1

Paediatric small round cell sarcomas – an update

International Society of Bone and Soft Tissue Pathology meeting, Baltimore, MD

March 3, 2013

David M. Parham, MD

Oklahoma University Health Science Center

Oklahoma City, OK

The past several decades have been marked by phenomenal improvement in our ability to diagnosis the small blue cell tumors of childhood, traditionally including rhabdomyosarcoma, Ewing sarcoma (including PNET), neuroblastoma, lymphoma/leukemia, small cell osteosarcoma, and mesenchymal chondrosarcoma. Newer entities have been added, including desmoplastic small round cell tumor, extrarenal rhabdoid tumor, and BRD/NUT carcinoma. To a large degree, our increased recognition of classical entities and description of newer ones has been driven by a switch from immunohistochemistry to genetics as an ancillary diagnostic technique. With these advances has come a corollary question: how do we deal with cases that have clinical and histological features of specific pathological entities but lack their characteristic genetic features?

Two major genetic tests - FOXO1 and EWSR1 FISH - have driven diagnosis of most pediatric round cell sarcomas in the past decade. This is to be expected, as their corresponding pathological entities, rhabdomyosarcoma and Ewing sarcoma (as well as a host of non-Ewing EWS fusion-associated sarcomas), comprise the majority of pediatric sarcomas. Therefore, I will review recent advances in pediatric sarcomas in the practical diagnostic context of what to consider when these tests are negative.

Fusion-negative rhabdomyosarcoma

Rhabdomyosarcoma, the most frequent pediatric sarcoma, comprises two major histological species, embryonal (ERMS) and alveolar (ARMS) and variants of these patterns such as botryoid RMS. Genetic features that are useful in the context of diagnosing ERMS continue to be sparse, although certain chromosomal duplications and deletions recur1, and loss of heterozygosity of 11p is typical2. On the other hand, it is now established that ERMS characteristically shows a variably heterogeneous myogenin expression pattern, usually contrasting with the typical diffuse strong myogenin expression of ARMS3.

Historically, histology has been the gold standard for diagnosing ARMS, beginning with the recognition of the classical septate pattern by Riopelle and Theriault4. The criteria for diagnosing ARMS were loosened by the recognition of the “solid variant”, which lacks fibrous septa but shows similar cytology5. Histological classification of pediatric RMS became globally standardized with publication of the International Pediatric Sarcoma Classification (IPSC) in 19956. Histological diagnosis of ARMS has been a critical component of standardized therapy, as they cannot be treated on low risk protocols and show more aggressive behavior in high and intermediate risk groups7. Sadly, no major improvements in ARMS treatment have been achieved since development of standard chemotherapy in the 1970s8. Thus, classification of RMS remains been a major concern for pediatric oncologists.

2

In spite of our inability to improve the survival of ARMS, in the past 30 years we have learned an amazing amount about its biology. These discoveries have followed the initial description of the ARMS-associated t(2;13) translocation and its genetic counterpart, the PAX3-FOXO1 (formerly PAX3-FKHR) fusion, broadened by findings of an alternate t(1;13) and associated PAX7-FOXO1 fusion in another subset. However, 20-30% of ARMS have persistently failed to display a translocation, to the dismay of finicky oncologists. Thus, the big question of the past decade has been, “Does the behavior of ARMS reflect its histology or its genetics?”

A giant step forward came with Davicioni et al.’s work on gene expression arrays9,10. In these landmark papers, RNA extracted from a series of RMS was hybridized with Affymetrix gene chips, and the resultant morass of expression data was compared using histology as one parameter and fusion status as another. The results clearly indicate that fusion-negative ARMS cannot be distinguished from ERMS, whereas PAX-fusion positive ARMS gene expression indicates a distinctive genetic signature. This finding is perhaps not surprising, since downstream gene expression is being driven by the promoter region dysregulation associated with PAX fusion proteins. The results were repeated in subsequent expression array studies, and data emerged indicating that fusion-negative ARMS behave similarly to ERMS but not fusion-positive ARMS11. The conclusions that have been reached are: 1) fusion-negative ARMS are clinically and molecularly indistinguishable from ERMS, and 2) PAX-fusion positive ARMS show a distinctive molecular signature unlike that of other tumors. However, questions persist: What exactly are fusion-negative ARMS? Does the diagnosis of “ARMS” depend on fusion-status, histology, or both?

In recent studies, possible sources of fusion-negative ARMS have emerged. These include: 1) ARMS with alternate fusions not detected with standard genetic tests; 2) new histological entities with features similar to ARMS, 3) tumors with mixed ARMS and ERMS histology, and 4) ERMS that look like ARMS.

