Xiaoke Wang

Anaplastic lymphoma kinase immunoreactivity correlates with ALK gene rearrangement and transcriptional up-regulation in non–small cell lung carcinomas

Recently, the fusion gene EML4-ALK was identified in non-small cell lung carcinoma, which could be a potential therapeutic target. We investigated the prevalence of anaplastic lymphoma kinase protein expression in these tumors by immunohistochemistry and correlated the results with data from ALK molecular studies. Gene expression profiling was performed on 35 adenocarcinomas to identify cases with ALK gene up-regulation, which was correlated with protein overexpression by immunohistochemistry. Immunohistochemistry was also performed on an independent cohort consisting of 150 adenocarcinomas and 150 squamous cell carcinomas to evaluate the utility of anaplastic lymphoma kinase immunostaining as a screening tool. Florescence in situ hybridization for the ALK locus and reverse transcriptase-polymerase chain reaction for EML4-ALK were performed on tumors positive for anaplastic lymphoma kinase by immunohistochemistry. Transcriptional up-regulation of ALK was identified in 2 (6%) of 35 adenocarcinomas by gene expression profiling. These 2 cases were positive for anaplastic lymphoma kinase by immunohistochemistry, whereas the remaining 33 cases were completely negative. In the independent cohort, anaplastic lymphoma kinase immunostaining was positive in 1 of 150 squamous cell carcinomas and in 3 of 150 adenocarcinomas. The 6 cases positive for anaplastic lymphoma kinase by immunohistochemistry showed evidence of ALK locus rearrangement by florescence in situ hybridization but were negative for EGFR and KRAS mutation. The presence of EML4-ALK fusion transcript was confirmed in 2 cases by reverse transcriptase-polymerase chain reaction. In conclusion, anaplastic lymphoma kinase immunoreactivity in non-small cell lung carcinomas was associated with transcriptional up-regulation, ALK locus rearrangement, and the presence of EML4-ALK fusion transcript. Anaplastic lymphoma kinase immunohistochemistry may have utility as a screening tool or as a surrogate marker for the molecular techniques to detect the EML4-ALK fusion gene in these tumors.

Nodular fasciitis: a novel model of transient neoplasia induced by MYH9-USP6 gene fusion.

Nodular fasciitis (NF) is a relatively common mass-forming and self-limited subcutaneous pseudosarcomatous myofibroblastic proliferation of unknown pathogenesis. Due to its rapid growth and high mitotic activity, NF is often misdiagnosed as a sarcoma. While studying the USP6 biology in aneurysmal bone cyst and other mesenchymal tumors, we identified high expression levels of USP6 mRNA in two examples of NF. This finding led us to further examine the mechanisms underlying USP6 overexpression in these lesions. Upon subsequent investigation, genomic rearrangements of the USP6 locus were found in 92% (44 of 48) of NF. Rapid amplification of 5′-cDNA ends identified MYH9 as the translocation partner. RT-PCR and direct sequencing revealed the fusion of the MYH9 promoter region to the entire coding region of USP6. Control tumors and tissues were negative for this fusion. Xenografts of cells overexpressing USP6 in nude mice exhibited clinical and histological features similar to human NF. The identification of a sensitive and specific abnormality in NF holds the potential to be used diagnostically. Considering the self-limited nature of the lesion, NF may represent a model of ‘transient neoplasia’, as it is, to our knowledge, the first example of a self-limited human disease characterized by a recurrent somatic gene fusion event.

Molecular Testing for Lipomatous Tumors: Critical Analysis and Test Recommendations Based on the Analysis of 405 Extremity-based Tumors

