed a diminution of bone formation, an overwhelming presence of fibrosis and inflammation, and an unusually high presence of osteoclasts within the extraction sockets when compared with controls. Moreover, the control samples specifically had increased bone present along with many osteoblasts and only minimal inflammation and fibrosis (Fig. 8).
4. Discussion ORN of the mandible has been recognized since 1922 [9]. Despite the awareness of this disease for more than 85 years, little is still known about its pathogenesis. The ambiguity about defining mandibular ORN has also limited its treatment options to clinical judgment with minimal objective criteria. The majority of patients with mandibular ORN will succumb to advanced-stage disease requiring microvascular reconstruction. In a recent series by the senior author (VN), these patients had increased woundrelated complications and risk for continuing recurrent disease (accepted for publication, Otolaryngol Head Neck Surg April 2010). A lack of understanding of the pathophysiology of the disease process is the reason that we have created this postextraction rat model of radiogenic bone damage to serve as a platform to further study and define mandibular ORN. Theories on the pathophysiology of ORN have evolved over the last 40 years, with different treatments proposed along the way. Originally, Meyer [10] believed that, in the setting of irradiated bone, a traumatic event could lead to a superimposed infection similar to osteomyelitis. This led to conservative treatment measures including local wound irrigation, oral hygiene improvement, curettage,
debridement, sequestrectomy, bone filling, and/or longterm antibiotic use. Approximately a decade later, Marx [2] proposed his 3-H theory of ORN where he described a sequence of events starting with radiation, formulation of hypoxic-hypocellularhypovascular (3-H) tissue, then tissue breakdown, and finally a chronic nonhealing wound as ORN. Marx showed that ORN is a problem of wound healing due to a complex metabolic and tissue homeostatic oxygen deficiency, leading to the use of hyperbaric oxygen (HBO) for treatment. More recently, Delanian and Lefaix [11] proposed their fibroatrophic theory to explain the radiation-induced changes that occur with ORN. In this theory, ORN lesions are ultimately due to an imbalance of bone resorption and bone deposition. Bone atrophy occurs in the setting of extensive fibrosis. The combination of osteoblast death in the mandible after irradiation, failure of the osteoblasts to repopulate, and the excessive proliferation of myofibroblasts results in the bony matrix being replaced by fibrous tissues. On the basis of this theory, pentoxyifylline and tocopherol have emerged as possible adjuvant treatments for ORN with promising results, leading to possible future clinical trials [12]. With a pathobiology that is undefined, a true definition of ORN cannot be made yet. Current attempts at defining ORN are based on clinical criteria describing radiogenic bone damage. Actual cellular necrosis after radiation, the types of cells involved, and the sequence of events have not been fully elucidated by any model to date. As such, some would argue that our current model may present what we know clinically as ORN; yet we believe our model has successfully demonstrated post–dental extraction radiogenic bone damage. In reviewing prior animal models, the terminology of actual ORN has thus been used loosely and oftentimes incorrectly in place of radiogenic bone damage. Many previous attempts at creating a model of ORN in the rat
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
297
Fig. 7. Double labeling (calcein green, demeclocycline orange) seen on the inferior cortical rim of radiated (A) compared with control (B) mandibles, with radiated samples demonstrating a reduction in mineral apposition. Radiated tooth extraction socket sites (C) further demonstrate a reduction in new bone formation with minimal labeling (calcein green only) and no second label as compared with increased presence of both labels in the controls (D). Nondecalcified toluidine blue–stained sections from tooth extraction socket sites show increased fibrosis and minimal bone in radiated samples (E) when compared with controls (F).
298
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
Table 2 Fluorochrome dynamic analysis of bone apposition rate at standardized site Animal no.
Distance between 2 labels (µm)
MAR (µm/d) total days = 28
Animal 1 (radiated) Animal 3 (radiated) Animal 9 (control) Animal 10 (control)
160 190 360 340
5.70 6.76 12.86 12.14
have been successful to some degree. Horn et al histologically demonstrated delayed wound healing of rat tooth sockets after maxillary molar tooth extraction and whole head x-ray radiation [13]. The rats received a dose of 15 Gy either immediately before extraction or 2 weeks prior. The authors also noted that the group that received radiation 2 weeks prior demonstrated a greater delay in wound healing as compared with the group that was irradiated immediately before tooth extraction. Radiationinduced bone necrosis was not demonstrated, and cell-type specific death was not described. Guglielmotti et al [14] studied rats that underwent extractions of all 3 mandibular molars followed by whole head irradiation. Irradiation occurred immediately after, 3 days, or 7 days after extraction. The rats were also subdivided to receive 15-, 20-, and 30-Gy irradiation. The main finding of the study was that radiation at the time of extraction delays wound healing to a greater extent compared with radiation 7 days after extraction. This study is in agreement with the consensus that patients who are to undergo radiation therapy for head and neck cancer treatment should have tooth extractions performed at least 2 weeks before treatment [15]. Again, delays in wound healing do not define bone necrosis after radiation. Lorrente et al [16] studied the effects of a controlled 4-mm defect created in a rat mandibular ramus followed by fractionated Co-60 radiation. They also assessed whether
demineralized bone powder improved wound healing. The authors found that the only group to have complete filling of the bone defect was the nonirradiated group that had nonradiated bone powder placed in wound. There was partial filling of the defects in the irradiated group that had nonirradiated bone powder as well. Radiation-induced filling defects in bone lack evidence for osteogenic cell death that can define ORN. A more recent animal model to study ORN was presented by Kurihashi et al [17]. They divided 18 male rats into 2 groups with one receiving a 10-Gy x-ray dose and the other as a control. The left first maxillary molars were extracted from all animals; and animals were subdivided to be killed 3, 7, and 14 days after tooth extraction. The specimens were analyzed by H&E histologic staining, by immunohistochemistry with alkaline phosphatase (ALP) and tartrate-resistant acid phosphatase (TRAP), and by micro-CT. The findings of their study demonstrated a slower wound healing time and an increase in bone resorption in the irradiated group as demonstrated by elevated TRAP levels. Although this study appears more promising because cell-type–specific activity relating to osteoclasts was described, necrosis was not demonstrated. Our radiogenic mandibular bone damage model has incorporated many of the virtues of these previous models as well as improved several other areas. Significant credit is duly given to Niehoff et al [5] for first introducing this model that uses focused HDR brachytherapy to irradiate a defined region of bone similar to intensity-modulated radiation therapy in treating human cancers. Adding dental trauma by way of extractions in our model also maximizes the likelihood of radiogenic bone damage and clinical ORN as is the case in human series of ORN [18]. Furthermore, mandibular, and not maxillary, tooth extractions were performed to better simulate ORN. It is well accepted that ORN is exceedingly more common to occur in the mandible vs the maxilla [7]. Finally, both clinic signs and radiologic
Fig. 8. The H&E sections of tooth extraction socket sites from a radiated (A) and control (B) animal at 10× magnification. The radiated sample demonstrates increased fibrosis, inflammation, and osteoclasts with a reduction in bone formation when compared with the control.