Alternate fusions not detected by standard genetic assays have indeed been discovered in recent years. However, they comprise less than 1% in reports published to date12-14 and include fusions of PAX3 with AFX1, NCOA1, and NCOA2. ARMS with the latter fusions may biologically be more akin to ERMS11, but more data are needed.

Sclerosing RMS, a new histological entity, resembles ARMS and so was initially considered a “micro-alveolar” variant. In Menzel et al.15 and Folpe et al.’s16 seminal descriptions, only 7 adult tumors were reported, but 13 pediatric cases were added in a review of IRSG material by Chiles et al17. Of note, only one of five tested cases contained a PAX fusion in Chile et al.’s series, and the single case tested in Folpe et al.’s series was also negative. Besides the microalveolar pattern, which lacks “floating clusters” and rows of septal-adherent cells, these tumors also differ from classical ARMS by their extensive osteoid-like sclerosis and low myogenin expression.

Tumors with mixed ARMS and ERMS features have posed a diagnostic quandary since early descriptions18. The IPSC solved this dilemma by defining all mixed ARMS-ERMS as ARMS, regardless of the relative proportion of elements. Problems arose from this solution, as subsequent Intergroup Rhabdomyosarcoma Study Group (IRSG) and Children’s Oncology Group (COG) studies showed a rise in fusion-negativity from ~20% to 45% of ARMS, and the overall percentage of ARMS among RMS cases rose from 25-30% to 41%. The COG has responded by reverting to earlier IRSG criteria requiring >50% ARMS for the diagnosis. However, many mixed ARMS are fusion-negative, regardless of percentages19, and occasional mixed cases with a minority ARMS component are fusion-positive (unpublished data).

The major problem in over-diagnosing ARMS can be surmised from IPSC definitions that indicate ERMS may occur as dense cellular sheets, so that the difference from solid ARMS becomes a relatively subtle

3

cytological one. Solid ARMS contain round, monomorphous nuclei whereas dense ERMS contain nuclei with spindly or irregular profiles, and some overlap may exist. Ancillary testing is helpful in this distinction, as myogenin expression is generally strong and diffuse in ARMS and heterogeneous in ERMS3. Some overlap occurs in cases with 50-90% myogenin positivity, but expression array studies have produced new markers like AP2β that may be superior20.

To summarize recent RMS findings, it appears that ARMS can easily be over-diagnosed, which has contributed to studies indicating that fusion-negative ARMS have biological similarity to ERMS. The clinical significance of fusion-negative ARMS has not been thoroughly tested, as little data exists for ARMS treatment with low risk therapy. The likelihood of a lesion being fusion-negative is higher with solid variant ARMS than with classical ARMS21, but in the absence of molecular testing, myogenin staining can be helpful in making this diagnosis3. Stratification in future protocols will be based on genetic classification rather than histology, as the exact definition of ARMS has become a source of controversy22,23.

Fusion-negative Ewing sarcoma

Although fusion-negative Ewing sarcomas/PNETs have been reported24,25, their existence is also questioned. In some cases, there is a discrepancy between RT-PCR and FISH results, the latter appearing to be a more sensitive (but less specific) technique24,26. This may result from the wide array of EWS fusion partners that occur with Ewing sarcomas25. In fusion-negative lesions with Ewing histology and CD99-positivity, it behooves one to be rule out other diagnoses, the most likely ones being rhabdomyosarcoma, neuroblastoma, and lymphoma. Rhabdomyosarcomas can be excluded with myogenin and desmin stains, with the caveat that rare PNETs are desmin-positive but show other features of neuroepithelial differentiation or genetic change27. Undifferentiated neuroblastomas lacking rosettes or neuropil should also be separated from Ewing sarcomas. Undifferentiated neuroblastomas are aggressive neoplasms, automatically unfavorable in Shimada grading28, but unlike Ewing sarcoma they show CD99-negativity and CD56-positivity29,30. Lymphomas can be particularly treacherous, as lymphoblastic tumors often show CD99-positivity and CD45-negativity, but CD3, CD43, and/or TdT are usually positive31.

Undifferentiated sarcoma

In current practice, fusion-negative Ewing-like neoplasms are most often diagnosed as undifferentiated sarcomas. In recent COG studies, undifferentiated sarcomas have been treated as non-rhabdomyosarcomatous soft tissue sarcomas, and preliminary results suggest that they respond to chemotherapy (unpublished data). FUS FISH testing is recommended, however, as a subset of Ewing sarcomas show alternate FUS-ERG or FUS-FEV fusions, and FUS and EWS are related genes25. A new fusion, CIC-DUX4, derived from a t(4;19)(q35;13.1) translocation32, is found in a substantial number of undifferentiated sarcomas. Italiano et al.33 found CIC rearrangements in 15 of 22 EWS fusion-negative Ewing-like cases.

