Christopher O’Conor, MD, PhD
Varsha Manucha, MD
August 2024—The Milan System for Reporting Salivary Gland Cytopathology (MSRSGC) is a widely endorsed and increasingly adopted standardized reporting system for salivary gland fine-needle aspiration biopsy specimens. The inaugural edition published in 2018 was inspired by the Bethesda cervicovaginal and thyroid reporting systems. The MSRSGC sought to facilitate clear and consistent communication between pathologists and the clinical team, thereby enhancing patient-centered decision-making and therapeutic interventions and assisting inter- and intralaboratory comparisons.
Like the first edition, the second edition atlas is organized into six general diagnostic categories: Non-Diagnostic, Non-Neoplastic, Atypia of Undetermined Significance (AUS), Neoplasm: Benign, Neoplasm: Salivary Gland Neoplasm of Uncertain Malignant Potential (SUMP), Suspicious for Malignancy, and Malignant. There is no change in the recommended reporting format from the previous edition. Each category, as in the previous edition, continues to be designated by a number, but using only the number without the diagnostic category is not advised. The risk of malignancy (ROM) for each category has been updated to reflect the most recent systematic reviews and meta-analyses in the literature; however, reporting of ROM for each diagnostic category is at the pathologist’s discretion.
New chapters on the latest ancillary studies, radiologic features, and clinical management of salivary gland lesions have been added, along with updates from the fifth edition of the WHO classification of head and neck tumors.
The following is a summary of the key updates within each diagnostic category.
Non-Diagnostic. In salivary gland FNA, the minimum number of epithelial/lesional cells or the amount of matrix required has not yet been established in the literature. Therefore, the second edition of the guidelines has eliminated the suggested threshold of 60 lesional cells for adequacy and highlights the importance of clinical and radiologic correlation before labeling a salivary gland aspirate as “non-diagnostic,” regardless of cellularity. Additionally, the second edition has an added explanatory note stating that fibrous stroma alone and cellular necrosis in the absence of inflammatory cells should not be considered diagnostic, as these findings can be seen in both neoplastic and non-neoplastic conditions. Finally, the estimated ROM used for reference has decreased from 25 percent to 15 percent.
Benign: Non-Neoplastic. There is a wide range of non-neoplastic lesions that can present as a mass-forming lesion in the salivary gland, thereby mimicking a neoplasm clinically and radiologically. Therefore, the “triple test” analogy—which requires a clear correlation between the clinical, radiologic, and cytologic findings—is required to avoid the pitfall of a false-negative result. Furthermore, the presence of reactive atypia, squamous or mucinous metaplasia, and degenerative changes in the setting of acute or chronic sialadenitis can be challenging and warrants a designation of “atypia of undetermined significance” instead of “non-neoplastic” to avoid a false-negative diagnosis. The estimated ROM for reference has been revised from 10 percent to 11 percent.
Atypia of Undetermined Significance (AUS). This category encompasses a small subset of problematic salivary gland FNAs that lack the quantitative or qualitative cytomorphologic features necessary for definitively diagnosing either non-neoplastic or neoplastic lesions, including aspirates with an insufficient degree of atypia warranting a suspicious or a definite diagnosis of malignancy. This category includes sparsely cellular specimens with basaloid/oncocytoid/atypical lymphoid cells, reactive atypia, metaplastic changes, matrix component only, or mucinous cystic material. It is suggested that this category should not represent more than 10 percent of cases to avoid overuse of this diagnosis. The estimated ROM for reference has increased from 20 percent to 30 percent. Given the heterogeneous nature of this category, the second edition of the MSRSGC recommends an accompanying note with the diagnosis clarifying the limitations of the specimens, interpretive challenge, potential differential diagnosis, and management recommendations.
Neoplasm: Benign. This category is characterized by neoplasms that have classic features of benign neoplasms, such as pleomorphic adenoma and Warthin tumor. The estimated ROM for reference has been changed from less than five percent to less than three percent. To maintain the low ROM, this diagnostic category should be used only when classic/unequivocal features of the specific benign neoplasm are present. In the absence of those features or with the presence of accompanying atypia or discordant clinical or imaging findings, the aspirate should be categorized as SUMP.
Salivary Gland Neoplasm of Undetermined Malignant Potential (SUMP). The diagnosis of SUMP is used when the specimen is diagnostic of a neoplasm but the differential diagnosis includes both benign and malignant entities. As in the previous edition, SUMP is further subcategorized based on the predominant cell type: basaloid, oncocytoid, clear cell-rich neoplasms, or mixed features. Each category includes a subset of benign and low-grade malignant salivary gland neoplasms, but a definitive diagnosis is precluded by the absence of either classic features of a benign tumor or by the presence of additional findings such as nuclear atypia, increased mitotic activity, or necrosis that would instead warrant a diagnosis of suspicious for malignancy. Utilization of ancillary testing (histochemistry, immunocytochemistry, and molecular) when possible is recommended for definitive categorization. The estimated ROM for reference has remained unchanged in the second edition at 35 percent.
