The Effects of Tumor Location on Diagnostic Criteria for Canine Malignant Peripheral Nerve Sheath Tumors (MPNSTs) and the Markers for Distinction Between Canine MPNSTs and Canine Perivascular Wall Tumors

Abstract

Canine malignant peripheral nerve sheath tumors (MPNSTs) occur not only in the peripheral nervous system (PNS) but also in soft tissue and various organs (non-PNS). The most important diagnostic criterion is proof of peripheral nerve sheath origin. This is difficult in non-PNS MPNSTs, and its differential diagnosis is challenging. Canine perivascular wall tumors (PWTs) also commonly arise in soft tissue. Their histopathological features are quite similar to those of canine MPNSTs, making their differential diagnosis challenging. To elucidate whether the morphological features are applicable to diagnose non-PNS MPNSTs and to demonstrate useful markers for distinction between canine MPNSTs and PWTs, the authors examined 30 canine MPNSTs and 31 PWTs immunohistochemically for S100, nestin, NGFR, Olig2, claudin-1, CD57, PRX, α-SMA, desmin, and calponin. Among canine MPNSTs, the PNS tumors displayed significantly higher S100 and Olig2 expression than the non-PNS tumors. The expression levels of the other markers did not differ significantly, suggesting that the same morphological diagnostic criteria are applicable regardless of their location. The PWT cells displayed significantly weaker immunoreactivity than MPNSTs to markers used except α-SMA and desmin. Cluster analysis sorted most canine MPNSTs and PWTs into 2 distinctly different clusters, whereas 3 MPNSTs and 6 PWTs were assigned to the opposing cluster. These 3 MPNSTs were negative for almost all markers, while these 6 PWTs were positive for only neuronal markers. In particular, NGFR and Olig2 were almost negative in the rest of PWT cases. These findings suggest that NGFR and Olig2 are useful to distinguish these 2 tumors.

Keywords

canine cluster analysis diagnostic criteria hemangiopericytoma immunohistochemistry malignant peripheral nerve sheath tumor perivascular wall tumor

Peripheral nerve sheath tumors (PNSTs) occur not only in the peripheral nervous system (PNS) but also in soft tissue and various organs (non-PNS). PNSTs are histologically composed of pleomorphic spindle-shaped or polygonal neoplastic cells, which are derived from Schwann cells, perineurial cells, or both and are arranged in irregularly interwoven bundles, nuclear palisades, whorls, storiform patterns, or sheets. In addition, PNSTs contain various amounts of collagenous stroma.^{12,13,19,20,31,45} Due to the poor prognosis of human malignant PNSTs (MPNSTs), their immunohistochemical features, pathogenesis, genetic basis, prognostic factors, and treatment strategies have been investigated extensively.^{9,17,18,26,31,36,37,45} On the contrary, although there have been many reports on PNSTs,^{2, 6, 11
–13, 19
–21, 28, 33, 34, 39, 41, 43, 44, 46}. their diagnostic criteria, histogenesis, subclassification, and immunohistochemical features are poorly understood in veterinary medicine. The most important diagnostic criterion for human PNSTs is proving that the neoplastic cells originate from peripheral nerve sheath components.⁴⁵ It is easy to detect the involvement of peripheral nerve sheath components in PNSTs in PNS tissue but not in non-PNS tissue. Without the proof of peripheral nerve sheath origin, the diagnosis of non-PNS MPNSTs should be made by morphological features. However, some spindle cell tumors have morphological features similar to those of MPNSTs, making the diagnosis of non-PNS MPNSTs challenging. Thus, immunohistochemical features should be confirmed based on those of PNS MPNSTs. In canine MPNSTs, it is not clear whether the morphological features are applicable to diagnose non-PNS MPNSTs without the proof of peripheral nerve sheath origin. Moreover, distinguishing MPNSTs from other spindle cell tumors that affect soft tissue such as fibrosarcomas or canine hemangiopericytomas (CHPs)^23,29—recently suggested to be included in perivascular wall tumors (PWTs)^1,27—is rather difficult. In such a case, it is very important to consider the morphological and immunohistochemical features based on those of PNS MPNSTs. Thus, it is important to determine useful diagnostic criteria and to clarify the effects of tumor location on morphological and immunohistochemical features to diagnose especially non-PNS MPNSTs.

Perivascular wall tumor is a novel classification for CHP suggested by a couple recent studies.^1,27 It is not included in the current World Health Organization international histological classification of tumors of domestic animals.¹³ Canine PWTs also commonly arise in soft tissue. They are suggested to be a mixed group of tumors, including hemangiopericytoma, angioleiomyoma, angioleiomyosarcoma, myopericytoma, angiomyofibroblastoma, and angiofibroma. Their origin is considered to be the different cellular components of the vascular wall, such as undifferentiated stem cells, perivascular fibroblasts, pericytes, myopericytes, smooth muscle cells, myofibroblasts, or transitional cells with overlapping features.^1,27 The neoplastic cells are spindle shaped and arranged in interlacing bundles or in a myxoid, solid, storiform, or verocay-like pattern, and tumors have the following vascular growth pattern: perivascular whorls, staghorn, or placentoid vessels.^1,27 CHP has been considered a malignant unclassified soft tissue tumor,¹³ and many PWTs are still diagnosed as CHPs based on histologic features, including perivascular whorls. Furthermore, the distinction from PNSTs is still under debate.²⁷

The histological features of PWTs resemble those of MPNSTs; therefore, distinguishing PWTs from MPNSTs arising in non-PNS tissue is difficult. There are several studies on the immunohistochemical or ultrastructural subclassification of PWTs^1,16,27 but very few studies on the comparison of immunohistochemical features for the differential diagnosis between canine MPNSTs and PWTs.^6,16 Furthermore, in companion animal practice, MPNSTs and CHPs have been treated as soft tissue sarcomas (STSs) together with other spindle cell tumors that arise in the soft tissue, such as fibrosarcomas, rhabdomyosarcomas, leiomyosarcomas, myxosarcomas, and malignant mesenchymomas, due to their similar histological features and biological behaviors.^22,38 However, Stefanello et al³⁸ recently suggested that PWTs (former CHPs) are a unique subgroup of STSs characterized by low to nonaggressive (ie, “benign”) behavior. Oversimplification sometimes prevents development of specific treatment. Research on the definite differentiation and prognostic factors for each STS is essential for the development of novel specific treatment strategies such as molecular targeted therapies.

In the present study, 30 canine MPNSTs and 31 PWTs were immunohistochemically investigated to determine useful diagnostic criteria and elucidate the effects of tumor location on the diagnostic criteria for canine MPNSTs and to demonstrate a useful immunohistochemical panel for the differential diagnosis between canine MPNSTs and PWTs.