BRD4-NUT carcinoma

An undifferentiated, high-grade carcinoma that often affects younger patient and resembles Ewing tumors has been recently described and is named for its characteristic gene fusion, BRD4-NUT. The fusion is formed by a t(15;19)(q13;q13) translocation34. Tumors of this type typically affect midline structures cervicothoracic structures such as the upper respiratory tract and mediastinum, although the list is widening35. They are highly aggressive tumors that pursue an almost invariable fatal course. NUT carcinomas resemble small cell carcinomas of adults and predominately comprise sheets of small

4

undifferentiated cells. Well-defined foci of squamous differentiation often abruptly appear but may be absent in limited biopsies. NUT carcinomas typically show strong cytokeratin differentiation, but cytokeratin may also be expressed by a minority of Ewing sarcomas27. Diagnosis requires either genetic or immunohistochemical NUT confirmation, and other poorly differentiated pediatric carcinomas, including EBV-associated nasopharyngeal carcinoma, mucoepidermoid carcinoma, and germ cell tumors, should be excluded.

Extrarenal rhabdoid tumor

Rhabdoid tumors, known for their epithelial or rhabdomyosarcoma-like appearance, on occasion form sheets of lymphoma-like cells that can resemble Ewing sarcoma. The lymphomatoid pattern was first described by Weeks et al36 in their description of the histological diversity of rhabdoid tumors of the kidney. In these lesions, characteristic hyaline inclusions may be inconspicuous or absent. A similar phenomenon occurs with proximal variants of epithelioid sarcomas, which exhibit identical genetic changes. These lesions are generally recognized today by INI1-negativity, which separates them from the strongly positive Ewing sarcomas37. The list of INI1-negative neoplasms grows ever longer38 and now includes renal medullary carcinoma, extraskeletal myxoid chondrosarcoma, epithelioid MPNST, myoepithelial carcinoma, small cell hepatoblastoma, and poorly differentiated chordoma, none of which are often included in the differential diagnosis of Ewing sarcoma. A subset of synovial sarcomas shows reduced INI1 expression39, and this diagnosis should always be considered with fusion-negative Ewing-like sarcomas40.

Synovial sarcoma

In soft tissue pathology, synovial sarcoma is the great imitator. It occurs in a wide variety of locations and can exhibit features identical to Ewing sarcoma, including monomorphous round cell histology and CD99 positivity40. Thus, it should always be considered in EWS fusion-negative sarcomas resembling Ewing sarcoma, even if cytokeratin and EMA stains are negative40. An illustrative case with extensive bony invasion is presented, in which diagnosis was accomplished via FISH demonstration of SYT rearrangement (courtesy of Dr. Zhongxin Yu, University of Oklahoma). In the post-chemotherapy resection specimen, abundant glands were present.

Summary

Adoption of genetic techniques in diagnosis of pediatric sarcomas has improved recognition of many lesions but raised diagnostic quandaries in fusion-negative ones. Small cell sarcomas may resemble ARMS or Ewing sarcoma but lack characteristic fusions, creating diagnostic dilemmas. With fusion-negative ARMS, future protocols will rely on genetics rather than histology for stratification, although fusion-associated immunomarkers may come to be used as substitutes. For fusion-negative Ewing-like sarcomas, it is wise to do a full panel of diagnostic markers, including SYT FISH, to exclude other lesions. Undifferentiated pediatric soft tissue sarcomas lacking EWS fusions are treated as non-RMS soft tissue sarcomas; many contain a unique CIC-DUX4 fusion.

References

1 Bridge, J. A. et al. Novel genomic imbalances in embryonal rhabdomyosarcoma revealed by comparative genomic hybridization and fluorescence in situ hybridization: an intergroup rhabdomyosarcoma study. Genes, chromosomes & cancer 27, 337-344 (2000).

2 Xia, S. J., Pressey, J. G. & Barr, F. G. Molecular pathogenesis of rhabdomyosarcoma. Cancer biology & therapy 1, 97-104 (2002).

5

3 Folpe, A. L. MyoD1 and myogenin expression in human neoplasia: a review and update. Advances in anatomic pathology 9, 198-203 (2002).

