Synovial cell sarcoma in a dog: A misnomer Cytologic and histologic findings and review of the literature

Transcription

1 DOI: /vcp REVIEW ARTICLE Synovial cell sarcoma in a dog: A misnomer Cytologic and histologic findings and review of the literature Paola Monti Darren Barnes Anna M. Adrian Roberta Rasotto Dick White Referrals, DWR Diagnostics, Cambridgeshire, UK Correspondence P. Monti, Dick White Referrals, DWR Diagnostics, Cambridgeshire, UK. paolamonti@hotmail.com Abstract A 4-year-old Irish Setter was presented with a history of progressive left pelvic limb lameness. Orthopedic examination revealed pain on manipulation of the left stifle. Radiographs showed an osteolytic lesion in the subchondral bone of the medial tibial condyle. Fine-needle aspirates were taken, and cytology revealed numerous cohesive clusters of plump, oval to spindloid cells often with perivascular distribution and moderate cellular atypia. A diagnosis of sarcoma was made with synovial cell sarcoma (SCS) and histiocytic sarcoma being the 2 main differentials. Histopathology confirmed the diagnosis of sarcoma and provided the same differentials. All neoplastic cells were positive for vimentin, and approximately 5% of them also stained with pan-cytokeratin using immunohistochemical staining methods. Neoplastic cells did not express CD18. The combination of this immunohistochemical profile and cell morphology was consistent with an SCS. Synovial cell sarcoma is a rare and poorly understood canine tumor entity. This is the first extensive description of the cytologic features of this neoplasm. The literature was also reviewed, focusing on comparative aspects of dogs and people, with a special emphasis on the cell of origin and diagnostic tools. Controversies regarding the nomenclature of this tumor are also presented. The authors propose a new term (cytokeratin-positive joint-associated sarcoma) for addressing this neoplasm until the cell of origin of this tumor is elucidated. 1 CASE PRESENTATION A 4-year-old, male entire, Irish Setter was referred for evaluation of a 2-week history of progressive left pelvic limb lameness. At the time of assessment, the dog was receiving nonsteroidal anti-inflammatory treatment (Rimadyl; Zoetis, NJ, USA), Gabapentin (Milpharm Limited, South Ruislip, UK), and Tramadol (Summit Veterinary Pharmaceuticals Limited, Kidlington, Oxfordshire, UK). At presentation, he was markedly lame (8/10) on his left hind limb. Orthopedic examination revealed pain on manipulation of the left stifle, mild muscle atrophy affecting the left quadriceps, and moderate stifle effusion. There was soft-tissue swelling medial to the stifle that was initially suggestive of a buttress, as seen with cruciate disease. Neurological examination was unremarkable. No significant changes were seen on hematology and clinical chemistry other than a moderate elevation in CK consistent with muscular injury (697 IU/ L; RI: IU/L). Radiographs of the stifle revealed a well-circumscribed, predominantly osteolytic lesion (1.1 cm), surrounded by an illdefined zone of bone sclerosis affecting the subchondral bone of the medial tibial condyle, as well as increased intracapsular softtissue opacity within the femorotibial joint. Distally, on the medial aspect of the tibial metaphyseal region, there was a discrete zone of mildly irregular periosteal new bone formation and associated soft-tissue swelling. There was no evidence of osteophytosis (Figure 1). Ultrasound-guided aspirates of the lesion were taken. Slides were stained with a modified Wright-Giemsa stain. Vet Clin Pathol. 2018;47: wileyonlinelibrary.com/journal/vcp 2018 American Society for Veterinary Clinical Pathology 181

