Abstract
The formalin-fixed, amber-colored right globe from a 12-week-old female silver Labrador Retriever dog was submitted to the Comparative Ocular Pathology Laboratory of Wisconsin for light microscopic evaluation. The clinical history described a collapsed anterior chamber and multifocal nodular lesions in the peripheral iris. Histologically, immunohistochemically, and ultrastructurally, the uveal mass was consistent with a malignant schwannoma; there was extension along peripheral nerves within the sclera. The signalment and behavior of the neoplasm distinguish it from the uveal schwannoma of blue-eyed dogs and bear some resemblance to the ocular lesions in human neurofibromatosis. The dilute color mutation may contribute to the cause. Six weeks later, the dog did not develop any additional masses.
Uveal schwannomas have never been reported in non-blue-eyed dogs nor in a dog as young as 12 weeks of age. Here we present a uveal schwannoma with peripheral nerve extension in a 3-month-old silver Labrador Retriever puppy. The established relationship between iris color and uveal schwannoma in dogs raises questions about the potential risk of this neoplasm in color-dilute phenotypes. Uveal schwannomas are rare but are reported in the human literature, and they are not associated with a blue-eyed phenotype. 3,11,12 Neural tumors in a young animal are reminiscent of human neurofibromatosis (NF), a genetic disease characterized in part by tumors of the nervous system that present in childhood, sometimes in infancy. A truly parallel canine homologue to NF has not been described, although canine cases with features similar to NF type 1 have been published. 1,4
The formalin-fixed, amber-colored, right globe from a 12-week-old female silver Labrador Retriever dog was submitted to the Comparative Ocular Pathology Laboratory of Wisconsin (COPLOW). The animal had an abnormal eye of “several weeks” duration, with blepharospasm, conjunctivitis, and a darkened ventral iris with a shallow anterior chamber. The eye was paraffin embedded, and 4-μm dorsoventral sections were stained with hematoxylin and eosin and labeled with immunohistochemical markers for GFAP, S100, neurofilament, and melan A. Representative samples of the uveal mass were immersed in 2.5% glutaraldehyde for 24 hours and submitted for transmission electron microscopy processing.
Grossly, the iris tissue was thickened, nodular, and tan. On cut section, the anterior uveal mass was smooth, tan, and homogeneous (Fig. 1). Although the lesion is grossly multinodular—suggesting multiple, coalescing tumors—the histologic section samples only a single mass. Histologically, the ventral iris base, ciliary body, and peripheral choroid were expanded and distorted by a densely cellular, fairly well demarcated, cellular infiltrate. The infiltrate also effaced multiple peripheral nerves in the ventral sclera and episclera (Figs. 2, 3). The infiltrate formed a mass lesion that regionally collapsed the anterior chamber and the drainage apparatus and bulged into the posterior chamber cradling the lens. The mass was multilobulated and supported by a moderately robust fibrovascular stroma with collagenous septae. It comprised plump spindle cells arranged in interlacing sheets and streams and, occasionally, fascicles. They had moderate amounts of amphophilic, sometimes vacuolated, cytoplasm and indistinct cell borders. Nuclei were oval to irregularly shaped. The chromatin pattern varied from dense to moderately stippled to clumped. Many nuclei contained 1 or 2 dense nucleoli. Anisocytosis and anisokaryosis were moderate, and there were 10 mitotic figures in 10 high-power fields. At the periphery of the mass in the iris stroma, tendrils of cells extended into the stroma, and few islands of similar cells were present in the iris leaflet. The tendrils at this irregular edge wrapped around and engulfed amorphous islands of fully differentiated cartilage (Fig. 2, inset). A myxomatous matrix expanded the pars plicata, but the remainder of the globe was histologically pristine. The expansile and infiltrative nature of the mass and the cellular features are consist with a neoplastic process.
Eye; dog. Smooth, tan, nodular thickening of anterior uvea.
