Available online at www.sciencedirect.com
Annals of Diagnostic Pathology 13 (2009) 405 – 412
Central granular cell odontogenic tumor: a histopathologic and immunohistochemical study Ana Terezinha Marques Mesquita, DDS, PhDa,⁎, Cássio Roberto Rocha Santos, DDS, PhDa , Ricardo Santiago Gomez, DDS, PhDb , Jacks Jorge, DDS, PhDc , Jorge Esquiche León, DDS, PhDc , Oslei Paes de Almeida, DDS, PhDc a
Department of Dentistry, University of Diamantina, UFVJM, CEP: 39100-000 Diamantina, Minas Gerais, Brazil Laboratory of Molecular Biology, Department of Oral Surgery and Pathology, School of Dentistry, Universidade Federal de Minas Gerais, Pampulha CEP 31.270-901, Belo Horizonte, Minas Gerais, Brazil c Department of Oral Pathology, Dental School, University of Campinas, Piracicaba-UNICAMP, Caixa Postal 52, CEP: 13414-903, Piracicaba-SP, Brazil b
Abstract
Keywords:
The central granular cell odontogenic tumor (CGCOT) is a rare lesion that usually affects the posterior region of the mandible of young adults. We present a case of CGCOT involving the mandible of a 20-year-old white woman, emphasizing the immunohistochemical characteristics using a large panel of antibodies. The lesion was removed surgically, and after 4 years of follow-up, there are no evidences of recurrences. The odontogenic epithelium (OE) showed positivity for cytokeratins (CKs) AE1/AE3, 34βE12, CK5, CK7, CK8, CK14, CK19, E-cadherin, β-catenin, CD138, and p63. The granular cells were positive for vimentin, CD68, lysozyme, muscle-specific actin, α-smooth muscle actin, calponin, neuron-specific enolase (NSE), CD138, and bcl-2. Dendritic-like cells surrounding the OE displayed positivity for vimentin, CD1a, S100, CD68, and bcl-2, but it was negative for factor XIIIa, supporting a Langerhans cell phenotype. Ki-67 labeling index was 1.8%, whereas p53 was negative. These data confirm the benign nature of CGCOT, the association of OE with Langerhans cells, and a variable phenotype of the granular cells. © 2009 Elsevier Inc. All rights reserved. Granular cell odontogenic tumor; Mandible; Langerhans cells; Immunohistochemistry
1. Introduction The central granular cell odontogenic tumor (CGCOT) is a rare benign lesion that was firstly described by Werthemann [1] in 1950, who used the term spongiocytic adamantinoma. To date, about 32 cases of CGCOT have been reported using varied terminology [2,3]. There is a great controversy regarding its nature and characterization; however, more recently, it has been considered that those tumors composed mainly of granular cells (GCs) and odontogenic epithelium (OE) should be named GCOT [2]. In the recent classification of odontogenic tumors of the World Health Organization [4], this entity was not considered. Granular cells may eventually be found in odontogenic tumors, such as ameloblastoma, ameloblastic fibroma, and odontogenic fibroma, or constitute ⁎ Corresponding author. Tel.: +55 38 35311811; fax: +55 38 35313283. E-mail address:
[email protected] (A.T.M. Mesquita). 1092-9134/$ – see front matter © 2009 Elsevier Inc. All rights reserved. doi:10.1016/j.anndiagpath.2009.04.001
Fig. 1. Clinical features of CGCOT showing a swelling in the left mandibular body.
