Kentaro Kikuchi

Enhanced expression of podoplanin in ameloblastomas

OBJECTIVE Podoplanin, a mucin-type transmembrane glycoprotein, is specifically expressed by lymphatic but not blood vascular endothelial cells, and is also widely expressed in various specialized cell types throughout the body. Recent studies have demonstrated that it mediates a pathway leading to collective cell migration and invasion in vivo and in vitro. In the present study, we carried out an immunohistochemical investigation of podoplanin to clarify whether it is expressed in human ameloblastomas (AMs), which are characterized by locally aggressive behavior with a high rate of recurrence. In addition, we examined the localization of the epithelial marker E-cadherin and the mesenchymal marker vimentin to clarify whether AMs show epithelial-mesenchymal transition (EMT). METHODS Paraffin-embedded tissue specimens of 38 AMs were examined immunohistochemically using antibodies against podoplanin, E-cadherin, and vimentin. RESULTS Immunohistochemical reactivity for podoplanin was detected in the cell membrane and cytoplasm of most odontogenic tumor epithelial cells in AMs. Podoplanin was expressed strongly in peripheral columnar cells and slightly in central stellate reticulum-like cells. E-cadherin was expressed in central stellate reticulum-like cells and showed decreased expression in peripheral columnar cells. Immunoreactivity for E-cadherin was weak or negative in keratinizing cells of acanthomatous AMs, suggesting terminal differentiation of the tumor cells. Immunohistochemical reactivity for vimentin was found in stromal cells, but partial or no reaction was observed in neoplastic cells. CONCLUSION Expression of podoplanin in AMs is considered to be associated with neoplastic odontogenic tissues; this molecule might play a role in the collective cell migration of tumor nests in AMs. The pattern of expression of E-cadherin and vimentin suggests that invasion in AMs occurs in the absence of EMT. The migration and invasion mediated by podoplanin in AMs could be related to cytoskeletal reorganization.

Methotrexate-Related Epstein-Barr Virus (EBV)-Associated Lymphoproliferative Disorder—So-Called “Hodgkin-Like Lesion”—of the Oral Cavity in a Patient with Rheumatoid Arthritis

Patients affected by autoimmune diseases (rheumatoid arthritis, psoriasis, dermatomyositis) who are treated with methotrexate (MTX) sometimes develop lymphoproliferative disorders (LPDs). In approximately 40% of reported cases, the affected sites have been extranodal, and have included the gastrointestinal tract, skin, lung, kidney, and soft tissues. However, MTX-associated LPD (MTX-LPD) is extremely rare in the oral cavity. Here we report a 69-year-old Japanese woman with rheumatoid arthritis (RA) who developed MTX-LPD resembling Hodgkin’s disease—so-called “Hodgkin-like lesion”—in the left upper jaw. Histopathologically, large atypical lymphoid cells including Hodgkin or Reed-Sternberg-like cells were found to have infiltrated into granulation tissue in the ulcerative oral mucosa. Immunohistochemistry showed that the large atypical cells were positive for CD20, CD30 and Epstein-Barr virus (EBV)-latent infection membrane protein-1 (LMP-1) and negative for CD15. EBV was detected by in situ hybridization (ISH) with EBV-encoded small RNA (EBER), and polymerase chain reaction (PCR) for LMP-1 and EBNA-2 in material taken from the formalin-fixed, paraffin-embedded specimen. To our knowledge, this is the first reported case of MTX-related EBV-associated LPD (MTX-EBVLPD), “Hodgkin-like lesion”, of the oral cavity in a patient with RA.

Significance of podoplanin expression in keratocystic odontogenic tumor.

BACKGROUND The most important clinical features of the keratocystic odontogenic tumor (KCOT) are its potential for locally destructive behavior, a tendency to recur, and its origin in the odontogenic epithelium. The clinical features of KCOT are similar to those of ameloblastoma (AM). Histologically, KCOT is distinguished from jaw cyst with keratinization (orthokeratinized odontogenic cyst; OOC). However, current scientifically based clinical parameters cannot predict any potential for neoplastic behavior, or aggressive and localized invasiveness, in patients with KCOT. We have shown that podoplanin, a lymphatic endothelial marker, is highly expressed in AM. The purpose of this study was to determine the usefulness of podoplanin for reclassification of the odontogenic keratocyst (OKC) from cyst to tumor status. METHODS Paraffin-embedded tissue specimens of 57 OKCs (46 KCOTs and 11 OOCs) and 15 dentigerous cysts (DCs) were immunohistochemically examined using antibody against podoplanin. RESULTS Immunohistochemical reactivity for podoplanin was detected in the cell membrane and cytoplasm of most of the basal and suprabasal layer, areas of budding basal cell proliferation, epithelial nests and peripheral cells of daughter cysts in the stromal connective tissue in KCOTs. In the case of OOC and DC, only cases associated with inflammation were positive for podoplanin. CONCLUSION Podoplanin is strongly expressed in KCOTs in comparison with OOCs. The pattern of staining for podoplanin in KCOT could be related to its neoplastic nature, and suggests a role of the protein in tumor invasiveness.