4 Raney, R. B., Oberlin, O. & Parham, D. M. An English Translation of Joseph Luc Riopelle, MD, (Hotel-Dieu of Montreal), and Jean Paul Theriault (Hopital General of Verdun, Quebec, Canada): Sur une forme meconnue de sarcome des parties molles: le rhabdomyosarcome alveolaire (concerning an unrecognized form of sarcoma of the soft tissues: alveolar rhabdomyosarcoma). annales d'anatomie pathologique 1956;1:88-111. Pediatric and developmental pathology : the official journal of the Society for Pediatric Pathology and the Paediatric Pathology Society 15, 407-416, doi:10.2350/12-05-1192-MISC.1 (2012).

5 Tsokos, M. et al. Rhabdomyosarcoma. A new classification scheme related to prognosis. Archives of pathology & laboratory medicine 116, 847-855 (1992).

6 Newton, W. A., Jr. et al. Classification of rhabdomyosarcomas and related sarcomas. Pathologic aspects and proposal for a new classification--an Intergroup Rhabdomyosarcoma Study. Cancer 76, 1073-1085 (1995).

7 Raney, R. B. et al. Results of the Intergroup Rhabdomyosarcoma Study Group D9602 protocol, using vincristine and dactinomycin with or without cyclophosphamide and radiation therapy, for newly diagnosed patients with low-risk embryonal rhabdomyosarcoma: a report from the Soft Tissue Sarcoma Committee of the Children's Oncology Group. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 29, 1312-1318, doi:10.1200/JCO.2010.30.4469 (2011).

8 Ognjanovic, S., Linabery, A. M., Charbonneau, B. & Ross, J. A. Trends in childhood rhabdomyosarcoma incidence and survival in the United States, 1975-2005. Cancer 115, 4218-4226, doi:10.1002/cncr.24465 (2009).

9 Davicioni, E. et al. Molecular classification of rhabdomyosarcoma--genotypic and phenotypic determinants of diagnosis: a report from the Children's Oncology Group. The American journal of pathology 174, 550-564, doi:10.2353/ajpath.2009.080631 (2009).

10 Davicioni, E. et al. Identification of a PAX-FKHR gene expression signature that defines molecular classes and determines the prognosis of alveolar rhabdomyosarcomas. Cancer research 66, 6936-6946, doi:10.1158/0008-5472.CAN-05-4578 (2006).

11 Williamson, D. et al. Fusion gene-negative alveolar rhabdomyosarcoma is clinically and molecularly indistinguishable from embryonal rhabdomyosarcoma. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 28, 2151-2158, doi:10.1200/JCO.2009.26.3814 (2010).

12 Sumegi, J. et al. Recurrent t(2;2) and t(2;8) translocations in rhabdomyosarcoma without the canonical PAX-FOXO1 fuse PAX3 to members of the nuclear receptor transcriptional coactivator family. Genes, chromosomes & cancer 49, 224-236, doi:10.1002/gcc.20731 (2010).

13 Barr, F. G. et al. Genetic heterogeneity in the alveolar rhabdomyosarcoma subset without typical gene fusions. Cancer research 62, 4704-4710 (2002).

14 Wachtel, M. et al. Gene expression signatures identify rhabdomyosarcoma subtypes and detect a novel t(2;2)(q35;p23) translocation fusing PAX3 to NCOA1. Cancer research 64, 5539-5545, doi:10.1158/0008-5472.CAN-04-0844 (2004).

15 Mentzel, T. & Katenkamp, D. Sclerosing, pseudovascular rhabdomyosarcoma in adults. Clinicopathological and immunohistochemical analysis of three cases. Virchows Archiv : an international journal of pathology 436, 305-311 (2000).

16 Folpe, A. L., McKenney, J. K., Bridge, J. A. & Weiss, S. W. Sclerosing rhabdomyosarcoma in adults: report of four cases of a hyalinizing, matrix-rich variant of rhabdomyosarcoma that may be confused with osteosarcoma, chondrosarcoma, or angiosarcoma. The American journal of surgical pathology 26, 1175-1183 (2002).

6

17 Chiles, M. C. et al. Sclerosing rhabdomyosarcomas in children and adolescents: a clinicopathologic review of 13 cases from the Intergroup Rhabdomyosarcoma Study Group and Children's Oncology Group. Pediatric and developmental pathology : the official journal of the Society for Pediatric Pathology and the Paediatric Pathology Society 7, 583-594, doi:10.1007/s10024-004-5058-x (2004).

18 Gonzalez-Crussi, F. & Black-Schaffer, S. Rhabdomyosarcoma of infancy and childhood. Problems of morphologic classification. The American journal of surgical pathology 3, 157-171 (1979).