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
examination of the specimens were used to document radiogenic bone damage as is done in the clinical setting. In addition to radiation-induced bone damage, the fluorochrome labeling patterns in our model are highly suggestive of cell necrosis occurring in the extraction site. The lack of any second label (demeclocycline, orange) in the extraction sockets, while being present in the lamellar inferior rim of the radiated mandibles, is open to interpretation. One explanation may be the absence of any bone metabolism in the extraction site suggestive of earlier osteogenic cell death after radiation. Another possibility is excess resorption of all of the second label within the extraction socket due to increased osteoclastogenesis. The latter is less likely, as the first label remains present in the radiated sockets and both labels are present in the sockets of all nonradiated mandibles. Using multiple different-colored fluorochromes on a more frequent and closer timeline interval to radiation and extractions may provide a chronological profile of bone metabolism as demonstrated in another animal model by the senior author [8]. We acknowledge that this model has several shortcomings. First, the difference in weight gain between the 2 experimental groups is a confounding variable. It is possible that the radiated animals' malnutrition may have led to the changes seen. The benefits of this advanced mode of HDRtype radiation allows for future intraindividual split-mouth comparisons. This can help clarify this variable by using micro-CT to compare differences between contralateral hemimandibles in the radiated group to control animals. Second, the small number of rats used in our experiment limits appropriate statistical analysis of our results. The intention of this project was to develop a preliminary model of radiogenic bone damage for understanding mandibular ORN. With this early platform for studying ORN, our current efforts are aimed at reproducing these results in large numbers to demonstrate statistically significant outcomes. Lastly, the 30-Gy single time dose of radiation is significantly higher than an equivalent dose that would be used in treating human oropharyngeal cancers. Moreover, the one mortality out of 6 radiated animals is too high and may be related to the higher radiation dose. Our reasoning for increasing the dose to 30 Gy compared with that in the article of Niehoff et al is similar to why we added dental extractions, to maximize radiogenic bone damage in a model set to define ORN. From a clinical standpoint, our goal was to create an advanced stage of ORN; and the highest radiation dose given in prior animal models was up to 30 Gy. Several staging systems for ORN have been constructed [2,19]. Most recently, Schwartz and Kagan [7] have developed an updated staging system that appears to better address both the clinical and pathophysiologic aspects of the disease. In stage I ORN, there are superficial involvement of the mandible only and minimal soft tissue ulceration; and only the exposed cortical bone is necrotic. In stage II disease, both the exposed cortical bone and underlying medullary bone are necrotic. The majority of stage II disease resolves
299
with conservative treatment. In stage III ORN, there is diffuse involvement of the mandible, including a fullthickness segment of bone. All of these patients require surgical intervention. By using a 30-Gy dose, our expectations were to create radiogenic bone damage analogous to stage III ORN. Future work with our model can compare different doses to test the clinical analogy of staging ORN.
5. Conclusions Our modifications of the HDR brachytherapy model incorporating postradiation dental extractions and using a higher radiation dose have successfully demonstrated reproducible radiogenic mandibular bone damage. These modifications are analogous to the clinical scenario culminating in advanced mandibular ORN. The use of fluorochrome labeling in this dental extraction model suggests changes consistent with cell necrosis. Radiated tooth extraction sites demonstrated greatly increased fibrosis and osteoclastic numbers that are concordant with recent theories and treatment trials for ORN [11,12]. Moreover, this imbalance at the cellular level with increased osteoclasts is currently serving as a basis for the senior author's current in vitro studies into the cellular basis of ORN. Although clinical criteria continue to be used today in describing ORN, we believe that our model can serve as a platform for future studies to define ORN and delineate its pathogenesis.