2 182 MONTI ET AL. A B FIGURE 1 Caudo-cranial view of the left stifle in a dog, medial is to the left. The red arrows indicate a well-circumscribed osteolytic lesion surrounded by sclerosis (green arrow) that is associated with a soft-tissue swelling (yellow arrow). Distally to this lesion is a mildly irregular periosteal reaction (blue arrow) The aspirates were highly cellular, and cytology revealed the presence of high numbers of atypical cells. They often showed a perivascular arrangement (Figure 2A,B), but were also occasionally found individually and frequently in large apparently cohesive clusters. Cells were associated with extracellular pink amorphous material in some areas (Figure 2C). When individualized, cells were plump to oval, with occasional small cytoplasmic projections. Anisokaryosis and anisocytosis were moderate. Cells were predominantly mononuclear and had medium-large, round-oval nuclei. Occasional multinucleated cells were observed. The chromatin was coarsely stippled with occasional and multiple small, round, and variably visible nucleoli. The cytoplasm was moderate, lightly to moderately basophilic, and contained sporadic small clear vacuoles or fine pink granules. A low number of mitoses were found. The cytologic interpretation was consistent with a sarcoma with a principal differential diagnosis of synovial cell sarcoma (SCS) or, less likely, histiocytic sarcoma. Following this cytologic diagnosis, abdominal ultrasound and thoracic radiography were performed for staging, which were normal. Five small biopsies were taken for histopathologic confirmation of sarcoma. Three of the 5 biopsies consisted of a neoplastic cell population arranged in sheets and short bundles, supported by a scant fibrovascular stroma. Rarely, neoplastic cells had a perivascular orientation. Cell morphology varied from oval-spindle in densely cellular areas to round-stellate in areas (Figure 3A) where cells were more widely separated by an edematous pale basophilic matrix. Cell borders were indistinct, and the cytoplasm was scant to moderate and C FIGURE 2 Fine-needle aspirates of a soft-tissue swelling at the level of the medial tibial metaphyseal region in a dog. Wright- Giemsa. A, 920 Objective, B, 950 Objective. C, Neoplastic cells are arranged in dense clusters, often in a perivascular fashion. When individualized, they are plump to oval with large, round to oval nuclei and scant to moderate, pale basophilic cytoplasm with illdefined margins. The cytoplasm contains occasional small punctate, clear vacuoles, 950 Objective. Neoplastic cells are occasionally associated with a small amount of pink amorphous material eosinophilic. Nuclei were round to oval with finely stippled chromatin and 1-2 nucleoli. Occasional binucleated and trinucleated neoplastic cells were observed. Anisocytosis and anisokaryosis were

3 MONTI ET AL. 183 moderate to marked. Mitoses were 2 per 10 HPFs. The other 2 biopsies were composed of granulation tissue and trabecular bone, the latter lined by increased numbers of osteoclasts within Howship s lacunae indicating bone resorption. Histopathology confirmed the cytologic diagnosis of sarcoma. Synovial cell sarcoma remained the primary differential diagnosis, followed by histiocytic sarcoma. For further confirmation, immunohistochemistry was performed with an antibody panel that included vimentin (Dako, Glostrup, Denmark), pan-cytokeratin (MNF116; Dako), CD18 (P. Moore/Leukocyte Antigen Biology lab, UC Davis), and smooth muscle actin (Dako). All of the neoplastic cells were strongly positive for vimentin and about 80% weakly stained for smooth muscle actin. Approximately 5% of the neoplastic cells strongly stained for pan-cytokeratin. Neoplastic cells did not express CD18. This staining pattern, with negative CD18 expression and positive pan-cytokeratin expression in small clusters of cells (Figure 2B), in conjunction with the cell morphology, was typical for an SCS. Although the cellular morphology suggested a monophasic SCS, the presence of pan-cytokeratin-positive epithelioid cells indicated a biphasic SCS. According to the grading system for canine SCS published by Vail et al, the tumor was considered low grade. 1 A After the diagnosis, leg amputation was elected. Following surgery, the dog received 4 doses of Doxorubicin (Pfizer, Walton Oaks, Surrey, UK) at 3 weekly intervals (30 mg/m 2 ). Ten months after diagnosis, the dog developed left forelimb lameness that showed an initial partial response to NSAIDs (Rimadyl; Zoetis). On physical examination, a swelling of all 3 remaining limbs was found with heat around the carpi and the remaining tarsus. Hematology and biochemistry were unremarkable except for a mild elevation in ALP (297 IU/L; RI: IU/L). Thoracic and limb radiographs and a thoracic CT scan were performed. Radiography revealed a polyostotic palisade-type periosteal new bone