BackgroundEnhancer of zeste homologue 2 (EZH2) is a polycomb group (PcG) family protein. Acting as a histone methyltransferase it plays crucial roles in maintaining epigenetic stem cell signature, while its deregulation leads to tumor development. EZH2 overexpression is commonly associated with poor prognosis in a variety of tumor types including carcinomas, lymphomas and soft tissue sarcomas. However, although the synovial sarcoma fusion proteins SYT-SSX1/2/4 are known to interact with PcG members, the diagnostic and prognostic significance of EZH2 expression in synovial sarcoma has not yet been investigated. Also, literature data are equivocal on the correlation between EZH2 expression and the abundance of trimethylated histone 3 lysine 27 (H3K27me3) motifs in tumors.MethodsImmunohistochemical stains of EZH2, H3K27me3, and Ki-67 were performed on tissue microarrays containing cores from 6 poorly differentiated, 39 monophasic and 10 biphasic synovial sarcomas, and evaluated by pre-established scoring criteria. Results of the three immunostainings were compared, and differences were sought between the histological subtypes as well as patient groups defined by gender, age, tumor location, the presence of distant metastasis, and the type of fusion gene. The relationship between EZH2 expression and survival was plotted on a Kaplan-Meier curve.ResultsHigh expression of EZH2 mRNA and protein was specifically detected in the poorly differentiated subtype. EZH2 scores were found to correlate with those of Ki-67 and H3K27me3. Cases with high EZH2 score were characterized by larger tumor size (≥ 5cm), distant metastasis, and poor prognosis. Even in the monophasic and biphasic subtypes, higher expression of EZH2 was associated with higher proliferation rate, larger tumor size, and the risk of developing distant metastasis. In these histological groups, EZH2 was superior to Ki-67 in predicting metastatic disease.ConclusionsHigh expression of EZH2 helps to distinguish poorly differentiated synovial sarcoma from the monophasic and biphasic subtypes, and it is associated with unfavorable clinical outcome. Importantly, high EZH2 expression is predictive of developing distant metastasis even in the better-differentiated subtypes. EZH2 overexpression in synovial sarcoma is correlated with high H3K27 trimethylation. Thus, along with other epigenetic regulators, EZH2 may be a future therapeutic target.

About 10% of epithelioid sarcomas have biallelic mutation of the SMARCB1 (SWI/SNF-related, matrixassociated, actin-dependent regulator of chromatin, subfamily b, member 1) gene resulting in a lack of this nuclear protein. It has been suggested that SMARCB1 may be silenced by epigenetic changes in the remaining 90% of tumors. Thus, we hypothesized that the promoter of SMARCB1 is hypermethylated. We also examined SMARCB1 mRNA level to determine if a post-translational change was possible. Thirty-six cases of epithelioid sarcomas were studied. Immunohistochemistry and mutation analysis of the SMARCB1 gene were performed to select appropriate cases. Methylation status was assessed by methylation-specific PCR. Laser capture microdissection of tumor cells followed by real-time PCR was applied to examine the expression of SMARCB1 mRNA. Of 36 epithelioid sarcomas, 31 (86%) displayed a lack of SMARCB1 nuclear protein. In all, 4 (13%) of 31 SMARCB1-negative cases harbored biallelic deletion while 9 (33%) cases showed single-allelic deletion. One (4%) frameshift deletion of exon 3 and one point mutation of exon 7 were also found. In 16 (59%) cases, both alleles were intact. Altogether, 25/31 (81%) SMARCB1-negative cases had at least one intact allele. None of these cases demonstrated promoter hypermethylation. Low levels of SMARCB1 mRNA were found in all cases with tumor tissue extracted RNA (because of the minimal normal cell contamination) but no mRNA could be detected in laser dissected cases (containing only tumor cells). Enhancer of zeste homolog 2 (EZH2) overexpression was not characteristic of epithelioid sarcoma. Thus, loss of SMARCB1 expression in epithelioid sarcoma is caused neither by DNA hypermethylation nor by post-translational modifications. Most likely it is the microRNA destruction of SMARCB1 mRNA but further investigations are needed to elucidate this issue. Keywords: epithelioid sarcoma; histone methylation; laser capture microdissection; methylation-specific PCR; promoter hypermethylation; SMARCB1The SMARCB1 (SWI/SNF-related, matrix-associated, actin-dependent regulator of chromatin, subfamily b, member 1; also known as INI1, BAF47 and hSNF5) gene, located on chromosomal position 22q11.23, encodes a core subunit of the SWI/SNF ATP-dependent chromatin remodeling complex. 1 Such observations show these complexes regulate gene expression, at least in part, by inducing a nucleosome conformation that is more approachable to the transcriptional machinery. 2 Recently, inactivation of both alleles of the SMARCB1 gene was identified in the majority of malignant rhabdoid tumors, suggesting that loss-of-function mutations of the gene contributed to the oncogenesis of this type of tumor.

The kidney is a relatively infrequent site for solitary fibrous tumor (SFT). Among the previously reported cases, only two cases of malignant renal SFT developing via dedifferentiation from a pre-existing benign SFT have been reported. Here we reported a case of de novo malignant renal SFT clinically diagnosed as renal cell carcinoma in a 50-year-old woman. The tumor was circumscribed but unencapsulated and showed obvious hemorrhagic necrosis. Microscopically, the tumor was composed of patternless sheets of alternating hypercellular and hypocellular areas of spindle cells displaying mild to moderate nuclear atypia, frequent mitoses up to 8 per 10 high power fields, and a 20% Ki-67 proliferative index. Immunohistochemical studies revealed reactivity for CD34, CD99 and vimentin, with no staining for all other markers, confirming the diagnosis of SFT. No areas of dedifferentiation were seen after extensive sampling. Based on the pathologic and immunohistochemical features, a diagnosis of de novo malignant renal SFT was warranted. Our report expands the spectrum of malignant progression in renal SFTs. Even though this patient has been disease-free for 30 months, long-term follow-up is still mandatory.

mutated cases were reclassified as spindle embryonal (n = 3), dense embryonal (n = 1) and unclassifiable (n = 1) RMSs. Conclusion: MYOD1-mutant RMSs are uncommonly mutated with PIK3CA and behave aggressively with an expanded morphological and genetic spectrum, including lipoblastic differentiation, multinucleated cells and the alternative p.E118K mutation.