19 Nishio, J. et al. Use of a novel FISH assay on paraffin-embedded tissues as an adjunct to diagnosis of alveolar rhabdomyosarcoma. Laboratory investigation; a journal of technical methods and pathology 86, 547-556, doi:10.1038/labinvest.3700416 (2006).

20 Wachtel, M. et al. Subtype and prognostic classification of rhabdomyosarcoma by immunohistochemistry. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 24, 816-822, doi:10.1200/JCO.2005.03.4934 (2006).

21 Parham, D. M. et al. Correlation between histology and PAX/FKHR fusion status in alveolar rhabdomyosarcoma: a report from the Children's Oncology Group. The American journal of surgical pathology 31, 895-901, doi:10.1097/01.pas.0000213436.99492.51 (2007).

22 Wexler, L. H. & Ladanyi, M. Diagnosing alveolar rhabdomyosarcoma: morphology must be coupled with fusion confirmation. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 28, 2126-2128, doi:10.1200/JCO.2009.27.5339 (2010).

23 Anderson, J. R. et al. Fusion-negative alveolar rhabdomyosarcoma: modification of risk stratification is premature. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 28, e587-588; author reply e589-590, doi:10.1200/JCO.2010.30.5466 (2010).

24 Ellison, D. A., Parham, D. M., Bridge, J. & Beckwith, J. B. Immunohistochemistry of primary malignant neuroepithelial tumors of the kidney: a potential source of confusion? A study of 30 cases from the National Wilms Tumor Study Pathology Center. Human pathology 38, 205-211, doi:10.1016/j.humpath.2006.08.026 (2007).

25 Barr, F. G. & Womer, R. B. Molecular diagnosis of ewing family tumors: too many fusions... ? The Journal of molecular diagnostics : JMD 9, 437-440, doi:10.2353/jmoldx.2007.070080 (2007).

26 Bridge, R. S., Rajaram, V., Dehner, L. P., Pfeifer, J. D. & Perry, A. Molecular diagnosis of Ewing sarcoma/primitive neuroectodermal tumor in routinely processed tissue: a comparison of two FISH strategies and RT-PCR in malignant round cell tumors. Modern pathology : an official journal of the United States and Canadian Academy of Pathology, Inc 19, 1-8, doi:10.1038/modpathol.3800486 (2006).

27 Folpe, A. L. et al. Morphologic and immunophenotypic diversity in Ewing family tumors: a study of 66 genetically confirmed cases. The American journal of surgical pathology 29, 1025-1033 (2005).

28 Tornoczky, T., Semjen, D., Shimada, H. & Ambros, I. M. Pathology of peripheral neuroblastic tumors: significance of prominent nucleoli in undifferentiated/poorly differentiated neuroblastoma. Pathology oncology research : POR 13, 269-275, doi:PAOR.2007.13.4.0269 (2007).

29 Abramowsky, C. R., Katzenstein, H. M., Alvarado, C. S. & Shehata, B. M. Anaplastic large cell neuroblastoma. Pediatric and developmental pathology : the official journal of the Society for Pediatric Pathology and the Paediatric Pathology Society 12, 1-5, doi:10.2350/07-04-0255.1 (2009).

30 Park, S. J. et al. Detection of bone marrow metastases of neuroblastoma with immunohistochemical staining of CD56, chromogranin A, and synaptophysin. Applied immunohistochemistry & molecular morphology : AIMM / official publication of the Society for Applied Immunohistochemistry 18, 348-352, doi:10.1097/PAI.0b013e3181d2ed4c (2010).

7

31 Lin, O., Filippa, D. A. & Teruya-Feldstein, J. Immunohistochemical evaluation of FLI-1 in acute lymphoblastic lymphoma (ALL): a potential diagnostic pitfall. Applied immunohistochemistry & molecular morphology : AIMM / official publication of the Society for Applied Immunohistochemistry 17, 409-412, doi:10.1097/PAI.0b013e3181972b6d (2009).

32 Kawamura-Saito, M. et al. Fusion between CIC and DUX4 up-regulates PEA3 family genes in Ewing-like sarcomas with t(4;19)(q35;q13) translocation. Human molecular genetics 15, 2125-2137, doi:10.1093/hmg/ddl136 (2006).

33 Italiano, A. et al. High prevalence of CIC fusion with double-homeobox (DUX4) transcription factors in EWSR1-negative undifferentiated small blue round cell sarcomas. Genes, chromosomes & cancer 51, 207-218, doi:10.1002/gcc.20945 (2012).

34 French, C. A. et al. BRD-NUT oncoproteins: a family of closely related nuclear proteins that block epithelial differentiation and maintain the growth of carcinoma cells. Oncogene 27, 2237-2242, doi:10.1038/sj.onc.1210852 (2008).