Materials and Methods

Cases

Sixty-one canine spindle cell tumors were submitted to the Department of Veterinary Pathology, Graduate School of Agricultural and Life Sciences, the University of Tokyo, between 2008 and 2011. Histologically, 30 cases were diagnosed as canine MPNSTs and 31 cases were diagnosed as canine PWTs. The case histories and tumor locations of the MPNSTs and PWTs are listed in Tables 1 and 2, respectively. The data of the cases between 2008 and 2010 were obtained retrospectively, and the data of the cases between 2010 and 2011 were obtained prospectively. All tumors were obtained by excisional biopsy or complete surgical excision. Among the MPNSTs, 16 occurred in the PNS (Nos. 1–16), and 14 occurred in non-PNS tissue (Nos. 17–30), including the subcutaneous tissue (6/14; Nos. 17–22), visceral organs (4/14; Nos. 23–26), gingiva (2/14; Nos. 27 and 28), nasal cavity (1/14; No. 29), and external soft tissue around the right scapula (1/14; No. 30). Nos. 11, 18, 26, and 27 were recurrent tumors. Nos. 25 and 26 had metastatic lesions and were obtained by laparotomy. All 31 PWTs occurred in subcutaneous tissue. The tumor locations of the PWTs included the head and neck region (2/31; Nos. 31 and 32), the forelimbs or hindlimbs (22/31; Nos. 33– 43 and 51–61), and the body trunk (7/31; Nos. 44–50). The mean (SD) ages of the MPNST and PWT patients were 10.3 (2.0) years (range, 7 years to 13 years and 8 months) and 11.1 (2.2) years (range, 7 years and 5 months to 15 years and 1 month), respectively. Among the MPNST patients, 19 were males (including 10 castrated males), and 11 were females (including 6 spayed females). Among the PWT patients, 13 were males (including 4 castrated males), and 17 were females (including 6 spayed females). Age and sex data were not recorded for case No. 48.

Table 1.

MPNST Cases Examined.

No.	Breed	Age^a	Sex^b	Tumor Location
PNS
1	Chihuahua	11Y	M	Medulla oblongata to C1 intradural-extramedullary
2	Miniature Schnauzer	10Y11M	F	C2 extramedullary
3	Mix	7Y	M	C2 within the vertebral canal
4	Miniature Schnauzer	7Y	F	C2–3 intradural-extramedullary
5	Miniature Dachshund	12Y2M	M	C5–6 nerve root
6	Miniature Dachshund	13Y1M	M	C6–7 right nerve root
7	Yorkshire Terrier	10Y4M	MC	C6–7 right nerve root
8	Miniature Dachshund	10Y	MC	C6–T1 right brachial plexus
9	Shetland Sheepdog	13Y6M	MC	Left brachial plexus
10	Welsh Corgi	9Y5M	M	Left ulnar nerve
11	Beagle	8Y	M	T8 right vertebral arch (recurrent tumor)
12	Welsh Corgi	10Y2M	FS	L2–3 intradural-extramedullary
13	Mix	13Y8M	FS	L2–3 extradural-extramedullary
14	Mix	13Y2M	MC	L4–5 intradural
15	American Cocker Spaniel	11Y	FS	L7 spinal nerve
16	Mix	7Y11M	FS	Intra-axial, nerve root
Non-PNS
17	Shih Tzu	8Y6M	F	Caudal surface of the left brachial region
18	Labrador Retriever	12Y	F	Left femoral region (recurrent tumor)
19	Shetland Sheepdog	10Y	M	Left femoral region
20	Golden Retriever	7Y4M	FS	Lower abdominal wall
21	Yorkshire Terrier	11Y6M	MC	Left exterior antebrachial region
22	Welsh Corgi	9Y9M	MC	Base of the tail
23	Welsh Corgi	12Y	MC	Spleen
24	Golden Retriever	11Y	M	Spleen
25	Beagle	13Y	MC	Spleen, liver, kidneys, and lungs
26	Beagle	9Y3M	M	Right adrenal gland, liver, ureter (recurrent tumor)
27	Golden Retriever	10Y2M	F	Gingiva (recurrent tumor)
28	Golden Retriever	8Y10M	MC	Gingiva
29	Beagle	8Y	MC	Nasal cavity
30	Great Pyrenees	8Y4M	FS	External soft tissue around right scapula

Abbreviations: MPNST, malignant peripheral nerve sheath tumor; PNS, peripheral nervous system.

^aY, years; M, months.

^bC, castrated; F, female; M, male; S, spayed.

Table 2.

CHP Cases Examined.

Location/No.	Breed	Age^a	Sex^b	Tumor Location
Head and neck
31	Welsh corgi	12Y	MC	Right regio buccalis
32	Shih Tzu	12Y	F	Left cervical region
Forelimb
33	Mix	9Y6M	FS	Left shoulder
34	Golden Retriever	12Y7M	F	Left axilla
35	Miniature Dachshund	9Y10M	MC	Left axilla to thoracic cavity (recurrent tumor)
36	Mix	14Y8M	M	Left elbow
37	Beagle	15Y	F	Left elbow
38	Boston Terrier	9Y2M	FS	Left elbow
39	Shiba Inu	13Y6M	M	Right elbow
40	Mix	15Y1M	M	Left elbow
41	Shetland Sheepdog	9Y10M	FS	Left elbow
42	Pomeranian	13Y11M	F	Left forelimb (recurrent tumor)
43	Labrador Retriever	10Y	F	Right forelimb
Body trunk
44	Boxer	8Y11M	M	Thoracic wall
45	Chihuahua	9Y	F	Right lumbar region
46	Welsh Corgi	13Y3M	F	Lower back
47	Labrador Retriever	9Y2M	MC	Left buttock
48	Shetland Sheepdog	ND	ND	Buttocks
49	Golden Retriever	13Y4M	FS	Left buttock
50	Mix	9Y3M	M	Left buttock
Hindlimb
51	Pomeranian	10Y2M	FS	Left lower hindlimb
52	Mix	13Y10M	M	Left femoral area (recurrent tumor)
53	Welsh Corgi	9Y7M	M	Right femoral area
54	Shetland Sheepdog	10Y11M	F	Left femoral area
55	Mix	10Y	M	Femoral area
56	Doberman	11Y3M	F	Femoral area
57	Welsh Corgi	7Y5M	FS	Femoral area
58	Golden Retriever	13Y	MC	Lumbar lymph nodes (metastasis from a right hindlimb tumor)
59	Labrador Retriever	8Y4M	M	Right tarsal joint area (recurrent tumor)
60	Golden Retriever	8Y6M	F	Left tarsal area (recurrent tumor)
61	Yorkshire Terrier	9Y2M	F	Between the second and third digits of the right hindlimb

^aY, years; M, months; ND, no data.

^bC, castrated; F, female; M, male; ND, no data; S, spayed.

Histopathology

All tissue samples were fixed in a 10% phosphate-buffered formalin solution. Then, 2- to 4-μm-thick paraffin-embedded sections were cut and stained with hematoxylin and eosin (HE). MPNSTs were diagnosed based on the following modified histopathological criteria that were derived from the findings of previous studies^{12,13,19,20,45}: (1) the tumor is composed of spindle-shaped or plump fusiform cells arranged in irregularly interwoven bundles, whorls, or sheets and/or displays nuclear palisades and contains various amounts of collagenous stroma; (2) peripheral nerve tributaries with neoplastic Schwann cells and/or perineurial cells can be observed, and the tumor appears to originate from them; and (3) the neoplastic cells display a certain degree of pleomorphism and invade the surrounding tissue, and the tumor also contains necrotic areas. PWTs were diagnosed using the histopathological criteria based on the current studies^1,27: the tumor is composed of spindle-shaped cells arranged in interlacing bundles, or in a myxoid, solid, storiform, or verocay-like pattern, and has the following vascular growth pattern: perivascular whorls, staghorn, or placentoid vessels.

