Giant Thoracic Schwannoma in a Rhesus Macaque (Macaca mulatta)

Abstract

A 15-y-old male rhesus macaque with a 3-d history of labored breathing, was culled from a nonhuman primate research colony after thoracic radiographs and exploratory surgery revealed a 10-cm, well-circumscribed space-occupying mass in the posterior thoracic cavity. The multilobulated cystic and necrotic neoplasm was composed of interlacing streams and fascicles of neoplastic spindle cells arranged in Antoni A, and less commonly, Antoni B patterns. Verocay bodies were present also. The neoplasm was encapsulated mostly, and histomorphologic features were benign. Immunohistochemistry indicated that neoplastic cells were positive for vimentin, S100, glial fibrillary acidic protein, and nerve growth factor receptor. Reticulin histochemical staining and immunohistochemical stains for collagen IV and laminin showed a prominent basal lamina surrounding the neoplastic cells. The histologic features and results of the immunohistochemical stains confirmed peripheral nerve origin and were consistent with schwannoma. To our knowledge, this is the first case of thoracic schwannoma in a rhesus macaque and the second reported case of schwannoma in a nonhuman primate.

Peripheral nerve sheath tumors (PNSTs) originate from the cells that comprise the sheaths of peripheral nerves. PNSTs that arise from Schwann cells are termed schwannomas, whereas 2 other types of cells, perineural cells and fibroblasts, are components of peripheral nerve sheaths and may also give rise to tumors.^6,17 In humans, schwannomas (synonyms include neurilemmoma and neurinoma) and neurofibromas are the most frequent benign PNSTs,²⁴ although malignant PNSTs, mostly arising from a peripheral nerve or neurofibromas, have also been described.²³ Although most schwannomas are sporadic, they occur in approximately 90% of human patients with neurofibromatosis (NF) type 2 due to alteration or loss of the product of the NF2 gene on chromosome 22, also known as Merlin (or schwannomin), a tumor suppressor gene.^23,30 Vestibular (acoustic) tumors arising from the vestibular nerve are diagnostic of NF2.^16,24 Of note, schwannoma is the most common nerve sheath tumor in the posterior mediastinum, and the majority of the patients with a tumor in this location are asymptomatic.¹⁵

PNSTs have been reported to occur in animals, including dogs^12,14,25,28 and cattle^3,5,18 most commonly, but also in other species such as cats,^12,26 goats,^20,31 pigs,²⁹ horses,^25,27 and birds.^4,19,25 Unlike those for humans, distinct histologic criteria that classify the benign PNSTs as schwannoma or neurofibroma in animals have not been established,^9,13 and all benign PNSTs are thought to have a similar prognosis.^9,25 Furthermore, NF2 and NF1 (von Recklinghausen disease) are not recognized disease entities in veterinary medicine. Therefore, many veterinary pathologists and authors of case reports in animals prefer the generic term ‘benign PNST’ instead of schwannoma or neurofibroma.^9,13,14,25

Two variants of PNST have been described in the dog: one of uncertain histogenesis that occurs in the skin and subcutis, has the tendency to recur but rarely metastasizes, and is termed hemangioperiocytoma by some pathologists^8,26 and another that resembles human malignant PNST occurring in the cranial and spinal nerves with metastatic potential.^26,28 Recently (2009), using morphologic features diagnostic for human neurofibroma such as growth patterns (localized, plexiform, and diffuse) and microscopic features (classic, cellular, collagenous, and pigmented), selective immunostaining of a subpopulation of neoplastic cells with S100 with the presence of intratumoral nerve fibers, and electron microscopic examination, investigators determined that identical subtypes of neurofibroma exist in certain animal and human cases.²⁵ Nonetheless, their findings reiterate the increased complexity of PNSTs in animals.