formation of the antebrachia, metacarpi, tibia/fibula, and tarsal bones consistent with hypertrophic osteopathy. On thoracic X-ray, a homogeneous soft-tissue mass (10 cm in diameter) was seen in the left caudal thorax, which on CT was visualized arising from the pleura and extending into the caudal mediastinum. Additionally, 2 smaller pleural masses, one in each hemithorax, and multiple pulmonary nodules were present. Ultrasound-guided fine-needle aspirates were obtained from the largest pleural mass, and cytology was consistent with metastasis of the previously diagnosed SCS. The cell morphology and arrangement were very similar to that of the primary mass. The dog was discharged on painkillers and NSAIDs and, shortly after, the owner elected euthanasia. 2 DISCUSSION B FIGURE 3 Synovial sarcoma in a dog. H&E, 940 Objective. A, Histologically, neoplastic cells vary from round to oval to spindle and have scant to moderate cytoplasm and round to oval nuclei. They are separated by an edematous matrix. B, Photomicrograph showing a small cluster of pan-cytokeratin-positive neoplastic cells. Pancytokeratin, 940 Objective The first aim of this paper was to provide a detailed cytologic description of an SCS in a dog, which, to the best authors knowledge, is scarce in veterinary publications. The most important cytologic features were the presence of numerous cells arranged in an epithelioid-type of clusters, while still showing individual cell morphology typical of a mesenchymal tumor. The second aim of this paper was to review the literature on canine SCS and compare it with the new discoveries on human synovial sarcoma. In veterinary medicine, the gold standard for the diagnosis of SCS is immunohistochemistry, which allows differentiation from other tumors that can develop in joints, including histiocytic sarcoma, synovial myxoma/myxosarcoma, and malignant fibrous histiocytoma. 2 The distinction between SCS and histiocytic sarcoma is particularly important, as these 2 tumors share many cytologic and histologic features, but have different biological behaviors and recommended treatments. 2 According to Craig, 2 the typical immunohistochemical profile of SCS in dogs is characterized by vimentin and pan-cytokeratin coexpression in a small percentage of cells, as observed in the case reported here. The use of CD18 (pan-leukocyte marker) is also considered crucial to rule out histiocytic sarcoma, especially in small biopsies where cells co-expressing pan-cytokeratin and vimentin might be few or not represented. Interestingly, Craig and colleagues showed that including CD18 in the diagnostic panel allowed 18 of 35 tumors (51.4%) originally diagnosed as SCS to be reclassified as

4 184 MONTI ET AL. histiocytic sarcoma; only 5 of the 35 tumors (14.3%) were identified to be true SCS by cell morphology and immunohistochemical staining, suggesting that SCS might not be the most common joint tumor in dogs as had been previously thought. 2 Canine SCS can be subclassified further using histopathology into either monophasic or biphasic forms according to the respective absence or presence of an epithelioid component. The monophasic variant is the most common form. 2 A histologic grading system was also published by Vail et al, but a subsequent study was unable to predict survival using this grading system. 1,3 Unfortunately, the study by Vail did not use CD18, and only 6 out of 20 cases were positive for both vimentin and pan-cytokeratin. The remaining 14 cases could have been histiocytic sarcomas or sarcomas of another origin; therefore, the prognostic utility of this grading system is now questioned. These findings could explain, in our case, the lack of correlation between the estimated low tumor grade and the aggressive biological behavior (intra-thoracic metastases that developed within 10 months of diagnosis). The pulmonary nodule, which based on CT scan was compatible with a metastasis, was likely responsible for the associated hypertrophic osteopathy. In human medicine, the current diagnostic gold standard for synovial sarcoma is the demonstration of a gene translocation specific for this tumor type. 4,5 This translocation (t(x;18)(p11;q11)) fuses the SYT gene from chromosome 18 to either one of 2 highly homologous genes at Xp11, SSX1, or