Immunohistochemistry

The primary antibodies used in the present study are listed in Table 3. To detect S100, claudin-1, CD57, PRX, desmin, and calponin, antigen retrieval was performed by autoclaving deparaffinized sections in citrate buffer (pH 6.0) at 121°C for 10 minutes, and for Olig2 detection, the sections were autoclaved in Target Retrieval Solution (DAKO Japan, Tokyo, Japan) (pH 9.0) at 121°C for 10 minutes. Antigen retrieval was not performed to detect nestin, NGFR, α-SMA, and neurofilament. The inactivation of endogenous peroxidase was performed by treating the sections with 3% hydrogen peroxide in methanol at room temperature (RT) for 5 minutes. To block nonspecific reactions, the sections were incubated in 8% skimmed milk in Tris-buffered saline (TBS) at 37°C for 30 minutes. Then, the sections were incubated with each primary antibody at 4°C overnight. After being washed 3 times with TBS, the sections were treated with Dako EnVision+ System horseradish peroxidase (HRP)–labeled polymer anti–mouse or anti–rabbit secondary antibodies (DAKO Japan) before being incubated at RT for 10 minutes, at 37°C for 40 minutes, and then at RT for a further 10 minutes. After being washed 3 times with TBS, 0.05% 3,3′-diaminobenzidine and 0.03% hydrogen peroxide in Tris-HCl buffer were used as the chromogen, and the sections were counterstained with Mayer’s hematoxylin. Tissue sections of the brachial plexus of a Pug dog (age 5 years and 4 months) without any neurological disorders were used as a positive control. Skeletal muscle and vascular wall were used as internal positive control for desmin and calponin, respectively. The primary antibodies were replaced with TBS to produce a negative control. The numbers of immunopositive neoplastic cells were counted within 10 randomly selected high-power (×400) fields, and the percentage of immunolabeled cells was calculated for each case.

Table 3.

Antibodies Used.

Antibody	Host	Dilution	Antigen Retrieval	Source
S100, polyclonal	Rabbit	1:4000	Autoclaved in citrate buffer (pH 6.0) at 121°C for 10 min	DAKO Japan (Tokyo, Japan)
Nestin, polyclonal	Rabbit	1:60	None	IBL (Gunma, Japan)
NGFR, monoclonal (ME20.4)	Mouse	4 μg/ml	None	Abcam (Cambridge, UK)
Olig2, polyclonal	Rabbit	1:500	Autoclaved in target retrieval solution (pH 9.0) at 121°C for 10 min	EMD Millipore (Billerica, MA)
Claudin-1, polyclonal	Rabbit	1:200	Autoclaved in citrate buffer (pH 6.0) at 121°C for 10 min	Abcam
CD57, monoclonal (VC1.1)	Mouse	1:2000	Autoclaved in citrate buffer (pH 6.0) at 121°C for 10 min	Sigma-Aldrich (St Louis, MO)
PRX, polyclonal	Rabbit	1:1000	Autoclaved in citrate buffer (pH 6.0) at 121°C for 10 min	Sigma-Aldrich
α-SMA, monoclonal (1A4)	Mouse	1:100	None	DAKO Japan
Desmin, monoclonal (D33)	Mouse	1:100	Autoclaved in citrate buffer (pH 6.0) at 121°C for 10 min	DAKO Japan
Calponin, monoclonal (hCP)	Mouse	1:30 000	Autoclaved in citrate buffer (pH 6.0) at 121°C for 10 min	Sigma-Aldrich
Neurofilament, monoclonal (2F11)	Mouse	Ready to use	None	DAKO Japan

Statistical Analyses

Statistical analyses were performed using the SPSS 15.0 J software (SPSS Japan, Tokyo, Japan) with the percentage of immunolabeled cells. Pearson’s correlation coefficient analysis was performed within each tumor group to assess whether there was any correlation between the expressions of each marker. The Student’s t test was performed to examine whether there were significant differences in the mean percentage of immunopositive cells (1) between MPNSTs that arose in different locations (PNS and non-PNS) and (2) between MPNSTs and PWTs. P values of less than .05 were considered statistically significant.

Cluster Analyses

To examine the immunohistochemical differences (1) between PNS and non-PNS MPNSTs and (2) between MPNSTs and PWTs, the percentage of immunopositive cells of each case was assessed for each marker using a multivariate cluster analysis method, the Ward method, with the SPSS 15.0 J software. A dendrogram was constructed by assessing the squared Euclidian distances between the clusters.

Results

Histopathology

All tumors investigated were diagnosed as MPNST or PWT according to the diagnostic criteria described above. The histopathological features of the MPNST cases were quite similar regardless of the tumor location. They were composed of pleomorphic spindle cells. The neoplastic cells were mostly arranged in interwoven bundles, palisades, or whorls, and they occasionally had a wavy cytoplasm and nucleus. Tumors had necrotic areas within the lesion and invaded into the surrounding tissues. The involvement of peripheral nerve sheaths was not found in non-PNS MPNSTs. PWT cases had quite a similar morphology to MPNST cases. They were also composed of pleomorphic spindle cells and were arranged in interlacing bundles (3/31) or in a myxoid (14/31), solid (18/31), storiform (12/31), or verocay-like pattern (1/31), and the tumors had the following vascular growth pattern: perivascular whorls (14/31), staghorn (1/31), or placentoid vessels (4/31). Seventeen PWTs invaded into surrounding tissues, and 22 cases had necrotic areas within the lesion.

Immunohistochemistry

Cellular location of immunoreactivity to each antibody in the normal peripheral nerve sheaths, MPNSTs, and PWTs is summarized in Table 4. The immunoreactivity of normal peripheral nerve sheath components to each marker is shown in Figs. 1 to 10. The normal canine peripheral nerve sheaths displayed positivity for S100, NGFR, claudin-1, and CD57, as reported in previous reports.^3,5,6,14 On the other hand, the present study also found that the normal canine peripheral nerve sheaths expressed nestin, Olig2, and PRX, which has never been reported previously to our knowledge. S100 was expressed in the nuclei and cytoplasm of Schwann cells (Fig. 1), nestin was expressed in the cytoplasm of vascular endothelial cells and a few Schwann cells (Fig. 2), and NGFR was expressed in the perineurial cell membrane (Fig. 3). In addition, Olig2 was expressed in the cytoplasm of Schwann cells (Fig. 4), claudin-1 was expressed in the perineurial cell membrane (Fig. 5), and CD57 was expressed in myelin sheaths and the Schwann cell membrane (Fig. 6). Furthermore, PRX was expressed in myelin sheaths and the nuclei and membranes of Schwann cells (Fig. 7), α-SMA was expressed in the perineurium and vascular smooth muscle cells (Fig. 8), and neurofilament was expressed in axons (data not shown). Desmin and calponin were not expressed in peripheral nerve sheaths (Figs. 9 and 10, respectively).

Figure 1.