In cattle, PNSTs are frequently multicentric, particularly arising from brachial, celiac, and coronary plexuses and in the intercostal nerves, causing variously sized neoplastic nodules to develop throughout the body.^3,5 However, despite having neoplastic cells with atypical features and the potential to metastasize (both of these characteristics are consistent with malignancy), PNSTs in cattle are rarely associated with clinical disease and are often noted in animals slaughtered at abattoirs.^3,5,18 Perhaps completely unrelated to the pathogenesis, a viral etiology has been suggested as the underlying cause in some cases of malignant schwannoma in cattle.¹⁸ This observation was based on a cluster of malignant schwannoma cases with a restricted geographical distribution, histopathologic and immunohistochemical evidence confirming the diagnosis, and the presence of virus-like particles in neoplastic Schwann cells from 3 of 5 cases examined by electron microscopy. However, even though virus-like particles were present, attempts to isolate a viral agent were unsuccessful.

Barring intestinal adenocarcinoma in aged rhesus monkeys, tumors of any type in immunocompetent macaques are rare;²¹ therefore, making the present case noteworthy. With the exception of an intestinal schwannoma reported in a rhesus macaque,² to our knowledge, no other published literature on PNSTs in nonhuman primates is available. Here, we describe the gross, histologic, and immunohistochemical features of thoracic schwannoma in a rhesus macaque (Macaca mulatta).

Case Report

A 15-y-old male rhesus macaque was presented to the Veterinary Medicine Division (US Army Medical Research Institute of Infectious Diseases, Fort Detrick, MD) after a physical confrontation with another rhesus macaque from the nonhuman primate colony. The facility is fully AAALAS-accredited. This animal had arrived at the facility approximately 1 y earlier from a commercial primate vendor (Osage Research Primates, Kaiser, MO) and had an insignificant clinical history. Previously assigned to a staphylococcal enterotoxin B and Bacillus anthracis study, this macaque was on a phlebotomy protocol when the incident occurred. Although the macaque was not assigned to an active scientific protocol, research at the facility is conducted in compliance with the Animal Welfare Act¹ and other federal statutes and regulations relating to animals and experiments involving animals and adheres to principles stated in the Guide for the Care and Use of Laboratory Animals.¹¹

On initial evaluation, the macaque was anesthetized, and a thorough physical examination was conducted. Multiple, superficial, and deep facial lacerations involving primarily the nose, mouth, and cheeks were the most significant clinical findings. The nasal septum was lacerated, but there was no involvement of the underlying nasal bone. Palpation of limbs and abdomen revealed no obvious injuries. Labored respirations, attributed to significant nasal bleeding associated with the traumatic injury to the nasal septum, was a noteworthy clinical finding. After placement of a nasal catheter, respirations improved, and the nasal septum and facial lacerations were closed surgically by using sutures or tissue glue, depending on the size of the wound. The macaque recovered uneventfully from anesthesia and received meloxicam and buprenorphine for inflammation and pain management, respectively, and broad-spectrum antibiotics (cefazolin and enrofloxacin) according to specification.

Over the next 48 to 72 h, the animal became dyspneic, which progressively worsened by 96 h (day 4). On day 5, the animal was anesthetized for physical examination and thoracic radiography. Although the facial lacerations were healing well, cardiovascular and thoracic auscultation revealed increased lung sounds in the midventral area bilaterally, but no crackles or rales were heard. Organomegaly was present in the cranial left abdomen. Lateral and ventrodorsal thoracic radiographs showed a large, circular, soft-tissue opacity that filled the thorax from the 5th to 12th rib space. The cardiac silhouette was barely perceptible, and the diaphragm was indistinguishable from the soft-tissue opacity. Exploratory surgery revealed a large, well-circumscribed, 8- to 10-cm, firm mass adjacent to the diaphragm. After thorough clinical evaluation and consideration, the animal was euthanized prior to recovery from anesthesia because of poor prognosis. The animal was submitted for a full necropsy, which was performed in a Biosafety Level 2 necropsy suite.

Gross necropsy findings.

