CCSST is rare, accounting for approximately 1% of all soft tissue sarcomas. It typically involves extremities, especially tendons and aponeuroses of the foot and ankle. The exact histogenesis is obscure, but the presence of intracellular melanin in two thirds of cases supports an origin from migrated neural crest cells that have the capacity to produce melanin.
The diagnosis of clear cell sarcoma is challenging. The differential diagnosis includes malignant peripheral nerve sheath tumor, monophasic synovial sarcoma, deep-seated epithelioid sarcoma, adult fibrosarcoma, psammomatous melanotic schwannoma, and metastatic malignant melanoma. All these tumors can involve deep soft tissues of extremities and bear similar microscopic morphology as CCSST. Immunohistochemical study plays an important role for the differentiation. CCSST is usually immunopositive for S-100, HMB-45, melan-A, microphthalmia transcription factor, bcl-2, and vimentin, viable positive for synaptophysin, CD56, and EMA, rarely positive for AE1/AE3, and immunonegative for smooth muscle actin (SMA), desmin, and CAM5.2 [5
]. Our case shows focal staining for AE1/AE3, which is seldom reported positive in the published data [5
]. The variable immunoprofile can confuse CCSST with other soft tissue tumors. Malignant peripheral nerve sheath tumor is immunopositive for S-100, but the staining is usually only focal. Moreover, it is negative for HMB-45 and melan-A. Synovial sarcoma is generally positive for cytokeratin, EMA, bcl-2, CD99, and calponin. S-100 protein may be detectable in 30% of synovial sarcomas. It is characterized by the SYT-SSX1 or SYT-SSX2 fusion, detectable by either RTPCR or FISH. Epithelioid sarcoma is immunoreactive for vimentin, low- and high-molecular-weight cytokeratins, and EMA, partial active for CD34, occasional active for SMA and S-100, and frequently negative for INI1. Adult fibrosarcoma is positive for vimentin and negative for S-100. The immunohistochemical appearance of CCSST resembles that of psammomatous melanotic schwannoma. However, psammomatous melanotic schwannoma has widespread psammoma bodies and usually affects spinal nerves, paraspinal ganglia, and autonomic nerves of viscera, which is not typical for CCSST. The most troubling differential is metastatic malignant melanoma, which has a similar deep soft tissue location as well as an identical microscopic morphology and immunohistochemical profile as CCSST. Moreover, metastatic melanoma, similar to CCSST, may not show a primary skin lesion (that may have been removed, sloughed off, or be in a difficult to find location such as within the mucosal membranes). The definitive diagnosis of CCSST relies on detecting its unique translocation by FISH study. More than 90% of CCSST are associated with a distinct t(12;22)(q13;q12) chromosomal translocation, resulting in the formation of a fusion protein, EWS-AFT1, which is absent in metastatic malignant melanoma [4
] and psammomatous melanotic schwannoma.
MR imaging is an important tool to diagnose soft tissue tumors. Because of T1 shortening caused by the paramagnetic effect of intralesional melanin, high signal intensity on T1-weighted images and low signal intensity on T2-weighted images has been the classic description in CCSST [6
]. However, a literature review of MR imaging for CCSST showed highly variable results. In a review of 31 CCSST cases [7
], MR images showed T1-hypointense signal in 33%, isointense signal in 19%, and T1-hyperintense signal in 48% of cases. The T2 characteristics showed T2 hypointensity in 22% of cases, isointense signal in 26%, and T2-hyperintense signal in 52% of these cases. The variable MR signal characteristics for CCSST cannot be explained only by melanin content. Tumors with similar melanin content can have quite different MR signals. In a case reported by Isoda et al. [9
], both T1- and T2-weighted images exhibited hypointense signals, and no intracellular melanin was observed microscopically. Our case also did not show melanin microscopically, but presented with mildly hyperintense signal on T1-weighted images and predominantly hyperintense T2 signal with scattered foci of low signal intensity. The role of MR imaging in the diagnosis of CCSST may therefore be limited.
The prognosis of metastatic malignant melanoma is dismal. Comparatively, the outcome in CCSST, although poor, is more protracted with high local recurrence, distant metastasis, and death from tumor. The survival rate for CCSST at 5, 10, and 20 years was 67, 33, and 10%, respectively, in one study [14
]. Patients with CCSST should therefore have a long follow-up.