35 French, C. A. et al. Midline carcinoma of children and young adults with NUT rearrangement. Journal of clinical oncology : official journal of the American Society of Clinical Oncology 22, 4135-4139, doi:10.1200/JCO.2004.02.107 (2004).

36 Weeks, D. A., Beckwith, J. B., Mierau, G. W. & Luckey, D. W. Rhabdoid tumor of kidney. A report of 111 cases from the National Wilms' Tumor Study Pathology Center. The American journal of surgical pathology 13, 439-458 (1989).

37 Hoot, A. C., Russo, P., Judkins, A. R., Perlman, E. J. & Biegel, J. A. Immunohistochemical analysis of hSNF5/INI1 distinguishes renal and extra-renal malignant rhabdoid tumors from other pediatric soft tissue tumors. The American journal of surgical pathology 28, 1485-1491 (2004).

38 Hollmann, T. J. & Hornick, J. L. INI1-deficient tumors: diagnostic features and molecular genetics. The American journal of surgical pathology 35, e47-63, doi:10.1097/PAS.0b013e31822b325b (2011).

39 Arnold, M. A. et al. A unique pattern of INI1 immunohistochemistry distinguishes synovial sarcoma from its histologic mimics. Human pathology, doi:10.1016/j.humpath.2012.08.014 (2012).

40 Folpe, A. L., Schmidt, R. A., Chapman, D. & Gown, A. M. Poorly differentiated synovial sarcoma: immunohistochemical distinction from primitive neuroectodermal tumors and high-grade malignant peripheral nerve sheath tumors. The American journal of surgical pathology 22, 673-682 (1998).

1/22/2013

Paediatric small round cell sarcomas – an update

David Parham , Oklahoma University Health Science Center,

Oklahoma City, OK

1/22/2013

Pediatric round cell sarcomas: classical types and differential diagnosis

• Rhabdomyosarcoma

• Ewing sarcoma (including PNET)

• Neuroblastoma

• Lymphoma/leukemia

• Small cell osteosarcoma

• Mesenchymal chondrosarcoma

• Undifferentiated sarcoma

1/22/2013

Pediatric small cell sarcomas – newer entities in the differential diagnosis

• Desmoplastic small round cell tumor

• Extrarenal rhabdoid tumor

• BRD/NUT carcinoma

1/22/2013

Diagnosis in the past decade: a shift from immunohistochemistry to genetics

1/22/2013

What do I do when the genetic test for a small round cell tumor is negative?

1/22/2013

Two major genetic tests for round cell sarcomas

• FOXO1 FISH: alveolar rhabdomyosarcoma (ARMS)

• EWSR1 FISH: Ewing sarcoma

1/22/2013

Rhabdomyosarcoma: the most frequent pediatric soft tissue sarcoma

Sarcoma incidence per million (source: WHO)

Rhabdomyosarcomas

Non-rhabdos

1/22/2013

Rhabdomyosarcoma: assignment of therapy

• Low risk

• Intermediate risk

• High risk

• Combination of stage (clinical), group (pathological), age, and histological subclassification

1/22/2013

Histology: the historical gold standard for classification

Classical ARMS as described by Riopelle and Theriault

1/22/2013

A 1980s change in paradigm: solid variant

1/22/2013

Current criteria for classification: the International Pediatric Sarcoma Classification

1/22/2013

Diagnosis of ARMS: a high stakes game

• Not treated on low risk protocols

• Worse outcome with high risk protocols

• May be aggressively treated with intermediate risk protocol.

• No truly significant improvements in therapy since 1970s (compared to embryonal RMS [ERMS]).

1/22/2013

Outcome in ARMS vs ERMS

Metastatic RMS, IRSG study IV

ERMS

ARMS

1/22/2013

RMS diagnosis: MyoD and myogenin

• Transcription factors that bind DNA and initiate myogenesis

• Nuclear staining is highly sensitive and specific for rhabdomyogenesis

• Cytoplasmic staining is non-specific

1/22/2013

Myogenin staining varies according to subtype

ERMS ARMS

1/22/2013

Another change in paradigm: genetics

1/22/2013

Fusion distribution in ARMS

• PAX3-FOXO1: 50%

• PAX7-FOXO1: 20%

• PAX fusion negative: 30%

1/22/2013

The question of the past decade: Does ARMS behavior reflect

histology or genetics?

