- Email: [email protected]
Validation of histologic changes induced by external irradiation in mandibular bone. An experimental animal model
Journal of Cranio-Maxillofacial Surgery (2010) 38, 47e53 Ó 2009 European Association for Cranio-Maxillofacial Surgery doi:10.1016/j.jcms.2009.07.011, available online at http://www.sciencedirect.com
Validation of histologic changes induced by external irradiation in mandibular bone. An experimental animal model* Matthias FENNER, MD, DDS1, Jung PARK, DDS, PhD1, Norbert SCHULZ1, Kerstin AMANN, MD, PhD2, Gerhard G. GRABENBAUER, MD, PhD3, Antje FAHRIG, MD, PhD3, Juergen KARG, MS3, Joerg WILTFANG, MD, DDS, PhD4, Friedrich W. NEUKAM, MD, DDS, PhD1, Emeka NKENKE, MD, DDS, PhD1 1
Department of Oral and Maxillofacial Surgery (Head: Prof. Dr. F.W. Neukam), University of Erlangen-Nuremberg, Glueckstrasse 11, 91054 Erlangen, Germany; 2 Department of Pathology (Head: Prof. Dr. A. Hartmann), University of Erlangen-Nuremberg, Krankenhausstrasse 8-10, 91054 Erlangen, Germany; 3 Department of Radiation Oncology (Head: Prof. Dr. R. Fietkau), University of Erlangen-Nuremberg, Universitaetstrasse 27, 91054 Erlangen, Germany; 4 Department of Oral and Maxillofacial Surgery (Head: Prof. Dr. Dr. J. Wiltfang), University of Kiel, Arnold-Heller-Strasse 16, 24105 Kiel, Germany
SUMMARY. The present experimental study sought to determine the effect of high-dose irradiation on the rat
mandible in order to establish an experimental model of radiogenic bone damage. The left mandibles of 20 adult Wistar rats were irradiated (single fraction 1500 cGy, total dose 60 Gy) by means of a hypofractionated stereotactic radiotherapy (hfSRT) over a period of 6 weeks. Follow-up was 6 weeks (group 1, n ¼ 10) and 12 weeks (group 2, n ¼ 10). The contralateral mandibles as well as 5 non-irradiated animals served as controls. Primary endpoints were fibrosis, loss of cell count, decreased immunohistochemical labelling for bone morphogenetic protein-2 (BMP-2) and osteocalcin as well as increased expression of transforming growth factor (TGF-b). Cell loss, progressive fibrosis, and focal necrosis were detected in all irradiated sites. Quantitative measurement revealed 32.0 ^ 8.7% and 37.3 ^ 9.5% empty osteocyte lacunae for groups 1 and 2 resp., compared to 16.3 ^ 4.7% and 18.9 ^ 4.9% on the contralateral side and 7.9 ^ 1.7% for unirradiated controls (ManneWhitney U test; p \ .01). BMP-2 and osteocalcin labelling showed a marked decrease in irradiated and contralateral sides while TGF-b was expressed strongly in irradiated sites only (for all p \.05). External hypofractionated irradiation with a total dose of 60 Gy is feasible in rats and yields all histologic changes attributed to osteoradionecrosis (ORN) after a follow-up of 6 weeks. The irradiation protocol is suitable for an assessment of regenerative options in severe radiogenic bone damage. As a split mouth design entails major inaccuracies healthy animals have to be used as controls. Ó 2009 European Association for Cranio-Maxillofacial Surgery Keywords: osteoradionecrosis, animal model, mandible, irradiation
(Ang et al., 2003; Teng and Futran, 2005). Also the regeneration of bone defects by application of osteoinductive proteins in irradiated sites has been reported in experimental settings and might offer an alternative to radical resection in the near future (Lorente et al., 1992; Wurzler et al., 1998; Springer et al., 2008). However, a prerequisite for an exact determination of the regenerative potential of any measure is the availability of an experimental irradiation model. The histopathologic effects of external irradiation with a total reference dose of up to 35 Gy on healthy bone have been characterized before in experimental trials (Aitasalo, 1986; Kalz et al., 1988). Yet in humans ORN is known to occur at doses of 60 Gy and beyond (Glanzmann and Gratz, 1995) which have not been applied in the aforementioned animal trials. Consequently an experimental model that may be suited to evaluate histologic and clinical changes attributed to mandibular ORN does not exist in the current literature. It has been the aim of the present experimental study to characterize the histologic changes attributable to
INTRODUCTION Osteoradionecrosis (ORN) of the mandible is a serious complication following radiation therapy with or without surgical intervention for malignancies of the head and neck (Thorn et al., 2000). The clinical findings include persistent pain and chronically exposed bone, pathologic fracture, or orocutaneous fistula (Ang et al., 2003). Tissue breakdown may be the result of radiogenic fibrosis, endarteriitis and subsequent tissue hypoxia, hypocellularity, and hypovascularity (Ewing, 1926; Marx, 1983). Yet the exact pathogenesis of ORN remains to be elucidated. The treatment of ORN comprises different modalities such as administration of antibiotics, hyperbaric oxygen therapy, debridement, as well as radical resection of affected tissues followed by microvascular reconstruction
*
The study was supported by grants of the Johannes und Frieda Marohn-Stiftung, Erlangen, Germany. 47
48 Journal of Cranio-Maxillofacial Surgery
radiogenic bone damage following external irradiation in adult rats with a total reference dose of 60 Gy. MATERIALS AND METHODS The study protocol was approved by the Animal Care Committee of the Regional Government of Mittelfranken (Ansbach, Germany, approval number 621-542531.3122/02). The goal was to induce a reproducible pattern of fibrosis and loss of cell labelling by irradiation with 60 Gy when compared to non-irradiated controls. Two groups of animals were formed according to the period of observation which started after completion of the irradiation protocol (experimental group 1, 10 animals, follow-up period of 6 weeks), (experimental group 2, 10 animals, follow-up period of 12 weeks). Five animals served as unirradiated controls (control group, 5 animals) (Fig. 1). Twenty-five adult Crl:(WI) BR Wistar rats (Charles River, Sulzfeld, Germany) with a body weight of approximately 300 g were maintained in airconditioned cages providing a constant temperature of 20e25 C with a relative humidity of 40e55% and a light-dark cycle of 12 h. Animals were given a standard pelleted rodent diet (No. 1320, Altromin, Lage, Germany) and water ad libitum. For radiation treatment, the animals were anaesthetized with a single intraperitoneal injection of 30 mg ketamine (Ketavet; Pharmacia & Upjohn, Erlangen, Germany) and 3.5 mg xylazine (Rompun; Bayer, Leverkusen, Germany). Native CT scans (1 mm, Siemens Volume ZoomÔ, Siemens Medical Solutions, Erlangen, Germany) were performed for each animal prior to the first fraction and the target volume was identified for stereotactic radiosurgery (NovalisÔ BrainScan Treatment Planning System, Version 5.31, BrainLAB AG, Feldkirchen, Germany) (Fig. 2). Posi-
CT-Scan
RT
RT
2 weeks
RT
2 weeks
tioning of the animals was performed using X-ray registration (ExacTracÔ System, BrainLAB AG, Feldkirchen, Germany). The treatment delivery system, dose calculation algorithm, and X-ray registration have been described previously in detail (Ernst-Stecken et al., 2007). For each fraction two anterioreposterior opposed fields of approximately 8 18 mm were applied using a 6 MV linear accelerator supplied with a mircomultileaf-collimator (Novalis, BrainLAB AG, Feldkirchen, Germany). The left mandible was covered by the 95% isodose, 15 Gy were prescribed to the isocentre (Fig. 2). All animals were irradiated with a fractionation scheme of 4 15 Gy (n ¼ 20 animals), targeted to the left mandible every other week. Animals were followed weekly and sacrificed 6 and 12 weeks after the last irradiation treatment. The mandible and surrounding soft tissues were harvested and fixed in 1.5% paraformaldehyde solution overnight. Primary endpoints for histological and immunohistochemical analysis were fibrosis, empty lacunae and reduced labelling index (LI) of bone morphogenetic protein-2 (BMP-2) and osteocalcin as well as expression of TGF-b (Aitasalo, 1986; Schultze-Mosgau et al., 2005). Briefly, the mandibles were cut perpendicular to the occlusal plane in the postmolar, molar, and canine region (Fig. 3). Specimens were decalcified using EDTA and embedded in paraffin as described previously (Park et al., 2003). Serial sections of approximately 4 mm were cut with a microtome, deparaffinized and rehydrated. From each tissue sample two consecutive sections were obtained and processed on microscopic slides, with one serving as a negative control in each case. A preparation known to be positive was also stained in each staining series as a positive control. Standard histological techniques were employed for haematoxylineeosin (HE) and toluidine blue staining. For determination of
RT
2 weeks
Histology
6 / 12 weeks
Fig. 1 e Flow-chart of study design.