406
A.T.M. Mesquita et al. / Annals of Diagnostic Pathology 13 (2009) 405–412
Fig. 2. Panorex (A) and occlusal (B) radiographs of CGCOT showing a lytic lesion with radiopaque focus involving the left mandibular body.
the bulk of the tumor as typically found in GC tumor and congenital epulis of the newborn [5-9]. There are 3 reported cases of plexiform GC odontogenic tumor, but probably, these lesions represent a different entity [10,11]. The exact histogenesis of the GCs is controversial, and whether these cells are neoplastic, reactive, degenerative, or metabolic in nature remains unclear [3]. In the present report, we emphasize the immunohistochemical (IHC) features of a case of CGCOT, which was previously reported by Gomes et al [12]. To our knowledge, this is the first case of CGCOT evaluated by a large panel of immunomarkers. 2. Case report A 20-year-old white woman was referred to the Oral Diagnosis Clinic of the Dental School, University of Diamantina, Minas Gerais, Brazil, complaining of a swelling on the left side of the face for about 3 months. Medical history was not contributory. Intraoral examination showed a firm, asymptomatic swelling on the left side of the mandible,
extending from the second premolar to the second molar region (Fig. 1). The associated teeth in the lesional area were vital and nonmobile. The panoramic radiograph showed a well-demarcated radiolucent lesion with radiopaque areas, and the occlusal radiography showed vestibular and lingual cortical expansion (Fig. 2). Needle aspiration was negative for cystic fluid. With the clinical diagnosis of odontogenic tumor, an incisional biopsy was performed under local anesthesia. A diagnosis of CGCOT was made. Surgical treatment included the complete resection of the tumor, and the histopathologic analysis of the resected specimen showed similar features of the incisional biopsy. The postoperative period was uneventful, and there were no signs of recurrence after 4 years of follow-up.
3. Pathologic findings 3.1. Microscopic findings Microscopically predominated large round cells with abundant cytoplasm containing eosinophilic granules and
Fig. 3. (A) Microscopic aspect of CGCOT showing odontogenic epithelial cords, GCs, and cementum-like calcified material (hematoxylin and eosin, original magnification ×10). (B) The former figure in high power (hematoxylin and eosin, original magnification ×40).
A.T.M. Mesquita et al. / Annals of Diagnostic Pathology 13 (2009) 405–412
eccentric nuclei, which showed in some areas mild nuclear atypia, were found. Islands and cords of inactive-looking OE formed by 2 or more rows of epithelial cells, some with clear cytoplasm permeated the GCs, were also found. The stroma was formed by scarce thin fibrous connective tissue, showing occasional small round cementum-like calcified structures. A diagnosis of GCOT was made (Fig. 3). The GCs were periodic acid-Schiff stain positive and diastase resistant, whereas the OE was negative after diastase treatment. 3.2. Immunohistochemical findings
407
negative (Fig. 4C and D). The GCs showed positivity for vimentin, CD68, neuron-specific enolase (NSE), musclespecific actin, α-smooth muscle actin, lysozyme, bcl-2, and CD138 (membranous and cytoplasmic pattern), and in scarce cells for calponin (Fig. 5). Ki-67 labeling index was 1.8%, calculated after analyzing about 1000 cells in 5 high-power fields in the region of the tumor with the greatest density of staining. The tumor was negative for all other immunomarkers used (Table 2).
4. Discussion
A large panel of antibodies was used for IHC evaluation (Table 1). The OE exhibited positivity for cytokeratins (CKs) AE1/AE3, 34βE12, CK5, CK7, CK8, CK14, CK19, Ecadherin, β-catenin, CD138, weakly for p63 and focally for CD56 (Fig. 4A and B). Nuclear and cytoplasmic positivity for S100 as well as cytoplasmic positivity for vimentin, CD1a, CD68, and bcl-2 was observed in scarce dendritic-like cells that surrounded OE cords, whereas factor XIIIa was Table 1 Antibodies used for IHC analysis of CGCOT affecting the mandible Antibody
