* We describe an ankle tumor arising in a 16-year-old girl. The tumor demonstrated histology typical of a malignant peripheral nerve sheath tumor (MPNST), but exhibited a variant form of the (X;18) translocation associated with synovial sarcoma. Immunohistochemical stains were positive for vimentin, CD57, collagen type IV, and Bcl-2. Routine and molecular cytogenetic studies showed an unbalanced 3-way chromosomal translocation that involved chromosomes X, 18, and 1. Electron microscopic findings were noncontributory. This unusual tumor raises the following questions and possibilities: (1) As the t(X;18) suggests, could this tumor be a monophasic synovial sarcoma with the histologic features of an MPNST? (2) Or, as the histology suggests, is this tumor an MPNST that has a t(X;18)? (3) Finally, could MPNST histology, a t(X;18), and no defining immunohistochemical or electron microscopic features represent an as yet unrecognized part of a spectrum that spans from synovial sarcoma to MPNST or other spindle cell tumors?
(Arch Pathol Lab Med. 2000;124:864-867)
Synovial sarcoma (SS) is the most common nonrhabdomyosarcoma soft tissue tumor in children. Synovial sarcoma accounts for approximately 20% and malignant peripheral nerve sheath tumor (MPNST) for approximately 10% of all nonrhabdomyosarcoma soft tissue tumors in childhood. The extremities are the most common sites for both tumor types. Malignant peripheral nerve sheath tumor is believed to be neural in origin, whereas the origin of SS continues to be debated. While MPNSTs by definition arise from a component of the peripheral nerve sheath, SSs are also documented to arise from peripheral nerves. While SS may have an anatomic relationship to a joint, the tumor has never been shown to arise from the synovium. Investigations of SS and MPNST using immunohistochemical and electron microscopic techniques have suggested that SS may be a carcinosarcoma of soft tissues or a tumor that has the potential to show neuroectodermal or nerve sheath differentiation.1-3
The histologic profiles of both MPNST and SS (monophasic fibrous type) are characterized by a spindle cell proliferation similar to fibrosarcoma. The characteristic chromosomal change of SS is t(X;18)(p1.2;q11.2), which is present in about 90% of cases. This translocation results in the fusion of the SYT gene on chromosome 18 with the SSX gene of the X chromosome, producing the SYT/SSX chimeric transcript.
Both neoplasms will recur locally if not completely excised. Synovial sarcoma, unlike most nonrhabdomyosarcoma soft tissue sarcomas, frequently spreads to regional lymph nodes.
REPORT OF A CASE
A 16-year-old white girl initially presented with a 4-month history of an ankle mass. Four months after surgical excision, the mass recurred. It appeared to be encapsulated and to originate from the medial plantar nerve. The margins of the resection were positive for tumor. Computed tomographic scan of the chest revealed 2 lung nodules (3 mm and 1-2 mm). A below-the-knee amputation was performed.
PATHOLOGIC FINDINGS
Gross Pathology
The re-excision specimen consisted of a skin ellipse with an attached deep mass. The mass measured 15.2 x 4.0 x 3.0 cm. Cross section of the mass revealed a tan-white, bulging cut surface with a whorled appearance and a semitransparent capsule.
Microscopic Histology
The tissue had a plexiform pattern (Figure 1, A) with alternating hypercellularity and hypocellularity (Figure 1, B). The areas of hypercellularity consisted of spindle cells that streamed in parallel. The cells had bland nuclei with a slightly wavy appearance and eosinophilic cytoplasm. The highest mitotic rate was 6 mitoses per 10 high-powered fields. Areas of hypocellularity contained individual spindle cells in a myxoid stroma. The histopathology was consistent with MPNST.
Electron Microscopy
The ultrastructural findings showed fibroblast-like cells with nonspecific features that contained rough endoplasmic reticulum. Intercellular collagen was present; however, no elongated cytoplasmic processes, basal lamina, or epithelial-like junctions were seen.
Immunohistochemistry and Histochemistry
Immunohistochemical stains for vimentin (Dako Corporation, Carpinteria, Calif), CD57 (Becton Dickinson, San Jose, Calif), collagen type IV (Dako), and Bcl-2 (Dako) were positive, while those for epithelial membrane antigen (Dako), S100 (Dako), HMB-45 (Dako), muscle-specific actin (Dako), desmin (Dako), and CD34 (Becton Dickinson) were negative. A reticulin stain showed intercellular fiber deposition without evidence of epithelial differentiation.