Development and Growth of Adenomatoid Odontogenic Tumor Related to Formation and Eruption of Teeth

The number of published cases of adenomatoid odontogenic tumor (AOT) has steadily increased, and about half were reported in Asian populations. Although the clinicopathologic profile of AOT has been extensively reported, factual knowledge of its pathogenesis is rudimentary at best, and there is controversy as to precisely which odontogenic epithelium it arises from. AOT is a successional tooth-associated lesion which develops during the mixed dentition. The permanent successor differs from its deciduous predecessor in that it has an eruptive pathway from the dental follicle to the gingiva, the gubernaculum dentis. With this background in mind, the present review focuses mainly on the early development of AOT. We successfully demonstrated a close spatial relationship between AOT and the gubernaculum dentis in a typical case. From further observations of the same AOT in which an enclosed permanent canine showed enamel hypoplasia, it is possible to pinpoint areas around the crown of a developing successional tooth in continuity with the gubernaculum dentis as starting points. In addition, we discuss the unifying histogenetic concept of three clinical variants, namely, pericoronal (follicular), extracoronal (extrafollicular), (see Article note) and peripheral.

Podoplanin expression during dysplasia–carcinoma sequence in the oral cavity

Human podoplanin, a type-1 transmembrane sialomucin-like glycoprotein, is involved in cell migration, tumor cell invasion, and metastasis. However, the role of the protein in squamous cell carcinoma (SCC) has been unclear and immunohistochemical reactivity for podoplanin differs from organ-to-organ. In the present study, immunohistochemical and molecular biological analyses were performed to examine the importance of podoplanin expression in oral precancerous and cancerous lesions and metastases. We immunohistochemically investigated the expression of podoplanin in 103 precancerous lesions, 69 primary oral squamous cell carcinomas (OSCCs), and 32 metastases, and that of E-cadherin and vimentin in primary OSCCs with metastasis. Furthermore, human OSCC-derived cell lines preincubated with fibrous growth factor-basic, epidermal growth factor (EGF), and tumor growth factor-β1 were subjected to real-time reverse transcription polymerase chain reaction. Immunoreactivity for podoplanin was detected in 89 (86.4%) of the precancerous lesions and the intensity was correlated with the degree of epithelial dysplasia (P = 0.016). Enhanced podoplanin expression was observed in 66 (95.7%) of the OSCCs and was significantly associated with a poor pathologic grade of differentiation (P = 0.020). Epithelial–mesenchymal transition was observed in 18 (58.1%) of the primary OSCCs with metastasis to regional lymph nodes. Messenger RNA for podoplanin was markedly increased after treatment with EGF in three OSCC cell lines. The present findings suggest that podoplanin is associated with tumor development via the oral dysplasia–carcinoma sequence and could be involved in a signaling pathway governing tumor growth and invasion in OSCC.

Primordial odontogenic tumour: is it truly novel?