1/22/2013

1/22/2013

Array studies: RMS gene expression clusters by fusion positivity,

not by histology

Red: ARMS; Blue: ERMS; Green: fusion-negative; Yellow: fusion-positive

1/22/2013

Survival correlates with fusion

• 5 year overall survival:

– ERMS: 75%

– Fusion negative ARMS: 75%

– Fusion positive ARMS: 25%

• Metastatic rate:

– ERMS: 15%

– Fusion negative ARMS: 20%

– Fusion positive ARMS: 45%

1/22/2013

Conclusion

• Fusion-negative ARMS is clinically and molecularly indistinguishable from ERMS

• Molecular results confirmed in four independent international studies

• Question: does diagnosis of ARMS depend on fusion, histology, or both?

1/22/2013

Possible sources of fusion-negative ARMS

• Alternate fusions not detected with standard tests

• New histological entities with similar features

• “Mixed ARMS-ERMS”: lesions with features of both ARMS and ERMS

• Embryonal rhabdomyosarcoma look-alikes

1/22/2013

Alternate fusions not detected with standard tests

• PAX3-AFX1

• PAX3-NCOA1

• PAX3-NCOA2

• All involve forkhead type proteins

• All negative with FOXO1 FISH

• Rare; <1% of ARMS

1/22/2013

New histopathological entity with similar features:

Sclerosing RMS

1/22/2013

Sclerosing RMS in adults: Folpe et al., Am J Surg Pathol 2002;26:1175.

• Four cases of adult RMS characterized by the production of abundant, hyalinized osteoid, or chondroid-like matrix

• Initially considered probable variants of ARMS because of occurrence in extremities of adults, microalveolar pattern, and primitive round cell morphology

1/22/2013

Pediatric sclerosing RMS: Chiles et al. Pediatr Devel Pathol 2004; 7:583-594.

• Thirteen patients had features of sclerosing RMS.

• 9 had been diagnosed as ARMS, 3 as ERMS, and 1 as spindle cell RMS.

• Only 1 of 5 tested cases had a PAX fusion demonstrated.

1/22/2013

Mixed ARMS: tumors with features of both ERMS and ARMS

1/22/2013

International Classification (1995)

• “Traditionally, these tumors have been called mixed alveolar/embryonal RMS or embryonal if the alveolar pattern was less prominent than the embryonal one.

• In the proposed classification, any alveolar features are sufficient to classify the tumor as alveolar RMS.”

1/22/2013

Relative proportions of fusion-status vs IRSG study period

Per

cen

t

1/22/2013

Percentage of ARMS increase, COG report

Year 1980-91

1992-97 1998-2004

Number of

RMS

2003 864 968

Number of

ARMS

511 262 400

Per cent 25.5 30 41

1/22/2013

Protocol adjustments

• IRS3: 50% ARMS required for diagnosis

• IRS4: ARMS diagnosis not used to stratify therapy.

• Post-IRS4: Any ARMS foci make diagnosis

• After 2005: Protocols revised to IRS3 standard.

1/22/2013

ERMS that look like ARMS Dense ERMS: The International Classification

• “The histologic pattern of embryonal RMS is predominantly that of a moderately cellular tumor with loose myxoid stroma, although some dense areas may occur frequently. “

• “Some tumors may consist exclusively of fields of closely packed cells. “

Dense ERMS pattern

1/22/2013

Solid ARMS: International Classification diagnosis

• “The ”solid” variant of alveolar RMS grows as solid masses of closely aggregated cells, with no or scarcely discernible alveolar arrangement.”

• “No marked compartmentalization and no distinct alveolar configuration is apparent.”

“Solid” variant ARMS

1/22/2013

A very subtle distinction

1/22/2013

Myogenin staining

ERMS ARMS

1/22/2013

Conclusions

• ARMS can easily be overdiagnosed. – Myogenin staining can be helpful.

• Solid variant “ARMS” are often fusion-negative.

• The clinical significance of fusion-negative ARMS remains to be determined, but current studies show no differences from ERMS

• Future COG stratification will be based on fusion status.

1/22/2013

What is fusion-negative “Ewing sarcoma”?

• Rhabdomyosarcoma, neuroblastoma, and lymphoma should be excluded

• Undifferentiated sarcoma

• BRD4-NUT carcinoma

• Extrarenal rhabdoid tumor

• Synovial sarcoma

1/22/2013

Ewing sarcoma histology

Classical Ewing sarcoma Primitive neuroectodermal tumor

1/22/2013

Ewing sarcoma genetics

• t(11;22)(q24;q12): EWSR1-FLI1 fusion

• t(21;22): EWSR1-ERG fusion

• Other EWSR1 fusion partners FEV, E1AF, ETV1

– New non-ETS genes POU5F1, ZNF278, SP3

• Rare FUS-FEV or FUS-ERG fusions

– Use FUS FISH probe in typical cases with negative EWSR1 fusion results

1/22/2013

Undifferentiated sarcoma, round cell type

• Negative for all phenotypic markers – Focal neural positivity?

