ASPSCR1-TFE3 fusion in a case of Xp11 translocation PEComa of the liver: are ASPSCR1-TFE3 fusion-bearing tumours Xp11 translocation PEComa or alveolar soft part sarcoma?

Abstract

Perivascular epithelioid cell tumour (PEComa) expresses melanocytic and smooth muscle markers, and often harbours tuberous sclerosis complex (TSC) abnormalities. Different from conventional PEComa, Xp11 translocation PEComa is considered a more aggressive subtype of PEComa with the feature of TFE3 rearrangement but lack of TSC alterations.1Argani P. Aulmann S. Illei P.B. et al.A distinctive subset of PEComas harbors TFE3 gene fusions.Am J Surg Pathol. 2010; 34: 1395-1406Crossref PubMed Scopus (288) Google Scholar,2Malinowska I. Kwiatkowski D.J. Weiss S. Martignoni G. Netto G. Argani P. Perivascular epithelioid cell tumors (PEComas) harboring TFE3 gene rearrangements lack the TSC2 alterations characteristic of conventional PEComas: further evidence for a biological distinction.Am J Surg Pathol. 2012; 36: 783-784Crossref PubMed Scopus (98) Google Scholar Some cases have rich melanin production. Recently, it was proposed to rename them as ‘melanotic Xp11 neoplasm’ instead of ‘Xp11 translocation PEComa’, in order to genetically distinguish them from conventional PEComas harbouring TSC rearrangement.3Rao Q. Shen Q. Xia Q.Y. et al.PSF/SFPQ is a very common gene fusion partner in TFE3 rearrangement-associated perivascular epithelioid cell tumors (PEComas) and melanotic Xp11 translocation renal cancers: clinicopathologic, immunohistochemical, and molecular characteristics suggesting classification as a distinct entity.Am J Surg Pathol. 2015; 39: 1181-1196Crossref PubMed Google Scholar,4Wang X.T. Xia Q.Y. Ni H. et al.Xp11 neoplasm with melanocytic differentiation of the prostate harbouring the novel NONO-TFE3 gene fusion: report of a unique case expanding the gene fusion spectrum.Histopathology. 2016; 69: 450-458Crossref PubMed Google Scholar However, the term ‘melanotic Xp11 neoplasm’ still remains controversial. In 2020, Wang et al.5Wang X.T. Fang R. Zhang R.S. et al.Malignant melanotic Xp11 neoplasms exhibit a clinicopathologic spectrum and gene expression profiling akin to alveolar soft part sarcoma: a proposal for reclassification.J Pathol. 2020; 251: 365-377Crossref PubMed Scopus (11) Google Scholar reported the largest series of 27 cases of melanotic Xp11 neoplasm, strengthened the malignant behaviour of the tumour, and suggested that melanotic Xp11 neoplasm and alveolar soft part sarcomas (ASPS) might belong to a spectrum of the same entity. To date, the reported TFE3 fusion genes in Xp11 translocation PEComa have included SFPQ (PSF),3Rao Q. Shen Q. Xia Q.Y. et al.PSF/SFPQ is a very common gene fusion partner in TFE3 rearrangement-associated perivascular epithelioid cell tumors (PEComas) and melanotic Xp11 translocation renal cancers: clinicopathologic, immunohistochemical, and molecular characteristics suggesting classification as a distinct entity.Am J Surg Pathol. 2015; 39: 1181-1196Crossref PubMed Google Scholar,6Argani P. Zhong M. Reuter V.E. et al.TFE3-fusion variant analysis defines specific clinicopathologic associations among Xp11 translocation cancers.Am J Surg Pathol. 2016; 40: 723-737Crossref PubMed Scopus (134) Google Scholar NONO,6Argani P. Zhong M. Reuter V.E. et al.TFE3-fusion variant analysis defines specific clinicopathologic associations among Xp11 translocation cancers.Am J Surg Pathol. 2016; 40: 723-737Crossref PubMed Scopus (134) Google Scholar,7McGregor S.M. Alikhan M.B. John R.A. et al.Melanotic PEComa of the sinonasal mucosa with NONO-TFE3 fusion: an elusive mimic of sinonasal melanoma.Am J Surg Pathol. 2017; 41: 717-722Crossref PubMed Scopus (14) Google Scholar RBM10,8Xia Q.Y. Wang X.T. Zhan X.M. et al.Xp11 translocation renal cell carcinomas (RCCs) with RBM10-TFE3 gene fusion demonstrating melanotic features and overlapping morphology with t(6;11) RCC: interest and diagnostic pitfall in detecting a paracentric inversion of TFE3.Am J Surg Pathol. 2017; 41: 663-676Crossref PubMed Scopus (37) Google Scholar RBMX,9Argani P. Zhang L. Sung Y.S. et al.A novel RBMX-TFE3 gene fusion in a highly aggressive pediatric renal perivascular epithelioid cell tumor.Genes Chromosomes Cancer. 