A rare case of mixed tumor of the elbow mimicking sarcoma in radiology: a diagnostic challenge

Abstract

The occurrence of mixed tumors in soft tissues or myoepithelial tumors is extremely rare. Herein, we report a case of a mixed tumor suspected to be a sarcoma based on imaging findings. A 64-year-old male was admitted to our hospital with a mass that had been present for 10 years and had gradually increased in size, prompting medical attention. Magnetic resonance imaging raised the suspicion of sarcoma, leading to a pathological diagnosis of myoepithelioma. Surgery was performed, and histopathological examination confirmed the presence of both epithelial and myoepithelial structures, resulting in a final diagnosis of a mixed tumor. Three years after the surgery, no recurrence was observed. In this case, considering the size and duration of the illness, a benign subcutaneous tumor was suspected. However, the heterogeneity in imaging findings raised concerns about sarcoma, making differential diagnosis challenging. Although the imaging characteristics of mixed tumors are currently unclear, their signals and internal characteristics may be heterogeneous. Therefore, clinical findings and imaging should not mislead the diagnosis, and histopathological examination remains a shortcut for diagnosis and treatment.

Introduction

Myoepithelial tumors are neoplasms with myoepithelial differentiation and are classified as myoepitheliomas, myoepithelial carcinomas, or mixed tumors. A myoepithelioma containing ductal structures is considered a mixed tumor [1]. Myoepithelial tumors are often found in the salivary glands, and their occurrence in soft tissues is rare [1]. Mixed tumors of soft tissues are extremely rare [2]. The clinical features of myoepithelial tumors are similar to those of benign tumors [1, 3]. Magnetic resonance imaging (MRI) is useful for diagnosing soft tissue tumors; however, analyses involving a large number of cases of soft-tissue-origin myoepithelial tumors are lacking, and their characteristics remain unclear [2, 4, 5]. Histologically, these tumors exhibit a morphological spectrum encompassing architectural and cytological heterogeneity, potentially presenting diverse MRI findings [1]. Based on these findings, a diagnosis based solely on imaging is challenging, making a pathological histological diagnosis essential. Moreover, the presence of myoepithelial carcinoma highlights the importance of distinguishing it from benign tumors, owing to the differences in recurrence and metastasis rates [3]. The expression of fusion genes, including EWSR1, is useful for diagnosis [6]. Herein, we report a case of a mixed tumor that occurred in the elbow and posed diagnostic challenges because it was clinically benign, but MRI suggested a soft tissue sarcoma.