Neoplastic cells were globally S100 positive (Dako, clone Z0311, 1:2000) and melan A positive (Dako, M-7196, 1:50). Regions of cells (approximately 40%) labeled positive for GFAP (Dako, Z0334, 1:1500). Cells were uniformly negative for neurofilament (Dako, M-0762, 1:200) (Fig. 4); however, neurofilament-positive axons were seen in the nerves that were invaded within the sclera. Transmission electron microscopy of the tumor revealed a homogeneous population of neoplastic cells embedded in a scant and fibrillar extracellular matrix. Neoplastic cells were oval to elongated with oval nuclei, peripheralized chromatin and prominent nucleoli, scant cellular cytoplasm with marked amounts of typical mitochondria, dense lysosomal granules, and moderate amounts of cytoplasmic filaments. Neoplastic cells had numerous, elongated cytoplasmic processes and cellular membranes forming intricate interdigitations that were laid over an intermittent basal lamina. Rarely, in the intercellular space, there were linear banded structures of about 3 μm in length with a periodicity of 110 to 130 nm in close contact with cellular membranes, and they were interpreted as long-spacing collagen (Luse bodies) (Figs. 5, 6). No melanosomes were identified by electron microscopy.
Eye; dog. (a) Nodular expansion of the iris and ciliary body by a spindle cell neoplastic infiltrate., transitioning to nonneoplastic iris tissue axially. Neoplastic infiltrate expands multiple peripheral nerves in adjacent sclera. HE. (b) Strong, global S100 positivity of neoplastic cells in uvea and affected scleral nerves. Immunohistochemical staining for S100. (c) Regional positivity of neoplastic cells in the uvea but not the peripheral nerves. Immunohistochemical staining for GFAP. (d) Faint but global positive labeling of neoplastic cells. Immunohistochemical staining for melan A.
Iris; dog. Electron micrograph of the neoplastic cells; note the large amounts of interdigitating, elongated, cytoplasmic processes and cellular membranes. TEM, uranyl acetate–lead citrate.
Six weeks after enucleation, the surgical site healed with no complications, and no additional tumors or clinical signs were noted.
The histologic and ultrastructural features of this uveal tumor are consistent with a malignant schwannoma—a peripheral nerve sheath tumor derived entirely from Schwann cells. The immunohistochemical profile largely supports Schwann cell origin (S100 and GFAP positivity), though the labeling with melan A is puzzling. Clinically, the tumor is vaguely described as a “visible darkened area in the ventral iris,” but it is not directly called black or pigmented. The fixed tumor was not grossly pigmented. The tumor cells lack visible pigment, but there are nonneoplastic melanocytes scattered throughout the supporting stroma of the mass. The human literature includes multiple descriptions of melanotic schwannomas—tumors comprising melanin-producing cells with ultrastructural features of Schwann cells. 5,7,14 That description fits our tumor nicely, except human melanotic schwannomas are reported to grossly have a tarlike consistency that varies from gray to pitch black. 7 The tumor reported here lacks tarlike color and texture. Schwann cells and melanocytes of the iris are both derived from the neural crest; colabeling for melan A, S100, and GFAP is not a theoretic improbability. The significance of this interesting finding will depend on the repeatability of the staining in subsequent tumors.
Uveal schwannomas of blue-eyed dogs have been reported in adult canines and were first described in 2007 with the former name spindle cell tumor of blue-eyed dogs. 13 The present case is similar but has distinct characteristics that we believe distinguish the 2 entities. The first is the young age. Presentation at 12 weeks of age is unusual for any ocular tumor, including uveal schwannoma of blue-eyed dogs. There are 93 cases of uveal schwannoma of blue-eyed dogs in the COPLOW collection, in dogs ranging from 16 months to 14 years old, with a mean age of 7.87 years and a median age of 8 years.
