Myoepithelial tumors (METs) found in bone and soft tissue are uncommon mesenchymal tumors that the World Health Organization (WHO) categorizes as having uncertain differentiation. The main subtypes include mixed tumor, myoepithelioma, and myoepithelial carcinoma (1). Mixed tumors have tubular differentiation akin to their salivary gland equivalents. Myoepitheliomas mainly consist of spindle-shaped or oval myoepithelial cells with minimal atypia, while moderate to marked atypia is a characteristic of myoepithelial carcinoma.

While most myoepithelial tumors express epithelial markers like cytokeratin and/or epithelial membrane antigen (EMA) along with S-100 protein, there is a significant amount of variability among individual cases (2, 3). Because of their marked morphological and immunophenotypic heterogeneity, molecular genetic profiling has emerged as a valuable method for accurate characterization. For example, Salivary gland and skin lesions often show alterations in PLAG1 and HMGA2 genes, while approximately 50% of myoepithelial tumors in bone, soft tissue, and visceral organs possess recurrent fusions related to EWSR1 or FUS. Currently, the confirmed fusion partners include POU5F1, PBX1, ZNF444, ATF1, and PBX3 (4–6).

The EWSR1–PBX3 fusion was first detected in skin lesions but is now more commonly found in soft tissue and intraosseous myoepithelial tumors. Typically, tumors containing this fusion exhibit minimal or no cytokeratin expression and are generally related to a good prognosis (6–8). There are few studies on intraosseous METs that involve EWSR1-PBX3 fusion and reported cases most frequently involve the maxilla, ilium, tibia, and fibula (5, 6).

This study presents a unique case of myoepithelial tumor involving both the rib and adjacent paraspinal soft tissue, characterized by the coexistence of EWSR1-PBX3 fusion and rearrangement. As far as we know, this is among the initial documented instances of a paraspinal soft-tissue myoepithelioma invading a nearby rib with this distinct genetic change.

While myoepithelial tumors are frequently documented in scientific literature, instances involving the EWSR1-PBX3 fusion are extremely uncommon. This report presents a distinctive case that demonstrates both the EWSR1-PBX3 fusion and rearrangement. In the absence of RNA evidence, we propose that the EWSR1::PBX3 (E9;P5) rearrangement represents the candidate driver event, constituting a fusion gene, whereas the EWSR1::PBX3 (E9;P4) rearrangement, due to its 5’-to-5’ configuration, likely represents a non-productive structural rearrangement without expression of a functional fusion transcript. In previously reported cases, around fifty percent of soft-tissue and bone myoepithelial tumors contain EWSR1 gene fusions. The EWSR1-PBX3 fusion partner was first identified by Agaram et al. in 2014 in two intraosseous and one soft-tissue tumor (5). Suurmeijer et al. subsequently reported 10 additional cases harboring EWSR1-PBX3 fusion in METs (6). Later, Yun and Haglund described intraosseous lesions in the metatarsal and ulna, respectively (9, 10). Gandhi et al. reported six further cases, demonstrating a spectrum of osseous myoepithelial tumors ranging from benign to malignant (11). To date, approximately twenty cases of bone myoepithelial tumors featuring the EWSR1-PBX3 fusion have been described in various skeletal and soft tissue sites, including the tibia, fibula, phalanges, navicular, radius, and ulna. The present case represents the first reported example arising in the rib.

Microscopically, the tumor displayed pinkish nodular tissue with hemorrhage and coagulative necrosis. It consisted of bland oval-to-spindle cells with eosinophilic cytoplasm and scattered perivascular lymphocytes. Immunohistochemically, both bone and soft tissue components were positive for S100, EMA, and Calponin, but negative for cytokeratin. The EWSR1-PBX3 fusion is most commonly found in cutaneous syncytial myoepithelioma (CSM), which is a benign tumor noted for its syncytial growth pattern and the co-expression of EMA and S-100, along with the repetitive occurrence of the EWSR1-PBX3 fusion. Compared with the broader group of myoepithelial tumors, CSM displays a more uniform immunophenotypic and genetic profile (12–14). A small subset of myoepithelial tumors, however, arises in soft tissue or bone. The distinction between benign and malignant tumors harboring this fusion is often based on mitotic count in hematoxylin and eosin (H&E)-stained sections. Malignant lesions typically show marked nuclear atypia, prominent nucleoli, increased mitotic activity and areas of necrosis. Although most osseous METs with EWSR1-PBX3 fusion are benign, a minor subset exhibits malignant features.

Morphologic and immunohistochemical evaluation remains the cornerstone for identifying METs; however, due to their marked heterogeneity and complexity, detection of EWSR1-PBX3 fusion provides substantial value for accurate clinical classification. Misdiagnosis can occur in challenging cases—for example, as Haglund noted, diffuse and strong SATB2 staining may lead to an erroneous diagnosis of osteosarcoma (10). Similarly, TLE1 positivity in our case could mimic synovial sarcoma; although TLE1 is considered a sensitive and specific marker for synovial sarcoma, its biological role remains uncertain (15, 16). In our case, the initial misdiagnosis was partly due to the absence of confirmatory SS18-SSX fusion testing, which is a critical diagnostic marker for synovial sarcoma (17). Subsequent NGS analysis not only failed to detect any SS18-related alterations but conclusively identified an EWSR1 rearrangement, thereby definitively establishing the diagnosis of myoepithelial tumor. Therefore, molecular characterization of EWSR1-PBX3 fusion is essential for accurate diagnosis of myoepithelial tumors of bone and soft tissue. In addition, we also tested positive for an ARID1B p.H172delH mutation. ARID1B is known to promote cancer progression primarily by disrupting chromatin remodeling, impairing DNA repair, and dysregulating the cell cycle (18). ARID1B is frequently mutated in cancers such as breast, ovarian, and lung carcinomas (19–21); however, its role in myoepithelial tumors remains unstudied. We hypothesize that it may lead to dysregulation of chromatin remodeling, compromise normal gene expression control, and potentially contribute to carcinogenesis in this context.

To summarize, we present a compelling case involving EWSR1-PBX3 fusion and rearrangement located in the twelfth rib and surrounding soft tissue. This finding contributes additional insights into the molecular mechanisms underlying myoepithelial tumors in both bone and soft tissue.