Fig. 2 e Identification of target volume for stereotactic radiosurgery (NovalisÔ BrainScan Treatment Planning System, Version 5.31, BrainLAB AG, Feldkirchen, Germany) in CT scans performed prior to the first fraction. Target volume and 100% isodose are marked pink. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Validation of histologic changes induced by external irradiation in mandibular bone 49
Fig. 3 e Illustration of rat mandible. Histologic sections indicated as section 1 (canine region), section 2 (molar region) and section 3 (postmolar region).
osteocalcin, BMP-2, and TGF-b LI the avidinebiotin peroxidase complex (ABCePOX) method was used. The slides were incubated with a polyclonal primary rabbit anti-BMP-2/4 IgG antibody (sc9003, Santa Cruz Biotechnology, Santa Cruz, CA, USA), a polyclonal rabbit anti-TGF-b IgG antibody (sc146, Santa Cruz Biotechnology, Santa Cruz, CA, USA), and a monoclonal mouse anti-osteocalcin IgG antibody (OC4-30, Takara, Madison, WI, USA) in a humid chamber at room temperature for 1 h. As secondary antibodies a biotinylated goat antirabbit IgG antibody (E0432, Dako, Glostrup, Denmark) was used for BMP-2 and TGF-b assays, and a biotinylated rabbit anti-mouse IgG antibody (E0464; Dako, Glostrup, Denmark) was used for detection of osteocalcin (room temperature, 30 min). The chromogenic assay was obtained with aminoethyl carbazole (AEC) (‘‘AEC+’’ Substrate-Chromogen System, K3469, Dako, Glostrup, Denmark). To enhance the contrast and improve evaluation, the nuclei were counterstained with haemalaun (S3309, DAKO, Glostrup, Denmark). Slides were examined qualitatively under a bright-field microscope (Axioskop, Zeiss, Oberkochen, Germany) with 25e400 magnification. In HE stained sections the degree of fibrosis and necrotic areas was estimated visually by a semiquantitative scoring method and expressed as absent (), slight (+), moderate (++), and intense (+++) (Aitasalo, 1986). Osteocytes and empty lacunae were counted in section 2 (Fig. 4) at 400 enlargement in four high power fields (Schultze-Mosgau et al., 2005). For quantification of osteocalcin, BMP-2, and TGF-b expression, the LI (%) was determined as ratio of positively expressing cells and total number of cells per visual field. The evaluation was carried out independently by two examiners. Statistics For description of continuous variables, mean values are given with standard deviations. The Wilcoxon test was used for comparison of paired samples because normality of distribution could not be assumed due to small case numbers. The ManneWhitney U test was chosen for comparison of unpaired samples. p-Values #.05 were considered significant. All analyses were conducted with SPSS v. 15.0 (Statistical Package for the Social Sciences, SPSS Inc., Chicago, IL, USA).
Fig. 4 e Mandibular bone and surrounding soft tissues in the molar region (section 2) in experimental group 2 (60 Gy, 12 weeks following irradiation). Staining: Elastica/van Giesson; original magnification: 12.5.
RESULTS Following irradiation the animals showed reduced food uptake between day 2 and 8 which resolved again during subsequent follow-up. Skin erythema of grade 1e2 CTC (Trotti et al., 2003) with alopecia was visible after an interval of 10e14 days. During the following weeks a patchy atrophy of the skin with induration of subcutaneous tissues and loss of hair were noted. Five animals died during the period of irradiation due to unclear reasons. Deaths occurred between irradiation fraction number 3 (week 5) and 4 (week 7) in all cases. Five more animals were included subsequently and irradiated by the same protocol. The remaining animals recovered well and gained weight throughout the period of observation. No animals were lost during follow-up. In experimental group 2 (follow-up 12 weeks) a clouding of both eyes, varying in degree from slight to complete opacity became apparent between week 10 and 12 in all animals. At the end of follow-up soft tissue changes were classified as grade 1e2 according to radiation therapy oncology group (RTOG)/european organization for research and treatment of cancer (EORTC) (Late radiation morbidity scoring system) (Cooper et al., 1995). Ulcerations of skin or mucosa as well as necrosis of subcutaneous tissues were not observed. Qualitative and quantitative assessment In the irradiated sites histopathological changes known to occur after radiation exposure were detected in all animals. Changes included cellular loss, hypovascularity, progressive fibrosis, and focal necrosis. Changes were
50 Journal of Cranio-Maxillofacial Surgery
most prominent in the molar region (section no. 2). Progressive circumscribed fibrosis of the bone marrow was seen in all irradiated samples, most prominent in the molar region as marrow spaces were largest (Figs. 4e7). Fibrosis was neither detected in contralateral sides, nor in unirradiated controls. Also signs of necrosis were seen in irradiated sites only (Fig. 8). Empty lacunae were detected in irradiated and non-irradiated animals alike as well as in non-irradiated sites though at a different level. Quantitative measurement of osteocyte cell loss in HE staining revealed empty osteocyte lacunae of 32.0 ^ 8.7% and 37.3 ^ 9.5% for groups 1 and 2 resp. compared to 16.3 ^ 4.7% and 18.9 ^ 4.9% on the contralateral side (ManneWhitney U test; p \.01) (Table 1). In unirradiated controls only 7.9 ^ 1.7% appeared empty. The differences between irradiated sites and controls turned out to be significant (p \ .001). A comparison of changes in cell count over time did not yield any significant differences (Wilcoxon test; p . .05).
Immunohistochemical staining Expression of BMP-2 was reduced in irradiated sites and strongest in unirradiated controls (BMP-2 LI; 29.8 ^ 7.9% vs. 85.5 ^ 6.1%; 12 week follow-up; p\ .001) (Table 2). Also unirradiated sides of experimental groups 1 and 2 showed a marked decrease in BMP-2 labelling when compared to healthy controls (BMP-2 LI; 52.0 ^ 12.5% vs. 85.5 ^ 6.1%; 12 week follow-up; p \.01). Analysis of osteocalcin LI revealed significant differences between irradiated samples and unirradiated controls (osteocalcin LI; 22.5 ^ 18.3% vs. 85.8 ^ 11.3%; 12 week follow-up; p\ .01 resp.), while a difference between irradiated and unirradiated sides of experimental groups 1 and 2 could not be detected (p . .05). Also a comparison of changes in BMP-2 and osteocalcin LI over time did not yield any significant differences (Wilcoxon test; p . .05). Expression of TGF-b was strong in irradiated sites but found also on the contralateral sides of groups 1 and 2, though at a much lower level. No immunoreactivity was observed in unirradiated controls.
