Diagnostic performance of Milan system for reporting salivary gland cytopathology: A prospective study

Background: Fine-needle aspiration (FNA) is a well-established modality for diagnosing salivary gland pathologies. The Milan System for Reporting Salivary Gland Cytopathology (MSRSGC) provides a standardized uniform framework leading to an evidence-based risk of malignancy (ROM). Based on the current literature, ROM in the 6-tier MSRSGC ranges from <5% for neoplasm-benign to >90% for the malignant category. Here, we report our institutional experience adopting MSRSGC. Methods: The cytopathology group at our institution implemented MSRSGC at the end of 2018. Through a query of our laboratory information system, we identified all salivary gland FNA cases from 27 November 2018 to 26 October 2020. The pertinent surgical pathology follow-up was also extracted. After manual curation, data was analyzed in Rv4.0.2. Results: Our cohort comprised of 315 patients undergoing 343 salivary gland FNA biopsies, predominantly on the parotid (90%), 162 with a surgical pathology follow-up. The risk of malignancy ranged from 3.2% in neoplasm-benign (IVA) to 100% in suspicious for malignancy (V) and malignant (VI) categories. ROM in the other categories was: 12.5% for non-diagnostic, 0 for non-neoplastic, 33.3% for atypia of undetermined significance, and 41.9% for salivary gland neoplasm of uncertain malignant potential (SUMP). Most SUMP cases had a basaloid or oncocytoid cytomorphology with similar ROM. In distinguishing benign and malignant salivary gland lesions, FNA had adequacy of 93.6%, a diagnostic yield of 62.2%, a sensitivity of 93.1% and a specificity of 100%. Conclusions: MSRSGC was successfully adopted by our cytology group and clinicians, with overall diagnostic performance similar to previous studies.