References [1] Beumer III J, Silverman Jr S, Benak Jr SB. Hard and soft tissue necroses following radiation therapy for oral cancer. J Prosthet Dent 1972;27:640-4. [2] Marx RE. Osteoradionecrosis: a new concept of its pathophysiology. J Oral Maxillofac Surg 1983;41:283-8. [3] Hutchinson IL. Complications of radiotherapy in head and neck: an orofacial surgeon's view. In: Tobias JS, Thomas PRM, editors. Current radiation oncology. London: Arnold; 1996. p. 144-77. [4] Harris M. The conservative management of osteoradionecrosis of the mandible with ultrasound therapy. Br J Oral Maxillofac Surg 1992;30:313-8. [5] Niehoff P, Springer IN, Acil Y, et al. HDR brachytherapy irradiation of the jaw—as a new experimental model of radiogenic bone damage. J Craniomaxillofac Surg 2008;36:203-9. [6] Beumer 3rd J, Harrison R, Sanders B, et al. Preradiation dental extractions and the incidence of bone necrosis. Head Neck Surg 1983;5:514-21. [7] Schwartz HC, Kagan AR. Osteoradionecrosis of the mandible: scientific basis for clinical staging. Am J Clin Oncol 2002;25:168-71. [8] Nabili V, Brodie HA, Neverov NI, et al. Chronology of labyrinthitis ossificans induced by Streptococcus pneumoniae meningitis. Laryngoscope 1999;109:931-5. [9] Regaud C. Sur la necrosedes os attente par un processus cancereux et traits par les radiaions. Compt Rend Soc Biol 1922;87:629. [10] Meyer I. Infectious diseases of the jaws. J Oral Surg 1970;28:17-26. [11] Delanian S, Lefaix JL. The radiation-induced fibroatrophic process: therapeutic perspective via the antioxidant pathway. Radiother Oncol 2004;73:119-31.
300
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
[12] Delanian S, Depondt J, Lefaix JL. Major healing of refractory mandible osteoradionecrosis after treatment combining pentoxifylline and tocopherol: a phase II trial. Head Neck 2005;27:114-23. [13] Horn Y, Sela MN, Shlomi B, et al. Effect of irradiation-timing on the initial socket healing in rats. Int J Oral Surg 1979;8:457-61. [14] Guglielmotti MB, Ubios AM, Cabrini RL. Alveolar wound healing after x-irradiation: a histologic, radiographic, and histometric study. J Oral Maxillofac Surg 1986;44:972-6. [15] Koga DH, Salvajoli JV, Alves FA. Dental extractions and radiotherapy in head and neck oncology: review of the literature. Oral Dis 2008;14:40-4.
[16] Lorente CA, Song BZ, Donoff RB. Healing of bony defects in the irradiated and unirradiated rat mandible. J Oral Maxillofac Surg 1992;50:1305-9. [17] Kurihashi T, Iwata H, Nasu M, et al. Experimental study on wound healing of alveolar bone sockets in the rat maxilla after x-ray irradiation. Odontology 2002;90:35-42. [18] Beumer J, Seto B. Dental extractions in the irradiated patient. Spec Care Dentist 1981;1:166-73. [19] Epstein JB, Wong FL, Stevenson-Moore P. Osteoradionecrosis: clinical experience and a proposal for classification. J Oral Maxillofac Surg 1987;45:104-10.
American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291 – 300 www.elsevier.com/locate/amjoto
Animal model of radiogenic bone damage to study mandibular osteoradionecrosis☆,☆☆ Marc Cohen, MDa , Ichiro Nishimura, DDS, DMSc, DMDb , Matthew Tamplen, BSa , Akishige Hokugo, DDS, PhDb , John Beumer, DDS, MSb , Michael L. Steinberg, MDc , Jeffrey D. Suh, MDa , Elliot Abemayor, MD, PhDa , Vishad Nabili, MDa,⁎ a
Division of Head and Neck Surgery, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA UCLA School of Dentistry, Weintraub Center for Reconstructive Biotechnology, Los Angeles, CA, USA c Department of Radiation Oncology, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA Received 27 April 2010 b
Abstract
Objective: The objective of the study was to create an animal model to study mandibular osteoradionecrosis (ORN) using high–dose rate (HDR) brachytherapy. Methods: Ten Sprague-Dawley male rats were used in this study. Six rats received a single dose of 30 Gy using an HDR remote afterloading machine via a brachytherapy catheter placed along the left hemimandible. The remaining 4 rats served as controls with catheter placement without radiation (sham). On the day following irradiation or sham, all 3 left mandibular molars were atraumatically extracted. Twenty-eight days after irradiation, mandibles were examined using nondecalcified histology with sequential fluorochrome labeling, decalcified histology, and micro–computed tomography scanning. Results: Irradiated rats demonstrated exposed bone at the extraction sockets, whereas the control animals had complete mucosalization. Alopecia was also seen in the irradiated group. Both histologic and radiologic analyses of the mandible specimens demonstrated a reduction in bone formation in the radiated mandibles as compared with controls. Conclusions: Our HDR brachytherapy model incorporating postradiation dental extractions has successfully demonstrated reproducible radiogenic mandibular bone damage analogous to the clinical ORN. Although clinical criteria continue to be used today in describing ORN, this model can serve as a platform for future studies to define ORN and delineate its pathogenesis. © 2011 Elsevier Inc. All rights reserved.
