Search

The diagnosis of malignant peripheral nerve sheath tumors can be challenging and other spindle cell sarcomas commonly enter in the differential diagnosis. Loss of trimethylation at lysine 27 of histone-H3 (H3K27me3) by immunohistochemistry was recently described in malignant peripheral nerve sheath tumors. However, its specificity remains controversial. We therefore studied 82 synovial sarcomas, 39 malignant peripheral nerve sheath tumors, and 10 fibrosarcomatous dermatofibrosarcoma protuberans for H3K27me3 loss by immunohistochemistry. The diagnoses were based on morphology, immunophenotype, and genetics based on WHO classification. H3K27me3 immunohistochemistry was scored by two pathologists based on fraction of cells with nuclear staining (score 0 to 3+). Loss of H3K27me3 (score 0) was seen in 44% of malignant peripheral nerve sheath tumors and 9% of synovial sarcomas yielding positive and negative predictive values of 71% and 77%, respectively. Loss of H3K27me3 was seen in 10% of fibrosarcomatous dermatofibrosarcoma protuberans, yielding positive and negative predictive values of 94 and 29% in the differential diagnosis of malignant peripheral nerve sheath tumor versus fibrosarcomatous dermatofibrosarcoma protuberans. Partial loss (score 1-2) was common in all three tumor types. Among malignant peripheral nerve sheath tumors, there was no significant association between H3K27me3 loss and gender, tumor site, or size, and progression-free or overall survival. Patients with tumors with H3K27me3 loss were younger than those with tumors with retained H3K27me3 expression (P=0.011). H3K27me3 expression was lost in 50 and 31% of sporadic and Neurofibromatosis-associated malignant peripheral nerve sheath tumors, respectively (P=0.25).Complete H3K27me3 loss is a moderately sensitive and relatively specific marker for the diagnosis of malignant peripheral nerve sheath tumor when the differential diagnosis includes synovial sarcoma and fibrosarcomatous dermatofibrosarcoma protuberans. Partial loss has limited diagnostic utility. H3K27me3 status does not show significant association with clinical outcome in malignant peripheral nerve sheath tumors.

Anastomosing hemangiomas are recently described benign vascular lesions that occur chiefly in the genitourinary tract and paravertebral soft tissues. Owing to their rarity and unusual cytoarchitectural features, anastomosing hemangiomas are frequently confused with low-grade angiosarcomas. The specific genetic alterations underlying these lesions are currently unknown. We performed capture-based next-generation DNA sequencing analysis on 13 anastomosing hemangiomas and identified frequent somatic mutations in the heterotrimeric G-protein alpha-subunit, GNAQ. Nine of 13 cases (69%) harbored a somatic mutation at GNAQ codon 209, a known hotspot that is commonly mutated in uveal melanoma and blue nevi, as well as various congenital vascular proliferations. No other pathogenic or likely pathogenic mutations were identified in these genetically simple lesions. The finding of a recurrent driver mutation in the G-protein signal transduction pathway provides strong evidence that anastomosing hemangiomas are indeed clonal vascular neoplasms.

Cellular schwannoma is an uncommon, but well-recognized, benign peripheral nerve sheath tumor, which can be misdiagnosed as malignant peripheral nerve sheath tumor. To develop consensus diagnostic criteria for cellular schwannoma, we reviewed 115 malignant peripheral nerve sheath tumor and 26 cellular schwannoma cases from two institutions. Clinical data were retrieved from the electronic medical records, and morphologic features, maximal mitotic counts, Ki67 labeling indices, and immunohistochemical profiles (SOX10, SOX2, p75NTR, p16, p53, EGFR, and neurofibromin) were assessed. Several features distinguish cellular schwannoma from malignant peripheral nerve sheath tumor. First, in contrast to patients with malignant peripheral nerve sheath tumor, no metastases or disease-specific deaths were found in patients with cellular schwannoma. More specifically, 5-year progression-free survival rates were 100 and 18%, and 5-year disease-specific survival rates were 100 and 32% for cellular schwannoma and malignant peripheral nerve sheath tumor, respectively. Second, the presence of Schwannian whorls, a peritumoral capsule, subcapsular lymphocytes, macrophage-rich infiltrates, and the absence of fascicles favored the diagnosis of cellular schwannoma, while the presence of perivascular hypercellularity, tumor herniation into vascular lumens, and necrosis favor malignant peripheral nerve sheath tumor. Third, complete loss of SOX10, neurofibromin or p16 expression, or the presence of EGFR immunoreactivity was specific for malignant peripheral nerve sheath tumor (P<0.001 for each). Expression of p75NTR was observed in 80% of malignant peripheral nerve sheath tumors compared with 31% of cellular schwannomas (P<0.001). Fourth, Ki-67 labeling indices ≥20% were highly predictive of malignant peripheral nerve sheath tumor (87% sensitivity and 96% specificity). Taken together, the combinations of these histopathological and immunohistochemical features provide useful criteria to distinguish between malignant peripheral nerve sheath tumor and cellular schwannoma with high sensitivity and specificity. Additional retrospective and prospective multicenter studies with larger data sets will be required to validate these findings.

