[Pleomorphic and dedifferentiated leiomyosarcoma: a clinicopathologic analysis].

Objective: To investigate the clinicopathologic features, differential diagnosis and biological behavior of pleomorphic leiomyosarcoma (PLMS) and dedifferentiated leiomyosarcoma (DLMS). Methods: Forty-nine cases were collected from November 2007 to December 2016, including eight that diagnosed at Fudan University Shanghai Cancer Center, and 41 consultation cases. The clinical findings and pathologic features were reviewed. Immunophenotype was obtained in 33 cases and follow-up information was available in 38 cases. Results: There were 22 males and 27 females with ages ranging from 24 to 83 years (mean 52.5 years). Fifteen cases occurred in extremities, 14 in deep body cavity, 11 in the trunk, 4 in the head and neck, 2 in the bladder, and 1 each in the inguinal region, perineum and femoral vein, respectively. Tumor sizes ranged from 3 to 30 cm (mean 9.1 cm). The tumors were composed of at least small foci of typical leiomyosarcoma (LMS) and areas of high-grade pleomorphic/undifferentiated sarcoma. The typical LMS component showed the characteristic morphology of smooth muscle differentiation and was low to intermediate grade in most cases. Pleomorphic areas were mainly composed of atypical spindle and polygonal cells admixed with variable large, bizarre atypical cells and multinuclear giant cells, mostly mimicking undifferentiated pleomorphic sarcoma. The pleomorphic and leiomyosarcomatous areas were usually intermixed, but the demarcation may be distinct or gradual in some cases. The classical LMS component was positive for at least one myogenic marker: α-SMA in 97.0%(32/33), desmin in 72.7%(24/33), H-caldesmon in 90.9% (20/22), MSA in 14/16, and calponin in 15/15 of cases. The pleomorphic sarcoma component was reactive for at least one myogenic marker in 87.9% (29/33) of cases, usually showing focal and less intense immunoreactivity than classical LMS component: α-SMA was positive in 81.8%(27/33), desmin in 48.5%(16/33), H-caldesmon in 72.7% (16/22), MSA in 12/16, and calponin in 11/15 of cases. Based on staining for muscle markers in the pleomorphic component, 29 cases were designated as PLMS, 4 as DLMS. Ki-67 index ranged from 15% to 70% (mean 40%). Follow-up data was available in 38 cases (77.6%), of which 11 patients (28.9%) died of disease, 12 patients were alive with unresectable or recurrent disease, 14 patients were alive with no evidence of disease and another one died of unrelated cause. The median disease-free and overall survival was 6 and 10 months respectively. Twelve patients exhibited local recurrence and 11 developed metastases. The median interval to progression was 8 months. Conclusions: The identification of areas of typical LMS is crucial for accurate diagnosis of PLMS and DLMS. Both PLMS and DLMS show more aggressive behavior and poorer prognosis than ordinary LMS.