Brachial plexus; Pug. Many Schwann cells (myelin sheaths) in the normal peripheral nerve sheath are immunopositive for S100. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 2. Brachial plexus; Pug. A few Schwann cells in the normal peripheral nerve sheath are immunopositive for nestin. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 3. Brachial plexus; Pug. The perineurial cells in the normal peripheral nerve sheath are immunopositive for NGFR. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 4. Brachial plexus; Pug. A few Schwann cells in the normal peripheral nerve sheath are immunopositive for Olig2. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 5. Brachial plexus; Pug. The perineurial cells in the normal peripheral nerve sheath are immunopositive for claudin-1. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 6. Brachial plexus; Pug. Many Schwann cells (myelin sheaths) in the normal peripheral nerve sheath are immunopositive for CD57. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 7. Brachial plexus; Pug. Many Schwann cells (myelin sheaths) in the normal peripheral nerve sheath are immunopositive for PRX. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 8. Brachial plexus; Pug. The perineurial smooth muscle cells in the normal peripheral nerve sheath are immunopositive for α-SMA. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 9. Brachial plexus; Pug. The normal peripheral nerve sheath is immunonegative for desmin. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 10. Brachial plexus; Pug. The normal peripheral nerve sheath is immunonegative for calponin. Immunohistochemistry. Counterstain: Mayer’s hematoxylin.

Table 4.

Cellular Location of Immunoreactivity to each Antibody in the Normal Peripheral Nerve Sheaths, MPNSTs, and PWTs.

	Cellular Location
Marker	Normal Peripheral Nerve Sheath Cell	MPNST Cell	PWT Cell	Positive Cells in the Normal Tissues	References
S100	Nucleus and cytoplasm	Nucleus and/or cytoplasm	Nucleus and/or cytoplasm	Neurons, glias, Schwann cells, melanocytes, etc	2, 4, 6, 11, 28, 30, 45
Nestin	Cytoplasm	Nucleus and/or cytoplasm	Nucleus and/or cytoplasm	Neuroectodermal stem cells, neural progenitor cells, Schwann cells	15, 35, 36, 40
NGFR	Membrane	Membrane	Membrane	Schwann cells, perineurial cells, perivascular fibroblasts, neurons, neural crest progenitor or stem cells, etc	4, 6, 36, 45
Olig2	Cytoplasm	Nucleus and/or cytoplasm	Nucleus and/or cytoplasm	Oligodendrocyte, oligodendroglial tumor cells, Schwann cells	3, 24, 47
Claudin-1	Membrane	Membrane	Membrane	Perineurial cells	7, 10, 16, 42, 45
CD57	Membrane	Membrane	Membrane	Myelinating Schwann cells	4, 45
PRX	Nucleus and cytoplasm	Nucleus and/or cytoplasm	Cytoplasm	Schwann cells	32, 43
α-SMA	Cytoplasm in perineurial area	Negative	Cytoplasm	Smooth muscle cells
Desmin	Negative	Cytoplasm	Cytoplasm	Skeletal muscle cells
Calponin	Negative	Negative	Cytoplasm	Smooth muscle cells	1, 16, 27

Abbreviations: MPNST, malignant peripheral nerve sheath tumor; PWT, perivascular wall tumor.

The staining patterns of each marker in MPNSTs and PWTs are shown in Figs. 11 to 34, and the percentages of MPNST and PWT cells that were positive for each marker are summarized in Tables 5 and 6, respectively. The staining intensities of the neoplastic cells were highly variable among the cases. In canine MPNSTs, S100 (29/30; 96.7%), nestin (28/30; 93.3%), Olig2 (27/30; 90.0%), and PRX (12/30; 40.0%) expression was observed in the nuclei and/or the cytoplasm of neoplastic cells, whereas NGFR (21/30; 70.0%), claudin-1 (21/30; 70.0%), and CD57 (18/30; 60.0%) expression was detected in the neoplastic cell membrane (Figs. 11–24). α-SMA, desmin, and calponin were not expressed in MPNSTs. Any peripheral neural axon was not detected in non-PNS MPNSTs with a neurofilament antibody. In PWTs, S100 (11/31; 35.5%), nestin (26/31; 83.8%), and Olig2 (6/31; 19.4%) expression was observed in the nuclei and/or the cytoplasm of the neoplastic cells, whereas NGFR (5/31; 16.1%), claudin-1 (17/31; 54.8%), and CD57 (1/31; 3.2%) expression was found in the neoplastic cell membrane. α-SMA (7/31; 22.6%), desmin (9/31; 29.0%), and calponin (7/31; 22.6%) expression was found in the cytoplasm of the neoplastic cells (Figs. 25–34). PRX was not expressed in PWTs.

Figure 11.

C2 extramedullary; Miniature Schnauzer, case No. 2. The malignant peripheral nerve sheath tumor (MPNST) cells are strongly immunopositive for S100 protein. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 12. C6–T1 right brachial plexus; Miniature Dachshund, case No. 8. The MPNST cells are strongly immunopositive for nestin. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 13. C6–7 right nerve root; Yorkshire Terrier, case No. 7. The MPNST cells are moderately immunopositive for NGFR. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 14. Intra-axial, nerve root; mix, case No. 16. The MPNST cells are moderately immunopositive for Olig2. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 15. T8 right vertebral arch; Beagle, case No. 11. The MPNST cells are strongly immunopositive for claudin-1. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 16. C6–T1 right brachial plexus; Miniature Dachshund, case No. 5. The MPNST cells are weakly immunopositive for CD57. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 17. Left brachial plexus; Shetland Sheepdog, case No. 9. The MPNST cells are moderately immunopositive for PRX. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 18. Lower abdominal wall; Golden Retriever, case No. 20. The MPNST cells are moderately immunopositive for S100. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 19. Scapula; Great Pyrenees, case No. 30. The MPNST cells are strongly immunopositive for nestin. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 20. Gingiva; Golden Retriever, case No. 27. The MPNST cells are moderately to strongly immunopositive for NGFR. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 21. Right adrenal gland; Beagle, case No. 26. The MPNST cells are moderately immunopositive for Olig2. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 22. Left exterior antebrachial region; Yorkshire Terrier, case No. 21. The MPNST cells are strongly immunopositive for claudin-1. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 23. Gingiva; Golden Retriever, case No. 27. The MPNST cells are moderately immunopositive for CD57. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 24. Lower abdominal wall; Golden Retriever, case No. 20. The MPNST cells are moderately immunopositive for PRX. Immunohistochemistry. Counterstain: Mayer’s hematoxylin.

Figure 25.

Lumbar lymph nodes (metastasis from a right hindlimb tumor); Golden Retriever, case No. 58. The perivascular wall tumor (PWT) cells are weakly immunopositive for S100. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 26. Left lower hindlimb; Pomeranian, case No. 51. The PWT cells are very weakly immunopositive for nestin. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 27. Left elbow; Beagle, No. 37. The PWT cells are immunonegative for NGFR. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 28. Left tarsal area; Golden Retriever, case No. 60. The PWT cells are moderately immunopositive for Olig2. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 29. Left buttock; mix, case No. 50. The PWT cells are moderately immunopositive for claudin-1. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 30. Right forelimb; Labrador Retriever, No. 43. The PWT cells are immunonegative for CD57. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 31. Right lumbar region; Chihuahua, No. 45. The PWT cells are immunonegative for PRX. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 32. Left axilla; Golden Retriever, No. 34. The PWT cells are moderately immunopositive for α-SMA. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 33. Left elbow; Beagle, No. 37. The PWT cells are weakly immunopositive for desmin. Immunohistochemistry. Counterstain: Mayer’s hematoxylin. Figure 34. Left buttock; mix, No. 50. The PWT cells are moderately immunopositive for calponin. Immunohistochemistry. Counterstain: Mayer’s hematoxylin.

