Disseminated carcinoma ex pleomorphic adenoma in an adolescent confirmed by application of PLAG1 immunohistochemistry and FISH for PLAG1 rearrangement

Insights into the molecular alterations of PLAG1 and HMGA2 associated with malignant phenotype acquisition in pleomorphic adenoma

Pleomorphic adenoma (PA) is the most common neoplasm of the salivary gland, presenting with a variety of histological features. In some cases, PA can undergo malignant transformation to carcinoma ex pleomorphic adenoma (CXPA). The transition from PA to CXPA is associated with complex molecular alterations, particularly involving the pleomorphic adenoma gene 1 (PLAG1) and high mobility group protein gene (HMGA2). This review investigates the molecular alterations of PLAG1 and HMGA2 in all domains in the malignant transformation of PA. Our analysis highlights that these markers are key alterations in the etiopathogenesis of PA and CXPA, with gene fusion and amplification being frequently reported mechanisms. Although the exact role of PLAG1 and HMGA2 in the oncogenic process remains unclear, further studies on the HMGA2 and PLAG1, are needed particularly in HMGA2-PLAG1-IGF2 which is proving to be a potential pathway for the development of clinically applicable therapies, especially for CXPA management.

Clinicopathological effect of PLAG1 fusion genes in pleomorphic adenoma and carcinoma ex pleomorphic adenoma with special emphasis on histological features

AIMS

Pleomorphic adenoma gene 1 (PLAG1) rearrangement is well known in pleomorphic adenoma (PA), which is histologically characterised by admixed epithelial and mesenchymal components. Multiple fusion variants of PLAG1 and HMGA2 have been reported; currently, however, little is known regarding the clinicopathological impacts of these fusion types METHODS AND RESULTS: We examined the PLAG1- and HMGA2-related fusion status in 105 PAs and 11 cases of carcinoma ex PAs (CXPA) arising from salivary glands and lacrimal glands to elucidate their correlation to the clinicopathological factors. Forty cases harboured PLAG1 fusion genes: CTNNB1-PLAG1 in 22 cases, CHCHD7-PLAG1 in 14 cases and LIFR-PLAG1 in four cases. Only two cases possessed HMGA2 fusion genes. The mean age of LIFR-PLAG1-positive cases was significantly higher than that of CTNNB1-PLAG1- and CHCHD7-PLAG1-positive cases (P = 0.0358). PAs located in the submandibular gland demonstrated CTNNB1-PLAG1 fusion at a significantly higher rate than other fusions (P = 0.0109). Histologically, PLAG1 fusion-positive cases exhibited chondroid formation and plasmacytoid features more commonly (P = 0.043, P = 0.015, respectively) and myxoid abundant feature less frequently (P = 0.031) than PLAG1 fusion-negative cases. For CXPAs, four CTNNB1-PLAG1 fusions were detected in two salivary duct carcinomas and two myoepithelial carcinomas. Ductal formation was observed frequently (90.9%) in residual PA.

CONCLUSIONS

The presence of PLAG1 fusion was associated with specific histological features in PA. Detecting the PLAG1 fusion gene and searching residual ductal formation in salivary gland malignant tumours with extensive hyalinisation could be useful for diagnosis.

Use of fluorescent in-situ hybridisation in salivary gland cytology: A powerful diagnostic tool

OBJECTIVE

Salivary gland cytology is challenging because it includes a diversity of lesions and a wide spectra of tumours. Recently, it has been reported that many types of salivary gland tumours have specific molecular diagnostic signatures that could be identified by fluorescent in-situ hybridisation (FISH). The aim of the present study was to demonstrate the feasibility and efficiency of FISH on routine cytological salivary gland smears.

METHODS

FISH was conducted on 37 cytological salivary gland smears from 34 patients. According to the cytological diagnosis suspected, MECT1/MAML2 gene fusion and rearrangements of PLAG1, MYB, or ETV6 were analysed. The presence and percentages of cells that had gene rearrangements were evaluated. Results were compared with the histological surgical samples, available from 26 patients.

RESULTS

The PLAG1 rearrangement was observed in 12/20 (60%) cases of pleomorphic adenoma. MECT1/MAML2 gene fusion was observed in 1:2 mucoepidermoid carcinomas but was not observed in five other tumours (two pleomorphic adenomas, one Warthin's tumour, one mammary analogue secretory carcinoma [MASC] and one cystic tumour). MYB rearrangement was observed in 4/4 adenoid cystic carcinomas. ETV6-gene splitting identified one MASC.

CONCLUSION

Overall, FISH had a specificity of 100% and a sensitivity of 66.7%. When FISH and cytological analyses were combined, the overall sensitivity was increased to 93.3%. It can thus be concluded that when the FISH analysis is positive, the extent of surgery could be determined with confidence pre-operatively without needing a diagnosis from a frozen section.