2020; 59: 58-63Crossref Scopus (0) Google Scholar etc. In this study, the unique case occurred in the liver with melanocytic immunophenotype (cathepsin K and HMB45 positive), and TFE3 rearrangement. Thus, it was diagnosed as Xp11 translocation PEComa. Interestingly, following RNA sequencing, it was found this case harboured an ASPSCR1-TFE3 fusion, which is considered the common fusion in ASPS. Thus, are ASPSCR1-TFE3 fusion-bearing tumours Xp11 translocation PEComa or ASPS? The answers might be debatable, but these results could provide evidence for the opinion that the two tumours represent different subtypes of the same entity. The patient was a 33-year-old Chinese female who presented with abdominal pain. A computed tomography (CT) scan and B-mode ultrasound (Fig. 1A) showed a mass approximately 15 cm in size in the right posterior lobe of the liver. It was mainly solid, well-defined, and rich in blood flow signals. The patient had a history of posterior peritoneal mass but the pathological specimens were unavailable. Partial hepatectomy was undertaken. Follow-up study showed no recurrence and metastasis. Grossly, the mass was relatively well-defined and measured 15 cm at the greatest diameter, with haemorrhage and necrosis in the centre of the tumour (Fig. 1B). Microscopically, the tumour was encapsulated with a pushing growth pattern (Fig. 1C). The tumour cells were arranged in a perivascular pattern (Fig. 1D). Under high magnification, the tumour cells were pleomorphic with an eosinophilic large nucleolus, with mitotic activity at about 3 per 50 HPF (Fig. 1E). There were scattered giant tumour cells and calcification, and haemorrhage and necrosis were found in some areas (Fig. 1F). No melanin deposition was found in the whole section. The tumour cells did not express epithelial markers such as AE1/AE3 (Fig. 2A), Cam 5.2 and EMA. Instead, the tumour cells were positive for cathepsin K (Fig. 2B) and HMB-45 (Fig. 2C), but negative for smooth muscle actin (SMA) (Fig. 2D), desmin and S-100. Moreover, the tumour cells showed diffuse and strong positivity of TFE3 (Fig. 2E). Total RNA was extracted from formalin fixed, paraffin embedded tissue sections using the AllPrep DNA/RNA FFPE Kit (Qiagen, USA). One hundred nanograms of RNA was used to generate RNA-seq libraries using a TruSeq RNA Sample Preparation kit (Illumina, Australia). RNA-seq libraries were multiplexed (two per lane) and sequenced on a HiSeq 2000 (Illumina) per the manufacturer's recommendations. We generated 30 million paired-end (150 bp) sequencing reads per sample. All short-read data were evaluated for quality control using the Bioconductor ShortRead package (https://bioconductor.org/packages/release/bioc/html/ShortRead.html). Bowtie2 (http://bowtie-bio.sourceforge.net/bowtie2/index.shtml) was used for alignment and mapping of short-read data to the human genome reference hg19. FusionSeq data analysis (http://info.gersteinlab.org/FusionSeq) identified an ASPSCR1-TFE3 fusion as the top candidate. Alignment of the reads suggested a fusion of ASPSCR1 exon 7 to TFE3 exon 5 (Fig. 3A), a transcript that was subsequently confirmed by RT-PCR. To further confirm the ASPSCR1-TFE3 fusion at the genomic level, fluorescence in situ hybridisation (FISH) was conducted using a two-colour fusion FISH assay ASPSCR1 (17q25, red)-TFE3 (Xp11.2, green) dual-colour probe (LBP, China), and the positive yellow signals were observed with the co-localisation of ASPSCR1 (red) and TFE3 (green) (Fig. 3B). The clinicopathological characteristics (young female), morphology (epitheloid morphology and perivascular pattern), immunophenotype (AE1/AE3 negative, cathepsin K positive, HMB45 positive), genetic changes (TFE3 rearrangement) were consistent with previous cases, which favoured the diagnosis of Xp11 translocation PEComa. As noted in the original description,1Argani P. Aulmann S. Illei P.B. et al.A distinctive subset of PEComas harbors TFE3 gene fusions.Am J Surg Pathol. 2010; 34: 1395-1406Crossref PubMed Scopus (288) Google Scholar compared with conventional PEComas, Xp11 translocation PEComa affects younger patients with predominantly epithelioid cell morphology and more aggressive behaviour. The patient had a history of posterior peritoneal mass, but unfortunately the pathological details were unavailable. It was uncertain whether the liver mass was primary or metastatic. If metastatic, the results prove the aggressive behaviour of this entity. Thus, it is proposed to rename it as ‘melanotic Xp11 neoplasm’ or ‘Xp11 neoplasm with melanocytic differentiation’ to differentiate it from conventional PEComa. Interestingly, our case harboured an unusual ASPSCR1-TFE3 fusion, not previously reported, and expanded the fusion spectrum of Xp11 translocation PEComa. To our knowledge, the ASPSCR1-TFE3 fusion has been most often observed in ASPS. So, are ASPSCR1-TFE3 fusion-bearing tumours Xp11 translocation PEComa or ASPS? ASPS is a malignant soft tissue tumour with