Case presentation

The patient was a 64-year-old male with a mass that had been present distal to his right elbow for 10 years. Owing to its gradual enlargement, the patient visited our hospital. The mass located dorsally distal to the right elbow was elastic-hard in consistency and showed no swelling, redness, or tenderness. The range of motion of the forearm and elbow joints was normal, with no Tinel’s sign. His medical history included a diagnosis of prostatic hyperplasia five years before, for which he had been receiving oral treatment. Blood test results were within the normal range. Plain radiography (Fig. 1) revealed a tumor shadow (Fig. 1 [yellow asterisk]), but no calcification or bone abnormalities were identified. MRI revealed the subcutaneous location of the tumor, showing an isointense signal to the muscle with areas of high intensity on T1-weighted images (T1WI) (Fig. 2a, e) and a faintly hyper-intense signal compared to the muscle on T2-weighted images (T2WI) (Fig. 2b, f). Although fat suppression was insufficient, no apparent fat within the tumor was observed, and high-signal areas were observed in fat-suppressed T1WI (Fig. 2c, g [red arrow]). On fat-suppressed contrast-enhanced T1WI, the tumor exhibited heterogeneous enhancement (Fig. 2d, h). Based on the MRI findings, the tumor was considered a solid subcutaneous tumor containing hemorrhagic or myxoid components (showing high-signal intensity on fat-suppressed T1WI and T2WI) and cystic or cartilaginous elements (showing low signal intensity on fat-suppressed T1WI and high-signal intensity on T2WI). The solid portion was primarily composed of fibrous tissues. Differential diagnoses included fibrosarcoma, fibromyxoid sarcoma, malignant peripheral nerve sheath tumor (MPNST), and clear cell sarcoma. For an accurate diagnosis, an incisional biopsy was performed, revealing a suspected myoepithelioma. The surgery involved a marginal excision. Macroscopically, the tumor exhibited white areas along with some dark red regions measuring 26 mm × 22 mm × 15 mm in size (Fig. 3a). Histopathological findings revealed that the epithelial cells formed ductal structures that transitioned into nests. The solid structures were predominantly composed of myoepithelial cells (Fig. 3b, c). Myxoid, cartilaginous (Fig. 3d), and osseous lesions (Fig. 3e) were observed, with associated fibrosis and hyalinization. Immunohistochemistry (IHC) revealed strong positivity for cytokeratin AE1/AE3 (Fig. 3f) in the epithelial components, whereas the myoepithelial components exhibited weak-to-moderate positivity. The myoepithelial components were positive for calponin (Fig. 3g). S-100 was mosaic-like positive, showing positivity in the mesenchymal-differentiated areas, which supported cartilaginous differentiation (Fig. 3h). p63 is positive (Fig. 3i). Ki67 (MIB-1) labeling index was approximately 10% (Fig. 3j). FISH revealed no split signals for EWSR1 (Fig. 4a) or FUS (Fig. 4b). Based on these findings, the diagnosis was confirmed as a mixed tumor. Three years post-surgery, there was no evidence of recurrence.

Fig. 1.

Plain radiography. Tumor silhouette (yellow asterisk). No soft tissue calcifications or abnormalities in the surrounding bone were observed

Fig. 2.

Magnetic resonance imaging. Sagittal (a–d) and Axial (e–h) views show a subcutaneous tumor. (a, e) T1-weighted images: Tumor appears isointense to muscle with areas of high signal intensity. (b, f) T2-weighted images: Tumor shows a faintly hyperintense signal compared to muscle. (c, g) Fat-suppressed T1-weighted images: High-signal areas are observed within the tumor (red arrows). Note the insufficient fat suppression. (d, h) Fat-suppressed contrast-enhanced T1-weighted images: Tumor demonstrates heterogeneous enhancement

Fig. 3.

Pathological findings. (a) Macroscopic findings of the excised tumor, showing white areas and dark red regions (measuring 26 × 22 × 15 mm). (b–e) Histological findings (Hematoxylin and Eosin, H&E stain): (b, c) Predominant myoepithelial cells forming solid structures, with epithelial cells forming ductal structures. (d) Cartilaginous differentiation. (e) Osseous differentiation. (f-i) Immunohistochemistry (IHC): (f) Cytokeratin AE1/AE3 staining (epithelial components strongly positive). (g) Calponin staining (myoepithelial components positive). (h) S-100 staining (mosaic-like positivity in mesenchymal-differentiated areas, supporting cartilaginous differentiation). (i) p63 staining. (j) Ki67 (MIB-1) labeling index (approximately 10%). Magnifications: (b, d–j) ×200, (c) ×400

Fig. 4.

Fluorescence in situ hybridization analysis. (a) EWSR1 and (b) FUS showing no split signals

Discussion

Myoepitheliomas, myoepithelial carcinomas, and mixed tumors are categorized under a single group of myoepithelial tumors. They share morphological, immunophenotypic, and genetic characteristics [1, 7]. Myoepithelioma and mixed tumors arising from soft tissues were first reported in 1997 [8], and Hornick et al. reported 61 benign and 40 malignant cases [3]. Clinically, myoepithelial tumors are painless and characterized by slow growth, with a predilection for the extremities and proximal regions. Subcutaneous occurrence is slightly more common than deep-seated growth [1–3]. In addition to the subcutaneous tissue, the sites of occurrence include the skin, bones, and internal organs. There is no sex predilection, and the age range for occurrence is broad, peaking at 20–40 s. Additionally, approximately 20% of the cases have been reported in children [7]. The tumor size ranges from 0.7 to 20 cm, with an average of 4.7 cm, demonstrating significant variability [3]. In our case, the tumor had been present for 10 years, with gradual enlargement, suggestive of a benign tumor. However, sarcomas with indolent courses resembling benign tumors have also been reported [9]. Thus, imaging and pathological examinations are essential, without relying solely on clinical findings, to determine whether a tumor is benign.

