Morphological Features of the Spectrum of Ghost Cell Odontogenic Lesions

According to the latest classification of the World Health Organization (WHO), ghost cell odontogenic lesions comprise the following entities: calcifying odontogenic cyst (COC), dentinogenic ghost cell tumor (DGCT), and ghost cell odontogenic carcinoma (OGCC). Histopathologic diagnosis of these lesions is established by the presence of ameloblastic/ameloblastomatous proliferation, clusters of ghost cells that can calcify, and juxtaepithelial dentinoid/osteodentin-like material. The classification also included adenoid ameloblastoma (AA) as a separate pathological entity. Microscopically, AA is distinguished by the combination of cribriform pattern, duct-like structures, whorls/morules, dentinoid, and the presence of isolated clear cells and ghost cells [1]. However, some authors consider AA a hybrid lesion since it has overlapping microscopic features with DGCT, conventional ameloblastoma (CA), and adenomatoid odontogenic tumor (AOT) [2, 3]. In addition, the similarity of features between AA and DGCT suggests they represent a spectrum of “WNT-activated odontogenic lesions with mutated or wild-type ctnnb1” [4, 5]. AA is, therefore, a controversial entity that remains to be clarified. With the aim to contribute to the characterization and comparison of ghost cell odontogenic lesions and AA, we report a case of COC with areas reminiscent of AA.

A 27-year-old male patient presented with a 2 cm swelling in the left maxilla and normal lining mucosa of eight months duration; he reported agenesis of upper left second premolar (Fig. 1a). Computed tomography sections revealed a unilocular radiolucid osteolytic lesion extending from upper left canine to upper left first molar and involving the maxillary sinus. Root resorption of the involved teeth, and expansion, thinning, and fenestration of the buccal plate was also observed (Fig. 1b, c). The clinical-radiographic diagnosis was radicular cyst. Endodontic treatment was performed on the involved teeth because they did not respond to pulp sensitivity tests. The lesion was then removed, and curettage of the bone surface and affected tooth roots was performed. Macroscopic examination showed a 2.5 cm open cystic specimen with no content. Microscopic examination revealed the presence of a fibrous cyst wall lined with cuboidal epithelium with few layers of cells in some sectors and with ameloblastic/ameloblastous epithelium with intraluminal projections showing a cribriform pattern, duct-like structures, and focal whorled cellular condensations reminiscent of morules in others. Juxtaepithelial dentinoid was seen in some sectors. Ghost cells and chronic inflammation with giant multinucleated cells were also observed (Fig. 1d, e, g). Immunohistochemical studies showed diffusely positive AE1-AE3 in the epithelium and ghost cells (Fig. 1h, i); positive Beta-catenin expression in the epithelium, diffuse staining in the membrane/cytoplasm, focal expression in nuclei, and highest positivity in morules (Fig. 1j, k); focally positive p63 and Vimentin in the epithelium; diffusely positive CD68 in ghost cells; Ki67 expression up to 20% in the cribriform areas; and negative BRAFV600E, S100, and CD1a expression. Single-nucleotide variants (SNV) were studied in the ctnnb1 gene, and no mutations were detected (Fig. 1l). Diagnosis was COC with cribriform pattern, duct-like structures and morules. No signs of recurrence were observed during the 14-month follow-up (Fig. 1m).

Fig. 1.

Calcifying odontogenic cyst with cribriform pattern, duct-like structures, and whorls/morules. a: Clinical photograph showing swelling in the left maxilla. b, c: Computed tomography sections showing a unilocular radiolucid osteolytic lesion extending from upper left canine to upper left first molar and involving the floor of the maxillary sinus. d, e, f, g: Photomicrographs of hematoxylin-eosin stained histological sections showing sectors of the cystic wall in detail: cribriform pattern∞, duct-like structures×, whorls/morules*, a nest of ghost cells+, and juxtaepithelial dentinoid#. h, i: Immunohistochemistry, diffuse positivity of AE1-AE3 expression in the epithelium. j, k: Immunohistochemistry, positive expression of β-catenin in the epithelium (diffuse expression in the membrane/cytoplasm and focal expression in nuclei). l: Sequencing chromatogram of ctnnb1-FW indicating no mutations at codons 33, 34 and 37. m: Computed tomography Sect. 14 months post-surgery; note maxillary bone repair

The differential diagnosis considered in the case reported here was unicystic AA (UAA). To our knowledge, six cases have been reported to date, and all cases microscopically showed cystic ameloblastomatous proliferation with cribriform pattern, duct-like structures, morules, and juxtaepithelial dentinoid [2, 6].