Ancillary molecular testing has been advocated for diagnostic accuracy in the differentiation of lipomas from atypical lipomatous tumors/well-differentiated liposarcomas (ALT/WDL); however, the implications and specific indications for use are not well-established in the current literature. Herein, we extend previous findings by quantitatively evaluating the impact of molecular testing of lipomatous neoplasms in our routine clinical practice, how it modifies the historical perspective of their clinical course, and the effect of distinct surgical procedures in modulating the risk of local recurrence for these tumors after molecular classification. On the basis of these analyses, we suggest a specific set of basic recommendations for complementary molecular assessment in the diagnosis of lipomatous tumors. Four hundred and five lipomatous neoplasms located in the trunk and extremities were analyzed histologically and for the presence of 12q13-15 amplification on paraffin-embedded tissues by assessing MDM2/CPM amplification. Survival analyses were calculated with Kaplan-Meier and compared with the log-rank. Multivariate analysis was evaluated by the Cox regression method. The 405 tumors were histologically classified as ordinary lipoma (n=324), intramuscular lipoma (n=29), and ALT/WDL (n=52). The level of agreement between the histologic diagnosis and the molecular diagnosis was high (96%) but pathologists showed a tendency to overestimate cytologic atypia and the diagnosis of ALT/WDL (precision, 79%; accuracy, 88%). Molecular assessment led to a major diagnostic reclassification in 18 tumors (4%). Eleven of the tumors histologically classified as ALT/WDL were reclassified as ordinary lipoma (n=5) and intramuscular lipoma (n=6); none of which recurred. Seven ordinary lipomas were reclassified as ALT/WDL, 6 of which were larger than 15 cm and deeply located; 2 recurred locally. After molecular data, the 5-year local recurrence rates for ordinary lipoma, intramuscular lipoma, and ALT/WDL were 1%, 12%, and 44%, respectively. Multivariate analyses after molecular assessment showed tumor type and type of resection to be associated with the risk of local recurrence. Complementary molecular testing refines the histologic classification of lipomatous tumors and better estimates the impact of surgical procedures on the risk of local recurrence. Pathologists tend to overestimate the degree of cytologic atypia and the indiscriminate use of molecular testing should be avoided, especially for extremity-based tumors. Molecular testing should be considered for “relapsing lipomas,” tumors with questionable cytologic atypia (even if widely excised), or for large lipomatous tumors (>15 cm) without diagnostic cytologic atypia.

Frequency and characterization of HMGA2 and HMGA1 rearrangements in mesenchymal tumors of the lower genital tract.

Mesenchymal tumors of the lower genital tract predominantly occur in women of reproductive age and are mainly represented by aggressive angiomyxoma (AAM) and angiomyofibroblastoma (AMF). Whether these tumors are different phenotypic expressions of the same biological entity is still debatable. Genetic rearrangements of HMGA2 have been reported in a few cases of AAM but its frequency and clinicobiological implications have not been studied systematically. We evaluated 90 cases of mesenchymal tumors of the lower genital tract that comprised 42 AAMs, 18 AMFs, 6 cellular angiofibromas, 5 fibroepithelial stromal polyps, 15 genital leiomyomas, 3 superficial angiomyxomas, and 1 spindle cell lipoma. Fluorescence in situ hybridization was used to identify rearrangements of HMGA2 and its homologue HMGA1. HMGA2 rearrangements were identified in 14 AAMs (33%) and in 1 vaginal leiomyoma. All other tumors were negative for HMGA2 rearrangements. HMGA1 rearrangement was not found in any of the cases. RT‐PCR confirmed transcriptional upregulation of HMGA2 only in tumors with HMGA2 rearrangements. Standard cytogenetic analyses were performed in two AAMs and one AMF. One AAM had a t(1;12)(p32;q15); the other tumors had normal karyotypes. Mapping and sequence analysis of the breakpoint showed fusion to the 3′ untranslated region of HMGA2 to genomic sequences derived from the contig NT 032977.8 on chromosome 1p32. Our findings support the hypothesis that AAM and AMF are distinct biological entities. The diagnostic usefulness of HMGA2 rearrangements to differentiate between AAM and other tumors of the lower genital tract may be limited due to the their low frequency.

Gains of COL1A1-PDGFB genomic copies occur in fibrosarcomatous transformation of dermatofibrosarcoma protuberans.