1. Introduction ☆
Presented at the American Academy of Otolaryngology Head and Neck Surgery Annual Meeting, October 6, 2009, San Diego, CA. ☆☆ Justification for additional authors: This project entailed expertise from multiple areas with each author playing a key role in providing input and resources including Head and Neck Surgical Oncology (Drs Cohen, Suh, and Abemayor), Microvascular Reconstruction (Dr Nabili), Dental biology (Drs Nishimura, Hokugo, and Beumer), and Radiation access and physicist input (Dr Steinberg). ⁎ Corresponding author. Division of Head and Neck Surgery, David Geffen School of Medicine at UCLA, 10833 LeConte Ave, CHS RM 62132, Los Angeles, CA 90095-1624, USA. Tel.: +1 310 709 7970; fax: +1 310 206 1393. E-mail address: [email protected] (V. Nabili). 0196-0709/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.amjoto.2010.06.001
Despite more target-specific advances in radiation technology for the treatment of head and neck cancer, including intensity-modulated radiation therapy and high– dose rate (HDR) brachytherapy, patients continue to suffer from adverse effects of this treatment modality. Significant morbidity can ensue from radiation therapy months to years after treatment. One specific adverse effect of radiation treatment for head and neck cancer patients is osteoradionecrosis (ORN) of the mandible. Being cured of their head and neck cancers, patients with advanced mandibular ORN may still succumb to extensive microvascular reconstructive surgery once reserved initially as an oncologic treatment
292
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
option. These heroic measures may still prove futile because ORN can recur and little is still known about the disease. ORN of the mandible is a pathologic condition that has been commonly described by clinical criteria, whereas a true definition has been elusive as the pathogenesis of this disease still remains undefined. Over the past 35 years, several authors have attempted to define ORN. Beumer et al [1] stated that ORN occurs “when bone in the radiation field [is] exposed for at least 2 months in the absence of local neoplastic disease.” Marx [2] defined ORN as “an area greater than one centimeter of exposed bone in a field of irradiation that had failed to show any evidence of healing for at least 6 months.” Hutchinson [3] defined ORN to only require 2 months of exposed bone, whereas Harris [4] used the 3-month marker. Despite the variability in these clinical descriptions, several commonalities are present in clinically defining ORN: (1) exposed bone for at least 2 to 6 months, (2) a history of radiation therapy to the region of exposed bone, (3) the presence of necrotic or devitalized bone, and (4) no evidence of tumor recurrence. Patients with mandibular ORN often have radiologic assessment in the form of computed tomography (CT) and/or panorex. Limitations in radiologic analysis of ORN are not due to technology but again more so due to lack of defined radiologic criteria for mandibular ORN. Common radiographic findings for mandibular ORN include but are not limited to radiolucencies, bony sequestration, and/or pathologic fracture(s). Posttreatment histopathologic analysis often shows chronic inflammation and necrotic bone, yet the pathogenesis is unclear. To investigate mandibular ORN further, we set out to first create an animal model as a platform to demonstrate the aforementioned criteria used today. Previous attempts at creating a model of ORN in a rat have been successful to some degree. However, these models were lacking in certain respects. Most of the models involved whole head irradiation, and this is not analogous to current radiation treatment for head and neck cancer. Recently, Niehoff et al [5] described using HDR brachytherapy to create an experimental model of radiogenic bone damage in a rat. The HDR brachytherapy is a technique that uses a relatively intense source of radiation, typically iridium-192, to deliver a therapeutic dose of radiation through temporarily placed catheters at a specified location such as a unilateral rat mandible. The authors successfully demonstrated mandibular bone damage after HDR radiation. Mandibular ORN, however, was not described or defined. It is known that dental trauma, especially tooth extractions in a radiated field, can increase the incidence of ORN [6]. Several previous animal models involved maxillary tooth extraction, rather than mandibular. This is not as clinically relevant because it is well established that ORN predominately occurs in the mandible and not in the maxilla [7]. Most studies used histologic examination of the specimens as the only evaluation tool. In reality, ORN is
characterized by clinical examination with the aid of histologic and radiologic assessments. The goal of our study was to improve and update the previous models of ORN not only using the most current technology, but incorporating the current knowledge of the disease process in our model. The rat model that has been designed involves irradiating rat mandibles with an extremely targeted device, extracting mandibular molars after radiation, and assessing outcomes using high-speed digital photography, micro-CT, and fluorescent and basic microscopy for histologic analysis. A thoroughly effective treatment plan with minimal morbidity does not exist for ORN and most likely stems from a lack of understanding of the pathophysiology of the disease process. It is for this reason that we have attempted to create an updated rat model of ORN. Our hope is to establish a reproducible and clinically relevant animal model of post–dental extraction radiogenic bone damage from which to define and study the pathogenesis of mandibular ORN.
2. Materials and methods 2.1. Experimental design Approval for the research protocol was obtained from the University of California Los Angeles Chancellor's Animal Research Committee. Ten male Sprague-Dawley rats of 7 weeks of age (192–224 g) were used in this study. Rats were obtained from Charles River Laboratories International, Inc (Wilmington, MA). The rats were kept in pairs and given a standard pelleted rodent diet and water ad libitum in accordance with the requirements of the United States of America Animal Welfare Act and the Public Health Service Policy on Humane Care and Use of Laboratory Animals. The 10 rats were divided into 2 groups: group 1—left mandibular irradiation followed by left molar extractions (n = 6) and group 2—left mandibular sham irradiation (placement of HDR catheter only) followed by left molar extractions (n = 4). 2.2. Irradiation Under inhalational isoflurane anesthesia, the left cheek skin of the rat was shaved and prepared with betadine. A sterile plastic HDR catheter (Alpha Omega Services, Long Beach, CA) was implanted along the lateral aspect of the left mandibular body. The catheter was inserted via a stab incision made just lateral to the left central incisor in the gingivolabial sulcus. The catheter was advanced submucosally along the inferior edge of the mandible. The catheter exited the oral cavity via a second stab incision at the posterior border of the mandibular ramus and advanced just inferior to the external auditory meatus. This advancement of the external portion of the catheter was consistently 10 mm and served as a reproducible length along with the above-
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
293
microscopy using UV light (Olympus BX51 Research Microscope, Tokyo, Japan) was used to analyze bone apposition at mandibular tooth sockets. 2.5. Killing of animals The animals were killed 28 days post–tooth extraction. Mandibles were extracted and photographed using a Canon Rebel Ti DSLR camera (Canon, Tokyo, Japan). Mandibles were then placed in 70% isopropyl alcohol. 2.6. Radiologic analysis Fig. 1. The HDR brachytherapy catheter is inserted along the lateral border of the left mandibular body.
