Encephalic fibrosarcoma in a dog

CNS sarcomas are thought to arise from mesenchymal cells in the dura mater, leptomeninges, meningeal or neuroparenchymal blood vessels, or choroid plexus stroma. ¹ These neoplasms are rarely reported in human and veterinary medicine. In humans, CNS fibrosarcoma is an exceedingly rare and highly aggressive tumor that accounts for <2% of all intracranial neoplasms. ¹ The average age of affected patients is 31 y, and the most common tumor location is the parietal lobe. ¹ Because of their poor prognosis, intracranial fibrosarcomas need to be differentiated from other CNS neoplasms, such as meningiomas and solitary fibrous tumors, which bear a more favorable prognosis. ² Although these tumors can often be distinguished by routine histology, immunohistochemistry (IHC) and molecular identification of specific genetic mutations are typically required for diagnostic confirmation.^1,2 In dogs, CNS sarcomas have been reported rarely and may occur in the brain or spinal cord.^{3,5 –7,11,12} However, unlike in humans, too few cases have been published to determine whether these tumors may have any predispositions related to age, breed, sex, or anatomic location.

A 9-y-old, castrated male mixed-breed dog was evaluated for a 1-wk history of lethargy and circling to the right. Physical examination revealed dullness, disorientation, and a right-sided head tilt, with normal menace response and spinal reflexes. Because of the progressive nature of the clinical signs, euthanasia was elected. Autopsy findings consisted of a pale-tan, firm, 1.5-cm mass that effaced ~60% of the right-ventral aspect of the pons near the medulla oblongata. All cranial nerve roots and associated nerves were examined, with no tumor involvement detected. No other tissue changes (including metastases) were observed.

Tissue sections were collected, fixed in 10% neutral-buffered formalin, embedded in paraffin, and processed routinely for histology; slides were stained with H&E. Histologically, the neoplasm was well-demarcated and primarily intraparenchymal ( Fig. 1 ). Neoplastic cells were arranged in closely apposed bundles supported by a scant-to-dense collagen stroma that occasionally prevailed over the neoplastic cell population (Figs. 2–5). Neoplastic cells were elongate and had abundant, eosinophilic cytoplasm with indistinct margins that often blended with the surrounding stroma. Round-to-oval nuclei had finely stippled chromatin and 1 or 2 nucleoli. Anisocytosis and anisokaryosis were moderate, with 11 mitoses in 2.37 mm² (10 FN22/40× fields). The surrounding neuroparenchyma was compressed by the tumor, and a few areas of infiltration were present ( Fig. 6 ).

OPEN IN VIEWER

Figures 1–6.

Encephalic fibrosarcoma in a dog. H&E. Figure 1. A well-demarcated, 1.5-cm mass effaces ~60% of the right-ventral aspect of the pons near the medulla oblongata and compresses the surrounding neuroparenchyma.

No cranial nerve involvement was observed grossly. However, histologic findings (infiltrative and densely cellular spindle-cell neoplasm with moderate cellular pleomorphism and high mitotic activity) raised suspicion of a malignant nerve sheath tumor (MNST), potentially arising from adjacent cranial nerves (abducens, facial, vestibulocochlear, glossopharyngeal, vagus, accessory, and hypoglossal). Alternatively, a spindle-cell sarcoma originating from the meninges or mesenchymal tissues within the neuroparenchyma was also possible. Tissue sections were subjected to IHC for vimentin, E-cadherin, S100, neuron-specific enolase (NSE), glial fibrillary acidic protein (GFAP), oligodendrocyte transcription factor 2 (OLIG2), myelin oligodendrocyte glycoprotein (MOG), myelin protein zero (MPZ), periaxin (PRX), smooth muscle actin (SMA), myoglobin, myogenin, desmin, factor VIII–related antigen (F8-Rag), and CD31 to further characterize the neoplasm ( Suppl. Table 1 ). Neoplastic cells had widespread, robust cytoplasmic immunolabeling for vimentin ( Fig. 7 ); patchy, moderate cytoplasmic immunolabeling for myoglobin ( Fig. 8 ); patchy, weak cytoplasmic immunolabeling for desmin ( Fig. 9 ); and rare, moderate cytoplasmic immunolabeling for myogenin ( Fig. 10 ). No immunolabeling was observed for the other immunomarkers. IHC findings were suggestive of a sarcoma with myogenic differentiation. The lack of immunolabeling for S100, NSE, MPZ, and PRX was not supportive of a MNST.

OPEN IN VIEWER

Figures 7–13.