DISCUSSION The results of the present study demonstrate that histologic changes attributed to ORN (i.e. empty lacunae, fibrosis, necrosis) can be reproducibly obtained in rat mandibular bone by stereotactic irradiation using a total dose of 60 Gy. In addition, the decreased LI of BMP-2 and osteocalcin as well as the strong expression of TGF-b resemble signs of impaired osseous healing and a decreased regenerative capacity of mandibular bone (Schultze-Mosgau et al., 2005). The general time- and dose-dependency of the effect of irradiation on mandibular bone were characterized before in different protocols and rat models with total reference doses of up to 35 Gy (Aitasalo, 1986; Kalz et al., 1988). It was shown, that major histologic changes occurred within a period of 5 weeks while the extent only gradually increased during subsequent follow-up (Aitasalo, 1986). Grimm reported on histologic changes in a rabbit model induced by doses of up to 50 Gy (Grimm, 1969). Also in the present study the 12 week follow-up did not yield any major additional findings when compared to a period of observation of 6 weeks. The question has thus to be raised, if the time lag between end of irradiation and any therapeutic intervention actually affects osseous healing in the present model. In a critical size defect model in rat femur it was shown that the interval between irradiation and surgery had no influence on the effectiveness of impairing osseous healing (Arnold et al., 1998). An effective dose to inhibit bone healing by 50% was determined to be 16.4 Gy. This effect of irradiation remained stable for more than 180 days which is beyond 20% of the entire life-span of the animal (Arnold et al., 1998). Consequently, a follow-up period of 6 weeks may be sufficient to assess the effectiveness of any therapeutic approach to ORN by use of the present animal model. One important finding is the fact, that the contralateral, non-irradiated side showed a decrease in cell count when compared to unirradiated controls. Interestingly an increase in the percentage of cell loss was also described after irradiation of the hippocampal region in a previous experimental model using hypofractionated stereotactic radiotherapy (hfSRT) with 40 Gy (Ernst-Stecken et al., 2007). Though a sophisticated repositioning method known to
Fig. 5 e Mandibular bone in the molar region (section 2) in experimental group 2 (60 Gy, 12 weeks following irradiation, left picture) shows changes in cancellous bone architecture as well as fibrous tissue formation when compared to the control group (right picture). Staining: Elastica/van Giesson; original magnification: 50.
Validation of histologic changes induced by external irradiation in mandibular bone 51
Fig. 6 e Mandibular bone in the molar region (section 2) in experimental group 2 (60 Gy, 12 weeks following irradiation, left picture) displays cellular loss and fibrosis while the control group shows regular bone structure and marrow spaces (right picture). Staining: Elastica/van Giesson; original magnification: 100.
Fig. 7 e Mandibular bone in the molar region (section 2) in experimental group 2 (60 Gy, 12 weeks following irradiation, left picture) and control group (right picture). Staining: Elastica/van Giesson; original magnification: 200.
Fig. 8 e Calcified focal necrosis (section 2) in experimental group 2 (60 Gy, 12 weeks following irradiation). Staining: haematoxylin/eosin; original magnification: 200.
yield a high degree of accuracy (Ernst-Stecken et al., 2007) was used in both studies, an exact application of irradiation limited to one side of the mandible does not seem to be feasible in a rat model. In the present study, the application of
an X-ray-based image-guided stereotactic positioning system was performed to avoid possible errors in positioning thus allowing for a precise comparison of histologic changes on both sides of the mandible. Concerning the feasibility of the experimental model to induce the histologic changes described above, no major advantages are seen in this particular method of irradiation. However, based on these results it has to be suggested to use non-irradiated controls rather than the contralateral side of irradiated animals when evaluating any mode of treatment by use of this model. Several authors have reported possible therapeutic approaches to radiogenic bone damage using various irradiation protocols and experimental animal models (Lorente et al., 1992; Khouri et al., 1996; Wurzler et al., 1998; Springer et al., 2008). Yet most of these reports provided neither clinical nor histologic evidence of the extent of radiation induced damage. Consequently the question remains to what extent mandibular bone has been compromised by irradiation and if the histologic changes resembled an ORN. Also some authors used the contralateral side as control which may further limit the comparability to other trials. Recently, Springer et al. (2008) described the application of basic fibroblast growth factor (bFGF) and bone morphogenetic protein-2
52 Journal of Cranio-Maxillofacial Surgery Table 1 e Quantitative assessment of histologic changes in experimental groups, contralateral sides of irradiated animals, and non-irradiated controls Follow-up [weeks]
Group I (60 Gy) Group I (control) Group II (60 Gy) Group II (control) Non-irradiated controls
6 6 12 12
n (animals)
10 10 10 10 5
Empty osteocyte lacunae
Qualitative assessment [number of sites]
[%]
SD
Min
Max
Fibrosis
Necrosis
32.0 16.3 37.3 18.9 7.9
8.7 4.7 9.5 4.9 17.3
26.8 8.2 26.0 11.0 4.9
55.6 22.9 54.1 25.1 10.2
10 0 10 0 0
10 0 10 0 0
Table 2 e Quantitative assessment of BMP-2 and osteocalcin LI as well as qualitative analysis of TGF-b expression in experimental groups, contralateral sides of irradiated animals, and non-irradiated controls Follow-up [weeks]
Group I (60 Gy) Group I (control) Group II (60 Gy) Group II (control) Non-irradiated controls
6 6 12 12
n (animals)
10 10 10 10 5
BMP-2 LI [%]
Osteocalcin LI [%]
TGF-b
[%]
SD
Min
Max
[%]
SD
Min
Max
33.7 51.1 29.8 52.0 85.5
7.0 9.0 7.9 12.5 6.1
22.1 39.0 17.3 40.8 72.8
43.0 62.0 33.0 70.3 92.3
32.0 37.9 22.5 37.0 85.8
20.2 15.9 18.3 12.1 11.3
8.0 15.9 7.3 18.7 62.0
75.2 62.3 43.2 49.7 94.8
(rhBMP-2) in a rat brachytherapy model that had been shown to be feasible for simulating radiogenic bone damage. While paraffin histology was reported to reveal only minimal changes in bone structure, a significant decrease in bone apposition rate as well as a significant reduction of the average diameter of the mandibular condyles were observed in irradiated mandibles (Niehoff et al., 2008). As the analysis focussed on growth delay and bone regeneration in growing animals, it may be an interesting alternative to the present ORN model. Yet the limited availability of brachytherapy as well as the need to perform individual CT scans following brachytherapy tube implantation have to be taken into account. Clinically ORN presents as persistent pain and chronically exposed bone, pathologic fracture or orocutaneous fistula (Ang et al., 2003). The main goal of management of ORN is the treatment of symptoms and restoration of form and function (Ang et al., 2003). Consequently radical resection of affected tissues and immediate reconstruction by free-flap transfer is the mainstay of therapy while conservative measures have a limited role in the treatment of mandibular ORN (Ang et al., 2003; Annane et al., 2004). In contrast to a patient situation, macroscopic follow-up in the present trial failed to display the clinical problem of mandibular ORN. Thus the present model may be suited best as a preliminary assessment of osseous regeneration in radionecrotic bone. Yet the question remains if the clinical picture of ORN can be reproduced at all in an experimental setting. Apart from the effect of irradiation, the regenerative potential of tissues can be compromised by local factors such as infection or surgical trauma as well as systemic factors (i.e. comorbidities, medication) (Glanzmann and Gratz, 1995; Curi and Dib, 1997; Goldwaser et al., 2007). Though the evidence is inconclusive on this issue (Goldwaser et al., 2007), the creation of a bone defect or tooth extraction might improve the comparability of the experimental model to a patient situation. As the onset of clinical ORN is known to occur more than 12 weeks after completion of radiation treatment, the period of observation may have been too short to see any effect in the present
+++ + +++ +
trial. On the other hand the metabolic rates in rodents are known to be 4e6 times higher as compared to humans (Schultze-Mosgau et al., 2005). The interval between the end of radiotherapy and histologic examination would thus be comparable to a follow-up of 24e36 or 48e72 weeks resp. in a patient situation. As ORN has been reported to develop in the first year after the completion of treatment in the majority of patient cases, a longer follow-up does not seem to yield any advantages in the present trial (Glanzmann and Gratz, 1995). Apart from the time-dependency, the relationship between radiation dose and the risk of ORN is well established in the literature (Goldwaser et al., 2007). Though an exact threshold does not seem to exist, ORN has been shown to occur following conventionally fractionated radiotherapy up to a total target dose of 66 Gy and higher (Glanzmann and Gratz, 1995). The fractionation schedule helps to minimize side effects of radiotherapy by allowing normal tissues to repair sublethal injuries between fractions. In a single-dose schedule the damage incurred by irradiation has been shown to be greater for tumour as well as normal tissues when compared to fractionation of the same dose (e.g. 15 Gy vs. 10 1.5 Gy) (Liu et al., 2003). The radiobiological effect is thus not a function of the total dose alone but also depends on the fractionation schedule as well as the type of tissues. As a consequence of this, the total dose has to be increased in fractionated radiotherapy to receive a radiobiologic effect equivalent to a given single-dose application. The radiobiologic effects of hypofractionated stereotactic treatment and of multiple-fraction regimen can be compared by use of the linear-quadratic (LQ) model (Liu et al., 2003). According to the LQ model, the biologic effect of 15 Gy applied at one time is equivalent to a total dose ranging from 29 Gy to 42 Gy in a multiple-fraction regimen using ten fractions (29 Gy for a/b ¼ 10 Gy i.e. early reacting tissues such as the skin, the intestinal epithelium, and tumours; 42 Gy for a/b ¼ 2 Gy i.e. late reacting tissues such as bone) (Liu et al., 2003). Thus the total dose applied in the present model would correspond
Validation of histologic changes induced by external irradiation in mandibular bone 53
to a total dose ranging between 116 Gy and 168 Gy in conventionally fractionated radiotherapy. In conclusion both the period of follow-up and the total reference dose should be sufficient to allow for the development of chronically exposed bone, pathologic fracture or orocutaneous fistula. However, for comprehensive modelling clinical symptoms of ORN, an experimental animal model remains to be established. CONCLUSION External irradiation with a total reference dose of 60 Gy is feasible in rats and yields all histologic changes attributed to ORN after a follow-up of 6 weeks. The irradiation protocol is suitable for an assessment of regenerative options in severe radiogenic bone damage. As a split mouth design entails major inaccuracies healthy animals have to be used as controls. ACKNOWLEDGEMENT The study was supported by grants of the Johannes und Frieda Marohn-Stiftung, Erlangen, Germany.