Table 5.

Percentages of MPNST Cells Immunopositive for Each Marker.^a

Location/No.	S100	Nestin	NGFR	Olig2	Claudin-1	CD57	PRX	α-SMA	Desmin	Calponin
PNS
1	81.1	0	6.8	78.0	0	8.2	0	0	0	0
2	66.8	39.2	32.0	21.2	47.3	25.1	0	0	0	0
3	49.4	34.3	22.3	45.5	16.4	0	0	0	0	0
4	55.0	0	51.0	29.0	0	25.8	0	0	0	0
5	80.3	42.5	14.3	46.0	19.3	4.0	0	0	0	0
6	50.9	42.1	0	76.5	7.1	38.8	55.6	0	0	0
7	78.1	48.0	48.1	32.2	53.5	10.1	11.4	0	0	0
8	41.3	61.7	5.7	14.2	8.3	8.6	20.9	0	0	0
9	60.9	75.6	30.9	25.8	51.2	42.0	37.8	0	0	0
10	57.5	71.2	25.9	41.7	20.5	13.5	28.6	0	51.6	0
11	59.1	1.5	0	17.7	56.8	0	0	0	0	0
12	67.2	6.2	0	7.4	6.1	0	0	0	0	0
13	69.3	19.6	36.0	32.2	42.7	0	0	0	0	0
14	46.6	9.4	47.4	39.2	24.6	9.6	9.4	0	0	0
15	67.6	74.6	8.1	55.9	29.5	0	0	0	0	0
16	85.1	37.7	40.4	20.6	41.2	8.8	13.8	0	0	0
Mean	63.5^b	35.2	23.0	36.4^b	26.5	12.2	11.1	0.0	3.2	0.0
Non-PNS
17	4.6	12.5	0	0	0	0	0	0	0	0
18	0	28.7	0	0	0	0	0	0	0	0
19	7.6	29.1	0	0	7.6	0	0	0	0	0
20	33.2	66.5	0	13.4	0	0	32.0	0	59.8	0
21	29.3	19.7	0	10.2	67.1	4.8	0	0	0	0
22	18.8	49.8	0	26.0	19.4	0	0	0	0	0
23	42.0	60.7	1.6	39.3	19.4	15.2	4.3	0	0	0
24	71.1	34.6	11.0	7.7	0	0	0	0	0	0
25	56.8	40.7	12.2	34.5	30.5	20.8	3.6	0	0	0
26	35.9	71.1	66.5	30.0	3.9	0	0	0	0	0
27	26.6	39.1	48.1	19.0	0	19.7	1.8	0	0	0
28	2.3	53.3	20.8	2.4	0	4.2	0	0	0	0
29	79.2	27.8	29.3	34.9	22.8	10.3	0	0	0	0
30	74.6	57.9	10.4	46.1	0	12.7	72.6	0	0	0
Mean	34.4^b	42.3	14.3	18.8^b	12.2	6.3	8.2	0.0	4.3	0.0

Abbreviations: MPNST, malignant peripheral nerve sheath tumor; PNS, peripheral nervous system.

^aThe numbers represent the percentage of immunopositive cells for each marker. The cases indicated by boldface were sorted into the perivascular wall tumor cluster by the cluster analysis.

^bStatistically significant (P < .05) difference between the mean values for different tumor locations.

Table 6.

Percentages of Perivascular Wall Tumor Cells Immunopositive for Each Marker.^a

Location/No.	S100	Nestin	NGFR	Olig2	Claudin-1	CD57	PRX	α-SMA	Desmin	Calponin
Head and neck
31	0	35.5	0	0	29.3	0	0	0	0	0
32	0	0	0	0	0	0	0	0	0	0
Mean	0.0	17.8	0	0	14.6	0	0	0	0	0
Forelimb
33	0	61.7	2.4	0	0	0	0	0	0	0
34	0	0	0	0	0	0	0	26.2	0	19.4
35	78.4	68.0	39.5	55.8	10.5	0	0	0	0	0
36	0	24.1	0	14.4	20.3	0	0	0	2.3	0
37	0	19.9	0	0	5.1	0	0	0	6.4	16.9
38	0	0	0	0	12.9	0	0	0	0	0
39	0	5.9	0	0	0	0	0	0	0	0
40	0	11.8	0	0	5.1	0	0	1.8	0	0
41	0	0	0	0	9.5	0	0	3.0	0	0
42	0	47.7	27.9	0	0	0	0	0	0	0
43	0	17.3	0	0	18.9	0	0	24.5	6.3	15.4
Mean	7.1	23.3	6.3	6.4	7.5	0	0	5.0	1.4	4.7
Body trunk
44	5.3	20.1	0	0	10.4	0	0	3.9	0	29.1
45	5.3	23.2	0	0	19.1	0	0	0	0	5.0
46	0	11.3	0	0	33.2	0	0	0	0	0
47	16.5	12.7	0	0	0	0	0	0	0	0
48	0	12.4	0	0	0	5.3	0	0	0	0
49	31.0	4.2	0	0	10.2	0	0	0	3.5	0
50	0	9.7	0	0	26.2	0	0	0	4.3	29.2
Mean	8.3	13.4	0	0	14.1	0.8	0	0.6	1.1	9.0
Hindlimb
51	0	6.1	0	0	0	0	0	5.9	10.0	6.5
52	7.7	14.8	0	0	29.7	0	0	0	0	0
53	5.7	2.1	0	0	3.5	0	0	0	7.6	0
54	0	30.0	0	0	25.7	0	0	0	4.9	0
55	0	8.1	0	0	0	0	0	3.9	0	0
56	22.5	16.2	0	54.5	12.0	0	0	0	0	0
57	11.1	85.9	43.7	19.3	0	0	0	0	0	0
58	4.9	5.3	0	0	0	0	0	0	3.1	0
59	12.6	60.5	6.5	66.7	0	0	0	0	0	0
60	0	28.3	0	22.1	0	0	0	0	0	0
61	0	0	0	0	0	0	0	0	0	0
Mean	5.9	23.4	4.6	14.8	6.4	0	0	0.9	2.3	0.6

^aThe numbers represent the percentage of immunopositive cells for each marker. The cases indicated by boldface were assigned to the malignant peripheral nerve sheath tumor cluster by the cluster analysis.

Statistical Analyses

The results of Pearson’s correlation coefficient analysis between each antibody within MPNST cases and within PWT cases are summarized in Tables 7 and 8. Within MPNST cases, moderate correlations were detected between S100 and Olig2, between nestin and PRX, and between CD57 and PRX. Within PWT cases, moderate correlations were detected between S100 and nestin, NGFR, and Olig2; between nestin and NGFR and Olig2; between NGFR and Olig2; and between α-SMA and calponin.

Table 8.