Suspicious for Malignancy (SM). The SM category is largely unchanged, and it is used when there are cytomorphologic features suggestive of malignancy but the impression is limited by the quantity or extent of the features present. The purpose of the SM category is to maintain the highest positive predictive value of the malignant category as possible. In a subset of cases, ancillary testing on limited samples can aid in definitive classification. In the second edition, the estimated ROM for reference has increased from 60 percent to 83 percent.
Malignant (M). The malignant category has remained largely unchanged. Entities include primary salivary gland malignancies, metastases, lymphomas, and other rare tumors such as malignant mesenchymal neoplasms. Once the diagnosis of malignancy is made, an attempt must be made to grade the cancer given the potential impact on clinical management. The estimated ROM for reference has changed from 90 percent to greater than 98 percent.
Ancillary testing. The identification of unique genetic alterations in different salivary gland neoplasms has enabled a definitive diagnosis in some cases using either molecular immunomarkers, fluorescence in situ hybridization, or next-generation sequencing. Some of the recently identified antibodies include PLAG1 and HMGA2 for the diagnosis of pleomorphic adenoma, NOR 1 (NR4A3) for the diagnosis of acinic cell carcinoma, and pan-TRK as a surrogate marker for NTRK fusion-positive secretory carcinoma. As a specific diagnosis is not always required for appropriate clinical management of salivary gland lesions, ancillary testing should be considered only in situations where the specific diagnosis would change the classification according to the MSRSGC or impact the clinical management, or both.
Imaging of the salivary glands. The new edition has a chapter that provides an overview of the different imaging modalities (computed tomography, magnetic resonance imaging, ultrasound, nuclear scintigraphy, and positron emission tomography) and the findings in the non-neoplastic and neoplastic salivary gland lesions. This addition supports the overarching objective of employing a multidisciplinary approach to interpreting salivary gland FNAs.
Histologic considerations and clinical management. The classification of salivary gland neoplasms is dynamic and often specific to the tumor type. Limitations of salivary gland cytology are well recognized. Salivary gland neoplasms are also notorious for having intertumoral heterogeneity, creating a diagnostic challenge in limited sampling. Furthermore, benign neoplasms and low-grade malignancies are often composed of a similar cell type. In some cases, differentiating low-grade carcinomas from benign salivary gland tumors can require inspection of the tumor-stromal interface, a feature not evaluable in fine-needle aspirations. An important consideration is that the treatment of low-grade salivary gland carcinomas is usually similar to that for benign neoplasms, requiring limited conservative excision. Similarly, once the diagnosis of a high-grade salivary gland malignancy is made, the management is usually tumor-type independent, with radical excision, possible nerve sacrifice, and selective neck lymph node dissection.
In summary, fine-needle aspiration biopsy of the salivary gland is an effective and reliable way to diagnose salivary gland lesions. Fortunately, many salivary gland neoplasms are benign Warthin tumors or pleomorphic adenomas, which can be diagnosed by fine-needle aspiration with high sensitivity and specificity. The MSRSGC provides an organized methodology to report salivary gland lesions, particularly those that do not represent a clear, singular diagnostic entity. Finally, the MSRSGC will continue to evolve with advances in ancillary methods, further refinement of salivary gland neoplasm classification, and future studies that assess and characterize the system’s performance.
- Faquin WC, Rossi ED, Baloch Z, et al., eds. The Milan System for Reporting Salivary Gland Cytopathology. 2nd ed. Springer; 2023.
- Rossi ED, Baloch Z, Barkan G, et al. Second edition of the Milan System for Reporting Salivary Gland Cytopathology: refining the role of salivary gland FNA. J Am Soc Cytopathol. 2024;13(1):67–77.
- Rossi ED, Baloch Z, Barkan G, et al. Second edition of the Milan System for Reporting Salivary Gland Cytopathology: refining the role of salivary gland FNA. Cancer Cytopathol. 2024;132(1):10–21.
Dr. O’Conor is assistant professor and associate director of quality in cytopathology, Vanderbilt University Medical Center, and director of cytology, VA Tennessee Valley Healthcare System. Dr. Manucha is professor of pathology, director of the cytology division, and section chief of genitourinary and head and neck pathology, University of Mississippi Medical Center, Jackson. Both are members of the CAP Cytopathology Committee.