Cytogenetics
Fresh tumor was disaggregated with scalpel blades and collagenase, cultured on 22-mm coverslips in minimal essential medium, harvested on days 4 and 5, and GTG-banded. Routine cytogenetic analysis showed an unbalanced 3-way translocation between chromosomes X, 18, and 1. The karyotype (Figure 2) shows a derivative chromosome 1 (arrow) with Xp (short arm of chromosome X) translocated to it. The distal part of lp is translocated to the derivative chromosome 18 (small arrowhead). One chromosome X (large arrowhead) is translocated with 18q (long arm of chromosome 18), that is, t(X;18). The complete karyotype is as follows: 43,X,der(X)t(X;18)(p11.2; qll.2),der(1)t(X;1)(p11.2; p13),-3,add(5)(qll.2),-10,-18,der(18)t(1;18)(p32;ql 1.2).
Molecular Cytogenetics: Fluorescence In Situ
Hybridization
To confirm the X;18 translocation, an Oncor whole chromosome 18 paint probe (Oncor, Gaithersburg, Md) was hybridized to a previously GTG-banded metaphase cell (Figure 3, A). The large arrowhead in Figure 3, A indicates the derivative chromosome X, t(X;18); the small arrowhead indicates the derivative chromosome 18, t(1;18); and an arrow indicates the derivative chromosome 1, t(X;1). Figure 3, B shows the same metaphase after hybridization. Arrows indicate the same chromosomes as described for the GTG-banded image. Bright signal is present on both the der(18)t(1;18) and der(X)t(X;18), but not on the der(1)t(X;1).
COMMENT
Spindle cell tumors can be difficult to differentiate from one another by histologic pattern alone. Immunohistochemical stains and electron microscopic studies are often employed to identify features characteristic of a particular tumor. Unfortunately, some soft tissue and bone tumors (eg, spindle cell and round cell tumors) still may not reveal distinguishing features, even by these methods. In those cases, cytogenetic analysis, molecular studies, or both may reveal the diagnosis by demonstrating a characteristic chromosomal abnormality or molecular rearrangement.
In our case, the patients age, tumor location, symptomatology, and features of the gross tumor failed to help distinguish between SS and MPNST. The patient had no history of neurofibromatosis I, a feature that would favor MPNST The negative immunostaining for keratin and epithelial membrane antigen, while not typical for SS, does not rule out SS. The S100, CD57, and collagen type IV stains may be positive in both MPNST and SS.4 Strong and diffuse Bcl-2 staining, as seen in this case, is more characteristic of SS than MPNST.5 The electron microscopic findings were not helpful in confirming either diagnosis. The histology, however, was characteristic for MPNST and did not suggest SS. Without cytogenetic analysis, SS as a diagnosis would not have been considered.
The finding of a variant of t(X;18) that is characteristic of approximately 90% of SS cases studied led to a review of this case and the literature. While re-examination of the tumor tissue and immunohistochemical stains did not reveal characteristic features of SS, there is precedent in the literature for this type of case. Mandahl et al6 and TurcCarel et al7 reported a fibrosarcoma and a recurrent malignant fibrous histiocytoma, respectively, with t(X;18). Faria et al8 described cases that were initially considered to be renal embryonal sarcomas, but that subsequently were thought to represent SS after the SSX/SYT gene fusion transcript was detected. Balazs et al9 reported finding SSX/SYT transcripts in 4 MPNSTs by reverse-transcriptase polymerase chain reaction. Hiraga et al,10 in a study of 84 soft tissue tumors analyzed by reverse-transcriptase polymerase chain reaction, found 10 of 10 SSs, 1 of 11 MPNSTs, and 3 of 5 unclassified tumors with SSX/SYT gene fusion transcripts.
SSX/SYT gene fusion results from the t(X;18)(pll.2;qll.2). The translocation fuses the SYT gene from the chromosome 18 long arm to either of 2 highly homologous genes on the chromosome X short arm, SSX1 or SSX2. This gene fusion and the t(X;18) have become the sine qua non for SS, leading investigators to conclude a tumor represents SS when the fusion product or the translocation is present. If this is indeed the case, then we need to consider that SS may have a larger histologic spectrum than previously thought. Within the recognized spectrum, there are biphasic, monophasic, and poorly differentiated forms of SS. There are reports of neural features in SS, SS arising within nerves, epithelial components in MPNST, as well as a blurring of immunohistochemical markers between spindle cell tumors. If some spindle cell tumors with histologic features of MPNST are in fact SS at one end of a spectrum, this supports a mesenchymal cell of origin with potential for differentiation in several directions, that is, epithelial, stromal, or neural. If this is true, cytogenetic or molecular analysis may be required to definitively differentiate these spindle cells tumors.
