Nerve Sheath Tumor in a Cynomolgus Macaque (Macaca fascicularis) and Diagnostic Differentiation from Amelanotic Melanoma

Abstract

There are few reports describing spontaneous neoplasms in cynomolgus macaques, despite the frequent use of this species in laboratory research. This report describes cytologic, histologic, and immunohistochemical findings of a cutaneous to subcutaneous nerve sheath tumor located within the haired skin of the abdomen of a 2.5-y-old, intact, female, captive Mauritius cynomolgus macaque. The nerve sheath tumor was well demarcated, partially encapsulated, densely cellular, and extended from the subcutis to the most superficial dermis, abutting the epidermis. Neoplastic cells formed intersecting streams and had a high mitotic count (18 per 2.37 mm²). Due to the substantial morphologic overlap of this neoplasm with amelanotic melanoma, particularly the close association with the epidermis, immunohistochemistry was required for definitive diagnosis. Neoplastic cells were immunoreactive to vimentin, S-100, SOX10, laminin, collagen IV, and CD56, and negative for melan-A, tyrosinase, MITF, and HMB45. This immunohistochemical profile is diagnostic for nerve sheath tumor based on human and canine criteria and rules out amelanotic melanoma. Despite incomplete excision, the nerve sheath tumor had not grossly recurred after 1 mo, at which point the animal was euthanized for unrelated reasons. This report underscores the importance of using an immunohistochemical panel in cases of cutaneous and subcutaneous spindle cell neoplasms, as there is substantial morphologic and immunohistochemical overlap between nerve sheath tumors and melanocytic neoplasms due to their shared neuroectodermal origin. To our knowledge, this is the first report of a nerve sheath tumor in a cynomolgus macaque, and one of the few reports of spontaneous neoplasia in this species.

Introduction

Cynomolgus macaques (CM; Macaca fascicularis) are nonhuman primates commonly used in pharmaceutical discovery and safety and efficacy toxicology studies.¹⁴ Even though this species is commonly used for laboratory research, there are few reports highlighting spontaneous neoplasms in these animals.^11,18 Most spontaneous neoplasms in CM are of epithelial origin, specifically of the endocrine, respiratory, and female reproductive systems. In a report describing spontaneous neoplasms in 2 populations of captive rhesus macaques (also commonly used in pharmacological studies), neural origin neoplasia represented less than 1% of all neoplasms.^1,18

Spindle cell neoplasms commonly encountered in veterinary medicine include both nerve sheath tumors (NSTs) and melanocytic neoplasms.⁵ Differentiating between these tumors may be challenging,²¹ as they often share similar morphologic features, particularly for pigmented variants of NSTs and amelanotic melanomas.⁷ In addition, NSTs and melanocytic neoplasms both arise from neural crest cells,² creating substantial immunohistochemical overlap.²⁰ The shared morphologic and immunohistochemical features of these neoplasms may present a diagnostic conundrum for pathologists.

The aim of this study was to review the cytologic and histologic morphology and immunohistochemical profiles of NSTs and melanomas that can be used to differentiate these neoplasms. The objective was to describe a case of a cutaneous to subcutaneous NST in a CM and discuss how it was differentiated from amelanotic melanoma.

Case Report

An approximately 2.5-y-old, intact, female, Mauritius CM was presented for veterinary assessment of an approximately 4-cm-diameter, cutaneous to subcutaneous, spherical, firm, slightly mobile, ulcerated, rapidly enlarging mass in the left abdominal region. The animal was captive bred in the United States and resided at a research facility. It resided with one social partner, had never been bred, and had never been used for a study. The animal had no additional clinical signs outside of the mass. Based on the young age of the animal and the rapidly growing nature of the mass, an abscess or foreign body was suspected. The animal was treated empirically with a subcutaneous dose of ceftiofur (Excede, 20 mg/kg SC; Zoetis Services). After 1 wk, there was no clinical improvement and mass continued to enlarge. Additional diagnostics were pursued, as the animal was not responding to empirical therapy. Fine-needle aspirates were performed for cytology, and swabs were submitted for bacterial and fungal cultures.