Source/clone
Dilution
bcl-2 β-Catenin Calponin CD1a CD56 CD57 CD68 CD138 CEA CK—cocktail CK—high molecular weight CK5 CK7 CK8 CK13 CK14 CK18 CK19 Desmin E-cadherin EMA Factor XIIIa GFAP h-Caldesmon Ki-67 Lysozyme MSA NSE p53 S100 SMA Vimentin
Dakoa, 124 Novocastrab, 17C2 Dakoa, CALP Dakoa, 010 Novocastrab, 1B6 Dakoa, NK1 Dakoa, PG-M1 Dakoa, My15 Dakoa, II-7 Dakoa, AE1/AE3 Dakoa, 34βE12 Novocastrab, XM26 Dakoa, OV-TL12/30 Dakoa, 35βH11 Novocastrab, KS-1A3 Novocastrab, NCL-L-LL002 Dakoa, DC10 Dakoa, RCK 108 Dakoa, D33 Dakoa, NCH-384 Dakoa, E29 Biocarec, AC-1A1 Dakoa, 6F2 Dakoa, N-CD Dakoa, MIB-1 Dakoa, Polyclonal Dakoa, HHF-35 Dakoa, BBS/NC/VI-H14 Dakoa, DO-7 Dakoa, polyclonal Dakoa, 1A4 Dakoa, Vim 3B4
1:50 1:100 1:600 1:400 1:50 1:1000 1:400 1:100 1:500 1:500 1:200 1:400 1:400 1:200 1:400 1:200 1:400 1:200 1:100 1:200 1:400 1:300 1:400 1:400 1:200 1:200 1:800 1:800 1:200 1:12 000 1:400 1:400
GFAP indicates glial fibrillary acidic protein; MSA, muscle-specific actin; SMA, α-smooth muscle actin. a Dako A/S, Glostrup, Denmark. b Novocastra Laboratories Ltda, Newcastle, England. c Biocare Medical, Walnut Creek, Calif.
Central GC odontogenic tumor was initially described with various terminologies, creating confusion with its classification. Couch et al [13] suggested the term granular cell ameloblastic fibroma because it resembled ameloblastic fibroma. Nevertheless, contrary to ameloblastic fibroma, the stromal cells of these lesions are granular and capable to produce cementum [14]. Furthermore, CGCOT differs from ameloblastic fibroma for the older age at presentation and lack typical ameloblastic differentiation and stellate reticulum-like, besides the absence of ectomesenchymal stroma [2,15]. The term central granular cell odontogenic fibroma was also used [6,16], considering it was a histologic variant of the odontogenic fibroma [5,17]. This is supported by the fact that the central odontogenic fibroma and CGCOT show similar clinicopathologic features such as age of presentation, location, and female predilection. Nevertheless, odontogenic fibroma presents a cell-rich fibroblastic stroma, which is lacking in GCOT [7]. It is interesting to mention that there are 3 cases reported as plexiform GC odontogenic tumor. In these cases, the epithelial strands showing a granular eosinophilic cytoplasm were loosely arranged against a fibrous/myxomatous stroma. Thus, the microscopic characteristics are quite different from GCOT and probably represent a distinct entity [10,11]. In 22 cases with radiographic features registered, most were radiolucent, showing a unicystic pattern, frequently with a sclerotic border. Nine cases were locally aggressive causing cortical expansion and teeth displacement. In the current case, the radiographic findings were similar, besides some radiopaque areas. Microscopically, calcifications have been found in 50% (16/32) of the reported cases of CGCOT [2,3]. Possibly, the amount of mineralized tissue is scarce and may not be detected by radiographic analysis. These calcifications have been considered to be dystrophic calcification, and it is possible that the intracellular degeneration of organelles in the GCs could start the formation of these mineralization foci [18]. Peripheral GCOT are exceptionally rare. We have found 4 reported cases, affecting the gingival tissue. Differential diagnosis include pyogenic granuloma, peripheral giant cell lesion, peripheral ossifying fibroma, and peripheral variants of others odontogenic tumors [17].
408
A.T.M. Mesquita et al. / Annals of Diagnostic Pathology 13 (2009) 405–412
Fig. 4. Immunohistochemical findings of CGCOT affecting the mandible. (A) Strong positivity for CK5 in the epithelial cords (original magnification ×20). (B) Immunoreactivity for CK19 in the OE (original magnification ×40). (C) Dendritic-like cells within epithelium positive for CD1a (original magnification ×40). (D) Nuclear and cytoplasmic positivity for S100 in scarce dendritic-like cells within the epithelial cords (original magnification ×40).