Pearson’s Correlation Coefficient Analysis Between Each Marker Within Perivascular Wall Tumor (PWT) Cases.^a

Within PWT Cases		Nestin	NGFR	CD57	Claudin1	Olig2	SMA	Desmin	Calponin
S100	Pearson r	.380669	.553707	–.08147	–.01425	.615893	–.15263	–.12521	–.16232
S100	P (2-tailed)	.037961	.001502	.668646	.940429	.000291	.42071	.509725	.391441
Nestin	Pearson r		.771813	–.07651	–.07896	.533494	–.2137	–.24491	–.17421
Nestin	P (2-tailed)		5.91E-07	.687797	.678335	.002398	.25684	.192099	.357188
NGFR	Pearson r			–.06586	–.19528	.417197	–.1291	–.20649	–.16875
NGFR	P (2-tailed)			.72951	.301076	.021808	.496543	.273621	.372715
CD57	Pearson r				–.16214	–.07973	–.0676	–.10811	–.08835
CD57	P (2-tailed)				.391976	.675361	.722656	.569589	.642458
Claudin1	Pearson r					–.11139	–.05357	.050778	.144861
Claudin1	P (2-tailed)					.557859	.778584	.789873	.445007
Olig2	Pearson r						–.15629	–.22817	–.20428
Olig2	P (2-tailed)						.409516	.225234	.278896
α-SMA	Pearson r							.195187	.467521
α-SMA	P (2-tailed)							.301303	.009187
Desmin	Pearson r								.283174
Desmin	P (2-tailed)								.129434

^aThe letters indicated by boldface represent statistically significant moderate correlation.

Table 7.

Pearson’s Correlation Coefficient Analysis Between Each Marker Within Malignant Peripheral Nerve Sheath Tumor (MPNST) Cases.^a

Within MPNST Cases		Nestin	NGFR	CD57	PRX	Claudin1	Olig2	Desmin
S100	Pearson r	–.04208	.259091	.22325	.184645	.348037	.536556	–.05881
S100	P (2-tailed)	.825241	.166805	.235672	.328668	.059473	.002238	.757572
Nestin	Pearson r		.072679	.177404	.430295	–.01403	.158544	.354792
Nestin	P (2-tailed)		.702705	.348315	.017621	.941357	.402707	.054378
NGFR	Pearson r			.245914	–.10205	.132455	.108349	–.09624
NGFR	P (2-tailed)			.190224	.591535	.485337	.568735	.612939
CD57	Pearson r				.489051	.155666	.355245	–.07345
CD57	P (2-tailed)				.006097	.41141	.054049	.699698
PRX	Pearson r					–.1011	.359965	.306777
PRX	P (2-tailed)					.595021	.050714	.099163
Claudin1	Pearson r						.016724	–.13697
Claudin1	P (2-tailed)						.930103	.470435
Olig2	Pearson r							.120201
Olig2	P (2-tailed)							.526934

^aThe letters indicated by boldface represent statistically significant moderate correlation.

Among MPNST cases, the mean percentages of neoplastic cells expressing S100 (63.5% in the PNS MPNSTs and 34.4% in the non-PNS MPNSTs) or Olig2 (36.4% in the PNS MPNSTs and 18.8% in the non-PNS MPNSTs) differed significantly between PNS and non-PNS (indicated in Table 5, Fig. 35). Between MPNSTs and PWTs, MPNSTs displayed significantly higher mean percentages of immunopositive neoplastic cells than did PWTs for all markers except α-SMA and desmin (Fig. 36).

Figure 35.

Mean percentages of immunopositive neoplastic cells in malignant peripheral nerve sheath tumors (MPNSTs). The peripheral nervous system (PNS) MPNSTs displayed significantly higher percentages of immunopositive cells than did non-PNS MPNSTs for S100 and Olig2 (mean ± SE). *Statistically significant (P < .05).

Figure 36.

Mean percentages of immunopositive neoplastic cells in malignant peripheral nerve sheath tumors (MPNSTs) and perivascular wall tumors (PWTs). The MPNSTs displayed significantly higher percentages of immunopositive cells than did PWTs for almost all markers except α-SMA and desmin (mean ± SE). *Statistically significant (P < .05).

Cluster Analyses

In the cluster analysis, the PNS and non-PNS MPNSTs were not separated into 2 different clusters (Fig. 37). On the other hand, except for the 9 cases described below, the MPNSTs and PWTs were separated into 2 distinctly different clusters (Fig. 38). Three MPNST cases (Nos. 17–19) were sorted into the PWT cluster. These 3 MPNST cases occurred in the brachial or femoral region and were negative for almost all antibodies. No. 17 was weakly positive for S100 and moderately positive for nestin. No. 18 was moderately positive for nestin only. No. 19 was weakly positive for S100 and claudin-1 and moderately positive for nestin. Six PWT cases (Nos. 33, 35, 42, 56, 57, and 59) were assigned to the MPNST cluster. All of these tumors occurred in the forelimb or hindlimb, and all were positive for S100, nestin, NGFR, Olig2, and/or claudin-1 and negative for α-SMA, desmin, and calponin.

Figure 37.

Cluster analysis between malignant peripheral nerve sheath tumor (MPNST) tumor locations. A dendrogram constructed using the Ward method demonstrates the relationship between MPNST tumor locations examined in the present study. Within MPNST tumor locations, cluster analysis did not sort them into 2 distinct groups.

Figure 38.

Cluster analysis between malignant peripheral nerve sheath tumors (MPNSTs) and perivascular wall tumors (PWTs). A dendrogram constructed using the Ward method demonstrates the relationship between MPNSTs and PWTs examined in the present study. MPNSTs and PWTs were sorted into 2 distinct groups. Three non-PNS MPNST cases (Nos. 17–19: gray shaded cases) were sorted into the PWT cluster, whereas 6 PWT cases (Nos. 33, 35, 42, 56, 57, and 59: black shaded cases) were assigned to the MPNST cluster.

Discussion

S100 and claudin-1 are markers of Schwann cells and perineurial cells, respectively.^10,35,42,45 Thus, the fact that we detected S100 and claudin-1 in canine MPNST cells indicates that neoplastic cells have a propensity to differentiate into Schwann cells and/or perineurial cells. Nestin has been used as a marker of neuroectodermal stem cells and cancer stem cells in human medicine.^35,36 Therefore, the detection of nestin expression in MPNST cells in the present study suggests that neural crest stem cells and/or cancer stem cell–like cells exist in canine MPNSTs. NGFR has been used as a Schwann cell differentiation marker^4,25 and a marker of neural crest progenitor cells or stem cells.^8,36 In 2 previous reports examining NGFR expression in MPNSTs,^6,14 10 of 10 human and 7 of 11 canine MPNSTs were NGFR positive. In the present study, 21 of 30 (70.0%) MPNSTs were NGFR positive, suggesting that undifferentiated Schwann cells and/or neural crest stem cells are present in canine MPNSTs as one of their components. Olig2 is a marker of the oligodendroglial lineage and some central nervous system progenitors.³ Olig2 is expressed in normal human oligodendrocytes and oligodendroglial tumor cells but not in normal human Schwann cells and schwannoma cells.⁴⁷ On the other hand, to our knowledge, the expression of Olig2 has never been investigated in normal canine Schwann cells or MPNSTs. In the present study, most of the canine MPNSTs (27/30; 90.0%) and several PWTs (6/31; 19.4%) possessed Olig2-positive tumor cells. These results suggest that Olig2 is a relatively useful marker for distinguishing MPNSTs from PWTs. CD57 is a cell type–specific marker for human Schwann cells,^4,45 and PRX has been used as a diagnostic marker for devil facial tumor disease in the Tasmanian devil (Sarcophilus harrisii),⁴³ which has been confirmed to originate from the Schwann cell. Their expression has been observed in myelinating Schwann cells in previous studies with dogs and Sprague-Dawley rats.^4,32 However, their expression has not been investigated in canine MPNSTs. In this study, half of the MPNST cases expressed CD57 (18/30; 60.0%) and/or PRX (12/30; 40.0%). The intensity was low or moderate. These results suggest that the degree of differentiation of canine MPNST cells into myelinating Schwann cells is low. α-SMA and calponin are smooth muscle–specific antibodies. Desmin is a striated muscle–specific antibody. These antibodies have been used to subclassify PWTs in recent studies.^1,6,16 In the present study, no MPNST cells were positive for α-SMA and calponin, but only 2 cases (Nos. 10 and 20) were positive for desmin. These results suggest that MPNSTs do not differentiate to smooth muscle, but several MPNSTs differentiate to striated muscle as well as triton tumors in humans.⁴⁵ On the other hand, in the present study, only a few PWTs expressed these markers. With these results and the expression of claudin-1, it is suggested that these cases include hemangiopericytomas (Nos. 34 and 44), myopericytoma (No. 43), angioleiomyosarcoma (No. 51), and angiofibroma (Nos. 32, 39, 47, 48, 60, and 61). Fifteen cases should be classified as PWTs of uncertain or undifferentiated classification because Nos. 31, 38, 46, and 52 were positive for only claudin-1; Nos. 36, 49, 53, and 54 were positive for claudin-1 and desmin; Nos. 37 and 50 were positive for claudin-1, desmin, and calponin; No. 45 was positive for claudin-1 and calponin; Nos. 40 and 41 were positive for claudin-1 and α-SMA; No. 58 was positive for only desmin; and No. 55 was positive for only α-SMA.