mentioned landmarks. The external distal portion of the catheter was secured to the non–hair-bearing neck and cheek skin using Steri-Strip adhesive skin closure tape (3M, St Paul, MN) (Fig. 1). A 3-dimensional treatment plan for irradiation was devised that encompassed a depth of 5 mm from a point 1 cm from the distal aspect of the catheter tip. The center of the radiation field was placed lateral to the midline of the left mandibular body. A single dose of radiation was applied with an HDR afterloading remote machine (GammaMed 12it; Varian Medical Systems, Charlottesville, Inc, VA) providing 30 Gy with predefined isodose lines (Fig. 2). The catheter was removed after irradiation was complete in group 1 and immediately removed after placement in group 2 without radiation. 2.3. Tooth extraction Under inhalational isoflurane anesthesia, all rats underwent atraumatic extraction of all 3 of the left mandibular molars 1 day following irradiation (or sham catheter placement). Extreme care was taken to avoid breaking the tooth roots from the crown. The teeth were examined after extraction with 3.5× surgical magnifying loupes to evaluate for completeness of extraction. Postoperative pain management was achieved with buprenorphine (Buprenex; Reckitt Benckiser Healthcare Ltd, Hull, England) at a dose of 0.03 mg/kg given subcutaneously. The diets were supplemented with sliced apples for the first 3 days after molar extractions.
CT of the extracted mandibles was performed using a desktop cone-beam micro-CT scanner (μCT 40 Scanco Medical, Brüttisellen, Switzerland). Three-dimensional reconstructions and volume analysis of the microradiographs were accomplished using μCT Evaluation Software v6.0 (Scanco Medical). The mandibular volume that was analyzed encompassed the region from the most superior aspect of the tooth socket where both medial and lateral cortices were seen (in an axial view) down to the nadir of the incisor root. Because of the relatively large volume that the incisor root socket normally encompasses in a rat mandible, these sockets were subtracted from volume calculations to limit the focus to the molar teeth regions. Bone volume (BV) and total volumes (TVs) were measured, and the ratio of these numbers was analyzed. 2.7. Histologic analysis After micro-CT analysis was performed, both the radiated and control specimens were randomized into 2 groups to allow for non–decalcified-based fluorochrome analysis and decalcified-based paraffin embedded analysis, respectively. The first group consisted of 4 nondecalcified specimens (2 irradiated and 2 controls). These samples were taken from
2.4. Fluorochrome labeling Sequential different-colored fluorochrome labeling of mineralizing bone was performed [8]. Two fluorochromes were used in this experiment: calcein (green fluorescence) (1% in 2% NaHCO3 solution, 20 mg/kg body weight; Sigma Aldrich, St Louis, MO) and demeclocycline (orange fluorescence) (1% in 2% NaHCO3 solution, 20 mg/kg body weight; Sigma Aldrich). All rats underwent intraperitoneal injection of calcein 2 days before irradiation and demeclocycline 2 days before being killed. Fluorescent
Fig. 2. The isodose lines of the 30-Gy radiation field as calculated by the GammaMed HDR brachytherapy device.
294
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
Fig. 3. Clinical manifestations of irradiation. Alopecia of the facial skin in the irradiated rat (A) and complete hair regrowth in the nonirradiated rat (B).
70% isopropyl alcohol and immediately embedded in methyl methacrylate. Five-micrometer–thick sagittal sections of left hemimandibles were cut on a Jung polycot microtome (Reichert-Jung, Heidelberg, Germany) and either stained with toluidine blue or left unstained for fluorochrome analysis. Bone formation and resorption were analyzed at 2 standardized sites using a Pyser-SGI 5-mm stage micrometer scale (Pyser-SGI, Kent, United Kingdom) and an infinity eyepiece with dimensions of 10 mm by 0.1 mm divisions (Fisher Scientific, Pittsburgh, PA) at 20× magnification. Bone mineralization was detected via fluorescent microscopy in the center of the tooth extraction socket immediately above inferior alveolar nerve at a standardized location. Bone mineral apposition rate (MAR) was calculated along the lower cortical rim of the mandible directly below the posterior molar extraction socket and the inferior alveolar nerve at a second standardized location. All
fluorochrome analysis was performed using an Olympus BX51 Research Microscope. The 5 remaining specimens (3 irradiated and 2 controls) were decalcified in 10% EDTA (pH 7.4) for 10 days, fixed in formalin overnight, and paraffin embedded by the UCLA Translational Pathology Core Laboratory. Four-micrometer sagittal sections of the left hemimandibles were stained with hematoxylin and eosin (H&E) and viewed for qualitative histologic analysis using camera-assisted light microscopy (Nikon Labophot-2; Nikon, Tokyo, Japan). 3. Results 3.1. Clinical Nine rats survived the study period. One rat from the irradiated group died 12 days after tooth extraction and was
Fig. 4. Clinical manifestations of irradiation. Retardation of central incisor growth in the irradiated rat (A) and uninhibited growth in the nonirradiated rat (B).