References Aitasalo K: Bone tissue response to irradiation and treatment model of mandibular irradiation injury. An experimental and clinical study. Acta Otolaryngol Suppl 428: 1e54, 1986 Ang E, Black C, Irish J, Brown DH, Gullane P, O’Sullivan B, Neligan PC: Reconstructive options in the treatment of osteoradionecrosis of the craniomaxillofacial skeleton. Br J Plast Surg 56: 92e99, 2003 Annane D, Depondt J, Aubert P, Villart M, Ge´hanno, Gajdos P, Chevret S: Hyperbaric oxygen therapy for radionecrosis of the jaw: a randomized, placebo-controlled, double-blind trial from the ORN96 study group. J Clin Oncol 22: 4893e4900, 2004 Arnold M, Kummermehr J, Trott KR: Radiation induced impairment of osseous healing: quantitative studies using a standard drilling defect in rat femur. Radiother Oncol 48: 259e265, 1998 Cooper JS, Fu K, Marks J, Silverman S: Late effects of radiation therapy in the head and neck region. Int J Radiat Oncol Biol Phys 31: 1141e1164, 1995 Curi MM, Dib LL: Osteoradionecrosis of the jaws: a retrospective study of the background factors and treatment in 104 cases. J Maxillofac Surg 55: 540e544, 1997 Ernst-Stecken A, Jeske I, Hess A, Ro¨del F, Ganslandt O, Grabenbauer G, Sauer R, Brune K, Blu¨mcke I: Hypofractionated stereotactic radiotherapy to the rat hippocampus. Determination of dose response and tolerance. Strahlenther Onkol 183: 440e446, 2007 Ewing J: Radiation osteitis. Acta Radiol 6: 399e412, 1926 Glanzmann C, Gratz KW: Radionecrosis of the mandibula: a retrospective analysis of the incidence and risk factors. Radiother Oncol 36: 94e100, 1995 Goldwaser BR, Chuang SK, Kaban LB, August M: Risk factor assessment for the development of osteoradionecrosis. J Oral Maxillofac Surg 65: 2311e2316, 2007 Grimm G: Animal experimental studies on the radiation injuries in the mandible of full grown rabbits. I. Morphological and histochemical findings. Dtsch Zahn Mund Kieferheilkd Zentralbl 53: 352e362, 1969
Kalz W, Buchali K, Dalluge KH, Lange KP, Magnus S, Seiler S, Topel M: Diagnosis of radiogenic jaw bone damage. The first results with an animal model. Zahn Mund Kieferheilkd Zentralbl 76: 835e837, 1988 Khouri RK, Brown DM, Koudsi B, Deune EG, Gilula LA, Cooley BC, Reddi AH: Repair of calvarial defects with flap tissue: role of bone morphogenetic proteins and competent responding tissues. Plast Reconstr Surg 98: 103e109, 1996 Liu L, Bassano DA, Prasad SC, Hahn SS, Chung CT: The linearquadratic model and fractionated stereotactic radiotherapy. Int J Radiat Oncol Biol Phys 57: 827e832, 2003 Lorente CA, Song BZ, Donoff RB: Healing of bony defects in the irradiated and unirradiated rat mandible. J Oral Maxillofac Surg 50: 1305e1309, 1992 Marx RE: Osteoradionecrosis: a new concept of its pathophysiology. J Oral Maxillofac Surg 41: 283e288, 1983 Niehoff P, Springer IN, Ac¸il Y, Lange A, Marget M, Rolda´n JC, Ko¨ppe K, Warnke PH, Kimmig B, Wiltfang J: HDR brachytherapy irradiation of the jaw e as a new experimental model of radiogenic bone damage. J Craniomaxillofac Surg 36: 203e209, 2008 Park J, Ries J, Gelse K, Kloss FR, van der Mark K, Wiltfang J, Neukam FW, Schneider H: Bone regeneration in critical size defects by cell-mediated BMP-2 gene transfer e a comparison of adenoviral vectors and liposomes. Gene Ther 10: 1089e1098, 2003 Schultze-Mosgau S, Lehner B, Ro¨del F, Wehrhan F, Amann K, Kopp J, Thorwarth M, Nkenke E, Grabenbauer G: Expression of bone morphogenic protein 2/4, transforming growth factor-beta1, and bone matrix protein expression in healing area between vascular tibia grafts and irradiated bone-experimental model of osteonecrosis. Int J Radiat Oncol Biol Phys 61: 1189e1196, 2005 Springer IN, Niehoff P, Ac¸il Y, Marget M, Lange A, Warnke PH, Pielenz H, Rolda´n JC, Wiltfang J: BMP-2 and bFGF in an irradiated bone model. J Craniomaxillofac Surg 36: 210e217, 2008 Teng MS, Futran ND: Osteoradionecrosis of the mandible. Curr Opin Otolaryngol Head Neck Surg 13: 217e221, 2005 Thorn JJ, Hansen HS, Specht L, Bastholt L: Osteoradionecrosis of the jaws: clinical characteristics and relation to the field of irradiation. J Oral Maxillofac Surg 58: 1088e1093, 2000 Trotti A, Colevas AD, Setser A, Rusch V, Jaques D, Budach V, Langer C, Murphy B, Cumberlin R, Coleman CN, Rubin P: CTCAE v3.0: development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol 13: 176e181, 2003 Wurzler KK, De Weese TL, Sebald W, Reddi AH: Radiationinduced impairment of bone healing can be overcome by recombinant human bone morphogenetic protein-2. J Craniofac Surg 9: 131e137, 1998
Dr. Matthias FENNER, MD, DDS University of Erlangen-Nuremberg Department of Oral and Maxillofacial Surgery Glueckstrasse 11 91054 Erlangen Germany Tel.: +49 9131 8533653 Fax: +49 9131 8534219 E-mail: [email protected] Paper received 26 November 2008 Accepted 29 July 2009
Validation of histologic changes induced by external irradiation in mandibular bone. An experimental animal model* Matthias FENNER, MD, DDS1, Jung PARK, DDS, PhD1, Norbert SCHULZ1, Kerstin AMANN, MD, PhD2, Gerhard G. GRABENBAUER, MD, PhD3, Antje FAHRIG, MD, PhD3, Juergen KARG, MS3, Joerg WILTFANG, MD, DDS, PhD4, Friedrich W. NEUKAM, MD, DDS, PhD1, Emeka NKENKE, MD, DDS, PhD1 1
Department of Oral and Maxillofacial Surgery (Head: Prof. Dr. F.W. Neukam), University of Erlangen-Nuremberg, Glueckstrasse 11, 91054 Erlangen, Germany; 2 Department of Pathology (Head: Prof. Dr. A. Hartmann), University of Erlangen-Nuremberg, Krankenhausstrasse 8-10, 91054 Erlangen, Germany; 3 Department of Radiation Oncology (Head: Prof. Dr. R. Fietkau), University of Erlangen-Nuremberg, Universitaetstrasse 27, 91054 Erlangen, Germany; 4 Department of Oral and Maxillofacial Surgery (Head: Prof. Dr. Dr. J. Wiltfang), University of Kiel, Arnold-Heller-Strasse 16, 24105 Kiel, Germany
SUMMARY. The present experimental study sought to determine the effect of high-dose irradiation on the rat
