Usefulness of cytokeratin subsets for distinguishing monophasic synovial sarcoma from malignant peripheral nerve sheath tumor.

Monophasic synovial sarcoma (MSS) and malignant peripheral nerve sheath tumor (MPNST) are spindle cell sarcomas with overlapping histologic features, and their immunophenotypes may overlap, since MPNSTs express S-100 protein in only 50% to 60% of cases and rarely express epithelial markers, whereas MSSs can express S-100 protein in up to 40% of cases. We immunostained 29 cases of MSS and 29 cases of MPNST with antibodies to AE1/AE3, CAM 5.2, epithelial membrane antigen (EMA), S-100 protein, and cytokeratin subsets 7 and 19. Inclusion criteria for MSS included a consistent histology with expression of at least 1 epithelial marker. Inclusion criteria for MPNST included a tumor with a consistent histology arising in a patient with neurofibromatosis type 1 and/or in a plexiform neurofibroma, or ultrastructural confirmation of clear-cut schwannian differentiation. By definition, all cases of MSS were positive for at least 1 epithelial marker. Ten cases showed focal S-100 protein immunoreactivity, and 26 cases stained for cytokeratins 7 and 19. Twenty-three cases stained for both antigens, whereas only 2 cases were negative for both cytokeratins. Twenty-two MPNSTs demonstrated immunoreactivity for S-100 protein, and 11 stained focally for AE1/AE3 or EMA. Two cases of MPNST stained for cytokeratin 7, and only 1 case stained for cytokeratin 19. No cases of MPNST stained for both cytokeratins. Antibodies to cytokeratins 7 and 19 are useful adjuncts for the separation of MSS from MPNST. The majority of MSSs stain for one or both of these antigens, whereas most MPNSTs, including those that are EMA- or AE1/AE3-positive, do not express these cytokeratin subsets.

Cytokeratin-negative desmoplastic small round cell tumor: a report of two cases emphasizing the utility of reverse transcriptase-polymerase chain reaction.

Desmoplastic small round cell tumor (DSRCT) is a unique, highly aggressive neoplasm that chiefly affects male adolescents and young adults. This tumor is characterized by nests of small undifferentiated cells that show immunohistochemical evidence of epithelial, mesenchymal, and neural differentiation. We report two cases of DSRCT that lacked immunohistochemical evidence of epithelial differentiation, but were found to have the fusion transcripts characteristic of this tumor. Both patients (a 41-year-old male and a 31-year-old female) presented with large intra-abdominal masses. After diagnostic biopsy, both were treated with multi-agent chemotherapy. One patient expired 18 days after diagnosis, and the other is currently alive 28 months later. Histologically, both tumors had the characteristic features of DSRCT and were composed of small round cells with hyperchromatic nuclei and scanty cytoplasm. In one of the cases, perinuclear intracytoplasmic hyaline inclusions were seen. Immunohistochemically, neither case expressed any of the epithelial markers tested, including AE1/AE3, CAM 5.2 and EMA. Both tumors were diffusely immunoreactive for desmin with a prominent globoid "dot-like" pattern of staining in one case. Both tumors stained for vimentin, neuron specific enolase, and synaptophysin, but were negative for CD99, muscle-specific actin, and myogenin. Reverse transcriptase-polymerase chain reaction revealed EWS-WT1 fusion transcripts characteristic of this neoplasm. In conclusion, we describe two cases of DSRCT that lacked immunohistochemical evidence of epithelial differentiation but had histologic and other immunohistochemical features which suggested this diagnosis. The ability to confirm the diagnosis of this rare tumor using molecular genetic techniques is particularly useful in those cases with unusual histologic or immunophenotypic features.

Synovial sarcoma of the extremities: a clinicopathologic study of 34 cases, including semi-quantitative analysis of spindled, epithelial, and poorly differentiated areas.