Cytologically, samples were moderately cellular with considerable hemocontamination. Nucleated cells were spindle shaped and presented individually or in cohesive aggregates (Figure 1). Spindle cells had mild to moderate anisocytosis and anisokaryosis with cells extending into long cytoplasmic tendrils. Nuclei were oval with fine granular chromatin and one or multiple small nucleoli. Rarely, binucleation and micronuclei were noted. Cells were occasionally associated with amorphous magenta material (matrix). Moderate numbers of mixed inflammatory cells were present with predominance of mostly nondegenerate neutrophils. Rare intracellular (in neutrophils) and extracellular cocci were identified. Occasional histiocytic erythrophagia was observed, and rare hematoidin crystals were seen. The findings were consistent with a spindle cell proliferation and septic mixed (primarily suppurative) inflammation with evidence of hemorrhage. No bacterial or fungal organisms were isolated in culture. Histology was recommended, and the lesion was surgically excised. Prior to surgical excision, chest and abdominal radiographs, complete blood count, and chemistry were performed; the results were within normal limits. For surgical removal of the mass, an aseptic technique was used. In brief, midazolam, buprenorphine, ketamine, dexmedetomidine, maropitant, and isoflurane were used for surgery. Bupivacaine was applied locally at the end of surgery. Meloxicam and buprenorphine were used for postoperative pain management. This work was done as part of standard veterinary care.⁴ The mass had not recurred at 1 mo after excision, at which point the animal was euthanized for unrelated reasons. A necropsy was performed and there was no gross evidence of regrowth of the mass or metastasis.

Figure 1.

Cytologic sections from a fine-needle aspirate of cutaneous to subcutaneous nerve sheath tumor in a macaque. Wright–Giemsa stain, 40× objective. (A) Neoplastic spindle cells form an aggregate with embedded pink matrix. (B, C) Individualized spindle cells have either veiling cytoplasm and oval nuclei (B), or bipolar cytoplasmic tails and elongate nuclei (C).

Histologically, expanding and effacing the dermis and subcutis, and elevating and extending to the epidermis, there was a round, partially encapsulated, well-demarcated, highly cellular neoplasm arranged in long streams and bundles of neoplastic spindle cells on a fine fibrovascular stroma (Figure 2). Neoplastic cells had indistinct cell borders and a small amount of eosinophilic slightly fibrillar cytoplasm. Nuclei were round to oval or fusiform with 1 to 2 distinct nucleoli and finely stippled chromatin. Mitotic count was up to 18 per 2.37 mm². Neoplastic cells had minimal anisocytosis and minimal to mild anisokaryosis. Neoplastic cells rarely formed vague Antoni A patterns. Few lymphocytes and plasma cells infiltrated the neoplasm and surrounding tissue, which had rare skeletal myocyte degeneration and regeneration and increased numbers of small-caliber blood vessels (neovascularization). A small amount of hemorrhage and edema expanded the neoplasm. The epidermis was mildly hyperplastic with blunted rete pegs, and focally ulcerated with superficial cocci. Neoplastic cells extended to the surgical margin (incomplete surgical excision). The neoplasm was consistent with a spindle cell neoplasm. Sarcoma was favored, but due to the close apposition to the epidermis, amelanotic melanoma was also considered. Immunohistochemical assessment was performed at Walter Reed National Military Medical Laboratory (Bethesda, MD) for further characterization.

Figure 2.

Histologic sections of cutaneous to subcutaneous nerve sheath tumor in a macaque. (A) A poorly demarcated and mildly invasive neoplasm expands the dermis and subcutis and extends to the epidermis, which is regionally ulcerated. Hematoxylin and eosin, 4× objective. (B) Neoplastic spindle cells are arranged in long streams. Hematoxylin and eosin, 10× objective. (C) Neoplastic cells are rarely separated by hemorrhage. Hematoxylin and eosin, 40× objective.

Immunohistochemical stains applied to the neoplasm are provided in Table 1. In summary, the neoplastic cell population exhibited strong diffuse immunoreactivity to vimentin, S100, CD56, and collagen IV; approximately 40% of cells had strong immunoreactivity to SOX10 (Figure 3). The neoplastic cells were diffusely immunonegative for melan-A, pan-actin, desmin, tyrosinase, microphthalmia-associated transcription factor, and human melanoma black. Each immunohistochemical stain had appropriate positive and negative external primate (human) controls and appropriate internal (CM) controls. Based on cellular morphology and immunohistochemical profile, the neoplasm was diagnosed as a nerve sheath tumor.

Table 1.