With the exception of an 8-cm ventral thoracic and abdominal incision made during exploratory surgery, the carcass was in good postmortem condition with normal amounts of subcutaneous and visceral adipose tissue. Multiple, healing, surgically repaired (sutured) lacerations were present, primarily on the right side of the face and on the nose. A large, 8- to 10-cm, firm, well-circumscribed, round, 447-g thoracic mass filled the posterior thoracic cavity (particularly the right hemithorax) and displaced the heart superiorly and anteriorly toward the sternum. The esophagus and caudal vena cava were tightly adhered but easily separated from the curvature of the mass. The tumor also adhered to the diaphragm and both the left and right thoracic walls by numerous fibrous adhesions. Despite its extensive size and numerous fibrous adhesions, the mass was easily removed (‘shelled out’) with the pluck (Figure 1 A). A focally extensive, 2- to 3-cm area adjacent to the thoracic vertebrae and spanning the 7th, 8th, and 9th intercostal spaces contained a poorly defined stalk to which the tumor was firmly attached and from which it did not dislodge as easily as in other areas. On cut section, the tumor had a 2- to 3-mm thick capsule that surrounded pale tan, multilobulated, parenchymal tissue with extensive coalescing cystic areas, hemorrhage, fibrin, and necrosis (Figure 1 B). The accessory lung lobe and caudal segment of the left cranial lung lobe were atelectic, congested, and adhered to the tumor; both lobes separated easily from the tumor. Although slightly displaced and compressed, remaining lung lobes appeared grossly normal, as did the trachea and bronchi. The mediastinal lymph nodes were multifocally brownish-black (confirmed histologically as anthracosilicosis) but normal in size and consistency on cut section. A smaller, 1- to 2-cm well-circumscribed tumor that resembled the larger thoracic tumor was present on the right thoracic wall at the level of the 8th rib along the axillary line. No other significant gross lesions were observed in other organs at necropsy.

Figure 1.

Thoracic schwannoma in a male rhesus macaque. (A) The expansile and compressive neoplasm (*) is well-circumscribed and firm and has a smooth capsular surface, except where it attached to the thoracic vertebral bodies (arrow). (B) On cut section, the encapsulated neoplasm is multilobulated and cystic, with extensive areas of hemorrhage and necrosis. Note the irregular, roughened areas on the capsular surface corresponding to the areas of attachment to the vertebral bodies (arrows).

Histopathologic findings.

Tissues were fixed in 10% neutral buffered formalin. The tissues were trimmed and processed according to standard protocol. Histology sections were cut at 5 to 6 µm on a rotary microtome, mounted on glass slides, and stained with hematoxylin and eosin. Histologically, both expansile tumors were multilobulated, often surrounded by a thick fibrous capsule, and composed of spindled cells arranged in short interlacing streams and fascicles with indistinct cell borders, small amounts of eosinophilic fibrillar cytoplasm, oval to elongate nuclei with finely to coarsely stippled chromatin and generally indistinct nucleoli. Depending on the section of tumor, the mitotic rate varied from less than 1 mitotic figure per 10 high-power (magnification, ×400) fields to 1 mitotic figure per high-power field; the significance of this finding was unknown. In central areas of the larger tumor, multifocal hemorrhage, necrosis, and mineralization were consistent findings, and larger caliber vessels often had thickened, hyalinized walls. There was conspicuous alternating cellular density; in the more predominant densely cellular areas (Antoni A, Figure 2 A), cells were separated by abundant collagenous matrix, whereas the less frequent sparsely cellular areas (Antoni B, Figure 2 B) cells were less orderly and separated by a loose collagenous matrix. Multifocally in the densely cellular areas, there was regimentation of nuclei, with 2 well-aligned compact rows of palisading nuclei separated by fibrillary cell processes (Verocay bodies, Figure 2 C). Neither tumor infiltrated the adjacent ribs or vertebral bodies. Neoplastic cells were not noted in the respiratory tract or remaining organ systems, and the tumor did not invade the adjacent vertebral bodies, ribs, lymphatics, or blood vessels. The histologic appearance of both tumors was most consistent with PNST.

Figure 2.