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
295
Fig. 5. Gross pathologic evaluation of rat mandibles. Exposed bone at the site of extraction in the irradiated rat (A) and complete wound mucosalization in the nonirradiated rat (B).
excluded from further evaluation; the cause of death was attributed to malnutrition. The remainder of the irradiated rats survived the study time of 4 weeks and continued to gain weight. However, the irradiated group average weight gain was significantly less than the control group (124.8 vs 196.8 g, P = .008). The rats from the irradiated group demonstrated adverse effects of target-specific HDR radiation as evidenced by unilateral left cheek skin alopecia in 100% of the group, whereas all of the animals in the control group had complete hair regrowth (Fig. 3). Likewise, ipsilateral incisor growth was also retarded in the radiated animals as compared with nonirradiated animals (Fig. 4). Evaluation of the intraoral inferior alveolar ridges using high-resolution digital photography demonstrated exposed bone at the site of dental extraction in 100% of the irradiated rats. Complete wound mucosalization was seen in all of the animals in the control group (Fig. 5). 3.2. Micro-CT evaluation Before using the analysis software for the microradiographs, the raw micro-CT data were visually analyzed by 2 independent observers to assess for efficacy of tooth extractions. Two rats (one from each group) had evidence of more than 2 retained tooth roots. Because of the variability in analysis caused by retained tooth elements, these 2 rats were excluded from microradiograph volumetric analysis. As described in the methods section, BV was measured in relation to the TV of analyzed mandible and expressed as a Table 1 Micro-CT data of hemimandible volume analysis
Rats 4 and 9 (shaded) were excluded from statistical analysis because of retained tooth roots.
ratio (BV/TV) (Table 1). The average BV/TV in the irradiated group was 0.559 as compared with the control group ratio of 0.745 (P = .009). Evaluation of the 3-dimensional CT reconstructions of the rat mandibles did allow for ultrastructural viewing of bone regrowth in the tooth extraction socket sites and qualitative comparisons between radiated and nonradiated samples (Fig. 6). Lack of bone in the exposed sockets in the irradiated mandibles is seen in these CT reconstructions. 3.3. Fluorescent microscopic analysis Bone MAR was measured at the second standardized location site along the inferior border of the mandible below the central incisor and inferior alveolar nerve for the 4 nondecalcified specimens (2 irradiated and 2 controls). This area demonstrated lamellar nontraumatized bone with presence of both calcein green and demeclocycline orange labeling in all samples. MAR for each sample were calculated by dividing the distance of bone deposition between the 2 labels by a total of 28 days between fluorochrome injections. Both irradiated samples showed far less bone deposition, averaging an MAR of 6.23 µm/d compared with a 12.5-µm/d average for the control samples (Fig. 7 and Table 2). The tooth extraction sockets in all of the animals demonstrated a disarray of fluorochrome labeling that prevented any quantitative analysis. Qualitatively, it was noted that the second label, demeclocycline (orange), was not present in any of the radiated extraction sites while still being present along the inferior cortical border of all mandibles as mentioned above. The control group showed clear evidence of double labeling within the extraction sockets indicative of new bone formation postextraction. Qualitative analysis of toluidine-stained slides from the nondecalcified specimens further demonstrated a lack of bone present in the extraction sites of irradiated samples when compared with controls (Fig. 7).
296
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
Fig. 6. Three-dimensional micro-CT reconstructions of rat mandibles. Irradiated mandible (A) with a lower bone density at extraction sites vs nonirradiated mandible (B).
3.4. Decalcified histologic analysis The radiated specimens showed a diminution of bone formation, an overwhelming presence of fibrosis and inflammation, and an unusually high presence of osteoclasts within the extraction sockets when compared with controls. Moreover, the control samples specifically had increased bone present along with many osteoblasts and only minimal inflammation and fibrosis (Fig. 8).
4. Discussion ORN of the mandible has been recognized since 1922 [9]. Despite the awareness of this disease for more than 85 years, little is still known about its pathogenesis. The ambiguity about defining mandibular ORN has also limited its treatment options to clinical judgment with minimal objective criteria. The majority of patients with mandibular ORN will succumb to advanced-stage disease requiring microvascular reconstruction. In a recent series by the senior author (VN), these patients had increased woundrelated complications and risk for continuing recurrent disease (accepted for publication, Otolaryngol Head Neck Surg April 2010). A lack of understanding of the pathophysiology of the disease process is the reason that we have created this postextraction rat model of radiogenic bone damage to serve as a platform to further study and define mandibular ORN. Theories on the pathophysiology of ORN have evolved over the last 40 years, with different treatments proposed along the way. Originally, Meyer [10] believed that, in the setting of irradiated bone, a traumatic event could lead to a superimposed infection similar to osteomyelitis. This led to conservative treatment measures including local wound irrigation, oral hygiene improvement, curettage,
debridement, sequestrectomy, bone filling, and/or longterm antibiotic use. Approximately a decade later, Marx [2] proposed his 3-H theory of ORN where he described a sequence of events starting with radiation, formulation of hypoxic-hypocellularhypovascular (3-H) tissue, then tissue breakdown, and finally a chronic nonhealing wound as ORN. Marx showed that ORN is a problem of wound healing due to a complex metabolic and tissue homeostatic oxygen deficiency, leading to the use of hyperbaric oxygen (HBO) for treatment. More recently, Delanian and Lefaix [11] proposed their fibroatrophic theory to explain the radiation-induced changes that occur with ORN. In this theory, ORN lesions are ultimately due to an imbalance of bone resorption and bone deposition. Bone atrophy occurs in the setting of extensive fibrosis. The combination of osteoblast death in the mandible after irradiation, failure of the osteoblasts to repopulate, and the excessive proliferation of myofibroblasts results in the bony matrix being replaced by fibrous tissues. On the basis of this theory, pentoxyifylline and tocopherol have emerged as possible adjuvant treatments for ORN with promising results, leading to possible future clinical trials [12]. With a pathobiology that is undefined, a true definition of ORN cannot be made yet. Current attempts at defining ORN are based on clinical criteria describing radiogenic bone damage. Actual cellular necrosis after radiation, the types of cells involved, and the sequence of events have not been fully elucidated by any model to date. As such, some would argue that our current model may present what we know clinically as ORN; yet we believe our model has successfully demonstrated post–dental extraction radiogenic bone damage. In reviewing prior animal models, the terminology of actual ORN has thus been used loosely and oftentimes incorrectly in place of radiogenic bone damage. Many previous attempts at creating a model of ORN in the rat
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
297
Fig. 7. Double labeling (calcein green, demeclocycline orange) seen on the inferior cortical rim of radiated (A) compared with control (B) mandibles, with radiated samples demonstrating a reduction in mineral apposition. Radiated tooth extraction socket sites (C) further demonstrate a reduction in new bone formation with minimal labeling (calcein green only) and no second label as compared with increased presence of both labels in the controls (D). Nondecalcified toluidine blue–stained sections from tooth extraction socket sites show increased fibrosis and minimal bone in radiated samples (E) when compared with controls (F).