Encephalic fibrosarcoma in a dog. Figure 7. Neoplastic cells have robust cytoplasmic immunolabeling for vimentin.

Transmission electron microscopy (TEM) was performed on sections of the neoplasm. Formalin-fixed, paraffin-embedded tumor samples (1–2 mm³) were deparaffinized and post-fixed in 2.5% glutaraldehyde in 0.1 M sodium cacodylate buffer and in 1% osmium tetroxide in 0.1 M sodium cacodylate buffer (Electron Microscopy Sciences). Samples were processed, infiltrated, and embedded in resin (Embed 812 resin; Electron Microscopy Sciences). Sections were obtained and stained with toluidine blue before identifying the region of interest. Thin sections (60–70 nm) were contrasted with 5% uranyl acetate and Santos lead citrate (Electron Microscopy Sciences). Ultrastructural findings consisted of oval-to-elongate neoplastic cells with small-to-medium aggregates of cytoplasmic rough endoplasmic reticulum cisterns. Intermediary filaments or specialized contractile muscle structures were not observed. Nuclei had a 1:7 heterochromatin:euchromatin ratio with central and peripheral distribution and one or more compact-to-reticulate nucleolus. Sagittal and cross-sections of collagen bundles were in intimate contact with the plasma membrane of neoplastic cells and formed an abundant extracellular matrix. Intercellular anchoring attachments and basal lamina were not present. Ultrastructural findings were consistent with neoplastic cells of fibroblastic origin (Figs. 11–13).

Our histologic, immunohistochemical, and ultrastructural findings were consistent with fibrosarcoma. The origin of the neoplasm could not be determined grossly because of the unclear location (extraparenchymal vs. intraparenchymal) of the mass. However, histologic features were consistent with an intraparenchymal encephalic neoplasm, as highlighted by the fact that the tumor was almost surrounded by the encephalic neuroparenchyma. A search on 2025 Sep 20 of Google, PubMed, CAB Direct, Web of Science, and Scopus for cases of intracranial fibrosarcoma in dogs revealed inflammatory fibrosarcomas,^9,12 a meningeal rhabdomyosarcoma, ⁷ a meningeal osteosarcoma, ¹¹ and a neuroparenchymal perivascular wall tumor. ³ In addition, a spinal meningeal sarcoma and a spinal meningeal rhabdomyosarcoma have been documented.^5,6

Inflammatory fibrosarcomas (also known as inflammatory pseudotumors or inflammatory myofibroblastic tumors) are solitary, benign masses that occur in many anatomic sites in humans and rarely in dogs, cats, and horses.^{8 –10,12,13} In the 2 documented cases of inflammatory myofibroblastic tumors with CNS involvement in dogs, one tumor was extraparenchymal and affected the olfactory and frontal cerebral lobes, and one tumor was extraparenchymal and intradural, occurring in the thoracolumbar spinal cord.^9,12 Unlike our case, inflammatory myofibroblastic tumors are composed of a uniform population of elongate myofibroblastic cells admixed with reactive lymphocytes, plasma cells, and macrophages.^8,9,13 These findings suggest that intracranial fibrosarcomas have not been described previously in dogs.

Ultrastructural features in our case were typical of fibrosarcoma. ⁴ The occasional immunolabeling for myoglobin and desmin, and the rare immunolabeling for myogenin, suggested myogenic differentiation. Although preservation and processing artifacts precluded a detailed TEM evaluation of the tumor, no ultrastructural evidence of intermediary filaments forming specialized structures (e.g., sarcomeres of dense cytoplasmic density) was identified, thereby ruling out a rhabdomyosarcoma or a leiomyosarcoma. The absence of basal lamina ruled out leiomyosarcoma and schwannoma. Further, no evidence of pinocytic vesicles was found, ruling out leiomyosarcoma and neurofibroma. Finally, the absence of intercellular anchoring attachments was not consistent with a meningioma. Our findings highlight the value of IHC and TEM in the diagnosis of sarcomas of unknown or uncertain histogenesis, underscoring the importance of blending immunolabeling and ultrastructural findings to support a more specific diagnosis.

Supplemental Material

sj-pdf-1-vdi-10.1177_10406387251410934 – Supplemental material for Encephalic fibrosarcoma in a dog

Supplemental material, sj-pdf-1-vdi-10.1177_10406387251410934 for Encephalic fibrosarcoma in a dog by Megan J. Crawford, Elena A. Demeter, Anibal G. Armién, Rebecca Radisic and Daniel R. Rissi in Journal of Veterinary Diagnostic Investigation