Serial sections of the thoracic schwannoma; rhesus macaque. (A) Streams of densely packed neoplastic cells (Antoni A) are separated by a collagenous matrix. Hematoxylin and eosin stain. (B) Neoplastic cells are frequently separated by a loose collagenous matrix in less cellular areas (Antoni B). Hematoxylin and eosin stain. (C) Nuclear and cellular regimentation separated by fibrillary cell processes (Verocay bodies) is common. Hematoxylin and eosin stain. (D) Neoplastic cells are diffusely immunoreactive for S100 protein. (E) There is multifocal cellular immunoreactivity with glial fibrillary acidic protein. (F) Diffusely, neoplastic cells are intensely immunoreactive for nerve growth factor receptor protein. (G) Neoplastic cells are individually surrounded by reticulin (black and linear) due to the presence of pericellular basement membranes. (H, I) Prominent basal lamina surrounding neoplastic cells is demonstrated with positive (H) collagen IV and (I) laminin immunohistochemical staining, respectively. Magnification, ×40.

Immunohistochemical findings.

A battery of immunohistochemical stains was performed to further characterize the tumors. By immunohistochemistry, neoplastic spindled cells were diffusely positive for vimentin and S100 protein (Figure 2 D) and multifocally positive for glial fibrillary acidic protein (Figure 2 E), indicating peripheral nerve origin. Diffusely neoplastic cells also expressed strong cytoplasmic immunoreactivity for nerve growth factor receptor protein (Figure 2 F). The reticulin stain and immunohistochemical stains for collagen IV and laminin demonstrated a prominent basal lamina surrounding neoplastic cells supporting the diagnosis of schwannoma (Figures 2 G to I). Laminin and collagen IV immunostaining was more predominant in the Antoni A areas, consistent with the staining pattern reported for schwannomas in humans.²⁴ In addition, this macaque case was studied in consultation with human pathologists (Departments of Neuropathology and Soft Tissue Pathology, Armed Forces Institute of Pathology, Washington, DC). Pathologists from both Departments concurred with diagnosis of PNST origin and indicated that the histologic features and immunohistochemical findings were most consistent with schwannoma.

Discussion

On the basis of the gross and histologic appearance, the large space-occupying thoracic tumor observed radiograpically and perioperatively was consistent with a PNST; the labored breathing observed clinically was attributed to the tumor. Prior to the behavioral confrontation with another rhesus monkey that precipitated the sudden physical exertion and subsequent respiratory distress, there were no outward clinical signs in the presented macaque to indicate such an extensive intrathoracic lesion as we describe in this report. In this case, we prefer the term ‘giant’ thoracic schwannoma. Schwannomas generally do not grow larger than 8 cm, and giant schwannomas differ only because of the substantial size they can reach.^7,32

In addition to characteristic histologic features, neoplastic cells exhibited an intense positive immunoreactivity for vimentin, S100, glial fibrillary acidic protein, and nerve growth factor receptor. Furthermore, the unique immunostaining pattern with collagen IV and laminin, demonstrating prominent pericellular basal lamina, distinguished this PNST from other soft tissue sarcomas and confirmed a diagnosis of primary thoracic schwannoma. In this case, the firm attachment adjacent to the 7th to 9th thoracic vertebrae suggested that the tumor originated from a thoracic spinal nerve. The smaller, 2-cm mass on the right thoracic wall also was consistent with schwannoma and likely originated from an intercostal nerve. Although the histologic and immunohistochemical features were confirmatory for a diagnosis of schwannoma, we considered evaluation by electron microscopy. Well-differentiated neoplastic Schwann cells with long, interlacing cytoplasmic processes that are partially to completely covered by an occasionally duplicated basal lamina are typical findings in human schwannoma.²⁴

