ⅠINTRODUCTION

Glandular odontogenic cyst (GOC) is a rare odontogenic cyst which has been proposed as a distinct entity by Gardener et al. in 1988¹⁾. GOC is characterized by a potentially aggressive behavior with a high recurrence rate,^2,3) and therefore an accurate diagnosis is critical for treatment and follow-up. However, due to similar histopathological features between GOC and other lesions such as botryoid odontogenic cyst, dentigerous cyst with mucous metaplasia or central mucoepidermoid carcinoma, the differential diagnosis can be challenging.⁴⁾

In a review of English language literature, Kaplan et al.⁴⁾ have proposed a set of major and minor criteria for histological diagnosis of GOC. Fowler et al.⁵⁾ have presented similar, but not identical, microscopic parameters for differential diagnosis between GOC and GOC mimickers. In addition to microscopic features, Macdonald-Jankowski⁶⁾ evaluated the clinical and radiological features of GOC, which can help distinguish GOC from other cystic lesions, through systematic review.

To date, case series of GOC have been studied regarding the demographics, radiological, histopathological and immunohistochemical features, treatment, and recurrence.^2-8) These accumulated data on GOC can help to make a correct diagnosis and to choose the best treatment. In an Asian population, however, a case series study of GOC has been carried out in a single institution, reporting 14 Chinese cases.^7,8) Furthermore, there has been only one case report in a Korean population.⁹⁾

The aim of this study was to present the clinical, radiological, and histopathological findings of 20 new Korean GOC cases. In addition, a brief review of the English literature was performed focusing on the demographic features, cyst location, and recurrence rate.

ⅡMATERIALS and METHODS

The present study included a total of 20 patients who were diagnosed with GOC and surgically treated in the Seoul National University Dental Hospital between 2006 to 2015. The demographic and clinical information was retrieved from the electronic medical record system. The radiological findings were analyzed using panoramic radiography and computed tomography (CT). Hematoxylin and eosin-stained slides were reviewed for histopathological evaluation.

The radiological analysis was done based on following eight radiological features mentioned in a previous systematic review:⁶⁾ radiolucency, shape, marginal definition, cortical status, expansion, tooth displacement, root resorption, and association with an impacted tooth. We evaluated the presence or absence of nine histological parameters which were previously suggested:^4,5) eosinophilic cuboidal cells, intraepithelial microcysts, clear cells in the basal or spinous layer, mucous cells, papillary projections, cilia, epithelial spheres, variations in thickness, and multiple cystic spaces.

A review and a case series including more than 20 GOC cases were selected for our literature review. The information on demographics, cyst location, and recurrence rate as well as radiological and histopathological features of each study was reviewed and summarized.

ⅢRESULTS

1Demographics

The age of the 20 GOC patients ranged from 29 to 73 years, with a mean age of 43.5 years (Fig. 1). Of the 20 GOC cases, 16 cases occurred in males and four in females, showing a strong male predilection (male-to-female ratio, 4:1).

2Location

All of the 20 GOC cases occurred in the tooth-bearing area. Nine cases (45.0%) were located in the maxilla and 11 cases (55.0%) in the mandible. Eight cases (40.0%) involved the anterior areas of the jaws and 12 cases (60.0%) were located in the posterior areas (Fig. 2).

3Radiological features

The results of our analysis of eight radiological features are presented in Fig. 3. Most of the cases (16/20, 80.0%) showed a well-defined unilocular radiolucency (Fig. 4A). Cortical perforation and bony expansion were found in 10 (55.6%) and 13 (72.2%) cases, respectively, out of 18 cases (Fig. 4B). Of the 20 GOC cases, only one case showed heterogeneous calcification within radiolucency (Fig. 4C). Eight cases (40.0%) were associated with an unerupted tooth, especially involving the crown of the tooth (Fig. 4D).

4Histopathological features

The results of our analysis of nine histological parameters are summarized in Table. 1. The most common histopathological feature was eosinophilic cuboidal cells, so-called “hobnail cells”, on the surface of the lining (Fig. 5A). The second most common feature was clear cells in the basal or spinous layer of the lining (Fig. 5B). More than 80% of the cases showed intraepithelial microcysts or duct-like structures, mucous cells, and variations in thickness of the lining epithelium (Fig. 5C-5E). Less commonly, papillary projections and epithelial spheres were found (Fig. 5F). Cilia and multiple cystic spaces were observed in half of the cases, respectively. One case showed intraluminal dystrophic calcification corresponding to its radiological finding.

