Mammary analogue secretory carcinoma (MASC) is a recently described salivary gland tumour that harbours the recurrent ETV6-NTRK3 translocation. This is the first series of MASC cases identified in the historic cohort of carcinomas of salivary glands with clinical/pathological correlation and follow-up data. We reviewed 183 primary carcinomas of major and minor salivary glands resected at the Medical University of Gdańsk, Poland, between 1992 and 2012. Based on morphology and immunohistochemistry, cases suspicious for MASC were selected, and the diagnosis was confirmed by fluorescence in situ hybridization (FISH) for ETV6 rearrangement and by RT-PCR for the ETV6-NTRK3 fusion transcript. Seven carcinomas met the criteria of MASC, as they exhibited a typical appearance with solid/microcystic and papillary architecture and intraluminal secretions, and cells completely devoid of basophilic cytoplasmic zymogen granules indicative of true acinar differentiation. The only paediatric case was an unencapsulated tumour composed of macrocystic structures covered by a mostly single but, focally, double layer of cells with apocrine morphology. In all cases, the neoplastic cells revealed immunoreactivity for S100, mammaglobin, cytokeratin CK7, CK8, STAT5a and vimentin. FISH for ETV6 gene rearrangement was positive in six out of seven cases, and RT-PCR was positive in three cases. MASC is a new entity of malignant epithelial salivary gland tumours not included in the 2005 WHO Classification of Head and Neck Tumours. There is a growing body of evidence that it is not as rare as was assumed, as is also indicated by our series (3.8 %). In most cases, MASC shares some microscopic features with AciCC, adenocarcinoma/cystadenocarcinoma NOS and low-grade MEC. In rare cases, MASC with high-grade transformation may mimic the morphological appearances of high-grade salivary gland malignancies, such as salivary duct carcinoma.

Mammary analogue secretory carcinoma of salivary gland origin (MASC) is a recently described tumor resembling secretory carcinoma of the breast characterized by strong S-100 protein, mammaglobin, and vimentin immunoexpression and which harbors a t(12;15) (p13;q25) translocation resulting in ETV6-NTRK3 fusion product. Histologically, conventional MASC displays bland histomorphology and a lobulated growth pattern and is often composed of microcystic, tubular, and solid structures with abundant eosinophilic homogenous or bubbly secretions. Colloid-like secretory material stains positively for periodic acid-Schiff with and without diastase as well as for Alcian Blue. We present for the first time, 3 patients with MASC of the parotid gland in which high-grade (HG) transformation developed in each case characterized by an accelerated clinical course and poor outcome. The HG component revealed strong membrane staining for EGFR and β-catenin, cytoplasmic/nuclear staining for S-100 protein, and nuclear staining for cyclin-D1, whereas HER-2/neu was absent. Analysis for the presence of the ETV6-NTRK3 fusion transcript revealed positivity in both HG and low-grade component of MASC in 2 of the 3 studied cases. The tumor in case 2 was negative in both its elements for the t(12;15) translocation, but ETV6 gene rearrangement was detected in both components in all 3 cases. Analysis of TP53 and CTNNB1 gene mutations in the HG component of MASCs as well as detection of copy number aberration of EGFR and CCND1 gene did not harbor any abnormalities. All 3 patients with HG-transformed MASC died of disseminated disease within 2 to 6 years after diagnosis. Recognizing HG-transformed MASC and testing for ETV6 rearrangement may be of potential value in patient treatment, because the presence of the ETV6-NTRK3 translocation may represent a therapeutic target in MASC.

Metagenomic sequencing revealed enrichment of microbes in single types of sarcoidosis samples but limited concordance across sample types. Statistical analysis accounting for environmental contamination was essential to avoiding false positives.