Synovial sarcoma and malignant peripheral nerve sheath tumor pose a significant diagnostic challenge given similar histomorphology. The distinction is further complicated by similar immunophenotype and especially by occasional synovial sarcomas that present as intraneural tumors. Although the presence of a t(X;18) rearrangement or expression of TLE1 can help confirm the diagnosis of synovial sarcoma, negative results for these tests are not diagnostic of malignant peripheral nerve sheath tumor. The SOX10 transcription factor, a putative marker of neural crest differentiation, may have diagnostic utility in this differential, but immunohistochemical data are limited. The goal of the present study was to determine the diagnostic utility of SOX10 to discriminate between synovial sarcoma and malignant peripheral nerve sheath tumor. Forty-eight cases of malignant peripheral nerve sheath tumor, all from patients with documented neurofibromatosis, and 97 cases of genetically confirmed synovial sarcoma, including 4 intraneural synovial sarcomas, were immunohistochemically stained for SOX10. The stain was scored for intensity and fraction of cells staining. Thirty-two of 48 malignant peripheral nerve sheath tumors (67%) were SOX10-positive. The majority of malignant peripheral nerve sheath tumors showed ≥2+ staining, but staining did not correlate with grade. By contrast, only 7/97 (7%) synovial sarcomas were SOX10-positive. Only three synovial sarcomas showed ≥2+ staining but, importantly, two of these were intraneural synovial sarcoma. Therefore, SOX10 is a specific (93%), albeit not very sensitive (67%), diagnostic marker to support a diagnosis of malignant peripheral nerve sheath tumor over synovial sarcoma. Furthermore, the stain needs to be interpreted with caution in intraneural tumors in order to avoid a potential diagnostic pitfall. It remains to be determined whether SOX10-positive cells in intraneural synovial sarcoma represent entrapped Schwann cells, synovial sarcoma cells or both.

The molecules Indian hedgehog (IHH), SP7 (also known as osterix), sex-determining region Y-box 9 (SOX9), runt-related transcription factor 2 (RUNX2) and TWIST1 regulate the normal differentiation of osteo- and chondrogenic cells from precursors during skeletal development and remodeling. The aberrant function of the same molecules has been implicated in the pathogenesis of bone tumors. Preliminary studies suggest that antibodies against these molecules have practical, diagnostic or prognostic utility in tumors. However, a comprehensive analysis of the expression of these molecules in a large, diverse set of bone tumors has yet to be reported. The goals of this study were to compare the immunohistochemical profiles of IHH, SP7, SOX9, RUNX2 and TWIST1 among bone tumors and to determine the optimum panel for diagnostic utility. Tissue microarrays prepared from 206 undecalcified tumors (71 osteosarcomas, 26 osteoblastomas/osteoid osteomas, 50 giant cell tumors, 5 chondromyxoid fibromas and 54 chondroblastomas) were stained with antibodies to IHH, SP7, SOX9, RUNX2 and TWIST1. The stains were scored for intensity (0-3+) and distribution. The results were analyzed by cluster analysis. Optimum antibody panels for diagnostic sensitivity and specificity were calculated. Analysis revealed six main clusters that corresponded well to tumor types and suggested a close relationship between the stromal cells of giant cell tumor and the osteoblasts of osteosarcoma. The expression profile of chondromyxoid fibroma and chondroblastoma also suggested related differentiation. The distribution of osteoblastomas and osteoid osteomas was more heterogeneous. RUNX2, SOX9 and TWIST1 represented the most sensitive and specific immunohistochemical panel to distinguish among these diagnoses with the limitation that no result could discriminate between chondroblastoma and chondromyxoid fibroma. IHH and SP7 did not yield additional utility.