Immunohistochemistry of CGCOT may be useful to elucidate the CK profile of the OE as well as the origin and nature of the GCs. It is suggested that CK19 is specific for OE; however, it is not a uniform finding [19]. The current case showed positivity for high molecular weight CK besides CK7, CK8, and CK19. Nevertheless, in the Machado de Sousa et al [14] study, CK7, CK8, and CK19 were negative in 2 cases of CGCOT tested, which probably represent variable expression for these CKs. Epithelial membrane antigen (EMA) was negative in the current case, as well as in the Mirchandani and Sciuba [20] study. Carcinoembryonic antigen (CEA) is a large surface cell glycoprotein that can be detected in adenocarcinoma of the colon, lung and breast, and squamous carcinoma of the bronchus and cervix. Carcinoembryonic antigen immunoreactivity was also found in cases of odontogenic keratocyst and focally in squamous odontogenic tumors [21]. There is only one case of GCOT showing positivity for CEA in the GCs [3]. It can be due to products of secretion of GCs, but a type of crossreaction cannot be excluded. The current case was negative for CEA.
β-Catenin binds to membrane-associated E-cadherin that maintains intercellular adhesiveness. In ameloblastomas, Ecadherin and α-catenin are expressed strongly in central polyhedral cells and slightly in peripheral columnar cells. Expression of these molecules in malignant ameloblastomas is greatly reduced. In calcifying epithelial odontogenic tumor and clear cell odontogenic carcinoma, the E-cadherin and αcatenin expression seems to be preserved [22]. In the current case, expression of E-cadherin was detected in the OE only, suggesting preservation of cell-cell interaction and adhesive function. β-Catenin is the major effector of the canonical Wnt signaling pathway. Mutations in components of Wnt signaling pathway stabilize β-catenin, which dislocates to the nucleus and causes the constitutive activation of TCF/ LEF1 gene transcription and expression of c-myc, tcf1, and cyclin D1 genes [23]. The nuclear and/or cytoplasmic expression of β-catenin found in cystic calcifying odontogenic tumor, benign and malignant ameloblastomas, and primary intraosseous odontogenic carcinoma indicates that Wnt pathway is involved in oncogenesis and cytodifferentiation of OE via deregulation of cell proliferation mediated by
A.T.M. Mesquita et al. / Annals of Diagnostic Pathology 13 (2009) 405–412
409
Fig. 5. Immunohistochemical findings of CGCOT affecting the mandible. (A) Granular cells marked for NSE (original magnification ×20). (B) Positive expression of CD138 in both GCs and OE (original magnification ×20). (C) Note positivity for α-smooth muscle actin in the GCs (original magnification ×40). (D) The GCs also were positive for specific muscle actin (original magnification ×40).
β-catenin [23,24]. In the present case, β-catenin was positive only in the OE in a membranous pattern, which probably indicates that this molecule is not implicated in the tumorigenesis of CGCOT. CD138 (syndecan-1) is a cell surface proteoglycan that binds cells to the extracellular matrix. It is present in epithelial, endothelial, connective, and muscular normal tissues, and also in plasma cells. In the epithelium, syndecan modulates epithelial-stromal interactions, cell-cell adhesion, and cell proliferation. Bologna-Molina et al [25] showed a reduced expression of CD138 in solid ameloblastoma when compared with unicystic ameloblastoma, and probably, this finding may be related to the more aggressive biologic behavior of solid ameloblastoma. In the current case, both OE and GCs showed positivity for CD138, the latter in a cytoplasmic pattern as well. These findings indicate cell-cell adhesion and OE-GCs reciprocal interactions and that, probably, CD138 expression may be involved in the development of CGCOT. Most studies show that CGCOT are consistently negative for S100 and NSE [2,3,6,14,18,20,26]. Only Chen [18],
evaluating 4 cases of CGCOT, reported slight positivity for S100 on the GCs only. Moreover, Mirchandani and Sciuba [20] showed positivity for Leu-7 in the GCs in one case of CGCOT. In the present case, S100 and CD57 were negative, whereas NSE was strongly positive on the GCs only. The isolated expression of NSE is not sufficient to confirm a neuroectodermic origin of the GCs as previously considered. CD56 is a transmembrane protein belonging to an immunoglobulin superfamily, and it can be found in several hematologic and nonhematologic neoplasms. The CD56 expression during molar tooth development in mice suggest an important morphoregulatory role of this molecule during odontogenesis [27]. The positivity for CD56 in cells at the periphery of ameloblastoma nests suggests an adhesive function [28]. The current case showed focal expression of CD56 on OE only. Probably, because of the ectodermic origin of the oral epithelium and interactions with cells derived from neural crest, CD56 expression should be better evaluated in odontogenic tumors, and its participation in the tumorigenesis should be clarified.