Sir: We would like to express our views regarding the recently published paper by Mosqueda-Taylor et al. on primordial odontogenic tumour (POT), which they considered to be a previously undescribed, benign ectomesenchymal neoplasm. We agree with the authors that all six of the lesions presented have many features of interest and importance with regard to classification and pathogenesis, and appear not to correspond to any of the accepted diagnostic categories. However, we question their claims of first description of this tumour type. Differences in nomenclature exist, and the previously published cases of ameloblastoma, ameloblastic fibroma, odontogenic fibroma and odontogenic myxoma are assumed to represent the same pathological process, nine of these accounts actually corresponding to previous descriptions of POT both clinically and pathologically. Although, at the time, none of these lesions were considered to be a distinct entity, odontogenic myxofibromas coated by a single layer of enamel epithelium have long been well known (References S1). Bernier, Tiecke and Gorlin, oral pathologists of great renown, illustrated a tumour compatible with POT in their seminal publications written approximately half a century ago. We understand that Mosqueda-Taylor et al. did not take into account the above reports because some of them were not indexed in PubMed, and their pathological repertoire did not include a superficial smattering of Japanese. Considering that six cases have been reported from Latin America and at least five from Japan, more examples of this type of tumour might be found in the local domestic scientific literature worldwide (References S1 and S2). Thus POT, previously unrecognized by Mosqueda-Taylor et al., may not be as rare as the authors suggest. At present, it is reasonable to consider that POT may arise during the course of embryonic, fetal or early postnatal development, not because of its marked structural resemblance to primitive dental tissues (papilla and follicle) but because of its usual association with anomalies of dental development, including missing, malformed or unerupted teeth and its limited period of onset before adult life. It may well be that the surface enamel epithelium of POT often invaginates into the mesenchymal component in a manner and pattern reminiscent of normal odontogenesis, leading to its characteristic multilobulated configuration. Such a continuous double epithelial layer is occasionally expanded at the free end to form a microcystic space. Also of note is the presence of several fragments of fibrous tumour lacking a surface epithelial lining. Other examples of this histogenetic interest are dentigerous ameloblastic fibroma and odontogenic fibroma with sparse epithelial strands along the periphery of the tumour (References S3). In summary, there are two tumour mesenchymal patterns of POT, cell-rich (ameloblastic fibromalike) and cell-poor (odontogenic fibromyxoma-like) types. Accordingly, some doubt exists as to the exact nature of POT and whether it is truly a newly recognized embryonal tumour of immature dental tissues exhibiting the neoplastic characteristic of progressive growth, or whether it is merely a particular subset (architectural morphological variant) of ameloblastic fibroma or odontogenic fibroma/myxoma of dentigerous type arising during active dental development. Moreover, consideration should be given to the possibility that the cell-rich type may represent the primitive stage of a developing ameloblastic fibroma. The unique combination of proliferating fibrous and covering epithelial elements, a key diagnostic feature, accounts for the biphasic growth pattern of POT, thus leaving its ectomesenchymal or mixed epithelial– ectomesenchymal origin an open question.

Sclerosing odontogenic carcinoma.

To the Editor: In the October 2010 issue of the Pathology International, a fairly Japanese group of oral pathologists from Tokyo, Hiroshima and Osaka described a rare case of sclerosing odontogenic carcinoma (SOC) with features of benign fibroosseous lesion (BFOL). In their paper, Irie et al. claimed that previous to their case only three other examples of SOC have been reported. They also concluded: ‘To the best of [their] knowledge, this report is the first case of SOC with BFOL’ and ‘. . . there are currently no reports demonstrating intragnathic manifestation of extragnathic adamantinoma . . .’ We would like to call attention to a similar case we reported in 2009. Our patient was a 47-year-old female with a mandibular radiolucent lesion causing cortical perforation and producing a soft-tissue mass (Fig. 1). Microscopic features could not be placed in any known tumor entity and the incisional biopsy from the gingiva was diagnosed in 2002, by consulting pathologists from the Armed Forces Institute of Pathology (AFIP), as a clear cell carcinoma, probably odontogenic in origin. At that time, we classified it into the category of primary intraosseous carcinoma (PIOC) according to the 2001 AFIP publication. After reading the 2008 article of Koutlas et al., we thought it looked for the most part exactly like SOC (Fig. 2a–c), yet not a perfect match (deposition of dentinoid and submucosal larger nests of clear cells). In our report, we discussed it as a diagnostically challenging case of PIOC, because the problem still remains whether SOC represents a specific clinical entity, a non-specific growth pattern, or both (see below). Koutlas et al. reported on what they considered to be a unique variant of PIOC under the name of SOC. Historically, one of three cases has been presented by Allen (the second author of the above paper) dating back to 1996. Histologically, it is composed of a mixture of both epithelial and fibrous components. The former lacks worrisome cytologic features (atypia and mitosis) and forms thin cords reminiscent of epithelial rests in central odontogeinc fibroma (COF). The latter varies in cellularity and density, ranging from completely cellular through mildly collagenized to sclerotic. Whatever the pattern, the epithelial proliferation is squeezed out by the profuse fibrous tissue, hence its name. The defining characteristic is the presence of infiltrative margins with muscular involvement or perineural spread. In our opinion, the features of SOC are not specific, shared by many, different odontogenic tumors (e.g. COF, ameloblastic carcinoma, clear cell odontogenic carcinoma, PIOC, etc. . . .) in varying proportions. Put in simple terms, the pathologists should first rule out all other mimics elsewhere in the lesion using multiple sections from the resected specimen, that is to say, a diagnosis of exclusion. We have been able to trace two possible cases, probably the first and second described, of SOC with BFOL. In 1989, Jones et al. published a report on controversial COF in which a 51-year-old female presented with paresthesia of the lower lip and an expanded chin prominence. The lesion was curetted under the provisional diagnosis of BFOL and recurred in 16 months. Finally, it was considered to be a hybrid and descriptively termed as an ossifying COF. Four years later, an ossifying variant of PIOC in the mandible of a 46-year-old female was reported by Bennett et al. The overall histomorphology (epithelial inclusion BFOL) appeared similar to the case of Jones et al., excluding the added presence of larger squamous nests in areas where the tumor fused with the surface gingival epithelium. Although reliable comparison is impossible, descriptions and illustrations would fit with what we now call SOC. It can be learned from these two cases that BFOL-like areas are an integral component of SOC. Figure 1 Radiographic image (partial view of panoramic film) showing an osteolytic lesion in the mandible. Pathology International 2011; 61: 259–261 doi:10.1111/j.1440-1827.2011.02648.x

