Salivary gland FNA: New markers and new opportunities for improved diagnosis

Salivary gland fine-needle aspiration (FNA) represents 1 of the most challenging areas of cytopathology in part because of the extraordinary diversity of both benign and malignant salivary gland tumors (SGTs), heterogeneity within the same tumor, and significant morphologic overlap between many of them. SGTs are 1 of the largest groups of human neoplasms, with over 3 dozen benign and malignant epithelial neoplasms in the 2005 World Health Organization classification. Many of these entities are rare, and additional entities have been recognized during the last decade. Nonetheless, FNA plays an important role in the evaluation of salivary gland lesions and reportedly has relatively high sensitivity (range, 86%-98%) for detecting carcinoma and high specificity (range, 97%-98%) in differentiating between benign and malignant tumors and between nonneoplastic and neoplastic tumors. This high degree of diagnostic accuracy is partially attributed to the finding that the majority of SGTs are pleomorphic adenoma (PA), Warthin tumor (WT), or high-grade carcinoma, including carcinoma that is metastatic to intraparotid and periparotid lymph nodes, for which the cytologic diagnosis is usually straightforward. However, the accuracy is more variable when cytology is used to specifically subtype a neoplasm (range, 48%-94%). Although precise classification of SGTs by FNA based on cytomorphology alone is possible for many of the commonly encountered lesions, it remains problematic for some entities. Indeed, most malignant SGTs are low grade (LG) and exhibit few overt cytologic features of malignancy. In addition, whereas high-grade (HG) malignancies are usually readily recognizable as malignant, they can be difficult to distinguish from 1 another. Finally, some benign SGTs (eg, basal cell adenoma or myoepithelioma) have a malignant counterpart that is often morphologically identical and can only be distinguished by growth pattern assessment; that is, it cannot be determined cytologically. Most SGTs presumably originate from acinar/ductal epithelial (luminal) or myoepithelial/basal (abluminal) cells and are composed of 1-cell types (eg, acinic cell carcinoma [AciCC]), 2-cell types (eg, PA and adenoid cystic carcinoma [AdCC]), 3-cell types (eg, mucoepidermoid carcinoma [MEC]), or more when considering superimposed metaplastic changes (eg, oncocytic and clear cells). Until recently, ancillary markers were of limited value for solving diagnostic problems in salivary gland neoplasia, primarily because they lacked specificity and because many SGTs have a similar composition of epithelial and myoepithelial cells with overlapping immunochemical profiles. However, the discovery of several novel and/or next-generation immunocytochemical and molecular