uncertain histogenesis. There is only one report of primary ASPS of the liver, by Shaddix et al. in 2008,10Shaddix K.K. Fakhre G.P. Nields W.W. Steers J.L. Hewitt W.R. Menke D.M. Primary alveolar soft-part sarcoma of the liver: anomalous presentation of a rare disease.Am Surg. 2008; 74: 43-46Crossref PubMed Google Scholar and one report of metastatic ASPS in the liver, by Silas et al. in 2010.11Silas O.A. Adoga A.A. Manasseh A.N. Echejoh G.O. Vhriterhire R.A. Mandong B.M. Persistent alveolar soft-part sarcoma with liver metastasis: a case report.J Med Case Rep. 2010; 4: 233Crossref PubMed Scopus (1) Google Scholar Immunophenotypes could be used for the differential diagnosis, since ASPS is always negative for melanotic markers such as cathepsin K and HMB-45, whereas these markers were positive in this case. ASPS has a similar morphology to Xp11 translocation PEComa, especially harbouring the same genetic changes of TFE3 translocation. The most common fusion is ASPSCR1-TFE3, as in this case. Other TFE3 fusion genes include HNRNPH3, DVL and PRCC,12Dickson B.C. Chung C.T. Hurlbut D.J. et al.Genetic diversity in alveolar soft part sarcoma: a subset contain variant fusion genes, highlighting broader molecular kinship with other MiT family tumors.Genes Chromosomes Cancer. 2020; 59: 23-29Crossref Scopus (11) Google Scholar of which the latter two fusions are also observed in renal cell carcinoma. Due to cluster analysis, Wang et al.5Wang X.T. Fang R. Zhang R.S. et al.Malignant melanotic Xp11 neoplasms exhibit a clinicopathologic spectrum and gene expression profiling akin to alveolar soft part sarcoma: a proposal for reclassification.J Pathol. 2020; 251: 365-377Crossref PubMed Scopus (11) Google Scholar speculated that ‘melanotic Xp11 neoplasm’ and ASPS might be a different variant of the same entity. However, strong supporting evidence was still lacking. In this study, the case showed the immunophenotype of so-called melanotic Xp11 neoplasm (cathepsis K and HMB45 positive), but harboured the ASPSCR1-TFE3 fusion, which might give direct proof for the same essence of Xp11 translocation PEComa (aka melanotic Xp11 neoplasm) and ASPS on gene expression level. ASPSCR1-TFE3 fusion is also the high-frequency fusion type of TFE3 translocation renal cell carcinoma (RCC), which should be added to the metastatic differential diagnosis. Both TFE3 translocation RCC and Xp11 translocation PEComa express TFE3 and melanotic markers. However, the former is positive for PAX8, which indicates the origin of renal tubular epithelium, while the latter is negative. There was no history of renal tumour in this case. Epithelioid haemangioendothelioma (EHE) with TFE3 rearrangement is another disease requiring consideration. Hepatic EHE also occurrs in young patients. The tumour cells are often histiocytoid in appearance with voluminous cytoplasm, and the tumour cells often lie along the hepatic sinusoids, which is a characteristic growth pattern. Immunoprofile of vascular markers such as ERG, CD34 could aid to distinguish EHE from Xp11 translocation PEComa. Malignant melanomas involving the liver are common for either primary or metastatic lesions. Lack of TFE3 rearrangements has been considered as the main difference between melanomas and Xp11 translocation PEComa. Additionally, S-100 expression is helpful for the diagnosis, since it is always strongly positive in melanomas but absent in Xp11 translocation PEComa. It is necessary to use additional melanocytic-associated markers in the differential workup to give a proper diagnosis. Clear cell sarcoma (CCS) was previously known as malignant melanoma (MM) of the soft tissue. The morphological features of this case were similar to CCS, showing nested to fascicular proliferation of obviously malignant appearing epithelioid to spindled cells, often with prominent nucleoli and interspersed neoplastic giant cells. However, CCS differs from Xp11 translocation PEComa due to the S-100 protein expression, the presence of EWSR1 and the absence of TFE3 rearrangements. In conclusion, this unique case reported an Xp11 translocation PEComa with ASPSCR1-TFE3 fusion in the liver, which expanded the spectrum of TFE3 fusion genes of this tumour. The shared ASPSCR1-TFE3 fusion with ASPS might provide direct evidence that Xp11 translocation PEComa and ASPS belong to the same entity due to the same genetic changes, and also for the possible renaming of it as ‘melanotic Xp11 neoplasm’ to differentiate from conventional PEComa. Nevertheless, it remains controversial, and more cases and further studies are required to clarify the nomenclature.