SSX2, 6 and can be detected by reverse transcription polymerase chain reaction, fluorescence in situ hybridization, or traditional karyotyping. Identification of this gene translocation, which is considered very specific, has allowed accurate investigation of the incidence and distribution of this tumor type in humans. Interestingly, it has been shown that less than 10% of human synovial sarcomas are intra-articular, while more than 80% arise in the deep soft tissue around the joints or tendons, 7 or other body sites such as kidney, lungs, mediastinum, and neck. 8 It has also been demonstrated that the cell of origin of this neoplasm is not a synoviocyte, but a pluripotent mesenchymal stem cell that occurs throughout the body, 9 which is why, in human medicine, the term synovial sarcoma is now preferred over synovial cell sarcoma. 9 Currently, in the literature, there are only 2 reported cases in dogs of nonarticular SCS, both identified in the mandibular region. 10,11 In veterinary medicine, little is known about the origin of SCS and some authors have suggested, as in human medicine, to use the term synovial sarcoma, although consensus on the most appropriate nomenclature has not yet been reached. However, it is our opinion that the use of this term might be misleading and perhaps incorrect until the exact origin of this tumor is proven. Until then, the new term of cytokeratin-positive joint associated sarcoma is proposed. Moreover, the authors suggest that joint sarcomas diagnosed on cytology or H&E histopathology should initially be classified as joint associated sarcomas. It is only with a complete IHC panel that these tumors can be further subclassified in histiocytic sarcoma, cytokeratin-positive joint associated sarcoma, and malignant fibrous histiocytoma. The IHC panel that would be required for subclassification of joint associated sarcomas is reported in Table 1. Using this panel, tumors that express only vimentin and that do not show specific morphological features to allow classification into specific histotypes, such as myxosarcoma, fibrosarcoma, or chondrosarcoma, will need to be diagnosed as undifferentiated joint associated sarcomas until their origin can be clarified with further research. Concerning the cytokeratin-positive joint associated sarcoma, studies that investigate a possible genetic alteration in dogs, similar to that identified in people, could help elucidate the exact origin, incidence, and anatomic distribution of this tumor in the canine species. Before the advent of the genetic analysis, synovial sarcoma diagnosis in people relied on immunohistochemistry. Due to the high costs and sample requirements (fresh or frozen tissue) for the genetic tests, 12 immunohistochemistry is still used in many human laboratories and immunohistochemical panels are continuously being implemented. Markers to differentiate human synovial sarcoma from other morphologically similar human sarcomas (eg, malignant peripheral nerve sheath tumor and Ewing sarcoma) include epithelial membrane antigen, a-smooth muscle actin, B-cell lymphoma 2 (bcl-2), CD117, and Transducin-Like Enhancer of split (TLE-1). 13,14 Little is known about the expression of these markers in canine cytokeratinpositive joint associated sarcoma (synovial cell sarcoma), and only one case in the literature reported a dog with a SCS that was positive for epithelial membrane antigen. 15 In the study by Craig et al, cytokeratin-positive joint-associated sarcomas were negative for a- smooth muscle actin, which was used to confirm the diagnosis of malignant fibrous histiocytoma. 2 In the present case, the neoplastic cells were weakly positive for a-smooth muscle actin, but the cell morphology and the positive pan-cytokeratin staining allowed us to exclude a malignant fibrous histiocytoma. Similarly, in human synovial sarcomas, a-smooth muscle actin was shown to stain a variable proportion (8%-21%) of tumors. 14 In the future, new immunohistochemical markers such as bcl2, CD117, and TLE-1 could become useful markers to add to vimentin, pan-cytokeratin, and CD18 in the diagnosis of canine SCS. TLE-1 seems to be the most interesting marker, as it has been shown in people to be sensitive, specific, and strongly correlated with the genetic translocation. 13 TABLE 1 The immunohistochemical panel required for subclassification of joint-associated sarcomas. Tumors that express only vimentin and that do not show specific morphological features to allow classification into specific histotypes (eg, myxosarcoma, fibrosarcoma, or chondrosarcoma) should be diagnosed as undifferentiated joint-associated sarcomas Smooth Vimentin CD18/ IBA1 Cytokeratin muscle actin Histiocytic sarcoma + + Cytokeratin-positive joint-associated sarcoma (synovial cell sarcoma) + + +/ Malignant fibrous histiocytoma + +