mandible in order to establish an experimental model of radiogenic bone damage. The left mandibles of 20 adult Wistar rats were irradiated (single fraction 1500 cGy, total dose 60 Gy) by means of a hypofractionated stereotactic radiotherapy (hfSRT) over a period of 6 weeks. Follow-up was 6 weeks (group 1, n ¼ 10) and 12 weeks (group 2, n ¼ 10). The contralateral mandibles as well as 5 non-irradiated animals served as controls. Primary endpoints were fibrosis, loss of cell count, decreased immunohistochemical labelling for bone morphogenetic protein-2 (BMP-2) and osteocalcin as well as increased expression of transforming growth factor (TGF-b). Cell loss, progressive fibrosis, and focal necrosis were detected in all irradiated sites. Quantitative measurement revealed 32.0 ^ 8.7% and 37.3 ^ 9.5% empty osteocyte lacunae for groups 1 and 2 resp., compared to 16.3 ^ 4.7% and 18.9 ^ 4.9% on the contralateral side and 7.9 ^ 1.7% for unirradiated controls (ManneWhitney U test; p \ .01). BMP-2 and osteocalcin labelling showed a marked decrease in irradiated and contralateral sides while TGF-b was expressed strongly in irradiated sites only (for all p \.05). External hypofractionated irradiation with a total dose of 60 Gy is feasible in rats and yields all histologic changes attributed to osteoradionecrosis (ORN) after a follow-up of 6 weeks. The irradiation protocol is suitable for an assessment of regenerative options in severe radiogenic bone damage. As a split mouth design entails major inaccuracies healthy animals have to be used as controls. Ó 2009 European Association for Cranio-Maxillofacial Surgery Keywords: osteoradionecrosis, animal model, mandible, irradiation
(Ang et al., 2003; Teng and Futran, 2005). Also the regeneration of bone defects by application of osteoinductive proteins in irradiated sites has been reported in experimental settings and might offer an alternative to radical resection in the near future (Lorente et al., 1992; Wurzler et al., 1998; Springer et al., 2008). However, a prerequisite for an exact determination of the regenerative potential of any measure is the availability of an experimental irradiation model. The histopathologic effects of external irradiation with a total reference dose of up to 35 Gy on healthy bone have been characterized before in experimental trials (Aitasalo, 1986; Kalz et al., 1988). Yet in humans ORN is known to occur at doses of 60 Gy and beyond (Glanzmann and Gratz, 1995) which have not been applied in the aforementioned animal trials. Consequently an experimental model that may be suited to evaluate histologic and clinical changes attributed to mandibular ORN does not exist in the current literature. It has been the aim of the present experimental study to characterize the histologic changes attributable to
INTRODUCTION Osteoradionecrosis (ORN) of the mandible is a serious complication following radiation therapy with or without surgical intervention for malignancies of the head and neck (Thorn et al., 2000). The clinical findings include persistent pain and chronically exposed bone, pathologic fracture, or orocutaneous fistula (Ang et al., 2003). Tissue breakdown may be the result of radiogenic fibrosis, endarteriitis and subsequent tissue hypoxia, hypocellularity, and hypovascularity (Ewing, 1926; Marx, 1983). Yet the exact pathogenesis of ORN remains to be elucidated. The treatment of ORN comprises different modalities such as administration of antibiotics, hyperbaric oxygen therapy, debridement, as well as radical resection of affected tissues followed by microvascular reconstruction
*
The study was supported by grants of the Johannes und Frieda Marohn-Stiftung, Erlangen, Germany. 47
48 Journal of Cranio-Maxillofacial Surgery
radiogenic bone damage following external irradiation in adult rats with a total reference dose of 60 Gy. MATERIALS AND METHODS The study protocol was approved by the Animal Care Committee of the Regional Government of Mittelfranken (Ansbach, Germany, approval number 621-542531.3122/02). The goal was to induce a reproducible pattern of fibrosis and loss of cell labelling by irradiation with 60 Gy when compared to non-irradiated controls. Two groups of animals were formed according to the period of observation which started after completion of the irradiation protocol (experimental group 1, 10 animals, follow-up period of 6 weeks), (experimental group 2, 10 animals, follow-up period of 12 weeks). Five animals served as unirradiated controls (control group, 5 animals) (Fig. 1). Twenty-five adult Crl:(WI) BR Wistar rats (Charles River, Sulzfeld, Germany) with a body weight of approximately 300 g were maintained in airconditioned cages providing a constant temperature of 20e25 C with a relative humidity of 40e55% and a light-dark cycle of 12 h. Animals were given a standard pelleted rodent diet (No. 1320, Altromin, Lage, Germany) and water ad libitum. For radiation treatment, the animals were anaesthetized with a single intraperitoneal injection of 30 mg ketamine (Ketavet; Pharmacia & Upjohn, Erlangen, Germany) and 3.5 mg xylazine (Rompun; Bayer, Leverkusen, Germany). Native CT scans (1 mm, Siemens Volume ZoomÔ, Siemens Medical Solutions, Erlangen, Germany) were performed for each animal prior to the first fraction and the target volume was identified for stereotactic radiosurgery (NovalisÔ BrainScan Treatment Planning System, Version 5.31, BrainLAB AG, Feldkirchen, Germany) (Fig. 2). Posi-
CT-Scan
RT
RT
2 weeks
RT
2 weeks
tioning of the animals was performed using X-ray registration (ExacTracÔ System, BrainLAB AG, Feldkirchen, Germany). The treatment delivery system, dose calculation algorithm, and X-ray registration have been described previously in detail (Ernst-Stecken et al., 2007). For each fraction two anterioreposterior opposed fields of approximately 8 18 mm were applied using a 6 MV linear accelerator supplied with a mircomultileaf-collimator (Novalis, BrainLAB AG, Feldkirchen, Germany). The left mandible was covered by the 95% isodose, 15 Gy were prescribed to the isocentre (Fig. 2). All animals were irradiated with a fractionation scheme of 4 15 Gy (n ¼ 20 animals), targeted to the left mandible every other week. Animals were followed weekly and sacrificed 6 and 12 weeks after the last irradiation treatment. The mandible and surrounding soft tissues were harvested and fixed in 1.5% paraformaldehyde solution overnight. Primary endpoints for histological and immunohistochemical analysis were fibrosis, empty lacunae and reduced labelling index (LI) of bone morphogenetic protein-2 (BMP-2) and osteocalcin as well as expression of TGF-b (Aitasalo, 1986; Schultze-Mosgau et al., 2005). Briefly, the mandibles were cut perpendicular to the occlusal plane in the postmolar, molar, and canine region (Fig. 3). Specimens were decalcified using EDTA and embedded in paraffin as described previously (Park et al., 2003). Serial sections of approximately 4 mm were cut with a microtome, deparaffinized and rehydrated. From each tissue sample two consecutive sections were obtained and processed on microscopic slides, with one serving as a negative control in each case. A preparation known to be positive was also stained in each staining series as a positive control. Standard histological techniques were employed for haematoxylineeosin (HE) and toluidine blue staining. For determination of
RT
2 weeks
Histology
6 / 12 weeks
Fig. 1 e Flow-chart of study design.