410
A.T.M. Mesquita et al. / Annals of Diagnostic Pathology 13 (2009) 405–412
Table 2 Immunohistochemical findings in OE and GCs of CGCOTs Immunomarkers
Cellular type OE
− − −
+ + NA
− − −
+ + −
CK8
− −
+ −
CK13
− −
− +/−
CK14
− −
+ +
CK18
− −
− −
CK19
CK-HMW 34βE12 HMW MNF116 EMA
E-cadherin β-Catenin CD138 CEA Vimentin
S100
NSE
CD57 (Leu-7)
− −
Immunomarkers
Reference
GC Pan-CK AE1/AE3 CK (cocktail of CK) Prekeratin (cocktail of CK) CK5 CK7
Table 2 (continued)
+ −
− − − − −
+ + + − NA
− − + − + + + + +
+ + + − − −a − − −
+ +
− NA
− − − −
−a − − −a
− +/− −
− − NA
− + − −
− − − NA
− +
− NA
Current case Yih et al [26] (1) Meer et al [3] (1) Current case Mirchandani and Sciuba [20] (1) Current case Current case Current case Current case Meer et al [3] (1) Current case Meer et al [3] (1) Brannon et al [2] (2) Machado de Sousa et al [14] (2) Yih et al [26] (1) Mirchandani and Sciuba [20] (1) Current case Meer et al [3] (1) Brannon et al [2] (3) Machado de Sousa et al [14] (2) Yih et al [26] (1) Chen [18] (4) Mirchandani and Sciuba [20] (1) Shiro et al [6] (1) Current case Yih et al [26] (1) Mirchandani and Sciuba [20] (1) Current case Mirchandani and Sciuba [20] (1)
Reference
GC
OE
− − − + + + +
scarce cells + − − −a − − −
AAT
+ + +
− − NA
AACT
+ + + + + + − −
− − − − − − − −
− − − + + + − − + + − − −
−a − − − − − − − − − − − −
−
NA
scarce cells + − − − + + − 1.8% −
− − − − −a + − − −
CD56 GFAP Current case Brannon et al [2] (2) Mirchandani and Sciuba [20] (1) Current case Current case Machado de Sousa et al [14] (2) Current case Machado de Sousa et al [14] (2) Current case Machado de Sousa et al [14] (2) Current case Machado de Sousa et al [14] (2) Current case Machado de Sousa et al [14] (2) Current case Machado de Sousa et al [14] (2)
Cellular type
CD68
Lysozyme
Factor XIIIa
CD1a
CD1 (OKT6) HLA-DR SMA Antiactin MSA
Calponin h-Caldesmon Desmin bcl-2 p53 Ki-67
Current case Current case Chen [18] (4) Current case Meer et al [3] (1) Brannon et al [2] (3) Machado de Sousa et al [14] (2) Brannon et al [2] (2) Yih et al [26] (1) Mirchandani and Sciuba [20] (1) Brannon et al [2] (2) Yih et al [26] (1) Current case Meer et al [3] (1) Brannon et al [2] (2) Yih et al [26] (1) Current case Machado de Sousa et al [14] (2) Current case Meer et al [3] (1) Brannon et al [2] (1) Chen [18] (4) Chen [18] (4) Current case Brannon et al [2] (2) Yih et al [26] (1) Chen [18] (4) Current case Meer et al [3] (1) Brannon et al [2] (2) Machado de Sousa et al [14] (2) Mirchandani and Sciuba [20] (1) Current case Current case Current case Yih et al [26] (1) Current case Meer et al [3] (1) Current case Current case Meer et al [3] (1)
Results of 36 antibodies and 16 reports, including the current case, are presented. The number of cases evaluated in each report is indicated in parentheses. AAT indicates α1-antitrypsin; AACT, α1-antichymotrypsin; GFAP, glial fibrillary acidic protein; MSA, muscle-specific actin; SMA, αsmooth muscle actin; HMW, high molecular weight; NA, not available; +/−, weak. a Dendritic-like cells within the epithelium were positive for vimentin, S100, CD1a, CD68, and bcl-2.
