Distant metastasis from salivary gland secretory carcinoma (SC) is rare, with lung and pleura being the most frequent site. While cytological features of SC on fine needle aspirates are well documented, its morphology in serous effusions has not been described. We describe the cytomorphological features on effusion cytology of two patients with ETV6::NTRK3 fusion-positive SC, who subsequently developed pleural metastases. Cytospin preparations of pleural fluid showed tightly cohesive, irregularly shaped and ball-like clusters of large tumor cells with scant to abundant uni-and multi-vacuolated cytoplasm. Nuclei were eccentrically placed, round to oval, vesicular, with finely granular chromatin, irregular nuclear membranes and conspicuous to prominent nucleoli. With these features, the tumors resembled an adenocarcinoma, indistinguishable from a lung primary. Cell blocks from both cases showed tumor fragments, some of which had the hollow appearance of transversely sectioned cell spheres as seen in lung and breast adenocarcinomas. Immunohistochemistry on cell blocks revealed nuclear pan-TRK positivity in both cases. Case 1 also showed focal mammaglobin staining, and TTF1 negativity. Pleural metastases from SC may mimic other adenocarcinomas. As targeted therapy, that is, selective TRK inhibitors are available for treatment of metastatic disease, NTRK3 fusion status is not only diagnostic, but also required to plan treatment. Pan-TRK immunohistochemistry serves as a viable cost-effective, easy to apply surrogate marker for NTRK3 fusion, particularly in diagnostic laboratories lacking easy access to molecular testing on cytological material.

Background Fine‐needle aspiration cytology (FNAC) is the first‐line diagnostic procedure for salivary gland masses. Secretory carcinoma (SC) is characterized by ETV6 and RET rearrangements detected by fluorescence in situ hybridization (FISH) or reverse transcriptase‐polymerase chain reaction optimized for paraffin‐embedded and fresh‐frozen tissue, respectively. The authors performed FISH on cytological material to assess its role in the diagnosis of SC. Methods FNACs with SC as a diagnostic consideration and cases diagnosed as SC on histology with a corresponding FNAC with any diagnosis were evaluated for ETV6 rearrangement by FISH. If acinic cell carcinoma (ACC) was a differential diagnosis and ETV6 rearrangement was absent, NR4A3 FISH was performed. FISH results were compared with those on histological specimens, where available. Results Fifteen cases were included. FISH initially performed on three cell blocks did not yield good results, was then performed on direct smears, and was interpretable in 14 cases (93.3%). An ETV6 rearrangement was identified in seven cases (50%), and an NR4A3 rearrangement was identified in three cases (21.4%), providing a confirmatory diagnosis in 10 of 15 cases (66.7%). The Milan System for Reporting Salivary Gland Cytopathology category was altered in two cases (6.7%). Complete correlation (100%) was seen with FISH on corresponding histological specimens. Conclusions With minor modifications, the FISH procedure can be optimized for FNAC smears with results comparable to those on histological specimens. ETV6 FISH testing on cytological smears in cases suspected as SC improves the diagnostic accuracy of FNAC and can help lower the proportion of the Milan categories salivary gland neoplasm of uncertain malignant potential and suspicious for malignancy, maximizing diagnostic information from less invasive samples and aiding in patient management.

No abstract