Epithelioid sarcoma is a rare, malignant, soft tissue neoplasm that can be classified into proximal, distal and fibroma-like subtypes. Regardless of subtype, epithelioid sarcoma often shows morphologic and immunophenotypic evidence of epithelial differentiation. Current therapeutic strategies include surgical resection, amputation, radiation or chemotherapy, although the overall prognosis remains poor. The epidermal growth factor receptor (EGFR) is a novel therapeutic target in carcinomas. In some carcinomas, EGFR kinase domain mutations or gene amplification may correlate with response to specific inhibitors. EGFR expression has been reported in some sarcoma types, but expression, amplification and mutations have not been studied in epithelioid sarcoma. We evaluated 15 cases of epithelioid sarcoma from 14 patients for EGFR expression using immunohistochemistry, EGFR copy number aberration using fluorescence in situ hybridization and screened for mutations in the tyrosine kinase domain of the EGFR gene using direct sequencing. In all, 14 of the 15 epithelioid sarcomas (93%) showed expression of EGFR by immunohistochemistry. A majority of the cases (n=11, 73%) showed strong (2+ to 3+) and homogeneous (>75% of cells) membrane staining. No amplification or polysomy of the EGFR gene or mutations of the tyrosine kinase domain of EGFR (exons 18-21) were detected. These results imply that although EGFR is expressed in most epithelioid sarcomas regardless of subtype, gene amplification and activating mutations in the tyrosine kinase domain appear to be rare or absent. Thus, the benefit of targeted therapy against EGFR in patients with epithelioid sarcoma remains to be determined.

Liposarcoma represents a unique model insofar as some well-differentiated liposarcomas progress to non-lipogenic, so-called 'dedifferentiated,' forms. The well-differentiated and dedifferentiated family of liposarcomas demonstrates amplification of the chromosome subregion 12q13-q15 with resultant amplification of the MDM2 and CDK4 genes. However, the specific genetic changes that distinguish between well-differentiated and dedifferentiated liposarcomas are less well understood. To study the genetic changes in dedifferentiated liposarcomas, paired well-differentiated and dedifferentiated components of 29 tumors were analyzed separately by array-based comparative genomic hybridization. A bacterial artificial chromosome array at approximately 1-Mb resolution was used. The genetic changes were compared with clinical presentation, grade of the dedifferentiated component and overexpression of MDM2 and CDK4. Most tumors (n=21, 72%) were retroperitoneal, with both components present at initial diagnosis (n=25, 86%). Eight tumors (28%) were classified as low-grade dedifferentiation. In four cases (14%), a well-differentiated liposarcoma preceded the presentation of the dedifferentiated tumor by 1-5 years. 12q13-q15 was amplified in all tumors. Using unsupervised hierarchical clustering of copy-number changes, all but two tumors showed close similarities between well-differentiated and dedifferentiated components, and segregated as pairs. Dedifferentiated components had more total amplifications (P=0.008) and a trend for gain at 19q13.2, but no genetic changes were significant in distinguishing between the two components. High-level amplifications of 1p21-32 (n=7, 24%), 1q21-23 (n=9, 31%), 6q23-24 (n=6, 21%) and 12q24 (n=3, 10%) were common, but none significantly correlated with differentiation. Presentation and grade correlated with the frequency of changes at a number of genetic loci (P<0.001), whereas CDK4 immunostaining showed negative correlation with 12q13.13 amplification. The genotypic similarity, at the limit of the array's resolution, between components implies that most genetic changes precede phenotypic 'progression,' early in tumorigenesis. The relationship between genetic changes and presentation or grade may reflect differences in factors that control genomic instability or the background genotype of the tumor.

Dedifferentiated liposarcoma can be readily diagnosed by the juxtaposition of a well-differentiated liposarcoma to a nonlipogenic sarcoma. However, if the lipogenic component is not abundant due to surgical sampling or small biopsy, dedifferentiated liposarcoma can be difficult to distinguish from other poorly different sarcomas. Peroxisome proliferator-activated receptor gamma (PPAR-gamma) is a nuclear hormone receptor that plays a critical role in adipocyte differentiation. Prior studies have not only demonstrated PPAR-gamma mRNA in various subtypes of liposarcoma but have also shown that adipocyte differentiation can be induced in some liposarcomas by a PPAR-gamma agonist. In the present study, we investigated whether immunostaining for PPAR-gamma can be used to distinguish dedifferentiated liposarcoma from other retroperitoneal sarcomas. We examined a series of 40 dedifferentiated liposarcoma and compared the staining for PPAR-gamma to a series of 24 retroperitoneal sarcomas that lacked lipogenic differentiation. A monoclonal antibody against PPAR-gamma was used to stain formalin-fixed paraffin-embedded tissue. Specific nuclear immunostaining was present in 37/40 (93%) of the dedifferentiated liposarcoma and 6/24 (25%) of the other sarcomas (two leiomyosarcomas and four undifferentiated sarcomas). Interestingly, immunostaining for CDK4 and/or MDM2 was identified in three of the four PPAR-gamma-positive undifferentiated sarcomas, raising the possibility that these may represent dedifferentiated liposarcoma. This is the first study demonstrating the utility of PPAR-gamma immunohistochemistry in the diagnosis of dedifferentiated liposarcoma in tissue sections. Although not completely specific, the presence of PPAR-gamma staining, in combination with histologic findings and other markers, can aid in the diagnosis of dedifferentiated liposarcoma, particularly on small biopsies that may not sample the well-differentiated component.