Fig. 2 e Identification of target volume for stereotactic radiosurgery (NovalisÔ BrainScan Treatment Planning System, Version 5.31, BrainLAB AG, Feldkirchen, Germany) in CT scans performed prior to the first fraction. Target volume and 100% isodose are marked pink. (For interpretation of the references to colour in this figure legend, the reader is referred to the web version of this article.)
Validation of histologic changes induced by external irradiation in mandibular bone 49
Fig. 3 e Illustration of rat mandible. Histologic sections indicated as section 1 (canine region), section 2 (molar region) and section 3 (postmolar region).
osteocalcin, BMP-2, and TGF-b LI the avidinebiotin peroxidase complex (ABCePOX) method was used. The slides were incubated with a polyclonal primary rabbit anti-BMP-2/4 IgG antibody (sc9003, Santa Cruz Biotechnology, Santa Cruz, CA, USA), a polyclonal rabbit anti-TGF-b IgG antibody (sc146, Santa Cruz Biotechnology, Santa Cruz, CA, USA), and a monoclonal mouse anti-osteocalcin IgG antibody (OC4-30, Takara, Madison, WI, USA) in a humid chamber at room temperature for 1 h. As secondary antibodies a biotinylated goat antirabbit IgG antibody (E0432, Dako, Glostrup, Denmark) was used for BMP-2 and TGF-b assays, and a biotinylated rabbit anti-mouse IgG antibody (E0464; Dako, Glostrup, Denmark) was used for detection of osteocalcin (room temperature, 30 min). The chromogenic assay was obtained with aminoethyl carbazole (AEC) (‘‘AEC+’’ Substrate-Chromogen System, K3469, Dako, Glostrup, Denmark). To enhance the contrast and improve evaluation, the nuclei were counterstained with haemalaun (S3309, DAKO, Glostrup, Denmark). Slides were examined qualitatively under a bright-field microscope (Axioskop, Zeiss, Oberkochen, Germany) with 25e400 magnification. In HE stained sections the degree of fibrosis and necrotic areas was estimated visually by a semiquantitative scoring method and expressed as absent (), slight (+), moderate (++), and intense (+++) (Aitasalo, 1986). Osteocytes and empty lacunae were counted in section 2 (Fig. 4) at 400 enlargement in four high power fields (Schultze-Mosgau et al., 2005). For quantification of osteocalcin, BMP-2, and TGF-b expression, the LI (%) was determined as ratio of positively expressing cells and total number of cells per visual field. The evaluation was carried out independently by two examiners. Statistics For description of continuous variables, mean values are given with standard deviations. The Wilcoxon test was used for comparison of paired samples because normality of distribution could not be assumed due to small case numbers. The ManneWhitney U test was chosen for comparison of unpaired samples. p-Values #.05 were considered significant. All analyses were conducted with SPSS v. 15.0 (Statistical Package for the Social Sciences, SPSS Inc., Chicago, IL, USA).
Fig. 4 e Mandibular bone and surrounding soft tissues in the molar region (section 2) in experimental group 2 (60 Gy, 12 weeks following irradiation). Staining: Elastica/van Giesson; original magnification: 12.5.
RESULTS Following irradiation the animals showed reduced food uptake between day 2 and 8 which resolved again during subsequent follow-up. Skin erythema of grade 1e2 CTC (Trotti et al., 2003) with alopecia was visible after an interval of 10e14 days. During the following weeks a patchy atrophy of the skin with induration of subcutaneous tissues and loss of hair were noted. Five animals died during the period of irradiation due to unclear reasons. Deaths occurred between irradiation fraction number 3 (week 5) and 4 (week 7) in all cases. Five more animals were included subsequently and irradiated by the same protocol. The remaining animals recovered well and gained weight throughout the period of observation. No animals were lost during follow-up. In experimental group 2 (follow-up 12 weeks) a clouding of both eyes, varying in degree from slight to complete opacity became apparent between week 10 and 12 in all animals. At the end of follow-up soft tissue changes were classified as grade 1e2 according to radiation therapy oncology group (RTOG)/european organization for research and treatment of cancer (EORTC) (Late radiation morbidity scoring system) (Cooper et al., 1995). Ulcerations of skin or mucosa as well as necrosis of subcutaneous tissues were not observed. Qualitative and quantitative assessment In the irradiated sites histopathological changes known to occur after radiation exposure were detected in all animals. Changes included cellular loss, hypovascularity, progressive fibrosis, and focal necrosis. Changes were
50 Journal of Cranio-Maxillofacial Surgery
most prominent in the molar region (section no. 2). Progressive circumscribed fibrosis of the bone marrow was seen in all irradiated samples, most prominent in the molar region as marrow spaces were largest (Figs. 4e7). Fibrosis was neither detected in contralateral sides, nor in unirradiated controls. Also signs of necrosis were seen in irradiated sites only (Fig. 8). Empty lacunae were detected in irradiated and non-irradiated animals alike as well as in non-irradiated sites though at a different level. Quantitative measurement of osteocyte cell loss in HE staining revealed empty osteocyte lacunae of 32.0 ^ 8.7% and 37.3 ^ 9.5% for groups 1 and 2 resp. compared to 16.3 ^ 4.7% and 18.9 ^ 4.9% on the contralateral side (ManneWhitney U test; p \.01) (Table 1). In unirradiated controls only 7.9 ^ 1.7% appeared empty. The differences between irradiated sites and controls turned out to be significant (p \ .001). A comparison of changes in cell count over time did not yield any significant differences (Wilcoxon test; p . .05).
Immunohistochemical staining Expression of BMP-2 was reduced in irradiated sites and strongest in unirradiated controls (BMP-2 LI; 29.8 ^ 7.9% vs. 85.5 ^ 6.1%; 12 week follow-up; p\ .001) (Table 2). Also unirradiated sides of experimental groups 1 and 2 showed a marked decrease in BMP-2 labelling when compared to healthy controls (BMP-2 LI; 52.0 ^ 12.5% vs. 85.5 ^ 6.1%; 12 week follow-up; p \.01). Analysis of osteocalcin LI revealed significant differences between irradiated samples and unirradiated controls (osteocalcin LI; 22.5 ^ 18.3% vs. 85.8 ^ 11.3%; 12 week follow-up; p\ .01 resp.), while a difference between irradiated and unirradiated sides of experimental groups 1 and 2 could not be detected (p . .05). Also a comparison of changes in BMP-2 and osteocalcin LI over time did not yield any significant differences (Wilcoxon test; p . .05). Expression of TGF-b was strong in irradiated sites but found also on the contralateral sides of groups 1 and 2, though at a much lower level. No immunoreactivity was observed in unirradiated controls.