Immunohistochemical antibodies applied to a nerve sheath tumor in a macaque and their expression by neoplastic cells

Expression by neoplastic cells	Antibody	Manufacturer (catalog number, clone)	Origin (dilution)	Control tissue (human)
Positive	CD56	Cell Marque (156R-96, MRQ-42)	RM (1 in 100)	Peripheral nerve
	SOX10	Biocare (API3099AA, BC34)	MM (proprietary)	Cutaneous melanocytes
	Vimentin	Ventana (790-2917, V9)	MM (proprietary)	Cartilage
	S100	Dako (Z0311, NA)	RP (1 in 200)	Peripheral nerve
	Collagen IV	Ventana (760-2632, CIV22)	MM (proprietary)	Skeletal muscle
Negative	Desmin	Ventana (760-2513, DE-R-11)	MM (proprietary)	Skeletal muscle
	Pan-actin	Ventana (760-2601, HHF 35)	MM (proprietary)	Small intestinal muscularis
	Melan-A	Ventana (790-2990, A103)	MM (proprietary)	Cutaneous melanocytes
	Tyrosinase	Ventana (790-4365, T311)	MM (proprietary)	Cutaneous melanocytes
	MITF	Ventana (790-4367, C5/D5)	MM (proprietary)	Cutaneous melanocytes
	HMB45	Ventana (790-4366, HMB45)	MM (proprietary)	Cutaneous melanocytes

All immunohistochemical stains were performed at Walter Reed National Military Medical Center Laboratory (Bethesda, MD) and were performed according to the manufacturers’ instructions and laboratory protocols. HMB45, human melanoma black 45; MITF, microphthalmia-associated transcription factor; MM, mouse monoclonal; NA, not available; RM, rabbit monoclonal; RP, rabbit polyclonal; SOX10, SRY-box transcription factor 10.

Figure 3.

Immunohistochemistry of a nerve sheath tumor in a macaque. (A) CD56 cytoplasmic immunoreactivity of neoplastic cells. 40× objective. (B) SOX10 nuclear immunoreactivity of neoplastic cells. 40× objective.

Discussion

Nerve sheath tumors (NSTs) are a category of soft tissue tumors that arise from cellular components of peripheral nerves, including Schwann cells, fibroblasts, and other perineural cells. To our knowledge, this is the first report of a NST in a cynomolgus macaque, which was a cutaneous to subcutaneous mass that did not recur grossly after 1 mo following incomplete surgical excision. However, lack or recurrence was determined based on gross necropsy examination, and microscopic recurrence at the surgical margin was not excluded. The mass was ulcerated but was not associated with substantial morbidity. This is in contrast to a previous report of an intrathoracic Schwannoma in a rhesus macaque that was euthanized due to labored breathing secondary to the space-occupying mass.¹ The rhesus macaque case also differed from our case in that cytology was not performed and melanoma was not a likely differential due to the intrathoracic location. Both this case and the previously referenced rhesus macaque had spontaneous NSTs, which are apparently rare in macaques, although multiple cases of radiation-induced NSTs have been reported.^10,17

NSTs are reported widely in people and many domestic species. They are most well characterized in people and dogs, who frequently develop subcutaneous NSTs.^9,19,22 NSTs are broadly categorized as benign nerve sheath tumors (i.e., Schwannoma, perineurioma, neurofibroma, or hybrid tumors) or malignant nerve sheath tumors, based on morphologic criteria and biologic behavior.³ Benign NSTs generally present as localized nodular masses that are well demarcated and partially to fully encapsulated. This is similar to our case, which was well demarcated and partially encapsulated. Benign NSTs lack vascular invasion but may have small areas of hemorrhage and/or necrosis; a small amount of hemorrhage was present in this case. Tumor cells in benign NSTs demonstrate mild to no atypia and have a low mitotic count (less than 3 per 2.37 mm²).^3,22 Benign features in this case included clear demarcation and partial encapsulation; however, the mitotic rate was much higher than is typical in benign NSTs.

