Clear cell changes in salivary gland neoplasms: A 20-year retrospective study

Background Clear cells are observed histopathologically in both benign and malignant neoplasms but their presence in salivary gland tumors has not been extensively documented. Material and Methods With IRB approval, the archive of the University of Florida College of Dentistry oral pathology biopsy service was retrospectively searched from 1994-2014 for all benign and malignant salivary tumors. Epidemiological data, tumor location and duration, and type of tumor were recorded. A four reviewer panel examined the original slides. Reviewers scaled each case as 0 (no clear cells present), 1 (few to focal clear cells), 2 (less than 50% clear cells), and 3 (greater than 50% clear cells). Results A total of 535 cases were included of which 48% of tumors displayed 0 clear cells (257/535), 31.4% (168/535) scored 1, 13.6% (73/535) scored 2, and 7% (37/535) scored 3. Of the 251 (47%) malignant neoplasms, 64% (160/251) demonstrated 0-1 clear cell change, while 36% (91/251) showed a score of 2-3. For the total 284 (53%) benign tumors, 93% (265/535) scored 0-1 and 7% (19/535) scored a 2-3 range. No statistical difference was noted for gender, age, or duration of time present in regards to presence or absence of clear cells. Statistically significant differences in clear cell presence were found between location groups, between benign and malignant diagnosis, and between specific diagnostic groups. Conclusions This study demonstrates the frequent presence of increased numbers of clear cells in oral salivary malignancies and highlights salivary gland differential diagnoses when presented with clear cell changes. Key words:Clear cell change, salivary tumors, benign tumors, malignant tumors.


Introduction
Clear cells may be found as incidental histologic findings in a multitude of benign or malignant tumors of many cell origins including epithelial, melanocytic, mesenchymal, or hematopoietic (1). They may be a result of many different processes, including artifact, degeneration of cellular organelles, or accumulation of substances within the cells -most commonly glycogen, but sometimes mucopolysaccharides, mucin, lipids, or foreign bodies (1,2). Within the head and neck region, clear cells are found most commonly in salivary gland tumors, but also may be seen in tumors of squamous or odontogenic epithelial origin, primary or metastatic carcinoma, benign or malignant melanocytic lesions, or benign or malignant mesenchymal tumors (1,3). Knowledge of the frequency and patterns of clear cell presentation in salivary gland tumors may be helpful in determining a diagnosis. This study serves to describe both the epidemiologic, clinical, and histologic features of a large series of salivary tumors and also to detail the occurrence and histologic appearance of clear cells found within this group of tumors.

Material and Methods
Institutional review board approval and permission for waiver of informed consent was granted from the University of Florida Institutional Review Board Ethics Committee (Approval #IRB201400598). We identified 641 cases of salivary tumors in the archive files of the University of Florida College of Dentistry's Oral Pathology Laboratory (Gainesville, FL) dating from January 1994-January 2014. Epidemiologic data was collected to include age, gender, diagnosis, site of occurrence, and time period the lesion was present. In instances where both an incisional and excisional specimen were available, the excisional specimen was chosen for examination in the study. Metastatic tumors and cases with insufficient tissue were excluded from the study. Hematoxylin and eosin stained slides were examined for each tumor by a panel of four Oral and Maxillofacial Pathologists (IB, DC, MI, SF). Clear cell changes were tabulated for each case with the following criteria: tumors with no clear cells present = grade 0 (negative); those with few to focal clear cells (less than 25%) = grade 1 (focal); tumors with moderate clear cells over 25% but less than 50% = grade 2 (moderate); tumors composed of greater than 50% clear cells = grade 3 (diffuse). Statistical analysis was performed to compare differences between groups in terms of clear cell composition with Pearson chi-square test using IBM SPSS version 22, and a p-value < 0.05 was considered statistically significant.

Results
The final number of cases included in the study was 535. Examples of specimens graded as levels 0-3 may be seen in figure 1. An overall comparison between    e280 was a statistical difference in comparison to benign tumors (p=0.000).

Discussion
The cases described in our series contribute one of the largest case series of salivary gland tumors described in the literature. For the most part, the demographics of the cases in our series are commensurate to the parameters previously reported. The mean age affected was 57.7 years which is similar to overall rates mentioned in the literature for salivary tumors (4). Females comprised 57% of the patients in the study, which is also consistent with most cited reports (4). Due to the nature of our biopsy service which is comprised mainly of submissions from private practice offices not involving a hospital setting, parotid tumor submissions were rare, and most cases represented minor salivary gland based tumors. The benign to malignant ratio in this study was close to equal, which is typical of minor salivary gland tumors (4). PA was the most common benign tumor in the study and MEC the most common malignancy, which fits with prior studies (2,5  (16). In addition, multiple benign and malignant adnexal eccrine and sebaceous tumors may show clear cell change (16). Squamous cell carcinoma may also show predominant clear cell populations in the skin in some cases (16). Myoepithelial tumors of soft tissue may show clear cell change in 45-55% of tumors but it is rarely predominant (17). Mesenchymal tumors with reported clear cell change include clear cell sarcoma, benign and malignant smooth muscle tumors, and perivascular epithelioid cell tumors (16). Finally, metastatic tumors such as renal cell carcinoma (RCC) are a significant consideration in evaluating clear cell tumors of the head and neck, but RCC is often distinguishable by a robust vascular network and distinct immunohistochemical profile (1).

Conclusions
In conclusion, in our study population, tumors of primarily ductal origin rarely showed clear cell change, and malignant tumors were significantly more likely than benign to exhibit more extensive clear cell change. Location was somewhat predictive of clear cell change in this study population, but explainable by prevalence of certain tumors in specific locations. Age, gender, and duration were not predictive of clear cell change. Clinicians and pathologists should carefully consider other clear cell tumors in the differential diagnosis when confronted with a salivary tumor with predominant clear cell change.