Dermatofibrosarcoma protuberans is a superficial low-grade sarcoma that rarely evolves into a high-grade fibrosarcoma. Dermatofibrosarcoma protuberans is genetically characterized by the unbalanced chromosomal t(17;22)(q21;q13), usually in the form of a supernumerary ring chromosome. The product of this chromosomal translocation is the chimeric gene COL1A1-PDGFB (collagen type I alpha I-platelet-derived growth factor beta), which is amplified at low levels in the ring chromosome. The aims of this study were to evaluate (1) whether genomic gains of this fusion gene occur during the clonal evolution of dermatofibrosarcoma protuberans into fibrosarcomatous dermatofibrosarcoma protuberans and (2) whether there is a difference between the number of genomic copies of COL1A1-PDGFB between classic dermatofibrosarcoma protuberans and dermatofibrosarcoma protuberans areas associated with fibrosarcomatous dermatofibrosarcoma protuberans. Eleven cases of fibrosarcomatous dermatofibrosarcoma protuberans with both dermatofibrosarcoma protuberans and fibrosarcomatous areas and 10 cases of classic dermatofibrosarcoma protuberans were studied. Genomic copies of COL1A1-PDGFB were evaluated by fluorescence in situ hybridization using a custom designed probe for the PDGFB locus on 4 μm thick paraffin-embedded tissue sections. Genomic gains of the COL1A1-PDGFB gene were observed in six (of 10) fibrosarcomatous dermatofibrosarcoma protuberans in the fibrosarcomatous areas when compared to the dermatofibrosarcoma protuberans areas of the same tumor (2–7 gene copies (median PDGFB copy gain, 2.8) versus 1–3 gene copies (median PDGFB copy gain, 1.7), respectively, P=0.004). Four fibrosarcomatous dermatofibrosarcoma protuberans did not show genomic gains of COL1A1-PDGFB fusion gene between the two areas. Essentially no difference in the copy number of COL1A1-PDGFB fusion gene was observed between dermatofibrosarcoma protuberans areas of classic dermatofibrosarcoma protuberans and dermatofibrosarcoma protuberans areas of fibrosarcomatous dermatofibrosarcoma protuberans (median PDGFB copy gain of 1.8 versus 1.7, respectively, P=0.36). Genomic gains of COL1A1-PDGFB fusion gene is possibly an oncogenic mechanism that is identified in the clonal evolution of a subset of dermatofibrosarcoma protuberans that evolves into fibrosarcomatous dermatofibrosarcoma protuberans. Since this finding was not observed in all cases of fibrosarcomatous dermatofibrosarcoma protuberans, other oncogenic mechanisms may be operating in this form of tumor progression. Copy number of COL1A1-PDGFB fusion gene in the classic dermatofibrosarcoma protuberans areas does not seem to be a major predisposing mechanism for fibrosarcomatous transformation.

Recurrent PAX3 - MAML3 fusion in biphenotypic sinonasal sarcoma

Biphenotypic sinonasal sarcoma (SNS) is a newly described tumor of the nasal and paranasal areas. Here we report a recurrent chromosomal translocation in SNS, t(2;4)(q35;q31.1), resulting in a PAX3-MAML3 fusion protein that is a potent transcriptional activator of PAX3 response elements. The SNS phenotype is characterized by aberrant expression of genes involved in neuroectodermal and myogenic differentiation, closely simulating the developmental roles of PAX3.

Carboxypeptidase M: a biomarker for the discrimination of well-differentiated liposarcoma from lipoma

The discrimination between well-differentiated liposarcomas/atypical lipomatous tumors and lipomas can be diagnostically challenging at the histological level. However, cytogenetic identification of ring and giant rod chromosomes supports the diagnosis of well-differentiated liposarcoma/atypical lipomatous tumor. These abnormal chromosomes are mainly composed of amplified genomic sequences derived from chromosome 12q13-15, and contain several genes, including MDM2, CDK4 (SAS), TSPAN31, HMGA2, and others. MDM2 is consistently amplified in well-differentiated liposarcomas/atypical lipomatous tumors, and up to 25% in other sarcomas. As part of a large genomic study of lipomatous neoplasms, we initially found CPM to be consistently amplified in well-differentiated liposarcomas/atypical lipomatous tumors. To further explore this initial finding, we investigated the copy number status of MDM2 and CPM by fluorescent in situ hybridization (FISH) on a series of 138 tumors and 17 normal tissues, including 32 well-differentiated liposarcoma/atypical lipomatous tumors, 63 lipomas, 11 pleomorphic lipomas, 2 lipoblastomas, 30 other tumors and 17 normal fat samples. All 32 well-differentiated liposarcoma/atypical lipomatous tumors showed amplification of MDM2 and CPM, usually >20 copies per cell. The other tumors lacked MDM2 and/or CPM amplification. Chromogenic in situ hybridization confirmed the above results on a subset of these tumors (n=27). These findings suggest that identification of CPM amplification could be used as an alternative diagnostic tool for the diagnosis of well-differentiated liposarcoma/atypical lipomatous tumors.

Molecular cytogenetic analysis for TFE3 rearrangement in Xp11.2 renal cell carcinoma and alveolar soft part sarcoma: validation and clinical experience with 75 cases