Regarding imaging findings, calcified areas, which are rarely observed in pathological findings, may be detectable on radiography; however, they were not detected in this case. Overall, it is highly likely that there were no characteristic findings. Regarding MRI findings, there are no reports of analyses of multiple cases of myoepithelial tumors, including mixed tumors originating from the soft tissue. Therefore, it is beneficial to examine the imaging characteristics of the salivary glands. For myoepithelioma, MRI findings typically show low intensity on T1 and slightly higher signal intensity than muscle on T2, with a homogeneous internal structure. Contrast enhancement is observed, although it varies to different degrees [4, 5]. In contrast, mixed tumors exhibit intensity patterns similar to those of myoepitheliomas but display a heterogeneous internal structure, consistent with our case [2]. A potential reason for this is the coexistence of various tissues, including the myxoid matrix, osseous and cartilaginous metaplasia, and glandular structures. Therefore, the differential diagnosis in this case was more toward sarcoma rather than benign lesions, such as fibrosarcoma, fibromyxoid sarcoma (Evans tumor), MPNST, and clear cell sarcoma. On further examination of fibrosarcoma, MRI signals often show low intensity on T1 but a heterogeneous mixture of high and low signals on T2, suggesting that the intensity may be similar to that in our case [10]. Other characteristics of fibrosarcomas include a lobulated shape, well-defined margins, the presence of fibrous septa, and deep-origin tumors. The average size is 6.7 cm, but cases as small as 1.2 cm have been documented, warranting caution [10]. Furthermore, it is challenging to differentiate it from low-grade myxoid fibrosarcoma, also known as Evans’ tumor. The MRI characteristics of an Evans tumor range from iso- to hypo-intense on T1, and from iso- to hyper-intense on T2, showing mixed signal intensities. These features are similar to those observed in our case [11]. Other findings of Evans tumors include deep-origin tumors, minimal calcification, and variable enhancement effects [11]; however, these are common features observed in other sarcomas and are not distinctive. The average size is 6.2 cm, with a range of 1.4–19 cm, which is quite broad [11]. Lastly, MPNST presents heterogeneous signals at T1 and T2, similar to fibrosarcoma, Evans tumor, and mixed tumors, as in our case [12]. Thus, differentiating myoepithelial tumors using imaging alone is extremely challenging, and a histopathological diagnosis is essential.