The gene p63 plays an essential role in epithelial development and the proliferation of limb and craniofacial structures. The expression of p63 is common in normal and neoplastic epithelial and myoepithelial cells. In odontogenic tumors, it was demonstrated that expression of p63 correlates
A.T.M. Mesquita et al. / Annals of Diagnostic Pathology 13 (2009) 405–412
with pathologic features and biologic behavior of these lesions. Malignant and benign locally aggressive odontogenic tumors show a more intense and diffuse expression of p63 when compared with other benign odontogenic tumors [29]. According to these findings, the expression of p63 in the present case was weak and restricted to OE. There are strong evidences that GCs of CGCOT are mesenchymal in origin, because they show uniform positivity for vimentin and negativity for CK in all cases tested [2,3,14,20,26]. Granular cells are also positive for CD68, possibly associated with the presence of autophagic and phagocytic vacuoles in these cells [2,3,8,14]. The expression of CD68 by GCs does not necessarily imply a histiocytic origin because other nonmacrophagic tumor cells also express CD68, such as in neurofibroma, schwannoma, carcinoma, angiosarcoma, hemangioma, and melanoma [30]. Other antibodies considered in the past makers of macrophagic lineage can be expressed by GCs, as HLA-DR, lysozyme, α 1 -antitrypsin, and α 1 -antichymotrypsin [2,3,14,20,26]. It is now known that these markers are unspecific, and few are used today as markers of macrophagic differentiation. Recently, a new marker that supports macrophagic differentiation is CD163, and it would be interesting to test it in GCOT [30]. Granular cell odontogenic tumor consistently show negativity for myofibroblastic and smooth muscle cells immunomarkers such as musclespecific actin, α-smooth muscle actin, and desmin [2,3,14,26]. In the current case, contrarily to these findings, muscle-specific actin and α-smooth muscle actin were positive, whereas calponin showed scarce cells positive, but h-caldesmon and desmin were negative. Only Chen [18] showed positivity for actin in the GCs in 4 cases of CGCOT. The presence of actin in these cells possibly reflects motility properties needed to accomplish phagocytosis, and it is not necessarily an indication of myogenous or myofibroblastic origin. Further studies with larger numbers of cases are needed to corroborate this hypothesis. Bcl-2 is an antiapoptotic protein that has been demonstrated in the epithelium of odontogenic lesions, supporting the view that it might have an important role maintaining these cells viable [31]. Meer et al [3] showed positivity for bcl-2 in both GCs and OE evaluating one case of CGCOT. In the present case, only the GCs were bcl-2 positive, indicating variability of expression of this protein. Because both oral epithelium and OEs originate from the same oral ectoderm, it is possible that Langerhans cells will also migrate into tumor odontogenic epithelial nests. Interestingly, in the current case, some dendritic-like cells within the OE was found. These cells were positive for vimentin, S100, CD1a, CD68, and bcl-2, whereas factor XIIIa was negative. Chen [18] showed positivity for CD1 (clone OKT6) in both GCs and dendritic cells, these latter considered as Langerhans cells. Interestingly, this is the only study that demonstrated positivity for CD1 affecting the GCs. In contrast, the present and other 2 studies [2,3] showed negative expression for CD1a in the GCs. Wang et al [32],