298
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
Table 2 Fluorochrome dynamic analysis of bone apposition rate at standardized site Animal no.
Distance between 2 labels (µm)
MAR (µm/d) total days = 28
Animal 1 (radiated) Animal 3 (radiated) Animal 9 (control) Animal 10 (control)
160 190 360 340
5.70 6.76 12.86 12.14
have been successful to some degree. Horn et al histologically demonstrated delayed wound healing of rat tooth sockets after maxillary molar tooth extraction and whole head x-ray radiation [13]. The rats received a dose of 15 Gy either immediately before extraction or 2 weeks prior. The authors also noted that the group that received radiation 2 weeks prior demonstrated a greater delay in wound healing as compared with the group that was irradiated immediately before tooth extraction. Radiationinduced bone necrosis was not demonstrated, and cell-type specific death was not described. Guglielmotti et al [14] studied rats that underwent extractions of all 3 mandibular molars followed by whole head irradiation. Irradiation occurred immediately after, 3 days, or 7 days after extraction. The rats were also subdivided to receive 15-, 20-, and 30-Gy irradiation. The main finding of the study was that radiation at the time of extraction delays wound healing to a greater extent compared with radiation 7 days after extraction. This study is in agreement with the consensus that patients who are to undergo radiation therapy for head and neck cancer treatment should have tooth extractions performed at least 2 weeks before treatment [15]. Again, delays in wound healing do not define bone necrosis after radiation. Lorrente et al [16] studied the effects of a controlled 4-mm defect created in a rat mandibular ramus followed by fractionated Co-60 radiation. They also assessed whether
demineralized bone powder improved wound healing. The authors found that the only group to have complete filling of the bone defect was the nonirradiated group that had nonradiated bone powder placed in wound. There was partial filling of the defects in the irradiated group that had nonirradiated bone powder as well. Radiation-induced filling defects in bone lack evidence for osteogenic cell death that can define ORN. A more recent animal model to study ORN was presented by Kurihashi et al [17]. They divided 18 male rats into 2 groups with one receiving a 10-Gy x-ray dose and the other as a control. The left first maxillary molars were extracted from all animals; and animals were subdivided to be killed 3, 7, and 14 days after tooth extraction. The specimens were analyzed by H&E histologic staining, by immunohistochemistry with alkaline phosphatase (ALP) and tartrate-resistant acid phosphatase (TRAP), and by micro-CT. The findings of their study demonstrated a slower wound healing time and an increase in bone resorption in the irradiated group as demonstrated by elevated TRAP levels. Although this study appears more promising because cell-type–specific activity relating to osteoclasts was described, necrosis was not demonstrated. Our radiogenic mandibular bone damage model has incorporated many of the virtues of these previous models as well as improved several other areas. Significant credit is duly given to Niehoff et al [5] for first introducing this model that uses focused HDR brachytherapy to irradiate a defined region of bone similar to intensity-modulated radiation therapy in treating human cancers. Adding dental trauma by way of extractions in our model also maximizes the likelihood of radiogenic bone damage and clinical ORN as is the case in human series of ORN [18]. Furthermore, mandibular, and not maxillary, tooth extractions were performed to better simulate ORN. It is well accepted that ORN is exceedingly more common to occur in the mandible vs the maxilla [7]. Finally, both clinic signs and radiologic
Fig. 8. The H&E sections of tooth extraction socket sites from a radiated (A) and control (B) animal at 10× magnification. The radiated sample demonstrates increased fibrosis, inflammation, and osteoclasts with a reduction in bone formation when compared with the control.