5 MONTI ET AL. 185 In conclusion, this paper provides a detailed description of the cytologic features of a cytokeratin-positive joint-associated sarcoma (canine synovial cell sarcoma). It also describes the limitations and risks associated with using the currently available grading system in veterinary medicine and highlights the need for new clinical studies to investigate the behavior and prognostic factors of cytokeratinpositive joint-associated sarcomas (synovial sarcomas) in dogs. Little is known about the cell origin, incidence, and anatomic distribution of this tumor in dogs. For this reason, the authors propose that the terms, synovial cell sarcoma and synovial sarcoma, should be discontinued in favor of cytokeratin-positive joint-associated sarcoma. Advanced genetic studies would also help to identify and characterize cytokeratin-positive joint-associated sarcomas (synovial cell sarcomas) better, which can allow more targeted investigations regarding the cytologic, histologic, immunohistochemical, and clinical features of this tumor type. REFERENCES 1. Vail DM, Powers BE, Getzy DM, et al. Evaluation of prognostic factors for dogs with synovial sarcoma: 36 cases ( ). J Am Vety Med Assoc. 1994;205: Craig LE, Julian ME, Ferracone JD. The diagnosis and prognosis of synovial tumors in dogs: 35 cases. Vet Pathol. 2002;39: Fox DB, Cook JL, Kreeger JM, Beissenherz M, Henry CJ. Canine synovial sarcoma: a retrospective assessment of described prognostic criteria in 16 cases ( ). J Am Anim Hosp Assoc. 2002;38: Terry J, Saito T, Subramanian S, et al. TLE1 as a diagnostic immunohistochemical marker for synovial sarcoma emerging from gene expression profiling study. Am J Surg Pathol. 2007;31: Jagdis A, Rubin BP, Tubbs RR, Pacheco M, Nielsen TO. Prospective evaluation of TLE1 as a diagnostic immunohistochemical marker in synovial sarcoma. Am J Surg Pathol. 2009;33: Kawai A, Woodruff J, Healey JH, Brennan MF, Antonescu CR, Ladanyi M. SYT-SSX gene fusion as a determinant of morphology and prognosis in synovial sarcoma. N Engl J Med. 1998;338: Changchien YC, Katalin U, Fillinger J, et al. A challenging case of metastatic intra-abdominal synovial sarcoma with unusual immunophenotype and its differential diagnosis. Case Rep Pathol. 2012;2012: Kritsaneepaiboon S, Sangkhathat S, Mitarnun W. Primary synovial sarcoma of the abdominal wall: a case report and literature review. J Radiol Case Rep. 2015;9: Craig LE, Dittmer KE, Thompson KG. Tumours and tumour-like lesions of joints. In: Grant Maxie M, ed. Jubb, Kennedy, Palmer s Pathology of Domestic Animals, 6th ed. Missouri: Elsevier; 2016: Griffith JW, Frey RA, Sharkey FE. Synovial sarcoma of the jaw in a dog. J Comp Pathol. 1987;97: Takimoto N, Suzuki K, Ogawa T, et al. A non-joint tissue biphasic synovial sarcoma in a dog. J Comp Pathol. 2014;150: Foo WC, Cruise MW, Wick MR, Hornick JL. Immunohistochemical staining for TLE1 distinguishes synovial sarcoma from histologic mimics. Am J Clin Pathol. 2011;135: Kn osel T, Heretsch S, Altendorf-Hofmann A, et al. TLE1 is a robust diagnostic biomarker for synovial sarcomas and correlates with t (X;18): analysis of 319 cases. Eur J Cancer. 2010;46: Plemus M, Guillou L, Hostein I, Sierankowski G, Lussan C, Coindre JM. Monophasic fibrous and poorly differentiated synovial sarcoma: immunohistochemical reassessment of 60 t(x;18)(syt-ssx)-positive cases. Am J Surg Pathol. 2002;26: Loukopoulos P, Heng HG, Arshad H. Canine biphasic synovial sarcoma: case report and immunohistochemical characterization. J Vet Sci. 2004;5: How to cite this article: Monti P, Barnes D, Adrian AM, Rasotto R. Synovial cell sarcoma in a dog: A misnomer Cytologic and histologic findings and review of the literature. Vet Clin Pathol. 2018;47: vcp.12590