DISCUSSION The results of the present study demonstrate that histologic changes attributed to ORN (i.e. empty lacunae, fibrosis, necrosis) can be reproducibly obtained in rat mandibular bone by stereotactic irradiation using a total dose of 60 Gy. In addition, the decreased LI of BMP-2 and osteocalcin as well as the strong expression of TGF-b resemble signs of impaired osseous healing and a decreased regenerative capacity of mandibular bone (Schultze-Mosgau et al., 2005). The general time- and dose-dependency of the effect of irradiation on mandibular bone were characterized before in different protocols and rat models with total reference doses of up to 35 Gy (Aitasalo, 1986; Kalz et al., 1988). It was shown, that major histologic changes occurred within a period of 5 weeks while the extent only gradually increased during subsequent follow-up (Aitasalo, 1986). Grimm reported on histologic changes in a rabbit model induced by doses of up to 50 Gy (Grimm, 1969). Also in the present study the 12 week follow-up did not yield any major additional findings when compared to a period of observation of 6 weeks. The question has thus to be raised, if the time lag between end of irradiation and any therapeutic intervention actually affects osseous healing in the present model. In a critical size defect model in rat femur it was shown that the interval between irradiation and surgery had no influence on the effectiveness of impairing osseous healing (Arnold et al., 1998). An effective dose to inhibit bone healing by 50% was determined to be 16.4 Gy. This effect of irradiation remained stable for more than 180 days which is beyond 20% of the entire life-span of the animal (Arnold et al., 1998). Consequently, a follow-up period of 6 weeks may be sufficient to assess the effectiveness of any therapeutic approach to ORN by use of the present animal model. One important finding is the fact, that the contralateral, non-irradiated side showed a decrease in cell count when compared to unirradiated controls. Interestingly an increase in the percentage of cell loss was also described after irradiation of the hippocampal region in a previous experimental model using hypofractionated stereotactic radiotherapy (hfSRT) with 40 Gy (Ernst-Stecken et al., 2007). Though a sophisticated repositioning method known to
Fig. 5 e Mandibular bone in the molar region (section 2) in experimental group 2 (60 Gy, 12 weeks following irradiation, left picture) shows changes in cancellous bone architecture as well as fibrous tissue formation when compared to the control group (right picture). Staining: Elastica/van Giesson; original magnification: 50.
Validation of histologic changes induced by external irradiation in mandibular bone 51
Fig. 6 e Mandibular bone in the molar region (section 2) in experimental group 2 (60 Gy, 12 weeks following irradiation, left picture) displays cellular loss and fibrosis while the control group shows regular bone structure and marrow spaces (right picture). Staining: Elastica/van Giesson; original magnification: 100.
Fig. 7 e Mandibular bone in the molar region (section 2) in experimental group 2 (60 Gy, 12 weeks following irradiation, left picture) and control group (right picture). Staining: Elastica/van Giesson; original magnification: 200.
Fig. 8 e Calcified focal necrosis (section 2) in experimental group 2 (60 Gy, 12 weeks following irradiation). Staining: haematoxylin/eosin; original magnification: 200.
yield a high degree of accuracy (Ernst-Stecken et al., 2007) was used in both studies, an exact application of irradiation limited to one side of the mandible does not seem to be feasible in a rat model. In the present study, the application of
an X-ray-based image-guided stereotactic positioning system was performed to avoid possible errors in positioning thus allowing for a precise comparison of histologic changes on both sides of the mandible. Concerning the feasibility of the experimental model to induce the histologic changes described above, no major advantages are seen in this particular method of irradiation. However, based on these results it has to be suggested to use non-irradiated controls rather than the contralateral side of irradiated animals when evaluating any mode of treatment by use of this model. Several authors have reported possible therapeutic approaches to radiogenic bone damage using various irradiation protocols and experimental animal models (Lorente et al., 1992; Khouri et al., 1996; Wurzler et al., 1998; Springer et al., 2008). Yet most of these reports provided neither clinical nor histologic evidence of the extent of radiation induced damage. Consequently the question remains to what extent mandibular bone has been compromised by irradiation and if the histologic changes resembled an ORN. Also some authors used the contralateral side as control which may further limit the comparability to other trials. Recently, Springer et al. (2008) described the application of basic fibroblast growth factor (bFGF) and bone morphogenetic protein-2
52 Journal of Cranio-Maxillofacial Surgery Table 1 e Quantitative assessment of histologic changes in experimental groups, contralateral sides of irradiated animals, and non-irradiated controls Follow-up [weeks]
Group I (60 Gy) Group I (control) Group II (60 Gy) Group II (control) Non-irradiated controls
6 6 12 12
n (animals)
10 10 10 10 5
Empty osteocyte lacunae
Qualitative assessment [number of sites]
[%]
SD
Min
Max
Fibrosis
Necrosis
32.0 16.3 37.3 18.9 7.9
8.7 4.7 9.5 4.9 17.3
26.8 8.2 26.0 11.0 4.9
55.6 22.9 54.1 25.1 10.2
10 0 10 0 0
10 0 10 0 0
Table 2 e Quantitative assessment of BMP-2 and osteocalcin LI as well as qualitative analysis of TGF-b expression in experimental groups, contralateral sides of irradiated animals, and non-irradiated controls Follow-up [weeks]
Group I (60 Gy) Group I (control) Group II (60 Gy) Group II (control) Non-irradiated controls
6 6 12 12
n (animals)
10 10 10 10 5
BMP-2 LI [%]
Osteocalcin LI [%]
TGF-b
[%]
SD
Min
Max
[%]
SD
Min
Max
33.7 51.1 29.8 52.0 85.5
7.0 9.0 7.9 12.5 6.1
22.1 39.0 17.3 40.8 72.8
43.0 62.0 33.0 70.3 92.3
32.0 37.9 22.5 37.0 85.8
20.2 15.9 18.3 12.1 11.3
8.0 15.9 7.3 18.7 62.0
75.2 62.3 43.2 49.7 94.8
(rhBMP-2) in a rat brachytherapy model that had been shown to be feasible for simulating radiogenic bone damage. While paraffin histology was reported to reveal only minimal changes in bone structure, a significant decrease in bone apposition rate as well as a significant reduction of the average diameter of the mandibular condyles were observed in irradiated mandibles (Niehoff et al., 2008). As the analysis focussed on growth delay and bone regeneration in growing animals, it may be an interesting alternative to the present ORN model. Yet the limited availability of brachytherapy as well as the need to perform individual CT scans following brachytherapy tube implantation have to be taken into account. Clinically ORN presents as persistent pain and chronically exposed bone, pathologic fracture or orocutaneous fistula (Ang et al., 2003). The main goal of management of ORN is the treatment of symptoms and restoration of form and function (Ang et al., 2003). Consequently radical resection of affected tissues and immediate reconstruction by free-flap transfer is the mainstay of therapy while conservative measures have a limited role in the treatment of mandibular ORN (Ang et al., 2003; Annane et al., 2004). In contrast to a patient situation, macroscopic follow-up in the present trial failed to display the clinical problem of mandibular ORN. Thus the present model may be suited best as a preliminary assessment of osseous regeneration in radionecrotic bone. Yet the question remains if the clinical picture of ORN can be reproduced at all in an experimental setting. Apart from the effect of irradiation, the regenerative potential of tissues can be compromised by local factors such as infection or surgical trauma as well as systemic factors (i.e. comorbidities, medication) (Glanzmann and Gratz, 1995; Curi and Dib, 1997; Goldwaser et al., 2007). Though the evidence is inconclusive on this issue (Goldwaser et al., 2007), the creation of a bone defect or tooth extraction might improve the comparability of the experimental model to a patient situation. As the onset of clinical ORN is known to occur more than 12 weeks after completion of radiation treatment, the period of observation may have been too short to see any effect in the present
+++ + +++ +
trial. On the other hand the metabolic rates in rodents are known to be 4e6 times higher as compared to humans (Schultze-Mosgau et al., 2005). The interval between the end of radiotherapy and histologic examination would thus be comparable to a follow-up of 24e36 or 48e72 weeks resp. in a patient situation. As ORN has been reported to develop in the first year after the completion of treatment in the majority of patient cases, a longer follow-up does not seem to yield any advantages in the present trial (Glanzmann and Gratz, 1995). Apart from the time-dependency, the relationship between radiation dose and the risk of ORN is well established in the literature (Goldwaser et al., 2007). Though an exact threshold does not seem to exist, ORN has been shown to occur following conventionally fractionated radiotherapy up to a total target dose of 66 Gy and higher (Glanzmann and Gratz, 1995). The fractionation schedule helps to minimize side effects of radiotherapy by allowing normal tissues to repair sublethal injuries between fractions. In a single-dose schedule the damage incurred by irradiation has been shown to be greater for tumour as well as normal tissues when compared to fractionation of the same dose (e.g. 15 Gy vs. 10 1.5 Gy) (Liu et al., 2003). The radiobiological effect is thus not a function of the total dose alone but also depends on the fractionation schedule as well as the type of tissues. As a consequence of this, the total dose has to be increased in fractionated radiotherapy to receive a radiobiologic effect equivalent to a given single-dose application. The radiobiologic effects of hypofractionated stereotactic treatment and of multiple-fraction regimen can be compared by use of the linear-quadratic (LQ) model (Liu et al., 2003). According to the LQ model, the biologic effect of 15 Gy applied at one time is equivalent to a total dose ranging from 29 Gy to 42 Gy in a multiple-fraction regimen using ten fractions (29 Gy for a/b ¼ 10 Gy i.e. early reacting tissues such as the skin, the intestinal epithelium, and tumours; 42 Gy for a/b ¼ 2 Gy i.e. late reacting tissues such as bone) (Liu et al., 2003). Thus the total dose applied in the present model would correspond
Validation of histologic changes induced by external irradiation in mandibular bone 53
to a total dose ranging between 116 Gy and 168 Gy in conventionally fractionated radiotherapy. In conclusion both the period of follow-up and the total reference dose should be sufficient to allow for the development of chronically exposed bone, pathologic fracture or orocutaneous fistula. However, for comprehensive modelling clinical symptoms of ORN, an experimental animal model remains to be established. CONCLUSION External irradiation with a total reference dose of 60 Gy is feasible in rats and yields all histologic changes attributed to ORN after a follow-up of 6 weeks. The irradiation protocol is suitable for an assessment of regenerative options in severe radiogenic bone damage. As a split mouth design entails major inaccuracies healthy animals have to be used as controls. ACKNOWLEDGEMENT The study was supported by grants of the Johannes und Frieda Marohn-Stiftung, Erlangen, Germany.