A tumor in a young animal suggests a congenital or genetic condition. A neural tumor in a young animal is reminiscent of the human condition NF: a genetic disease characterized in part by the development of neural tumors at a young age. There are 3 recognized forms of the disease—NF types 1 and 2 and schwannomatosis. Each is a clinical diagnosis based on having a certain number and/or combination of lesions and affected relatives. Despite the perseverance of strictly clinical diagnoses, the different variants of NF are well matched with acquired or inherited genetic mutations at specific loci. This canine case does not fulfill the diagnostic criteria for any of the 3 variants, but it has some similarities with each, including the young age of presentation and anterior uveal involvement (neurofibromatosis type 1–like), 8 divergent mesenchymal differentiation (neurofibromatosis type 2–like), 7 and intratumoral myxoid changes and intraneural growth pattern (schwannomatosis-like). 9 The clinical progression of this animal as it ages will be of interest, as it may reveal additional NF-like lesions. If it does, genetic testing for mutations at the loci identified in human disease could reveal the presence or absence of homologous changes.
The extension into surrounding peripheral nerves is another feature that separates this case from classic uveal schwannoma of blue-eyed dogs. Local invasion of the sclera is present in 11.5% of the COPLOW cases, and there is 1 case of distant metastasis to the lung, liver, and mesenteric lymph nodes, but none of the COPLOW cases have histologic evidence of peripheral nerve involvement or extension. 2
The presence of well-differentiated cartilage is a phenomenon not observed in uveal schwannoma of blue-eyed dogs. Divergent osteogenic, myxomatous, and myoblastic differentiation have been described in a malignant peripheral nerve sheath tumor in a Golden Retriever dog, and divergent differentiation of human conventional schwannomas—usually in the setting of NF2—rarely includes cartilage and/or bone formation. 6,7
Finally, the most obvious distinction between this case and those previously reported is that this is not a blue-eyed dog. We believe there is some merit in exploring the role of pigmentation as a risk factor in schwannoma or uveal schwannoma in dogs. Previous studies in goniodysgenic (nonneoplastic) canine eyes demonstrate a prominent difference between blue- and brown-eyed dogs, whereby blue-eyed dogs show a significant increase in linear iridal GFAP positivity consistent with nonmyelinated nerves. 13 This blue-eyed GFAP positivity is absent in the nonneoplastic iris tissue in our case, which resembles the GFAP-positive distribution in a brown-eyed dog (data not shown). We believe that the preponderance of GFAP-positive cells in the blue uvea predisposes those animals to schwannoma formation, and we believe that the animal in the present case does not share the predisposition, because it lacks this immunohistochemical feature.
However, we are open to the possibility that the dilute color mutant genotype may play a role. The gray (or blue) dilute condition in dogs is believed to result from a mutation in the melanophilin (MLPH) gene. 10 The gene product is involved in shuttling of melanin in melanosomes and is required for visible pigmentation in hair and skin. There are multiple mutations within or near the MLPH gene that appear to result in the dilute phenotype in dogs, but none are currently known to be associated with neoplasia. However, the deeper that one delves into the genetics of coat color, the more complex that it becomes. We accept the possibility that this animal’s genotype may predispose it to the development of neural tumors by a yet undescribed mechanism.
This is the first report of uveal schwannoma in a non-blue-eyed dog and the first in a nonadult dog. The genotype of this animal may have contributed to the development of juvenile tumor, whether it be due to the color-dilute mutation or a possible genetic mutation that mimics human NF. Alternatively, this may be a spontaneously arising tumor, but the distribution of nonneoplastic GFAP-positive cells in the uvea that mimics the brown-eyed situation and the unusual peripheral nerve extension suggest that this is a distinct entity from the uveal schwannoma of blue-eyed dogs.
Footnotes
Acknowledgements
We thank Shariar Salamat, MD, PhD (Department of Pathology and Laboratory Medicine, University of Wisconsin–Madison), and Diane O’Brien, BS (School of Veterinary Medicine, University of Wisconsin–Madison) for their contributions.
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.