Malignant diffuse-type tenosynovial giant cell tumors - A series of 7 cases comparing with 24 benign lesions with review of the literature

Background: Malignant diffuse-type tenosynovial giant cell tumor (D-TSGCT), an unusual sarcoma with concurrent or previous benign D-TSGCTs, poses challenges to diagnosis and prognostication. Methods: We described the radiologic, clinicopathologic, and immunophenotypical findings of 5 primary and 2 metachronous malignant D-TSGCTs and reviewed published cases to better delineate their morphologic spectrum and behavior. Twenty-four benign D-TSGCTs were also statically compared to analyze the diagnostic values of various variables. Results: The 7 malignant cases affected 4 females and 3 males aged 45 to 78 (mean, 60.9) years, which included 1 intra-articular and 6 extra-articular lesions. These tumors were 5 to 17 cm (mean, 9.4) and located within or near the large joints of extremities. Magnetic resonance imaging revealed expansile or infiltrative masses with frequent lobulation and heterogeneous signals. Histologically, areas of benign D-TSGCTs blended abruptly or gradually with frank sarcomas composed of pleomorphic, spindle, or enlarged oval cells, forming malignant fibrous histiocytomalike (n = 4), fibrosarcomatous (n = 1), myxosarcomatous (n = 1), or giant cell tumorlike (n = 1) patterns. One patient experienced recurrences twice, and another 3 developed metastases to the lymph nodes (n = 2), lung (n = 1), or vertebrae (n = 1), with 1 dying from disseminated diseases. An older age (P = 0.003), a larger size (P = 0.036), tumor necrosis (P < 0.001), atypical mitoses (P<0.001), and Ki-67 overexpression (P<0.001) appeared preferentially in malignant lesions, but these parameters had overlap between few benign and malignant tumors. Conclusions: Malignant D-TSGCTs are a distinct sarcoma with considerable morphologic variability, metastatic propensity, and lethality. Altered architecture with anaplastic cells represents an important distinguishing feature, while abnormalities of other parameters should not be directly equated with malignancy.