Giant-cell tumor of bone is considered a benign, locally aggressive and rarely metastasizing neoplasm of bone. Specific microscopic or radiographic findings that reliably predict behavior have remained elusive. However, recent evidence suggests that activity of the telomerase enzyme complex correlates with recurrence in giant-cell tumor, although the subset of cells with telomerase activity in these heterogeneous tumors has not been defined. In the present study, we investigated whether immunostaining for human telomerase reverse transcriptase, a component of the telomerase complex, correlates with outcome in giant-cell tumor and the distribution of telomerase reverse transcriptase staining in these tumors. We analyzed 58 cases of giant-cell tumor for the presence and pattern of telomerase reverse transcriptase immunostaining, presence of soft tissue involvement and the type of initial surgery, and correlated these findings with recurrence-free survival and metastasis-free survival. Specific staining with telomerase reverse transcriptase was present in 20 out of 58 tumors (35%) in the nuclei of mononuclear cells and, occasionally, osteoclast-like giant cells. Furthermore, positive telomerase reverse transcriptase immunohistochemistry correlated with recurrence-free survival (P=0.02), whereas the presence of soft tissue extension (P=0.3) and the type of initial surgery (P=0.2) did not. Only soft-tissue extension significantly correlated with metastasis-free survival (P=0.003). Therefore, telomerase reverse transcriptase expression may predict recurrence in giant-cell tumor insofar as positive immunostaining correlates with shorter recurrence-free survival and may be a useful prognostic marker to stratify patients to more aggressive treatment protocols.

Clear cell sarcoma of soft parts is a rare soft tissue malignancy that shows phenotypic overlap with cutaneous melanoma but can be distinguished by the presence of a t(12;22) translocation. Microsatellite instability (MSI), a variation in the lengths of short repeat DNA segments in the genome, has been implicated in melanoma tumorigenesis, but is rare or absent in clear cell sarcoma. Defects in the mismatch repair (MMR) enzyme complex correlate with MSI in some tumor types, allowing the use of immunohistochemistry for the MMR proteins hMLH1 and hMSH2 to predict the presence of MSI. To determine if the association between MMR defects and MSI extends to clear cell sarcoma, we compared a group of nine clear cell sarcomas to 11 metastatic melanomas on the basis of MSI and the expression of MMR proteins. MSI was studied using fluorescence-based multiplexed PCR of five loci. Immunohistochemistry was evaluated on formalin-fixed paraffin-embedded tissue for hMLH1, hMHS2 and hMSH6. MSI was present in only 1/9 (11%) clear cell sarcoma case and in 8/11 (73%) melanoma cases. Immunostaining for hMLH1 and hMSH2 was preserved in all the clear cell sarcomas but loss of immunostaining for one or both proteins was seen in 6/11 melanomas (55%). hMSH6 was detected in 7/9 (78%) clear cell sarcomas and 10/11 (91%) of melanomas. Clear cell sarcoma and metastatic melanoma differed significantly with respect to the presence of MSI (P=0.010) and staining for hMLH1 and/or hMSH2 (P=0.014) but not hMSH6 (P=0.57). Mismatch repair, and consequently genomic instability may contribute to tumorigenesis in melanoma but not clear cell sarcoma. Immunostaining for hMLH1 and hMSH2 and MSI analysis may be helpful in the differential diagnosis of large soft tissue or visceral malignancies with melanocytic differentiation.

Overexpression of KIT protein (CD117), the product of the c-kit gene, has been shown to have important prognostic and therapeutic implications for a number of malignant neoplasms. Previous studies have shown conflicting results regarding the expression of c-kit in malignant mesothelioma. To determine whether malignant mesothelioma expresses KIT, immunohistochemistry and RT-PCR were used to analyze archived tissue from 37 cases of mesothelioma. Although a subset of mesotheliomas demonstrated specific staining with the DAKO anti-KIT antibody, in each case staining was nuclear. We could not detect c-kit mRNA by a sensitive RT-PCR assay, even in cases with strong nuclear staining. Furthermore, a second anti-KIT antibody (Cell-Marque) only demonstrated staining in a single mesothelioma case and in none of the cases that demonstrated nuclear staining. We conclude that immunoreactivity for KIT in mesothelioma does not represent expression of the c-kit gene and may represent antibody cross-reaction with nuclear proteins. Our results raise doubt about previously reported expression of KIT in mesothelioma and consequently, the applicability of therapeutic agents that target the kinase activity of KIT.