Sclerosing epithelioid mesenchymal neoplasm of the pancreas: a rare mimicker of common pancreatic neoplasms

A 59-year-old female presented to the general surgery outpatient clinic at our hospital for evaluation of a pancreatic incidentaloma. Abdominal magnetic resonance imaging obtained at local hospital showed a lobulated pancreatic tail mass, exhibiting low signal intensity on T1-weighted images, intermediate to high signal intensity on T2-weighted fat-saturated images and high signal intensity on diffusion-weighted imaging. Contrast-enhanced scanning revealed patchy area of significant enhancement in the arterial phase and persistent enhancement in the portal venous and delayed phases. Physical examination was unremarkable. The patient denied any significant past medical history or family history of relevant diseases. The serum levels of carcinoembryonic antigen, carbohydrate antigen 19-9, and carbohydrate antigen 125 were within normal ranges. Abdominal ultrasound showed an ill-defined hypo-echoic lesion with irregular morphology and color Doppler flow imaging detected no positive blood flow signal. Thin-section contrast-enhanced computed tomography revealed no dilatation of the main pancreatic duct. A hypodense lesion measuring 2.0 cm × 1.8 cm without calcification, cystic change, and bleeding (Figure 1A), which exhibited progressive enhancement (arterial phase: Figure 1B; portal venous phase: Figure 1C; delayed phase: Figure 1D) was observed in the pancreatic tail. Distal pancreatectomy was performed. Microscopically, the lesion was hypercellular (Figure 1E). The neoplastic cells were arranged in cord-like, trabecular, nested, and sheet-like structures (Figure 1F). The background stroma was hypocellular and sclerotic, containing entrapped pancreatic parenchyma (Figure 1G), lymphoid aggregates and follicles. The lesion lacked both characteristic epithelial architectures and secondary changes. The neoplastic cells were epithelioid or spindle-shaped with scant to moderate eosinophilic cytoplasm. The nuclei were of two types: round to ovoid nuclei with open chromatin and occasional inconspicuous nucleoli, and spindle-shaped nuclei with condensed chromatin. Mitosis was rarely observed. Immunohistochemically, the neoplastic cells were diffusely positive for AE1/AE3 (cytokeratin) (Figure 1H), vimentin (Figure 1I), CD99 (Figure 1J), and somatostatin receptor 2 (SSTR2) (Figure 1K); focally/patchily positive for CK8 and CK19; and negative for CD10, LEF-1, chromogranin A, synaptophysin, bcl-10, trypsin, smooth muscle actin, desmin, myo-D1, myogenin, CD31, CD34, ERG, S-100, MUC4, p40, and p63. No abnormal expression of β-catenin and STAT6 was observed. No common gene mutations including KRAS, TP53, SMAD4, CDKN2A, MEN1, DAXX, ATRX, GNAS, RNF43, CTNNB1, and VHL), specific gene fusions/amplifications, or common germline mutations were detected. Somatic missense mutations (in ZNF521, ZNF217, and NCOR1) of uncertain clinical significance were identified.

Figure 1.

Contrast-enhanced CT (A-D), HE staining (E-G), and immunohistochemical staining (H-K) images of SEMN. (A) CT revealed a hypodense lesion in the tail of the pancreas. (B-D) Progressive enhancement was observed [arterial phase (B); portal venous phase (C); delayed phase (D)]. (E) The lesion was hypercellular (20× magnification). (F) The neoplastic cells were arranged in cord-like, trabecular, nested, and sheet-like structures (200× magnification). (G) The stroma was sclerotic, containing entrapped pancreatic parenchyma (200× magnification). (H-K) The neoplastic cells were diffusely positive for AE1/AE3 (H), vimentin (I), CD99 (J), and SSTR2 (K) (200× magnification). CT, computed tomography; HE, hematoxylin-eosin; SEMN, sclerosing epithelioid mesenchymal neoplasm; SSTR2, somatostatin receptor 2.