411
evaluating one case of calcifying epithelial odontogenic tumor, found positivity for both CD1a and S100 in scattered dendritic cells, likely, Langerhans cells, within the epithelial cell sheets. The true pathologic significance of the presence of Langerhans cells in CGCOT and other odontogenic tumors needs further investigations. Recent studies have demonstrated that Langerhans cells are CD207 positive and factor XIIIa negative, whereas reactivity only for S100 and CD1a can be found in both indeterminated dendritic cells and Langerhans cells [33]. The findings of the present case support that the dendritic cells in the OE are of Langerhans cell lineage. The current case showed p53 negativity and Ki-67 index of 1.8%, confirming the benign nature of the lesion, which is different from the malignant variant of GCOT reported by Piatelli et al [34] where Ki-67 index was 21%. In summary, CGCOT is a rare benign odontogenic neoplasm reported with various terms and not included in the last World Health Organization odontogenic tumor classification. It shows a predilection for women, usually presenting as an asymptomatic swelling of the mandible. The IHC findings of the current case reinforced the odontogenic nature of the epithelium islands, whereas the GCs showed mesenchymal characteristics with a variable phenotype. Considering this case and those published in the literature, we suggest that CGCOT should be considered as a distinctive odontogenic tumor entity. Acknowledgments This work was supported by the State of São Paulo Research Foundation (FAPESP). References [1] Werthemann A. Uber spongiocytares adamantinoma. Oncologie 1950; 3:193-207. [2] Brannon RB, Goode RK, Eversole LR, et al. The central granular cell odontogenic tumor: report of 5 new cases. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2002;94:614-21. [3] Meer S, Altini M, Coleman H, et al. Central granular cell odontogenic tumor: immunohistochemistry and ultrastructure. Am J Otolaryngol 2004;25:73-8. [4] Barnes L, Eveson JW, Reichart P, et al, editors. Histological typing of odontogenic tumours. Lyon: IARC Press; 2005. p. 308. [5] Ruhl GH, Akuamoa-Boateng E. Granular cells in odontogenic and non-odontogenic tumors. Virch Arch A Pathol Anat 1989;415:403-9. [6] Shiro BC, Jacoway JR, Mirmiral SA, et al. Central odontogenic fibroma, granular cell variant. A case report with S-100 immunohistochemistry and a review of the literature. Oral Surg Oral Med Oral Pathol 1989;67:725-30. [7] Gardner DG. Central odontogenic fibroma current concepts. J Oral Pathol Med 1996;25:556-61. [8] Calvo N, Alonso D, Prieto M, et al. Central odontogenic fibroma granular cell variant: a case report and review of the literature. J Oral Maxillofac Surg 2002;60:1192-4. [9] Sathi GS, Han PP, Tamamura R, et al. Immunolocalization of cell signaling molecules in the granular cell ameloblastoma. J Oral Pathol Med 2007;36:609-14.