References Aitasalo K: Bone tissue response to irradiation and treatment model of mandibular irradiation injury. An experimental and clinical study. Acta Otolaryngol Suppl 428: 1e54, 1986 Ang E, Black C, Irish J, Brown DH, Gullane P, O’Sullivan B, Neligan PC: Reconstructive options in the treatment of osteoradionecrosis of the craniomaxillofacial skeleton. Br J Plast Surg 56: 92e99, 2003 Annane D, Depondt J, Aubert P, Villart M, Ge´hanno, Gajdos P, Chevret S: Hyperbaric oxygen therapy for radionecrosis of the jaw: a randomized, placebo-controlled, double-blind trial from the ORN96 study group. J Clin Oncol 22: 4893e4900, 2004 Arnold M, Kummermehr J, Trott KR: Radiation induced impairment of osseous healing: quantitative studies using a standard drilling defect in rat femur. Radiother Oncol 48: 259e265, 1998 Cooper JS, Fu K, Marks J, Silverman S: Late effects of radiation therapy in the head and neck region. Int J Radiat Oncol Biol Phys 31: 1141e1164, 1995 Curi MM, Dib LL: Osteoradionecrosis of the jaws: a retrospective study of the background factors and treatment in 104 cases. J Maxillofac Surg 55: 540e544, 1997 Ernst-Stecken A, Jeske I, Hess A, Ro¨del F, Ganslandt O, Grabenbauer G, Sauer R, Brune K, Blu¨mcke I: Hypofractionated stereotactic radiotherapy to the rat hippocampus. Determination of dose response and tolerance. Strahlenther Onkol 183: 440e446, 2007 Ewing J: Radiation osteitis. Acta Radiol 6: 399e412, 1926 Glanzmann C, Gratz KW: Radionecrosis of the mandibula: a retrospective analysis of the incidence and risk factors. Radiother Oncol 36: 94e100, 1995 Goldwaser BR, Chuang SK, Kaban LB, August M: Risk factor assessment for the development of osteoradionecrosis. J Oral Maxillofac Surg 65: 2311e2316, 2007 Grimm G: Animal experimental studies on the radiation injuries in the mandible of full grown rabbits. I. Morphological and histochemical findings. Dtsch Zahn Mund Kieferheilkd Zentralbl 53: 352e362, 1969
Kalz W, Buchali K, Dalluge KH, Lange KP, Magnus S, Seiler S, Topel M: Diagnosis of radiogenic jaw bone damage. The first results with an animal model. Zahn Mund Kieferheilkd Zentralbl 76: 835e837, 1988 Khouri RK, Brown DM, Koudsi B, Deune EG, Gilula LA, Cooley BC, Reddi AH: Repair of calvarial defects with flap tissue: role of bone morphogenetic proteins and competent responding tissues. Plast Reconstr Surg 98: 103e109, 1996 Liu L, Bassano DA, Prasad SC, Hahn SS, Chung CT: The linearquadratic model and fractionated stereotactic radiotherapy. Int J Radiat Oncol Biol Phys 57: 827e832, 2003 Lorente CA, Song BZ, Donoff RB: Healing of bony defects in the irradiated and unirradiated rat mandible. J Oral Maxillofac Surg 50: 1305e1309, 1992 Marx RE: Osteoradionecrosis: a new concept of its pathophysiology. J Oral Maxillofac Surg 41: 283e288, 1983 Niehoff P, Springer IN, Ac¸il Y, Lange A, Marget M, Rolda´n JC, Ko¨ppe K, Warnke PH, Kimmig B, Wiltfang J: HDR brachytherapy irradiation of the jaw e as a new experimental model of radiogenic bone damage. J Craniomaxillofac Surg 36: 203e209, 2008 Park J, Ries J, Gelse K, Kloss FR, van der Mark K, Wiltfang J, Neukam FW, Schneider H: Bone regeneration in critical size defects by cell-mediated BMP-2 gene transfer e a comparison of adenoviral vectors and liposomes. Gene Ther 10: 1089e1098, 2003 Schultze-Mosgau S, Lehner B, Ro¨del F, Wehrhan F, Amann K, Kopp J, Thorwarth M, Nkenke E, Grabenbauer G: Expression of bone morphogenic protein 2/4, transforming growth factor-beta1, and bone matrix protein expression in healing area between vascular tibia grafts and irradiated bone-experimental model of osteonecrosis. Int J Radiat Oncol Biol Phys 61: 1189e1196, 2005 Springer IN, Niehoff P, Ac¸il Y, Marget M, Lange A, Warnke PH, Pielenz H, Rolda´n JC, Wiltfang J: BMP-2 and bFGF in an irradiated bone model. J Craniomaxillofac Surg 36: 210e217, 2008 Teng MS, Futran ND: Osteoradionecrosis of the mandible. Curr Opin Otolaryngol Head Neck Surg 13: 217e221, 2005 Thorn JJ, Hansen HS, Specht L, Bastholt L: Osteoradionecrosis of the jaws: clinical characteristics and relation to the field of irradiation. J Oral Maxillofac Surg 58: 1088e1093, 2000 Trotti A, Colevas AD, Setser A, Rusch V, Jaques D, Budach V, Langer C, Murphy B, Cumberlin R, Coleman CN, Rubin P: CTCAE v3.0: development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol 13: 176e181, 2003 Wurzler KK, De Weese TL, Sebald W, Reddi AH: Radiationinduced impairment of bone healing can be overcome by recombinant human bone morphogenetic protein-2. J Craniofac Surg 9: 131e137, 1998
Dr. Matthias FENNER, MD, DDS University of Erlangen-Nuremberg Department of Oral and Maxillofacial Surgery Glueckstrasse 11 91054 Erlangen Germany Tel.: +49 9131 8533653 Fax: +49 9131 8534219 E-mail: [email protected] Paper received 26 November 2008 Accepted 29 July 2009