Histopathological examination is critical because of the difficulty in differentiating between the clinical and imaging findings. Myoepithelial tumors exhibit diverse histological features, with tumor cells often exhibiting a mixture of spindle and epithelioid cells in varying proportions. Conversely, clear plasmacytoid-like cells are uncommon [7]. Tumor cells proliferate in reticular and trabecular patterns. The stroma is diverse and includes myxoid, collagenous, and hyaline-like components [7]. Osseous and cartilaginous differentiation is relatively rare, occurring in approximately 10–15% of cases, whereas squamous and adipocytic metaplasia are rarer [1]. Determining the benignity or malignancy is particularly important for the diagnosis of myoepithelial tumors. Myoepitheliomas inherently exhibit moderate nuclear atypia, whereas myoepithelial carcinomas present with severe nuclear atypia, high mitotic counts, and necrosis [3]. While the basic histological features are consistent between cases arising from the salivary glands and those from soft tissue, in myoepitheliomas with minimal atypia, diagnostic criteria exist to distinguish benign from malignant cases in salivary gland; however, such criteria are absent for those arising from soft tissue [3]. In soft tissue myoepithelioma, severe nuclear atypia and vascular invasion are the reported hallmarks of malignant potential, and Ki67 is useful for differentiating between benign and malignant diagnoses [13]. In immunohistochemistry (IHC), AE1/AE3, EMA, S-100, and calponin reportedly show high positivity rates, with these rates in soft tissues mirroring those observed in the salivary gland [13]; however, in recent years, their molecular characteristics have been elucidated [1]. In myoepithelial tumors, EWSR1 gene rearrangements have been observed in approximately 38–45% of cases, with POU5F1 and PBX1 being the most common fusion partners, while ZNF444 and KLF17 are less common [6, 13]. Additionally, cases without EWSR1 rearrangements are often characterized by a subcutaneous origin, benign nature, and glandular differentiation, as observed in the present case [6]. Additionally, EWSR1 rearrangements are absent in salivary gland myoepithelial carcinomas [6]. Additionally, rearrangements in genes other than EWSR1, such as FUS and PLAG1, have been reported. PLAG1 rearrangements are observed in approximately 15% of cases, with a higher frequency in mixed tumors [14]. Finally, in terms of pathological differential diagnoses, extraskeletal myxoid chondrosarcoma (EMC) and synovial sarcoma (SS) were considered, and their key points are briefly discussed [7]. EMCs are often negative for cytokeratin and positive for S100 in IHC [7], whereas SS is positive for SSX, SS18, and TLE-1, which is useful for distinguishing them from myoepithelial tumors [7, 15]. Ewing sarcoma (ES) should be considered from the perspective of EWSR1 rearrangement. Although ES commonly originates in the bone, it can also occur in soft tissues, necessitating caution [16]. Important differential characteristics of Ewing’s sarcoma include histological features of a predominantly epithelioid or rhabdoid appearance, characterized by abundant eosinophilic or clear cytoplasm and eccentric nuclei, CD99 positivity, and fusion genes, such as EWSR1-FLI1 and EWSR1-ERG [7]. Moreover, distinguishing between adenoid cystic carcinoma and eccrine spiradenomas can sometimes be challenging [17, 18].

Approximately 18% of benign tumors show recurrence, but no metastases have been reported. In contrast, malignant cases show recurrence in 40–50% and metastasis in 32% of cases, with reported metastatic sites including the lungs and lymph nodes [1, 3]. Malignant cases exhibit significantly higher frequencies of both recurrence and metastasis than benign cases, underscoring the importance of differentiating between benign and malignant forms [3]. Additionally, pediatric cases disproportionately involve malignant occurrences and require caution during diagnosis and follow-up [19]. The principal treatment is surgery; however, there is no correlation between margin status and recurrence rates. While chemotherapy and radiotherapy have been employed for malignant cases, radiotherapy has shown potential effectiveness, whereas chemotherapy appears to have limited efficacy [3, 20]. Increasing the number of case studies in the future is highly desirable to improve the treatment approaches. As this is a single case report, the findings may not be generalizable to all mixed tumors of soft tissue.

Conclusions

We present a challenging case of a mixed tumor identified through imaging. Although the imaging suggested an indolent sarcoma, histopathological evaluation was essential. Malignant characteristics in myoepithelial tumors impact prognosis, necessitating a thorough assessment of both imaging and pathology to prevent misdiagnosis.

Author contributions

Conception and design: JI, TK, KO, MW, TH, SO, TH and HH. Provision of study materials or patients: JI, MW, TH, SO, and HH. Data Collection and Analysis: JI, TK, and KO. Manuscript writing: JI, TK, KO, MW, TH, SO, TH and HH. All authors have read and approved themanuscript.

Funding

None.

Data availability

The datasets generated and/or analyzed in the current study are available from the corresponding author upon reasonable request.

Declarations

Ethics approval and consent to participate

This study was conducted after obtaining informed consent in accordance with the Declaration of Helsinki and was approved by the Ethics Committees of University of Yamanashi (CS0073). Written informed consent was obtained from the patient for participation in this study.

Consent for publication

Written informed consent was obtained from the patient for publication of this paper.

Competing interests

The authors declare no competing interests.