Advancing Education, Research, and Quality of Care for the Head and Neck oncology patient.
Background: The Milan system for reporting salivary gland cytopathology (MSRSGC) was recently updated in 2023 using the data published since the original guidelines were published in 2018. This update focused on refining the malignancy rate for each of the six diagnostic categories to further improve patient counseling and surgical decision-making. We published our initial data in 2022. Our update seeks to re-examine the adoption of MSRSGC within the same community setting and compare the recent performance to the new benchmarks and our previous findings.
Methods: Between May 2021 and June 2024, a total of 266 salivary gland ultrasound-guided fine needle biopsies were performed in an Otolaryngology practice. A total of six pathology departments at the affiliated hospitals reported the results. Either MSRSGC or descriptive reporting was used by the pathologist to communicate cytology results to the treating surgeon. Frequency of MSRSGC reporting was calculated. The rate of sialoadenectomy and diagnostic accuracy were calculated for MSRSGC and descriptive reporting. Rates of malignancy were calculated for each Milan category for direct comparison against benchmark rates of malignancy.
Results: A total of 228 FNA cytopathology reports were used for surgical decision-making. Approximately 96% of all MSRSGC results came from three pathology departments (62.8%, 18.2%, and 14.9%). Sialoadenectomy rate was 37% for MSRSGC compared to 33% for descriptive cytopathology reports (p = 0.51). There were 79 surgical pathology specimens available, 37 of which utilized MSRSGC and 42 that used descriptive cytopathology. After excluding non-diagnostic FNA results from each category, the diagnostic accuracy of MSRSGC FNAs was 81% compared to 69% for descriptive reports (p = 0.22). The rates of malignancy for Milan 6, Milan 5, Milan 4b categories were similar to benchmarks mentioned in the recent update (100% vs. >98%, 100% vs. 83%, and 22% vs. 35% respectively). Our Milan 4a: Pleomorphic adenoma was in line with benchmarks (0% vs. <3%), however, our Milan 4a: Warthin Tumor category had an abnormally high rate of malignancy compared to benchmarks (25% vs. <3%). Only 4 patients, 29%, of the Milan 4a: Warthin Tumor had surgery. One of those patients had final pathology of high-grade neuroendocrine Merkel cell tumor. Milan 1, Milan 2, and Milan 3 categories all demonstrated much lower rates of malignancy compared to the updated benchmarks (0% vs. 15%, 0% vs. 11%, 17% vs. 30% respectively).
Conclusions: There has been an improvement in the adoption of MSRSGC reporting across pathology departments in our community compared to our prior study. In our cohort, there remains no statistically significant difference in the rate of sialoadenectomy or diagnostic accuracy between MSRSGC and traditional reporting. The rates of malignancy for MSRSGC reported pathology were similar to updated benchmark rates, except for the 4a: Warthin tumor category, which had one of four samples with malignant histopathology. Use of immunohistochemistry or molecular testing in the Milan 4a and 4b categories would be helpful in surgical decision-making. The MSRSGC brings uniformity and standardization to the FNA reporting process for salivary gland cytopathology and continues to be an overall accurate tool for predicting malignancy rates.