Kyu-Young Oh et al. reported similar findings and observed areas reminiscent of AA in 8/11 COC (72.7%). In addition, they found positive nuclear β-catenin in 10/11 cases, ctnnb1 mutation in six cases, and frameshift mutations in suppressor genes involved in the Wnt pathway (APC, NEDD4L) in the remaining four [5].

In their systematic review of 130 DGCT, de Sousa Vieira et al. observed a male predilection, a peak incidence in the second decade and between the fourth and eighth decades of life, prevalent localization in the mandible, and radiographically appearing as a mixed lesion with ill-defined margins involving the cortical plates and teeth. Immunohistochemistry showed positive expression of AE1/AE3, CK14, and CK19 in neoplastic epithelium and positive β-catenin in 4/4 cases, whereas mutations in the ctnnb1 gene were only observed in one case [7].

With regard to AA, the review by Jayasooriya et al. included approximately 40 cases. The authors observed no sex predilection, a peak incidence in the fourth decade of life, prevalent localization in the mandible, and radiographic appearance as a radiolucent lesion, occasionally with radiopaque foci, and with ill-defined margins and cortical perforations in some cases [2].

Loyola et al. performed immunohistochemical studies in five cases of AA and found the tumors stained positive for CK7, 8, 14, 18 and 19, p16, and p53 [8]. Bastos et al. observed focal nuclear positive immunoreactivity for β-catenin in 7/9 and ctnnb1 mutations in 4/9 AA cases [9].

Jiang Xue et al. compared the genetic characteristics of AA, DGCT, AOT, and CA; they observed ctnnb1 mutation in 1/13 AA cases, BRAFp.V600E mutation in 17/18 CA, 2/13 AA, 1/14 DGCT, and 2/11 AOT cases, and detected k-ras mutations in 7/11 AOT cases. Their results distinguish AA from CA and AOT but do not conclusively differentiate between AA and DGCT [10]. Although some cases of AA and ghost cell odontogenic lesions (COC and DGCT) have been found to harbor ctnnb1 mutations, the existence of Wnt activated ctnnb1 wild type tumors suggests the involvement of this molecular pathway in the development of these lesions [4].

The clinical-radiographic and microscopic features, and immunohistochemical and molecular profiles of the few cases of AA (including UAA) reported in the literature show similar and overlapping features with DGCT and COC. We therefore posit that the lesion defined as AA may not be a true entity but rather a variant of the wide spectrum of ghost cell odontogenic lesions. Further studies are necessary to clarify the nature of these lesions and reach a consensus on their nomenclature.

Author Contributions

L.F.P: Conceptualization, Investigation, Immunohistochemical Study, Molecular Study, Writing - Original Draft, Review and Editing. E.S: Molecular Study, Resources, Writing - Review and Editing. F.M.S: Immunohistochemical Study, Resources, Writing - Review and Editing. L.G, P.G, E.G and S.P: Clinical-radiographic Study, Treatment, Follow-up. Manuscript Review and Editing. R.B.M: Resources, Writing - Review and Editing. M.L.P: Conceptualization, Methodology, Investigation, Writing - Original Draft, Resources, Supervision, Review and Editing. All authors read and approved the final manuscript.

Funding

This study was funded by a grant from the Facultad de Odontología-Universidad de Buenos Aires, FOUBA Res (CD) N° 330/19 − 01.

Data Availability

No datasets were generated or analysed during the current study.

Code Availability.

Not applicable.

Declarations

Ethical Approval

This report has obtained ethics approval from the Ethics Committee at Facultad Odontología-Universidad de Buenos Aires.

Informed Consent

For this type of study informed consent is not required.

Consent for publication

For this type of study consent for publication is not required.

Stats Declaration

No statistical analysis was performed in the preparation of this manuscript.

Competing Interests

The authors declare no competing interests.