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
examination of the specimens were used to document radiogenic bone damage as is done in the clinical setting. In addition to radiation-induced bone damage, the fluorochrome labeling patterns in our model are highly suggestive of cell necrosis occurring in the extraction site. The lack of any second label (demeclocycline, orange) in the extraction sockets, while being present in the lamellar inferior rim of the radiated mandibles, is open to interpretation. One explanation may be the absence of any bone metabolism in the extraction site suggestive of earlier osteogenic cell death after radiation. Another possibility is excess resorption of all of the second label within the extraction socket due to increased osteoclastogenesis. The latter is less likely, as the first label remains present in the radiated sockets and both labels are present in the sockets of all nonradiated mandibles. Using multiple different-colored fluorochromes on a more frequent and closer timeline interval to radiation and extractions may provide a chronological profile of bone metabolism as demonstrated in another animal model by the senior author [8]. We acknowledge that this model has several shortcomings. First, the difference in weight gain between the 2 experimental groups is a confounding variable. It is possible that the radiated animals' malnutrition may have led to the changes seen. The benefits of this advanced mode of HDRtype radiation allows for future intraindividual split-mouth comparisons. This can help clarify this variable by using micro-CT to compare differences between contralateral hemimandibles in the radiated group to control animals. Second, the small number of rats used in our experiment limits appropriate statistical analysis of our results. The intention of this project was to develop a preliminary model of radiogenic bone damage for understanding mandibular ORN. With this early platform for studying ORN, our current efforts are aimed at reproducing these results in large numbers to demonstrate statistically significant outcomes. Lastly, the 30-Gy single time dose of radiation is significantly higher than an equivalent dose that would be used in treating human oropharyngeal cancers. Moreover, the one mortality out of 6 radiated animals is too high and may be related to the higher radiation dose. Our reasoning for increasing the dose to 30 Gy compared with that in the article of Niehoff et al is similar to why we added dental extractions, to maximize radiogenic bone damage in a model set to define ORN. From a clinical standpoint, our goal was to create an advanced stage of ORN; and the highest radiation dose given in prior animal models was up to 30 Gy. Several staging systems for ORN have been constructed [2,19]. Most recently, Schwartz and Kagan [7] have developed an updated staging system that appears to better address both the clinical and pathophysiologic aspects of the disease. In stage I ORN, there are superficial involvement of the mandible only and minimal soft tissue ulceration; and only the exposed cortical bone is necrotic. In stage II disease, both the exposed cortical bone and underlying medullary bone are necrotic. The majority of stage II disease resolves
299
with conservative treatment. In stage III ORN, there is diffuse involvement of the mandible, including a fullthickness segment of bone. All of these patients require surgical intervention. By using a 30-Gy dose, our expectations were to create radiogenic bone damage analogous to stage III ORN. Future work with our model can compare different doses to test the clinical analogy of staging ORN.
5. Conclusions Our modifications of the HDR brachytherapy model incorporating postradiation dental extractions and using a higher radiation dose have successfully demonstrated reproducible radiogenic mandibular bone damage. These modifications are analogous to the clinical scenario culminating in advanced mandibular ORN. The use of fluorochrome labeling in this dental extraction model suggests changes consistent with cell necrosis. Radiated tooth extraction sites demonstrated greatly increased fibrosis and osteoclastic numbers that are concordant with recent theories and treatment trials for ORN [11,12]. Moreover, this imbalance at the cellular level with increased osteoclasts is currently serving as a basis for the senior author's current in vitro studies into the cellular basis of ORN. Although clinical criteria continue to be used today in describing ORN, we believe that our model can serve as a platform for future studies to define ORN and delineate its pathogenesis.
References [1] Beumer III J, Silverman Jr S, Benak Jr SB. Hard and soft tissue necroses following radiation therapy for oral cancer. J Prosthet Dent 1972;27:640-4. [2] Marx RE. Osteoradionecrosis: a new concept of its pathophysiology. J Oral Maxillofac Surg 1983;41:283-8. [3] Hutchinson IL. Complications of radiotherapy in head and neck: an orofacial surgeon's view. In: Tobias JS, Thomas PRM, editors. Current radiation oncology. London: Arnold; 1996. p. 144-77. [4] Harris M. The conservative management of osteoradionecrosis of the mandible with ultrasound therapy. Br J Oral Maxillofac Surg 1992;30:313-8. [5] Niehoff P, Springer IN, Acil Y, et al. HDR brachytherapy irradiation of the jaw—as a new experimental model of radiogenic bone damage. J Craniomaxillofac Surg 2008;36:203-9. [6] Beumer 3rd J, Harrison R, Sanders B, et al. Preradiation dental extractions and the incidence of bone necrosis. Head Neck Surg 1983;5:514-21. [7] Schwartz HC, Kagan AR. Osteoradionecrosis of the mandible: scientific basis for clinical staging. Am J Clin Oncol 2002;25:168-71. [8] Nabili V, Brodie HA, Neverov NI, et al. Chronology of labyrinthitis ossificans induced by Streptococcus pneumoniae meningitis. Laryngoscope 1999;109:931-5. [9] Regaud C. Sur la necrosedes os attente par un processus cancereux et traits par les radiaions. Compt Rend Soc Biol 1922;87:629. [10] Meyer I. Infectious diseases of the jaws. J Oral Surg 1970;28:17-26. [11] Delanian S, Lefaix JL. The radiation-induced fibroatrophic process: therapeutic perspective via the antioxidant pathway. Radiother Oncol 2004;73:119-31.
300
M. Cohen et al. / American Journal of Otolaryngology–Head and Neck Medicine and Surgery 32 (2011) 291–300
[12] Delanian S, Depondt J, Lefaix JL. Major healing of refractory mandible osteoradionecrosis after treatment combining pentoxifylline and tocopherol: a phase II trial. Head Neck 2005;27:114-23. [13] Horn Y, Sela MN, Shlomi B, et al. Effect of irradiation-timing on the initial socket healing in rats. Int J Oral Surg 1979;8:457-61. [14] Guglielmotti MB, Ubios AM, Cabrini RL. Alveolar wound healing after x-irradiation: a histologic, radiographic, and histometric study. J Oral Maxillofac Surg 1986;44:972-6. [15] Koga DH, Salvajoli JV, Alves FA. Dental extractions and radiotherapy in head and neck oncology: review of the literature. Oral Dis 2008;14:40-4.
[16] Lorente CA, Song BZ, Donoff RB. Healing of bony defects in the irradiated and unirradiated rat mandible. J Oral Maxillofac Surg 1992;50:1305-9. [17] Kurihashi T, Iwata H, Nasu M, et al. Experimental study on wound healing of alveolar bone sockets in the rat maxilla after x-ray irradiation. Odontology 2002;90:35-42. [18] Beumer J, Seto B. Dental extractions in the irradiated patient. Spec Care Dentist 1981;1:166-73. [19] Epstein JB, Wong FL, Stevenson-Moore P. Osteoradionecrosis: clinical experience and a proposal for classification. J Oral Maxillofac Surg 1987;45:104-10.