412
A.T.M. Mesquita et al. / Annals of Diagnostic Pathology 13 (2009) 405–412
[10] Altini M, Hille JJ, Buchner A. Plexiform granular cell odontogenic tumor. Oral Surg Oral Med Oral Pathol 1986;61:163-7. [11] Siar CH, Ng KH, Jalil NA. Plexiform granular cell odontogenic tumor: unicystic variant. Oral Surg Oral Med Oral Pathol 1991;72: 82-5. [12] Gomes CC, Naves MD, Pereira MV, et al. Granular cell odontogenic tumour: case report and review of literature. Oral Oncol EXTRA 2006; 42:277-80. [13] Couch RD, Morris EE, Vellios F. Granular cell ameloblastic fibroma: report of 2 cases in adults, with observations on its similarity to congenital epulis. Am J Clin Pathol 1962;37:398-404. [14] Machado de Sousa SO, Soares de Araújo N, Melhado RM, et al. Central odontogenic granular cell tumor: immunohistochemical study of two cases. J Oral Maxillofac Surg 1998;56:787-91. [15] Takeda Y. Granular cell ameloblastic fibroma, ultrastructure and histogenesis. Int J Oral Maxillofac Surg 1986;15:190-5. [16] Vincent SD, Hammond HL, Ellis GL, et al. Central granular cell odontogenic fibroma. Oral Surg Oral Med Oral Pathol 1987;63: 715-21. [17] Rinaggio J, Cleveland D, Koshy R, et al. Peripheral granular cell odontogenic fibroma. Oral Surg Oral Med Oral Pathol Oral Radiol Endod 2007;104:676-9. [18] Chen SY. Central granular cell tumor of the jaw: an electron microscopic and immunohistochemical study. Oral Surg Oral Med Oral Pathol 1991;72:75-81. [19] Crivelini MM, de Araújo VC, de Sousa SOM, et al. Cytokeratins in epithelia of odontogenic neoplasms. Oral Diseases 2003;9:1-6. [20] Mirchandani R, Sciuba JJ. Granular cell lesions of the jaws and oral cavity: a clinicopathologic, immunohistochemical and ultrastructural study. J Oral Maxillofac Surg 1989;47:1248-55. [21] Howell RE, Handlers JP, Aberle AM, et al. CEA immunoreactivity in odontogenic tumors and keratocysts. Oral Surg Oral Med Oral Pathol 1988;66:576-80. [22] Kumamoto H, Ooya K. Expression of E-cadherin and α-catenin in epithelial odontogenic tumors: an immunohistochemical study. J Oral Pathol Med 1999;28:152-7.
[23] van Es JH, Barker N, Clevers H. You Wnt some, you lose some: oncogenes in the Wnt signaling pathway. Curr Opin Genet Dev 2003; 13:28-33. [24] Miyake T, Tanaka Y, Kato K, et al. Gene mutation analysis and immunohistochemical study of beta-catenin in odontogenic tumors. Pathol Int 2006;56:732-7. [25] Bologna-Molina R, Mosqueda-Taylor A, Lopez-Corella E, et al. Syndecan-1 (CD138) and Ki-67 expression in different subtypes of ameloblastomas. Oral Oncol 2008;44:805-11. [26] Yih WY, Thompson C, Meshul CK, et al. Central odontogenic granular cell tumor of the jaw: report of case and immunohistochemical and electron microscopic study. J Oral Maxillofac Surg 1995;53:453-9. [27] Obara N, Takeda M. Expression of neural cell adhesion molecule (NCAM) during second and third-molar development in the mouse. Anat Embryol 1993;188:13-20. [28] Nuray ER, Atilla D, Fügen T, et al. Neural cell adhesion molecule and neurothelin expression in human ameloblastoma. J Oral Maxillofac Surg 2001;59:900-3. [29] Lo Muzio L, Santarelli A, Caltabiano R, et al. p63 expression correlates with pathological features and biological behaviour of odontogenic tumours. Histopathology 2006;49:211-4. [30] Nguyen TT, Schwartz EJ, West RB, et al. Expression of CD163 (hemoglobin scavenger receptor) in normal tissues, lymphomas, carcinomas, and sarcomas is largely restricted to the monocyte/ macrophage lineage. Am J Surg Pathol 2005;29:617-24. [31] Fregnani ER, Pires FR, Quezada RD, et al. Calcifying odontogenic cyst: clinicopathology features and immunohistochemical profile of 10 cases. J Oral Pathol Med 2003;32:163-70. [32] Wang YP, Lee JJ, Wang JT, et al. Non-calcifying variant of calcifying epithelial odontogenic tumor with Langerhans cells: case report. J Oral Pathol Med 2007;36:436-9. [33] Rezk SA, Spagnolo DV, Brynes RK, et al. Indeterminate cell tumor: a rare dendritic neoplasm. Am J Surg Pathol 2008;32:1868-76. [34] Piatelli A, Rubini C, Goteri G, et al. Central granular cell odontogenic tumour: report of the first malignant case and review of the literature. Oral Oncol 2003;39:78-82.