A malignant myoepithelioma (MM) shows a predominantly reticular growth pattern with cords of epithelioid, ovoid, or spindled cells deposited in a variably collagenous stroma, and the presence of sever nuclear atypia with easily discerned nucleoli (1). Although the most frequent site is the salivary glands, it rarely occurs in soft tissue (2-5), and MM of the upper arm has been reported in two cases (6, 7). Here, we report an extremely rare case of MM arising in the upper arm.

A 73-year-old Japanese man presented with a growing mass in his left upper arm one year prior to surgery. His past history included treatment for malignant lymphoma and bladder carcinoma and has been in remission for several years.

Clinical examination revealed a non-mobile elastic hard mass measuring approximately 10×13 cm. Laboratory data were within normal limits. Plain radiographs and computed tomography (CT) revealed no calcified or ossified lesion and no apparent invasion to humerus. Magnetic resonance images (MRI) revealed a well-defined mass in the medial aspect of the upper arm adjacent to neurovascular bundle. In general, this lesion showed slightly high signal intensity on T1-weighted images (Figure 1A), and high signal characteristics and partial inhomogeneity with small areas of loss of signal intensity on T2-weighted images (Figure 1B). After contrast medium administration, heterogeneous enhancement was observed (Figure 1C). Therefore, our differential diagnosis included malignant soft tissue tumor, recurrence of malignant lymphoma, metastatic carcinomas and others.

We performed needle biopsy to obtain a definite pathological diagnosis. Pathological findings indicated myxoid sarcoma but did not allow definite diagnosis. An excision was performed as a malignant tumor. The profunda brachii artery was ligated, the brachial artery and vein were preserved, and almost the entire triceps brachii muscle was resected. The radial nerve was strongly adherent to the tumor, however, the radial nerve, the median, ulnar, and radial nerves could be preserved (Figure 2, Figure 3).

Histopathological examination revealed oval to short spindle-shaped tumor cells with mild pleomorphism proliferating in a fascicular pattern. In some areas, marked nuclear enlargement and increased nuclear chromatin were observed. Immunohistochemically, the tumor cells were positive for CK (AE1/AE3), vimentin, S-100, CD99, and BCL-2, and negative for CK (CAM5.2), CD34, desmin, myogenin, p53, GFAP, and SOX10. The Ki-67 labeling index was 16.4% (Figure 4, Figure 5). Based on these findings, MM and synovial sarcoma were considered in the differential diagnosis. Fusion gene analysis was performed for further differentiation, but no SS18-SSX fusion gene was detected, ruling out synovial sarcoma. The final diagnosis was MM.

As adjuvant therapy, postoperative intensity-modulated radiation therapy (IMRT) was administered, consisting of a total dose of 70 Gy delivered in 35 fractions. At the current follow-up, four years after surgery, the Enneking functional score of the affected limb is 90%. No local recurrence or metastasis has been observed.

MM primarily arises in the salivary glands; however, cases have also been reported in the nasopharynx, larynx, breast, and lung, and it rarely occurs in soft tissue and the bone (2-5). As in the present case, MM arising in soft tissue is extremely rare, accounting for only 1% of all soft tissue tumors (8). Soft tissue-origin malignant myoepithelioma is a relatively newly recognized tumor that was included in the WHO classification in 2002 and is categorized as a tumor of uncertain differentiation.

Kabarriti et al. summarized 58 cases of soft tissue-origin MM. In their report, the mean age was 28 years (range=3 months-82 years), indicating that the tumor occurs across a wide age range. The primary sites were the head and neck (13 cases), trunk (2 cases), pelvis (12 cases), upper extremities (12 cases), lower extremities (11 cases), and others (eight cases). Approximately 20% occurred in the upper extremities, with only two cases reported in the upper arm (9). Tumors arising in the upper arm are rare even among upper extremity cases, which may have been one reason for the diagnostic difficulty in the present case.

Histopathologically, it demonstrates highly diverse features, including epithelioid, spindle-shaped, plasmacytoid, and clear cell morphologies (10). The stromal characteristics are also variable, with myxoid, chondroid, and hyalinized changes observed (6). A characteristic finding is the sheet-like or cord-like proliferation of tumor cells within these stromal backgrounds (3). Immunohistochemically, the tumor frequently expresses cytokeratin and S-100 protein (6, 10, 11) and may also express EMA and GFAP (5). Genetically, EWSR1 rearrangement has been reported in 45% of cases (12). Regarding SS18-SSX, which was used in this case for differentiation from synovial sarcoma, one report found that all 33 examined cases were negative (3). Diagnosis may be difficult based solely on histopathological and immunohistochemical findings, and genetic testing should therefore be considered.

The mainstay of treatment is surgical resection, and the efficacy of radiotherapy and chemotherapy has not been established (9). Regarding chemotherapy, some reports have described responses to carboplatin and paclitaxel (13), whereas others have reported no response (6), and thus no consensus has been reached. However, platinum-based chemotherapy regimens including doxorubicin and paclitaxel are frequently used in clinical practice (3). Postoperative radiotherapy has been reported to reduce the risk of local recurrence (14); it was administered in the present case, and no local recurrence has been observed during the four years postoperative follow-up. In the report by Kabarriti et al., preoperative radiotherapy resulted in extensive treatment-related changes, with only 10% residual myoepithelial carcinoma identified in the surgical specimens (9).

Success in the management of soft tissue sarcomas is often limited by the extension of lesions to neurovascular structures, because of the difficulty in dissecting the neurovascular bundle from the tumor without compromising the function and local recurrence of residual lesions. Patients with soft tissue sarcoma involving or extending to neurovascular structures may be sometimes advised to undergo an amputation and nerve resection (15). In the present case, tumor bed irradiation, following conservative surgery has been shown to provide excellent local control and limb function. Adjuvant radiation therapy may be appropriate for MMs involving the neurovascular bundle after surgical resection with marginal or negative margins. However, reports regarding preoperative and postoperative radiotherapy remain limited, and these approaches have not yet been established as standard treatment. Further investigation is warranted.

Regarding prognosis, Aiba et al. reported local recurrence in 37.5% of 32 patients who underwent wide resection. Metastasis was observed in 60.0% of patients with localized disease, and the 5-year survival rate was 62.6% in patients without metastasis compared with 12.5% in those with metastasis (3). Although resection and radiation were performed in the present case, the risk of local recurrence may still remain, and careful periodic follow-up is essential. The most common sites of metastasis are the lungs, lymph nodes, bone, and soft tissue (10). Given the high metastatic rate and the markedly reduced 5-year survival rate in patients with metastasis, early diagnosis and early treatment are considered crucial.

We experienced an extremely rare case of MM arising in the upper arm. Surgical resection followed by postoperative radiotherapy was performed, and no recurrence or metastasis has been observed to date. Adjuvant radiation therapy may be appropriate for MMs involving the neurovascular bundle after surgical resection.

The Authors have no financial conflicts of interest to disclose concerning this presentation.

Conception/design: HT, TM, TN, TH, and NA. Provision of study materials and patients: TM, KN, RM, MY, and SA. Manuscript writing: HT, TM, YH, and NA. All Authors have read and approved the manuscript.

During the preparation of this manuscript, a large language model (ChatGPT, OpenAI) was used solely for language editing and stylistic improvements in select paragraphs. No sections involving the generation, analysis, or interpretation of research data were produced by generative AI. All scientific content was created and verified by the authors. Furthermore, no figures or visual data were generated or modified using generative AI or machine learning–based image enhancement tools.