The results of Pearson’s correlation coefficients between each antibody within MPNSTs are considered due to the Schwannian differentiation and the similarity of antigen (both CD57 and PRX recognize myelin), and the results within PWTs are considered due to the existence of MPNSTs misdiagnosed as PWTs.

Diagnosing canine non-PNS MPNST is challenging without the proof of peripheral nerve sheath origin due to not well-established diagnostic criteria. Thus, the immunohistochemical features of non-PNS MPNSTs should be confirmed based on those of PNS MPNSTs. In the present case, the canine MPNSTs that arose in the PNS displayed significantly higher expression levels of S100 and Olig2 than the non-PNS canine MPNSTs. The immunoreactivity levels of the other markers did not differ significantly between the PNS and non-PNS canine MPNSTs. Furthermore, cluster analysis revealed that the PNS and non-PNS MPNSTs were not sorted into 2 different clusters. These findings suggest that PNS and non-PNS canine MPNSTs display similar immunohistochemical and morphological features, and the same diagnostic criteria as for human MPNSTs are applicable regardless of their location. When it comes to distinguishing MPNSTs and other spindle cell tumors, especially PWTs, it is still difficult to do so with only morphological features. It is very important to consider the morphological and immunohistochemical features based on those of PNS MPNSTs. There are only few studies on the distinction between MPNSTs and PWTs.^6,16 In the present study, immunohistochemistry using neural and muscular markers demonstrated useful markers for distinction between MPNSTs and PWTs. The results revealed that the MPNSTs displayed significantly higher numbers of immunopositive cells than the PWTs for almost all markers except α-SMA and desmin. Furthermore, cluster analysis sorted the MPNSTs and PWTs into 2 distinct clusters. However, 3 non-PNS MPNSTs were sorted into the PWT cluster, and 6 PWT cases that expressed NGFR and/or Olig2 were assigned to the MPNST cluster. These 3 MPNST cases should be considered undifferentiated PWT, while the 6 PWT cases should be considered MPNSTs that arose in non-PNS tissue because of their immunohistochemical features. The above results suggest that diagnosing non-PNS MPNST and PWT with only HE-stained sections sometimes results in errors in diagnosis. Furthermore, the fact that PWT cases assigned into the MPNST cluster expressed NGFR and/or Olig2 and S100, nestin, and claudin-1 were positive in several or many PWT cases suggests that NGFR and Olig2 are the most useful markers to distinguish these 2 tumors immunohistochemically.

To develop novel strategies for the treatment of MPNSTs and PWTs such as molecular targeting therapies, it is very important to confirm the origins of these tumors and to study the cellular signaling pathways involved in their proliferation and invasion.^5,35 The expression levels of the neural crest stem cell markers nestin and NGFR observed in the present study suggest that canine MPNSTs contain neural crest stem cells or cancer stem–like cells, which might be useful treatment targets. Classifying MPNSTs and PWTs as “soft tissue sarcomas” together with other spindle cell tumors that arise in soft tissue will prevent treatments for MPNSTs and PWTs from being developed and, hence, their prognosis will never improve. Further studies are necessary to aid the development of novel treatment strategies for these tumors.

Footnotes

Acknowledgements

We thank Animal Clinic Kobayashi, Ishikawa Animal Hospital, Japan Animal Referral Medical Center, Kudo Animal Hospital, Ozawa Animal Hospital, Sanritsu Zelkova Veterinary Laboratory, Tajika Animal Hospital, Tamura Animal Hospital, Togasaki Animal Hospital, Umino-Nakamichi Animal Hospital, and the Veterinary Medical Center of the University of Tokyo for their cooperation with the sample collection.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Avallone

Helmbold

Caniatti

. The spectrum of canine cutaneous perivascular wall tumors: morphologic, phenotypic and clinical characterization. Vet Pathol. 2007;44(5):607–620.

Bergmann

Burgener

Roccabianca

. Primary splenic peripheral nerve sheath tumour in a dog. J Comp Pathol. 2009;141 (2–3):195–198.

Binder

Rukavina

Hassani

. Peripheral nervous system progenitors can be reprogrammed to produce myelinating oligodendrocytes and repair brain lesions. J Neurosci. 2011;31(17):6379–6391.

Bock

Beineke

Techangamsuwan

. Differential expression of HNK-1 and p75 (NTR) in adult canine Schwann cells and olfactory ensheathing cells in situ but not in vitro. J Comp Neurol. 2007;505(5):572–585.

Borrego-Diaz

Terai

Lialyte

. Overactivation of Ras signaling pathway in CD133+ MPNST cells. J Neurooncol. 2012;108(3):423–434.

Chijiwa

Uchida

Tateyama

. Immunohistochemical evaluation of canine peripheral nerve sheath tumors and other soft tissue sarcomas. Vet Pathol. 2004;41(4):307–318.

Cornelis

Chiers

Maes

. Claudin-1 and glucose transporter 1 immunolabelling in a canine intraneural perineurioma. J Comp Pathol. 2012;147 (2–3):186–190.

Crane

Trainor

. Neural crest stem and progenitor cells. Annu Rev Cell Dev Biol. 2006;22:267–286.

Demestre

Herzberg

Holtkamp

. Imatinib mesylate (Glivec) inhibits Schwann cell viability and reduces the size of human plexiform neurofibroma in a xenograft model. J Neurooncol. 2010;98(1):11–19.

10.

Folpe

Billings

McKenney

. Expression of claudin-1, a recently described tight junction–associated protein, distinguishes soft tissue perineurioma from potential mimics. Am J Surg Pathol. 2002;26(12):1620–1626.

11.

Gaitero

Añor

Fondevila

. Canine cutaneous spindle cell tumours with features of peripheral nerve sheath tumours: a histopathological and immunohistochemical study. J Comp Pathol. 2008;139(1):16–23.

12.