Many clinicopathologic studies of synovial sarcoma have grouped together tumors from different sites. The goal of this study was to identify clinical and pathologic features that correlate with a poor outcome in patients with extremity synovial sarcoma. Thirty-four cases of synovial sarcoma of the extremities were studied. Inclusion criteria included a consistent histology, the immunohistochemical expression of at least one epithelial marker (AE1/3, CAM 5.2, or epithelial membrane antigen), and adequate clinical follow-up. Features evaluated included the presence and extent of spindled, epithelial, and poorly differentiated areas, the presence and extent of calcification and necrosis, nuclear grade, the presence or absence of cells with a rhabdoid morphology, and the number of mitotic figures (MFs) per 10 high power fields (HPFs). Patients were considered to have an adverse outcome if they developed metastatic disease or died from tumor. The cohort included 15 males and 19 females with a median age 36 years (range, 11-82 years). There were 22 lower extremity tumors and 12 located on the upper extremities. Tumor size ranged from 1.2 to 16 cm (median, 6 cm). Follow-up ranged from 9 to 108 months (median, 38 months). Eleven (32%) patients had an adverse outcome, all with metastatic disease. Features associated with an adverse outcome included increasing age (p = 0.04), tumor size of 5 cm or greater (p = 0.03), tumor location on the lower extremities (p = 0.04), the presence of poorly differentiated areas (p = 0.04), grade 3 nuclei (p = 0.005), cells with a rhabdoid morphology (p = 0.003), and more than 10 MFs/10 HPFs (p = 0.005). Patients whose tumors were composed of at least 20% poorly differentiated areas were significantly more likely to have an adverse outcome (p < 0.001). In conclusion, a variety of clinical and pathologic features are associated with an adverse outcome in patients with synovial sarcoma of the extremities. These features include increasing age, tumor size of 5 cm or more, lower extremity tumor location, the presence of poorly differentiated areas, particularly when at least 20% of the tumor, grade 3 nuclei, rhabdoid cells, and more than 10 MFs/10 HPFs.

Utility of cytokeratin subsets for distinguishing poorly differentiated synovial sarcoma from peripheral primitive neuroectodermal tumour.

AIMS: Poorly differentiated synovial sarcoma (PDSS) is a round cell sarcoma that may be difficult to distinguish from other round cell sarcomas, such as peripheral primitive neuroectodermal tumour (pPNET), on histological examination alone. Furthermore, these two tumours may show overlapping immunophenotypes, as some cases of PDSS express CD99, and, on the other hand, pPNET may express epithelial markers. The goal of this study was to determine the utility of cytokeratin (CK) subsets in distinguishing between these two lesions. METHODS AND RESULTS: We evaluated 13 cases of pPNET with RT-PCR detected EWS/FLI-1 fusion transcript and 21 cases of PDSS for the immunohistochemical expression of CK7 and CK19. All cases of PDSS had areas of recognizable monophasic or biphasic synovial sarcoma that expressed at least one epithelial marker. Thirteen of 21 (62%) PDSS stained with AE1/AE3. CK7 and 19 were expressed in 11 (52%) and nine (43%) cases, respectively. Although six of 19 (32%) PDSS demonstrated cytoplasmic staining for CD99, none showed a membranous pattern of immunoreactivity. In contrast, 12 of 13 (92%) pPNET showed strong membranous immunoreactivity for CD99. Four tumours (31%) showed focal staining with AE1/AE3, two of which (15%) stained for CK19. CK7 was not detected in any of the pPNETs. CONCLUSIONS: Although AE1/AE3 may be found in up to 31% of cases of pPNET, the expression of CK7 makes this diagnosis less likely.

Cyclin D1 and MIB-1 immunohistochemistry in pilocytic astrocytomas: a study of 48 cases.

Pilocytic astrocytoma is an infrequently encountered, generally low-grade neoplasm. No study has extensively looked at both cyclin D1 and MIB-1 labeling indices in pilocytic astrocytoma and their relation to clinical outcome. This study retrospectively examines the clinicopathologic features of 48 patients with pilocytic astrocytoma including MIB-1 (cell proliferation marker) and cyclin D1 (protein that regulates progression from G1 to S phase of the cell cycle) immunohistochemistry. Of 48 patients (27 females and 21 males; mean age, 12.7 years; age range, 2 to 57 years), 26 initially underwent gross total resection; 17, subtotal resection; four, biopsy alone; in one patient, the extent of tumor resection was unknown. Histological features observed included Rosenthal fibers (83.3%), granular bodies (75%), vascular sclerosis (56.2%), vascular proliferation (56.2%), prominent nuclear pleomorphism (14.6%), necrosis (10.4%), and identifiable mitotic figures (2.1%). MIB-1 labeling indices (n=45) (positive staining tumor nuclei per 1,000 nuclei evaluated) ranged from 0 to 3.5% (mean, 0.6%); seven tumors had a labeling index greater than 1.0%. Cyclin D1 labeling indices (n=45) ranged from 0 to 0.8% (mean, 0.1%). Most tumors (N=29, 66.7%) had no immunostaining. At last known follow-up, 27 patients were alive with no evidence of disease (mean, 49.2 months), 17 patients were alive with evidence of disease (mean, 36.8 months), three died with tumor at 2, 22, and 156 months, and one patient was lost to follow-up. Eight patients had at least one tumor recurrence requiring additional surgery; seven of these patients had an initial subtotal resection. In summary, MIB-1 labeling indices were generally low (mean, 0.6%) and are reflective of the slow growth of the tumors. Cyclin D1 immunostaining does not appear to be significantly increased in pilocytic astrocytoma. Adverse outcome in patients with pilocytic astrocytoma may be related to extent of surgical resection and does not seem to correlate with histology, MIB-1 labeling indices, or cyclin D1 immunoreactivity.