• Treated as non-rhabdo sarcoma by COG

• High prevalence of new fusion type: CIC-DUX4 – t(4;19)(q35;13.1)

– Memorial study: CIC rearrangements found in 15/22 EWSR1 fusion-negative cases • CD99 diffuse positive, patchy positive, or negative

• Mostly negative for other markers

1/22/2013

BRD4-NUT carcinoma

• Predominately affects young patients, but wide age range

• Usually involves midline structures: upper respiratory tract, mediastinum

• Characterized by t(15;19)(q13;p13.1) translocation

• Almost invariably fatal course

1/22/2013

BRD4-NUT carcinoma

• Resembles small cell carcinoma.

• May not show obvious epithelial differentiation on H and E stain, but should be cytokeratin-positive (as are 15% of Ewing sarcoma).

• Squamous differentiation is often abrupt.

1/22/2013

Undifferentiated BRD4-NUT carcinoma

Courtesy of Sara Vargas, MD, Boston Children’s Hospital

1/22/2013

Differentiating BRD4-NUT carcinoma

Courtesy of Sara Vargas, MD, Boston Children’s Hospital

1/22/2013

BRD4-NUT carcinoma: cytokeratin

Courtesy of Sara Vargas, MD, Boston Children’s Hospital

1/22/2013

BRD4-NUT carcinoma NUT stain

Courtesy of Sara Vargas, MD, Boston Children’s Hospital

1/22/2013

Also consider:

• EBV-associated carcinoma (nasopharyngeal carcinoma)

• Mucoepidermoid carcinoma

• Keratin-positive carcinomas, such as rhabdoid tumor

• Poorly differentiated germ cell tumor

1/22/2013

Extrarenal rhabdoid tumor

• Wide variety of primary sites, including soft tissue and skin

• Primarily affects young children, with overlap with epithelioid sarcoma in older ones

• Diverse histologies, including “lymphomatoid” pattern described by Weeks and Beckwith.

1/22/2013

Extrarenal rhabdoid tumor

1/22/2013

INI1

1/22/2013

Tumors with negative INI1

• Rhabdoid tumor

• Epithelioid sarcoma

• Renal medullary carcinoma

• Extraskeletal myxoid chondrosarcoma

• Epithelioid MPNST

• Myoepithelial carcinoma

• Small cell hepatoblastoma

• Poorly differentiated chordoma

• Reduced in synovial sarcomas

1/22/2013

Final illustrative case Courtesy of Zhongxin Yu, MD

University of Oklahoma

• 14 years old with left hip pathological fracture

1/22/2013

1/22/2013

1/22/2013

CD99: negative

1/22/2013

INI-1: negative (occasional positive)

1/22/2013

Vimentin: negative

1/22/2013

Other IHC

• Lymphoma markers: negative – (CD3, CD5, CD20, CD43, CD79a, PAX5, TdT, CD117,

CD34)

• Myeloperoxidase: weakly positive, focal

• Myogenin and desmin: negative

• S-100: negative

• NSE and synaptophysin: negative

• CD68, CD163: negative

• EMA and pan-cytokeratin: negative

1/22/2013

Preliminary Diagnosis

• Small round blue cell tumor of the bone

– ? Ewing’s sarcoma

• Pending EWS FISH result

– ? Ewing’s – like small cell osteosarcoma

• Pending FISH to rule out Ewing’s sarcoma

• ?rebiopsy to get more tissue with osteoid

1/22/2013

Final diagnosis

• FISH is negative for EWSR1 rearrangement

• FISH is positive for SYT(SS18) rearrangement

• Final diagnosis—Synovial sarcoma

1/22/2013

Resection specimen histology (post chemotherapy)

1/22/2013

Summary

• Genetics improves diagnosis of many pediatric sarcomas but raises diagnostic quandaries in others.

• Lesions that resemble ARMS or Ewing sarcoma may lack fusions, creating diagnostic problems and delay.

1/22/2013

Summary

• Future protocols will rely on genetics of RMS rather than histology for stratification

– Fusion-associated immunomarkers may be used as substitutes.

• Fusion-negative Ewing-like sarcomas require a full panel of diagnostic markers, including SYT FISH.

• Undifferentiated pediatric sarcomas are treated as non-RMS soft tissue sarcomas

– Many contain a unique CIC-DUX4 fusion.