Myoepithelial carcinoma of salivary glands is an underrecognized and challenging entity with a broad morphologic spectrum, including an EWSR1-rearranged clear cell variant. Myoepithelial carcinoma is generally aggressive with largely unknown genetic features. A retrospective review of Salivary Gland Tumor Registry in Pilsen searching for the key words “clear cell myoepithelial carcinoma,” “hyalinizing clear cell,” and “clear cell malignant myoepithelioma” yielded 94 clear cell myoepithelial carcinomas (CCMCs) for molecular analysis of EWSR1 rearrangement using fluorescence in situ hybridization (FISH). Tumors positive for EWSR1 gene rearrangement were tested by next-generation sequencing (NGS) using fusion-detecting panels. NGS results were confirmed by reverse-transcription polymerase chain reaction or by FISH. Twenty-six tumors originally diagnosed as CCMC (26/94, 27.6%) revealed split signals for EWSR1 by FISH. Six of these tumors (6/26, 23%) displayed amplification of the EWSR1 locus. Fifteen cases were analyzable by NGS, whereas 9 were not, and tissue was not available in 2 cases. None of the CCMCs with EWSR1 rearrangements detected by FISH had an EWSR1 fusion transcript. Fusion transcripts were detected in 6 cases (6/15, 40%), including LIFR-PLAG1 and CTNNB1-PLAG1, in 2 cases each, and CHCHD7-PLAG1 and EWSR1-ATF1 fusions were identified in 1 case each. Seven cases, including those with PLAG1 fusion, were positive for PLAG1 rearrangement by FISH, with notable exception of CHCHD7-PLAG1, which is an inversion not detectable by FISH. One single case with EWSR1-ATF1 fusion in NGS showed ATF1 gene rearrangement by FISH and was reclassified as clear cell carcinoma (CCC). In addition, another 4 cases revealed ATF1 rearrangement by FISH and were reclassified as CCC as well. Moreover, 12/68 (17%) CCMCs with intact EWSR1 gene were selected randomly and analyzed by NGS. PLAG1 fusions were found in 5 cases (5/12, 41.6%) with LIFR (2 cases), FGFR1 (2 cases), and CTNNB1 (1 case) as partner genes. Overall, PLAG1 gene rearrangements were detected in 10/38 (26%) tested cases. None of the tumors had SMARCB1 loss by immunohistochemistry as a possible explanation for the EWSR1 abnormalities in FISH. Novel findings in our NGS study suggest that EWSR1-FISH positive CCMC is a gene fusion-driven disease with frequent oncogenic PLAG1 fusions, including LIFR-PLAG1 and CTNNB1-PLAG1 in most cases. Productive EWSR1 fusions are found only in a minority of EWSR1-ATF1-rearranged cases, which were in part reclassifiable as CCCs. Detectable EWSR1-FISH abnormality in CCMCs without gene fusion perhaps represents a passenger mutation with minor or no oncologic effect.

This study examines the presence of the EWSR1 rearrangement in a variety of clear cell salivary gland carcinomas with myoepithelial differentiation. A total of 94 salivary gland carcinomas with a prominent clear cell component included 51 cases of clear cell myoepithelial carcinomas de novo (CCMC), 21 cases of CCMCs ex pleomorphic adenoma (CCMCexPA), 11 cases of epithelial-myoepithelial carcinoma (EMC), 6 cases of EMC with solid clear cell overgrowth, and 5 cases of hyalinizing clear cell carcinoma of minor salivary glands. In addition, 10 cases of myoepithelial carcinomas devoid of clear cell change and 12 cases of benign myoepithelioma were included as well. All the tumors in this spectrum were reviewed, reclassified, and tested by fluorescence in situ hybridization (FISH) for the EWSR1 rearrangement using the Probe Vysis EWSR1 Break Apart FISH Probe Kit. The EWSR1 rearrangement was detected in 20 of 51 (39%) cases of CCMC, in 5 of 21 (24%) cases of CCMCexPA, in 1 of 11 (9%) cases of EMC, and in 4 of 5 (80%) cases of hyalinizing clear cell carcinoma. The 25 EWSR1-rearranged CCMCs and CCMCexPAs shared similar histomorphology. They were arranged in nodules composed of compact nests of large polyhedral cells with abundant clear cytoplasm. Necrosis, areas of squamous metaplasia, and hyalinization were frequent features. Immunohistochemically, the tumors expressed p63 (96%), cytokeratin CK14 (96%), and S100 protein (88%). MIB1 index varied from 10% to 100%, with most cases in the 20% to 40% range. Clinical follow-up information was available in 21 cases (84%) and ranged from 3 months to 15 years (mean 5.2 y); 4 patients were lost to follow-up. Ten patients are alive with no evidence of recurrent or metastatic disease in the follow-up period from 3 months to 15 years (mean 5 y), 3 patients are alive with recurrent and metastatic disease, and 8 died of disseminated cancer 9 months to 16 years after diagnosis (mean 6 y). Lymph node metastasis appeared in 5 patients within 5 months to 4 years after diagnosis (mean 22 mo), distant metastases were noted in 7 patients with invasion of orbit (2 cases), and in 1 case each metastasis to the neck soft tissues, liver, lungs, mediastinum, and thoracic vertebra was noted. We describe for the first time EWSR1 gene rearrangement in a subset of myoepithelial carcinomas arising in minor and major salivary glands. The EWSR1-rearranged CCMC represents a distinctive aggressive variant composed predominantly of clear cells with frequent necrosis. Most EWSR1-rearranged CCMCs of salivary glands are characterized by poor clinical outcomes.