Renal cell carcinoma with TFE3 rearrangement at Xp11.2 is a distinct subtype manifesting an indolent clinical course in children, with recent reports suggesting a more aggressive entity in adults. This subtype is morphologically heterogeneous and can be misclassified as clear cell or papillary renal cell carcinoma. TFE3 is also rearranged in alveolar soft part sarcoma. To aid in diagnosis, a break-apart strategy fluorescence in situ hybridization (FISH) probe set specific for TFE3 rearrangement and a reflex dual-color, single-fusion strategy probe set involving the most common TFE3 partner gene, ASPSCR1, were validated on formalin-fixed, paraffin-embedded tissues from nine alveolar soft part sarcoma, two suspected Xp11.2 renal cell carcinoma, and nine tumors in the differential diagnosis. The impact of tissue cut artifact was reduced through inclusion of a chromosome X centromere control probe. Analysis of the UOK-109 renal carcinoma cell line confirmed the break-apart TFE3 probe set can distinguish the subtle TFE3/NONO fusion-associated inversion of chromosome X. Subsequent extensive clinical experience was gained through analysis of 75 cases with an indication of Xp11.2 renal cell carcinoma (n=54), alveolar soft part sarcoma (n=13), perivascular epithelioid cell neoplasms (n=2), chordoma (n=1), or unspecified (n=5). We observed balanced and unbalanced chromosome X;17 translocations in both Xp11.2 renal cell carcinoma and alveolar soft part sarcoma, supporting a preference but not a necessity for the translocation to be balanced in the carcinoma and unbalanced in the sarcoma. We further demonstrate the unbalanced separation is atypical, with TFE3/ASPSCR1 fusion and loss of the derivative X chromosome but also an unanticipated normal X chromosome gain in both males and females. Other diverse sex chromosome copy number combinations were observed. Our TFE3 FISH assay is a useful adjunct to morphologic analysis of such challenging cases and will be applicable to assess the growing spectrum of TFE3-rearranged tumors.

Genomic gains of COL1A1-PDFGB occur in the histologic evolution of giant cell fibroblastoma into dermatofibrosarcoma protuberans

Giant cell fibroblastoma (GCF) is a subcutaneous mesenchymal neoplasm characterized by the chromosomal t(17;22), which results in the formation of the fusion gene COL1A1‐PDGFB. This same fusion gene is also seen in the supernumerary ring chromosome of dermatofibrosarcoma protuberans (DFSP). Several studies have addressed the molecular genetics of DFSP but molecular cytogenetic characterization of individual areas and cell components in pure GCF and GCF/DFSP hybrids have not been performed. Herein, we studied the frequency and genomic copy number of COL1A1‐PDGFB in pure GCF and GCF/DFSP hybrids, and identified the molecular cytogenetic signatures in individual cells in each component. Four pure GCF and nine GCF/DFSP hybrids were studied. All tumors exhibited classical histological features and CD34 expression. COL1A1 and PDGFB rearrangements were evaluated by fluorescence in situ hybridization (FISH) using probes for COL1A1 and PDGFB on paraffin‐embedded thin tissue sections. All GCF and GCF/DFSP hybrids showed unbalanced rearrangements of COL1A1‐PDGFB at the molecular cytogenetic level. Genomic gains of COL1A1‐PDGFB were found predominantly in the DFSP component of GCF/DFSP hybrids but in none of the pure GCF, suggesting that these gains are associated with the histologic evolution of GCF into DFSP. The molecular cytogenetic abnormalities were found not only in the spindle/stellated cells but also in individual nuclei of the multinucleated giant cells, suggesting that these cells may result from the fusion of individual neoplastic cells.

Frequency of USP6 rearrangements in myositis ossificans, brown tumor, and cherubism: molecular cytogenetic evidence that a subset of “myositis ossificans-like lesions” are the early phases in the formation of soft-tissue aneurysmal bone cyst

ObjectiveUSP6 rearrangements with several partner genes have been identified recently in primary but not in secondary aneurysmal bone cysts (ABCs). Several lesions show histologic features that may overlap with ABC, including myositis ossificans (MO), brown tumor, and cherubism. The objective of this study was to assess whether these lesions harbored USP6 rearrangements.Materials and methodsTwelve patients with classic radiologic and histologic features of MO, 6 with brown tumors, and 5 with cherubism diagnosed at our institution were studied for the presence of USP6 rearrangements using fluorescence in situ hybridization with probes flanking the USP6 locus on chromosome 17p13. In addition, conventional cytogenetic analysis was performed in 2 patients with cherubism.ResultsUSP6 rearrangements were identified in 2 patients with radiologic and histologic features consistent with MO. None of the patients with brown tumor or cherubism demonstrated USP6 rearrangements. Cytogenetic analysis of the cherubism patients demonstrated normal karyotypes.ConclusionThese findings indicate that a subset of cases with apparent classic histologic and imaging features of MO are rather better classified as being soft-tissue ABC with clonal USP6 rearrangements. In contrast, no USP6 rearrangements were found in patients with cherubism or brown tumor, supporting the prevailing view that these lesions are distinct biologic entities.