5Treatment and Follow-up

All of the 20 GOC cases were surgically treated by enucleation without marginal resection. Follow-up information was obtained for 16 patients. Follow-up periods ranged from 1 month to 6 years, with a mean of 1.9 years. All the cases showed no evidence of recurrence on follow-up radiography.

6A brief review of the literature on GOC

Data on clinical, radiographic and microscopic features of GOC were collected from two reviews^4,6) and one case series⁵⁾ and are summarized in Table. 2.

ⅣDISCUSSION

Compared to the mean age (range of 45-51 years) and the gender predilection (male/female ratio, range of 1 to 1.8) previously reported in GOC patients,^4-6,10) our patients were slightly younger (43.5 years) and had a stronger predilection for male (male/female ratio, 4).

It has been generally known that GOC has a strong predilection for the mandible and the anterior areas of the jaws.^5,6,10) However, almost half (9/20, 45.0%) of our GOC cases occurred in the maxilla and the posterior regions (12/20, 60.0%) were involved more frequently than the anterior regions (8/20, 40.0%).

Although GOC was referred to as “sialo-odontogenic cyst” at first in 1987,¹¹⁾ its name was changed to “glandular odontogenic cyst” because of the lack of evidence of salivary gland origin.¹⁾ An immunohistochemical study was performed to investigate expression of cytokeratins (CKs) in GOC specimens.⁸⁾ CK14 was detected in the basal and suprabasal layer, and CK19 was expressed in all the layers.⁸⁾ CK14 is the main intermediate filament of odontogenic epithelium,¹²⁾ and CK19 is a useful marker for identifying odontogenic epithelial componets.¹³⁾ In addition, similar to a previous report,⁷⁾ all of our GOC cases occurred in the tooth-bearing area. Consequently, the odontogenic origin of GOC can be supported by the immunohistochemical result and its close proximity to the tooth.

According to previous studies, most GOCs showed a well-defined radiolucency with a single case accompanying calcification.^2,4-6) Similarly, most of our GOC cases demonstrated a well-defined radiolucent lesion and one case showed heterogeneous calcification within radiolucency. High et al.¹⁴⁾ have suggested that dystrophic calcification within the cyst lumen is a form of heterotopic calcification in degenerating tissue and correlates with age changes of the cyst. Of the reported GOC cases, 53.8-65.6% were unilocular,^4-6) and 11.1-19.5% of the cases were associated with an unerupted tooth.^5,6) Our case series showed higher rates of cases exhibiting a unilocular lesion (85.0%) and an association with an impacted tooth (40.0%).

Kaplan et al.⁴⁾ have proposed five major microscopic criteria which is mandatory for diagnosis of GOC. However, three major criteria were observed in not all of our GOC cases: variations in thickness (85.0%), mucous cells (85.0%), and intraepithelial glandular structures (80.0%). Although Kaplan et al.⁴⁾ have proposed clear cells as one of the minor criteria, most of our cases (95.0%) showed clear cells in the basal or spinous layer. Our results of analysis of each histological parameter were generally similar to those Fowler et al.⁵⁾ have reported. The presence of eosinophilic cuboidal cells on the surface of the lining (Fowler, 100%; our, 100%) is the most common histopathological finding, whereas multiple cystic spaces (Fowler, 63.0%; our, 50.0%) and cilia (Fowler, 21.7%; our, 50.0%) are least frequently observed in the lining epithelium of GOC.

GOC is characterized by a high recurrence rate which is comparable to that of odontogenic keratocyst.^2,3,15) However, no recurrence was found in all of our cases. This can be explained by two possibilities. One is related to the fact that a majority of our cases were unilocular in shape. Based on an association of a high frequency of multilocularity with a high recurrence rate, Kaplan et al.^2,4) have suggested that locularity is a significant factor related to the tendency to recur. The other possibility is associated with short follow-up periods in our case series. Fowler et al.⁵⁾ have reported that the average time interval from initial treatment to first recurrence was 8 years, with a range of 3-13 years. Considering the mean follow-up period of 1.9 years in our case series, continuous follow-ups are required for detection of recurrence.