An expanded and revised early history of the adenomatoid odontogenic tumor

Adenomatoid odontogenic tumor (AOT) is not a new entity and has long been a subject of fascination after a century of recorded observation because of its unique biological profile. Hundreds of publications have covered a variety of aspects, ranging from demographic data to clinicopathologic features to treatment modalities. It is widely acknowledged that the first detailed reports of AOT were initially published in 1915 by Harbitz in Norway and later in 1916 by Wohl in the United States. However, it is very likely that earlier textbook and journal descriptions of AOT exist. This prompted us to review the extensive world literature on AOT dating back to the early part of the 19th century. Here we present a revised historical perspective of AOT, with a view to closing gaps in earlier knowledge about this entity and also encouraging further research.

Coexpression of an unusual form of the EWS–WT1 fusion transcript and interleukin 2/15 receptor βmRNA in a desmoplastic small round cell tumour

Background: The β chain of the interleukin 2/15 receptor (IL-2/15Rβ) is induced by the expression of the EWS–WT1. A case of desmoplastic small round cell tumour (DSRCT) expressing only an unusual EWS-WT1 treated by us is reported here. Aim: To characterise an unusual form of EWS–WT1. Methods: Frozen tissue sections of the axillary tumour were examined using a laser-assisted microdissection technique and reverse transcriptase polymerase chain reaction. Results: The novel fusion of exon 8 of EWS and the defective exon 10 of WT1 (−KTS) was detected. Although it was an unusual form, the coexpression of the present EWS–WT1, IL-2/15Rβ and Janus kinase (JAK1) mRNA was detected in the tumour cells. IL-2 and signal transducers and activators of transcription (STAT5) mRNA were detected in both tumour and stromal cells. Conclusion: The induction of the IL-2/15 receptor signalling pathway may contribute to tumorigenesis in DSRCT through a paracrine or an autocrine system, even though the EWS–WT1 was an unusual form.

Differentiation and apoptosis in pilomatrixoma.

We carried out a histopathologic study of pilomatrixoma, a benign skin tumor, and also examined apoptosis and hair differentiation with the aim to understand the presence of amorphous debris and cyst formation in the tumor. Among 16 cases of pilomatrixoma examined, 11 were at the early regressive stage and 5 were at the late regressive stage according to the classification by Kaddu et al. In the former cases, tumor nests were basically composed of basophilic, transitional, and shadow cells. Cyst formation was evident in all cases and squamoid epithelium was observed in 4 cases at the early regressive stage. Amorphous debris was found in all cases including those at the late regressive stage. Immunohistochemical analysis revealed positive reaction products for β-catenin and Lef-1 in basophilic and transitional cells, although their distribution differed. Immunoreactivity for β-catenin was observed in the lower transitional cells, whereas immunoreactivity for Lef-1 was also evident in the upper transitional cells. Positive reactions for hair keratins were found in the cytoplasm of transitional and shadow cells, but not in the amorphous debris. Examination by the terminal deoxynucleotidyl transferase-mediated dUTP-biotin nick end-labeling (TUNEL) method revealed positive reactions in transitional and some shadow cells. These results suggest that in pilomatrixoma, production of hair keratin and induction of apoptosis may occur at the same time, and that unlike the normal hair follicle irregular expression of β-catenin and Lef-1 results in the appearance of amorphous debris and cyst formation.