A novel cell line derived from pleomorphic adenoma expresses MMP2, MMP9, TIMP1, TIMP2, and shows numeric chromosomal anomalies

Pleomorphic adenoma is the most common salivary gland neoplasm, and it can be locally invasive, despite its slow growth. This study aimed to establish a novel cell line (AP-1) derived from a human pleomorphic adenoma sample to better understand local invasiveness of this tumor. AP-1 cell line was characterized by cell growth analysis, expression of epithelial and myoepithelial markers by immunofluorescence, electron microscopy, 3D cell culture assays, cytogenetic features and transcriptomic study. Expression of matrix metalloproteinases (MMPs) and their tissue inhibitors (TIMPs) was also analyzed by immunofluorescence and zymography. Furthermore, epithelial and myoepithelial markers, MMPs and TIMPs were studied in the tumor that originated the cell line. AP-1 cells showed neoplastic epithelial and myoepithelial markers, such as cytokeratins, vimentin, S100 protein and smooth-muscle actin. These molecules were also found in vivo, in the tumor that originated the cell line. MMPs and TIMPs were observed in vivo and in AP-1 cells. Growth curve showed that AP-1 exhibited a doubling time of 3.342 days. AP-1 cells grown inside Matrigel recapitulated tumor architecture. Different numerical and structural chromosomal anomalies were visualized in cytogenetic analysis. Transcriptomic analysis addressed expression of 7 target genes (VIM, TIMP2, MMP2, MMP9, TIMP1, ACTA2 e PLAG1). Results were compared to transcriptomic profile of non-neoplastic salivary gland cells (HSG). Only MMP9 was not expressed in both libraries, and VIM was expressed solely in AP-1 library. The major difference regarding gene expression level between AP-1 and HSG samples occurred for MMP2. This gene was 184 times more expressed in AP-1 cells. Our findings suggest that AP-1 cell line could be a useful model for further studies on pleomorphic adenoma biology.

An analysis of PLAG1 and HMGA2 rearrangements in salivary duct carcinoma and examination of the role of precursor lesions

AIMS

Salivary duct carcinoma (SDC) often arises in pleomorphic adenoma (PA). Putative precursors, including low-grade cribriform cystadenocarcinoma (LGCCC) and ductal carcinoma in-situ (DCIS), are more controversial. Rearrangement of PLAG1 or HMGA2 is seen in 50-70% of PAs, but this has not been investigated in SDC. Using a large collection of SDCs from a single institution, we aimed to study these genes by fluorescence in-situ hybridization (FISH), and to correlate the presence of precursor lesions/intraductal proliferations with gene alterations.

METHODS AND RESULTS

Forty-four SDCs were stained for smooth muscle actin, CK14, and p63, and examined with PLAG1 and HMGA2 FISH. Eight cases were SDC ex-PA; ten had a hyalinized nodule (HN), which is suspicious for PA; six arose in association with LGCCC; and twenty were 'de-novo' SDCs. Ten cases had PLAG1 rearrangement/amplification (22.7%) and eight had HMGA2 (18.2%) rearrangement/amplification. The positive cases were four SDC ex-PAs, eight SDCs with an HN, and five 'de-novo' SDCs. Twenty-three SDC ex-PAs were present in total (52.3%). All six SDC ex-LGCCCs were FISH-negative. Myoepithelial staining surrounded all LGCCCs, and demonstrated DCIS in 17 cases. Eleven DCIS lesions were in SDC ex-PAs or FISH-positive 'de-novo' SDCs. These cases represent 'cancerization' of ducts. Only six FISH-negative 'de-novo' SDCs showed DCIS.

CONCLUSIONS

A large proportion of SDCs arise in PAs (with or without residual evidence of a PA). A small proportion of SDCs arise in LGCCCs. Cases showing DCIS often represent cancerization.

PLAG1 alteration in carcinoma ex pleomorphic adenoma: immunohistochemical and fluorescence in situ hybridization studies of 22 cases

Carcinoma ex pleomorphic adenoma (CA-ex-PA) may arise with nearly any histologic subtype of carcinoma of the salivary gland. In the absence of recognizable residual pleomorphic adenoma (PA) or a prior history of PA, distinction of CA-ex-PA from morphologically similar de novo carcinomas may be difficult. Oncogenic rearrangement of PLAG1 (pleomorphic adenoma gene 1) has been established in PA; however, it has not yet been proven that PLAG1 alteration persists in carcinomas developed from preceding PA. We evaluated 22 histologically diverse CA-ex-PA by immunohistochemistry for PLAG1, and/or by FISH targeting PLAG1. Of these, 17 cases were immunoreactive (1+ to 3+) and 5 were immunonegative/rare positive for PLAG1. For comparison, 39 various salivary gland neoplasms were immunostained for PLAG1, of which all scored negative/rare positive. Twelve of 19 CA-ex-PA analyzed by PLAG1 FISH (63 %) were positive for gene rearrangement, 2 showed only a trisomy/polysomy profile, and 5 had a normal pattern. One FISH-positive tumor showed amplification of PLAG1. One of 3 cases analyzed for HMGA2 FISH was positive for gene rearrangement. In our series, the majority of CA-ex-PA harbored altered PLAG1 or HMGA2 genes detectable by FISH. While PLAG1 immunostain was specific for CA-ex-PA against other carcinomas, its application as a standalone discriminatory test was limited by variable expression. We conclude that most CA-ex-PA, regardless of morphologic subtype, carry altered PLAG1 or HMGA2 genes, and that FISH for PLAG1, along with immunohistochemistry for PLAG1, may help discriminate CA-ex-PA from its de novo carcinoma counterpart.