The differential diagnosis of a pancreatic mass includes mass-forming inflammatory diseases and neoplasms (epithelial or mesenchymal). For a middle-aged woman with no significant history, signs, or symptoms, the possibility of primary neoplasm should be highlighted. The radiological findings suggested a hypervascular lesion, possibly a pancreatic neuroendocrine tumor (PanNET) or solid pseudopapillary neoplasm (SPN). However, based on the evaluation of its morphology, immunohistochemistry, and gene mutation profile, the diagnoses of pancreatic ductal adenocarcinoma (PDAC), PanNET, SPN, pancreatic acinar cell carcinoma (PACC), and pancreatoblastoma were ruled out. Moreover, the lesion could not be classified as any of the mesenchymal neoplasms known to occur in or involve the pancreas. After thorough evaluation, this lesion was consistent with sclerosing epithelioid mesenchymal neoplasm (SEMN) of the pancreas. There was no evidence of disease recurrence for the patient at the last regular follow-up, 18.5 months after surgery. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the institutional ethics committee of Peking Union Medical College Hospital (approval No. S-K1593) and individual consent for this retrospective analysis was waived.

First described by Basturk et al. in 2020 (1), SEMN is a novel entity with only nine cases reported in the English-language literature. A comparison of the present case and previously reported cases is provided in Table S1. Of note, all SEMN cases have no defined lineage differentiation or recurrent genetic mutations (1,2). The neoplastic cells were epithelioid or spindle-shaped, with fibrotic stroma (1,2). All reported SEMN cases expressed AE1/AE3, vimentin, and CD99, but lack specific lineage differentiation (1,2). The radiological findings in our case were suggestive of a PanNET. Interestingly, the neoplastic cells demonstrated strong SSTR2 expression, suggesting that SEMN, similar to PanNET, may also show increased uptake on somatostatin receptor positron emission tomography/computed tomography (3).

The differential diagnosis of a solid pancreatic mass includes PDAC, PanNEN, SPN, and PACC, followed by rare mesenchymal neoplasms. Particularly, Basturk et al. listed several key morphological mimics with epithelioid neoplastic cells and sclerotic stroma for differential diagnosis, which include pancreatic sarcomatoid undifferentiated carcinoma (PSUC), sclerosing epithelioid fibrosarcoma (SEF), myoepithelial neoplasms of soft tissue, and angiomatoid fibrous histiocytoma (AFH) (Table S2). PSUC is a histological subtype of PDAC. At least 80% of the neoplasm is composed of spindle cells, with or without heterologous differentiation, whereas SEMN is more epithelioid and without heterologous component. Immunohistochemically, it expresses AE1/AE3 and vimentin as SEMN. However, PSUC contains common mutations observed in conventional PDAC (i.e., KRAS, TP53, SMAD4, and CDKN2A) which are not detected in SEMN. SEF is a (myo)fibroblastic neoplasm characterized by epithelioid neoplastic cells embedded in sclerosing collagenous stroma. It always occurs in low extremity and trunk. Involvement of pancreas is rare and only several cases were reported (4). MUC4 expression is sensitive and useful marker for diagnosis of SEF, which is not expressed in SEMN (5). There are no myoepithelial cells in normal pancreas. No myoepithelial neoplasm has been reported to involve the pancreas. Moreover, myoepithelial neoplasms express S-100 and p63, which is not observed in SEMN. AFH commonly occurs in extremities, trunk, or head and neck of young people. Pancreatic involvement is not reported in English literature. Morphologically, it contains variable dilated pseudo-vascular-like spaces and peripheral lymphoplasmacytic infiltration, which is not observed in SEMN. Neoplastic cells express desmin and CD99 but not express AE1/AE3, which is different from SEMN.

The absence of recurrence in all nine reported cases suggests indolent behavior but with uncertain biological potential (1,2). However, the biological nature of SEMN remains undefined given the limited data and follow-up. Moreover, there are several gaps in the diagnostic criteria, including: no specific marker, a mainly exclusion-based diagnosis, and limited experience with biopsy diagnosis. Therefore, reporting additional SEMN cases and their diagnostic workup by pathologists is critical to increase awareness and facilitate a better understanding of this rare entity.

Supplementary

The article’s supplementary files as

Ethical Statement: The authors are accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. The study was conducted in accordance with the Declaration of Helsinki and its subsequent amendments. The study was approved by the institutional ethics committee of Peking Union Medical College Hospital (approval No. S-K1593) and individual consent for this retrospective analysis was waived.