Malignant NSTs are generally poorly circumscribed, unencapsulated, and demonstrate infiltrative growth. They are typically highly cellular, with small to moderate amounts of stroma.^3,13,22 In human pathology, malignant NSTs most commonly occur on the extremities, trunk, and head; the sciatic nerve, brachial plexus, and sacral plexus are the most commonly affected large nerves. Histologically, classic malignant NSTs in people are defined by hypercellularity and a mitotic count of greater than or equal to 4 per 2.37 mm², both of which were present in this case.⁹ In people, malignant NSTs are divided into low and high grade. Low-grade malignant NSTs lack tumor necrosis and have a mitotic rate between 4 and 9 per 2.37 mm². High-grade malignant NSTs have either a combination of necrosis and 4 to 9 mitoses per 2.37 mm², or they have greater than 9 mitoses per 2.37 mm² irrespective of necrosis.¹³ Despite a lack of necrosis, the high mitotic rate in this case (i.e.,18 per 2.37 mm²) would classify this is a high-grade malignant NST according to the human criteria. A three-tier system is also sometimes used in malignant nerve sheath tumors in people, although this scheme is less specific, as it is applied across all soft tissue sarcomas, regardless of cellular origin.¹² In domestic animals, the criteria for malignancy in NSTs are less well defined. Some authors prefer to separate NSTs into benign and malignant, while others separate NSTs into low and high grade without specifying benign or malignant.^8,9 Histologic features that have been ascribed to malignant or high-grade NSTs in domestic animals include high cellular density, moderate pleomorphism, “extremely large nuclei,” and/or areas of necrosis and hemorrhage. Of these features, high cellular density and hemorrhage were present, the latter of which was a minor component. As there are different criteria for malignancy in NSTs in domestic animals and humans, and prognostic criteria for NSTs are not available in CM, we could not definitively further classify this NST. Although most veterinary pathology reports include only veterinary criteria of malignancy, human criteria were also applied in this case, as CM and humans are both primates. Regardless of histologic criteria, given the history of rapid growth and the high mitotic rate, the diagnosis of malignant NST is favored.

The primary differential diagnosis for this case was amelanotic melanoma. Cytologically, differentiation between sarcoma and amelanotic melanoma is often impossible. The cytologic morphology of malignant melanomas ranges from epithelioid to spindloid, and poorly differentiated melanomas may lack cytoplasmic melanin granules.¹⁶ Neoplastic melanocytes are often anaplastic and spindloid, making them impossible to differentiate from a sarcoma or poorly differentiated carcinoma.¹⁵ In veterinary pathology, there are reports of amelanotic melanomas misdiagnosed as sarcomas using cytology, and, conversely, sarcomas misdiagnosed as melanomas.^6,15 Cytologic criteria of malignancy in veterinary NSTs are not established. Both benign and malignant NSTs may have moderate to high numbers of spindle cells with variably distinct nucleoli. Benign NSTs can have cytologic features of atypia, mimicking malignant NSTs.¹⁶ Conversely, malignant NSTs can have mixed cytomorphology, and malignant features may be lacking depending on the region sampled for cytology. For these reasons, histology and immunohistochemistry are required to distinguish between NSTs and melanoma, and to screen for malignancy in NSTs.

Histologically, melanocytic neoplasms are assessed based on nuclear atypia, predominant cell type (spindloid, epithelioid, mixed), cellular pleomorphism, ulceration, level of infiltration or invasion, necrosis, and inflammation.²¹ They are most often cutaneous,²³ but can extend into the subcutis. Amelanotic and spindloid variants, particularly those extending into the subcutis, mimic sarcomas.²¹ Due to this histologic overlap, including lack of pigmentation and presence of neoplastic cells in both the skin and subcutis, melanoma was the primary differential in this case, necessitating immunohistochemistry.

There is substantial immunohistochemical overlap between NSTs and melanoma due to their shared neuroectodermal origin.^9,21 Vimentin, S-100, and SOX10 are immunoreactive in both NSTs and melanoma, and they were immunoreactive in this case. Antigens that are present in NSTs, but lacking in melanomas, include laminin, collagen IV, and CD56, all of which were immunoreactive in this case. Antigens that are present in melanoma, but lacking in NSTs, include melan-A, tyrosinase, MITF, and HMB45, all of which were negative in this case. This immunohistochemical profile was used to confirm a NST and rule out amelanotic melanoma.

In conclusion, to our knowledge, this case is the first description of a NST in a CM. The NST was in the skin and subcutis and did not recur grossly for 1 mo following incomplete excision. There is substantial cytologic, histologic, and immunohistochemical overlap between NSTs and amelanotic melanoma, especially for spindle cell neoplasms that are present in both the skin and subcutis. Immunohistochemistry was required to confirm this as a NST and rule out amelanotic melanoma.

Conflict of Interest

The views expressed in this manuscript are those of the authors and do not necessarily reflect the official policies of the U.S. Government or Department of Defense.

Funding

This work was internally funded.