ETV6 gene abnormalities are well described in tumor pathology. Many fusion partners of ETV6 have been reported in a variety of epithelial and hematological malignancies. In salivary gland tumor pathology, however, the ETV6-NTRK3 translocation is specific for mammary analogue secretory carcinoma (MASC), and has not been documented in any other salivary tumor type. The present study comprised a clinical and molecular analysis of 25 cases morphologically and immunohistochemically typical of MASC. They all also displayed the ETV6 rearrangement as visualized by fluorescent in situ hybridization but lacked the classical ETV6-NTRK3 fusion transcript by standard reverse-transcriptase-polymerase chain reaction. In 4 cases, the classical fusion transcript was found by more sensitive, nested reverse-transcription-polymerase chain reaction. Five other cases harbored atypical fusion transcripts as detected by both standard and nested reverse-transcription-polymerase chain reaction. In addition, fluorescent in situ hybridization with an NTRK3 break-apart probe was also performed; rearrangement of NTRK3 gene was detected in 16 of 25 cases. In 3 other cases, the tissue was not analyzable, and in 2 further cases analysis could not be performed because of a lack of appropriate tissue material. Finally, in the 4 remaining cases whose profile was NTRK3 split-negative and ETV6 split-positive, unknown (non-NTRK) genes appeared to fuse with ETV6 (ETV6-X fusion). In looking for possible fusion partners, analysis of rearrangement of other kinase genes known to fuse with ETV6 was also performed, but without positive results. Although numbers were small, correlating the clinico-pathologic features of the 4 ETV6-X fusion tumors and 5 MASC cases with atypical fusion transcripts raises the possibility of that they may behave more aggressively.

The present study was undertaken to evaluate real probability and pattern of cervical occult lymph node metastases (OLNM) in primary parotid carcinoma (PPC). We carried out a retrospective analysis of 66 patients treated in years 1992–2010 due to PPC, who underwent elective neck dissection (END). In search of risk factors for OLNM, we analysed the following parameters: age, sex, pT-Status, tumour size, skin invasion, facial nerve palsy, tumour fixation, extraparotid extension, localization, grade, histology, intra/periparotid LN metastases (IPLNM). OLNM was observed in 30.3% of patients. In a univariate analysis statistical significance was found for IPLNM, extraparotid extension and high risk histology. A multivariate analysis showed statistical significance only for the first variable. The most common location of cervical OLNM was level II (80%), then III (45%) and V (30%). In a compilation of our own material with data from the literature (5 series), we obtained a group of 80 patients with OLNM, selected out of 650 patients with cN0 (12.3%). The proportion of metastases to particular levels was the following: 69%—II, 22.5%—III, 20%—I,16%—V, 7.5%—IV. END should be carried out in case of all T3/T4a carcinomas with minimal range of levels II and III. Removal of levels Ib and Va is recommended as well. In the T1/T2 carcinomas with high grade/high risk histology, END should be performed including levels II and III.

The recurrent translocations t(11;19) and t(11;15) resulting in CRTC1-MAML2 or CRTC3-MAML2 fusion oncogenes, respectively, are identified in a large proportion of mucoepidermoid carcinomas (MECs) of the salivary gland and have impact on prognosis. However, there are conflicting data on the specificity of this translocation, in particular, on its putative occurrence in Warthin tumor (WT) of the parotid gland as reported in few previous cases. It was speculated that extensive squamous metaplasia could explain the presence of t(11;19) translocation in a subset of WTs. We evaluated 76 salivary gland tumors, including 16 cases of metaplastic WT and 8 cases of pleomorphic adenoma (PA) with squamous and/or mucinous metaplasia, extensive enough morphologically to mimic MEC. Detection of CRTC1-MAML2 and CRTC3-MAML2 fusion transcripts and MAML2 gene break was performed using nested reverse transcription-polymerase chain reaction and fluorescence in situ hybridization (FISH), respectively. None of 16 analyzed metaplastic WTs showed positivity for fusion transcripts CRTC1-MAML2 or CRTC3-MAML2, and none showed rearrangement of the MAML2 gene by FISH. Similarly, we did not detect these transcripts or break of MAML2 gene in any case of PA with extensive squamous/mucinous metaplasia. For comparison, 40 cases of low-grade MEC were also evaluated. CRTC1-MAML2 and CRTC3-MAML2 fusion transcripts were detected in 17 and 5 cases, respectively. The FISH method using break-apart probe demonstrated the MAML2 gene rearrangement in 25 cases of low-grade MEC. In contrast to low-grade MEC, neither metaplastic WTs nor metaplastic PAs harbored translocations t(11;19) and anticipated t(11;15) resulting in CRTC1-MAML2 and CRTC3-MAML2 fusion transcripts, respectively, and/or MAML2 gene rearrangement.