Due to tumor size, multiple anatomic locations, and various specific histologic features, we considered the possibility of malignant transformation of the PNSTs in this macaque. Humans diagnosed with malignant PNSTs have a poor prognosis, and the following general microscopic features characterize most malignant PNSTs:^23,26 high-grade, with pattern variation but, very uncommonly, nuclear palisading; mitotic index of at least 4 per 10 high-power fields; areas of geographic necrosis with or without pseudopalisading (perivascular sparing of tumor cell cuffs); patchy S100 positivity that varies in intensity; variable staining with glial fibrillary acidic protein; and direct soft tissue spread in addition to intraneural extension or vascular and osseous invasion. In the case of this rhesus macaque, both tumors were well-circumscribed and did not appear to infiltrate surrounding tissues, vessels, or lymphatics; cellular and nuclear atypia were negligible; and the cystic and necrotic areas were present only in the larger tumor and most likely resulted from expansile neoplastic growth that gradually overwhelmed the vascular support network. Although the mitotic rate was increased in some areas of the neoplasm, mitotic activity and degenerative atypia can occur in longstanding or ‘ancient’ tumors and thus are not considered exclusionary for the diagnosis of schwannoma.^24,26

Because thoracic schwannoma in nonhuman primates had not been reported previously, recommendations regarding clinical or surgical management, treatment, and prognosis are not available but likely resemble those for human cases. This clinical history (or lack thereof) is consistent with human cases, in which the majority of patients with thoracic schwannomas are asymptomatic.¹⁵ In humans, these slow-growing masses generally are recognized only after they have reached a very conspicuous size.²⁴ Our case closely resembles that of an asymptomatic giant thoracic schwannoma in a 59-y-old woman, which was found incidentally during chest radiographs.¹⁵ The patient recovered uneventfully following surgical removal of the tumor through a posterolateral thoracotomy and right lower lobectomy. When clinical signs occur in human cases, the patient generally improves markedly postoperatively, thereby suggesting that early diagnosis and treatment may lead to an improved clinical outcome.^15,32In light of the clinical and perioperative findings of our case, the initial prognosis was deemed poor, with minimal likelihood of the macaque's inclusion on a future animal protocol. Therefore, after thorough clinical consultation, a decision was made in favor of euthanasia. However, early diagnosis of thoracic schwannoma in nonhuman primates likely would improve the prognosis.

Schwannoma and neurofibroma, the 2 most common benign PNSTs, may occur sporadically or in association with NF1 (von Recklinghausen disease) or NF2. As discussed earlier, a strong association exists between schwannomas and patients with NF2.^23,30 In contrast, patients with NF1 are more prone to development of neurofibromas associated with a germline mutation in the NF1 gene.²² A third major form of neurofibromatosis, known as schwannomatosis, causes multiple schwannomas without the vestibular tumors diagnostic of NF2. Epidemiologic studies suggest that schwannomatosis is as common as NF2 but that familial occurrence is rare.^16,22 Over the years, multiple approaches, particularly involving mouse models of neurofibromatoses, have been developed to further characterize the disease process and tumorigenesis seen in human cases of both NF1 and NF2.¹⁰ First-generation knockout mouse modeling strategies such as NF1^+/– and NF2^+/– knockout mice fail to develop the tumor spectra observed in humans, thereby necessitating more elaborate approaches such as tissue-specific and conditional mutations.¹⁰ However, several limitations have also been noted in some of the newest generations of mouse models, thereby indicating the necessity of an animal model that closely mimics the human conditions. We do not know whether neurofibromatoses (NF1, NF2, schwannomatosis) are specific disease entities in rhesus macaques given that, to our knowledge, scientific literature and data are unavailable. Nevertheless, a nonhuman primate model could be extremely valuable in understanding the molecular pathogenesis of NF-associated tumors.

In conclusion, we here report the first case of giant thoracic schwannoma in a rhesus macaque, highlighting clinical, gross, histologic, and immunohistochemical features that most closely resemble the condition that occurs in humans. Although likely sporadic in nature, we could not definitively exclude an underlying genetic predisposition or, even possibly, an underlying viral etiology. Nonetheless, thoracic schwannoma should be considered as a possible cause for any space-occupying mass in the thoracic cavity of a nonhuman primate.