Assessing the diagnostic accuracy for pleomorphic adenoma and warthin tumor by employing the Milan System for Reporting Salivary Gland Cytopathology: An international, multi-institutional study.

BACKGROUND The Milan System for Reporting Salivary Gland Cytopathology (MSRSGC) has established distinct diagnostic categories for reporting cytopathological findings, and each is associated with a defined risk of malignancy (ROM). However, the ROM is applied at the overall category level and is not specific for particular morphological entities within a category. Here, the diagnostic performance of the MSRSGC for pleomorphic adenoma (PA) and Warthin tumor (WT) is reported. METHODS The pathology archives of 11 institutions from 4 countries were retrospectively searched to identify all salivary gland fine-needle aspiration (FNA) biopsies with a differential or definitive diagnosis of PA or WT and all resection specimens with a diagnosis of PA or WT; only paired cases were included. All FNA diagnoses were retrospectively classified according to the MSRSGC. RESULTS A total of 1250 cases met the inclusion criteria, and they included 898 PA cases and 352 WT cases. The ROM in the benign neoplasm category was 3.0% and 1.3% for cases with a differential or definitive diagnosis of PA and WT, respectively. The ROM in the salivary gland neoplasm with uncertain malignant potential (SUMP) category was 2.7% and 18.8% for PA and WT, respectively (P = .0277). The diagnostic accuracy for PA and WT was 95.1% and 96.1%, respectively. CONCLUSIONS The diagnostic accuracy for PA and WT on FNA is high. Furthermore, these findings highlight the difference in the ROMs associated with 2 specific differential diagnoses in the SUMP category: basaloid neoplasms and oncocytoid neoplasms.