Goldschmidt

Hendrick

. Tumors of the skin and soft tissues. In: Meuten

, ed. Tumors in Domestic Animals. 4th ed. Ames: Iowa State Press; 2002:45–117.

13.

Hendrick

Mahaffrey

Moore

. WHO International Histological Classification of Tumors of Domestic Animals Histological Classification of Mesenchymal Tumors of Skin and Soft Tissues of Domestic Animals. 2nd series, vol 3. Washington, DC: Armed Forces Institute of Pathology; 1999.

14.

Hoshi

Hiraki

Yamaki

. Frequent expression of 75 kDa nerve growth factor receptor and phosphotyrosine in human peripheral nerve tumours: an immunohistochemical study on paraffin-embedded tissues. Virchows Arch. 1994;424:563–568.

15.

Ide

Uchida

Kikuta

. Immunohistochemical characterization of canine neuroepithelial tumors. Vet Pathol. 2010;47(4):741–750.

16.

Jakab

Gálfi

Jerzsele

. Expression of claudin-1 in canine peripheral nerve sheath tumours and perivascular wall tumours: immunohistochemical study. Histol Histopathol. 2012;27(7):905–917.

17.

Jouhilahti

Peltonen

Callens

. The development of cutaneous neurofibromas. Am J Pathol. 2011;178(2):500–505.

18.

Jouhilahti

Peltonen

Heape

. The pathoetiology of neurofibromatosis 1. Am J Pathol. 2011;178(5):1932–1939.

19.

Koestner

Bilzer

Fatzer

. WHO International Histological Classification of Tumors of Domestic Animals Histological Classification of Tumors of the Nervous System in Domestic Animals. 2nd series, vol 5. Washington, DC: Armed Forces Institute of Pathology; 1999.

20.

Koestner

Higgins

. Tumors of the nervous system. In: Meuten

, ed. Tumors in Domestic Animals. 4th ed. Ames: Iowa State Press; 2002:697–738.

21.

LeCouteur

Withrow

. Tumors of the nervous system. In: Withrow

Vail

, eds. Withrow and MacEwen's Small Animal Clinical Oncology. 4th ed. Philadelphia, PA: Saunders; 2007:659–685.

22.

Liptak

Forrest

. Soft tissue sarcomas. In: Withrow

Vail

, eds. Withrow and MacEwen's Small Animal Clinical Oncology. 4th ed. Philadelphia, PA: Saunders; 2007:425–454.

23.

Mazzei

Millanta

Citi

. Haemangiopericytoma: histological spectrum, immunohistochemical characterization and prognosis. Vet Dermatol. 2002;13(1):15–21.

24.

Millán

Mascort

Blanco

. Hypomyelination in three Weimaraner dogs. J Small Anim Pract. 2010;51(11):594–598.

25.

Miller

Rangwala

Williams

. Large-scale molecular comparison of human Schwann cells to malignant peripheral nerve sheath tumor cell lines and tissues. Cancer Res. 2006;66(5):2584–2591.

26.

Miller

Jessen

Mehta

. Integrative genomic analyses of neurofibromatosis tumours identify SOX9 as a biomarker and survival gene. EMBO Mol Med. 2009;1(4):236–248.

27.

Palmieri

Avallone

Cimini

. Use of electron microscopy to classify canine perivascular wall tumors. Vet Pathol. 2013;50(2):226–233.

28.

Park

Woo

Jee

. Malignant peripheral nerve sheath tumour in the liver of a dog. J Comp Pathol. 2011;144 (2–3):223–226.

29.

Pérez

Bautista

Rollón

. Immunohistochemical characterization of hemangiopericytomas and other spindle cell tumors in the dog. Vet Pathol. 1996;33(4):391–397.

30.

Rodriguez

Folpe

Giannini

. Pathology of peripheral nerve sheath tumors: diagnostic overview and update on selected diagnostic problems. Acta Neuropathol. 2012;123(3):295–319.

31.

Scheithauer

Woodruff

Erlandson

. Primary malignant tumors of peripheral nerve. In: Rosai

, ed. Atlas of Tumor Pathology. 3rd series, fascicle 24. Washington, DC: Armed Forces Institute of Pathology; 1997:303–421.

32.

Scherer

Bannerman

. Periaxin expression in myelinating Schwann cells: modulation by axon-glial interactions and polarized localization during development. Development. 1995;121(12):4265–4273.

33.

Schöniger

Valentine

Fernandez

. Cutaneous schwannomas in 22 horses. Vet Pathol. 2011;48(2):433–442.

34.

Schöniger

Summers

. Localized, plexiform, diffuse, and other variants of neurofibroma in 12 dogs, 2 horses, and a chicken. Vet Pathol. 2009;46(5):904–915.

35.

Shimada

Tsuzuki

Kuroda

. Nestin expression as a new marker in malignant peripheral nerve sheath tumors. Pathol Int. 2007;57(2):60–67.

36.

Spyra

Kluwe

Hagel

. Cancer stem cell–like cells derived from malignant peripheral nerve sheath tumors. PLoS One. 2011;6(6):e21099. doi:10.1371/journal.pone.0021099.

37.

Staser

Yang

Clapp

. Mast cells and the neurofibroma microenvironment. Blood. 2010;116(2):157–164.

38.

Stefanello

Avallone

Ferrari

. Canine cutaneous perivascular wall tumors at first presentation: clinical behavior and prognostic factors in 55 cases. J Vet Intern Med. 2011;25(6):1398–405.

39.

Stoica

Tasca

Kim

H-T

. Point mutation of neu oncogene in animal peripheral nerve sheath tumors. Vet Pathol. 2001;38(6):679–688.

40.

Sugawara

Kurihara

Negishi

. Nestin as a marker for proliferative endothelium in gliomas. Lab Invest. 2002;82(3):345–351.

41.

Sugiyama

Morita

Shimada

. Primary malignant peripheral nerve sheath tumor with eosinophilic cytoplasmic globules arising from the greater omentum in a dog. J Vet Med Sci. 2008;70(7):739–742.

42.

Thway

Fisher

Debiec-Rychter

. Claudin-1 is expressed in perineurioma-like low-grade fibromyxoid sarcoma. Hum Pathol. 2009;40(11):1586–1590.

43.

Tovar

Obendorf

Murchison

. Tumor-specific diagnostic marker for transmissible facial tumors of Tasmanian devils: immunohistochemistry studies. Vet Pathol. 2011;48(6):1195–1203.

44.

Uchida

Nakayama

Sasaki

. Malignant schwannoma in the spinal root of a dog. J Vet Med Sci. 1992;54(4):809–811.

45.

Weiss

Goldblum

. Malignant tumors of the peripheral nerves. In: Enzinger

Weiss

Goldblum

, ed. Enzinger and Weiss’s Soft Tissue Tumors. 5th ed. St Louis, MO: Mosby Elsevier; 2008:903–944.

46.

Wohlsein

Cichowski

Baumgärtner

. Primary endocardial malignant spindle-cell sarcoma in the right atrium of a dog resembling a malignant peripheral nerve sheath tumour. J Comp Pathol. 2005;132(4):340–345.

47.

Yokoo

Nobusawa

Takebayashi

. Anti-human Olig2 antibody as a useful immunohistochemical marker of normal oligodendrocytes and gliomas. Am J Pathol. 2004;164(5):1717–1725.