Karyotyping of MPNST has not shown a recurrent chromosomal abnormality. Malignant peripheral nerve sheath tumors most often have a complex karyotype with multiple chromosomal aberrations. Our case, to our knowledge, is the first MPNST-like tumor with a t(X;18) documented by karyotypic analysis.
In summary, we describe a tumor that has histology consistent with MPNST and a cytogenetic abnormality diagnostic of SS. Data derived from this case, as well as from 5 other reported cases of MPNST with the SSX/SYT gene fusion of t(X;18),9,10 suggest the spectrum of SS is broader than previously recognized. These findings warrant further cytogenetic and molecular investigation of spindle cell tumors to elucidate this spectrum and to possibly solve the SS cell-of-origin enigma.
References
1. Folpe AL, Schmidt RA, Chapman D, Gown AM. Poorly differentiated synovial sarcoma. Am J Surg Pathol. 1998;22:673-682.
2. Lopes JM, Bjerkehagen B, Sobrinho-Simoes M, Nesland JM. The ultrastructural spectrum of synovial sarcomas: a study of the epithelial type differentiation of primary tumors, recurrences, and metastases. Ultrastruct Pathol. 1993;17:137151.
3. Noguera R, Navarro S, Cremades A, et al. Translocation (X;18) in a biphasic synovial sarcoma with morphologic features of neural differentiation. Diagn Molec Pathol. 1998;7:16-23.
4. Leong AS-Y, Vinyuvat S, Suthipintawong C, Leong FJ. Patterns of basal lamina immunostaining in soft-tissue and bony tumors. Appl Immunohistochem. 1997;5:1-7.
5. Suster S, Fisher C, Moran CA. Expression of bcl-2 oncoprotein in benign and malignant spindle cell tumors of soft tissue, skin, serosal surfaces, and gastrointestinal tract. Am J Surg PathoL 1998;22:863-872.
6. Mandahl N, Heim S, Arheden K, Rydholm A, Willen H, Mitelman F. Multiple karyotypic rearrangements, including t(X;18)(pll;ql I) in a fibrosarcoma. Cancer Genet Cytogenet 1988;30:323-327.
7. Turc-Carel C, Dal Cin P, Limon J, et al. Involvement of chromosome X in primary cytogenetic change in human neoplasia: non-random translocation in synovial sarcoma. Proc Natl Acad Sci U S A. 1987;84:1981-1985.
8. Faria P, Argani P, Epstein JI, Reuter VE, Beckwith B, Ladanyi M. Primary synovial sarcoma of the kidney: a molecular reappraisal of a subset of so-called embryonal renal sarcoma [abstract]. Mod PathoL 1999;12:94A.
9. Balazs L, Parham D, Shurtleff S, Pappo A, Downing J. Identification of the t(X;18) as a diagnostic tool in the differential diagnosis of synovial sarcomas versus peripheral nerve sheath tumors [abstract]. Mod PathoL 1997;10:8A.
10. Hiraga H, Nojima T, Abe S, et al. Diagnosis of synovial sarcoma with the reverse transcriptase-polymerase chain reaction: analyses of 84 soft tissue and bone tumors. Diagn Molec Pathol. 1998;7:102-110.
Russell Vans MD; David A. Biddle, MD; Wilbur R. Harrison, MS; Kent Heck, MD; Linda D. Cooley, MD
Accepted for publication October 18, 1999.
From The University of Texas Medical School, Department of Pathology and Laboratory Medicine, Houston, Tex.
Presented as a poster at the Texas Society of Pathologists, Dallas, Tex, January 29-30, 1999.
Reprints: Linda D. Cooley, MD, The University of Texas Medical School, Department of Pathology, 6431 Fannin, Room 2.292, Houston, TX 77030.
Copyright College of American Pathologists Jun 2000
Provided by ProQuest Information and Learning Company. All rights Reserved