Bilateral hibernomas in the femoral regions of a dog

Abstract

A 14-year-old intact male Chihuahua dog was presented with masses located between the biceps femoris and adductor muscles in both hind limbs. Based on histopathological, immunohistochemical, and ultrastructural findings, we diagnosed these masses as bilateral hibernomas in the femoral regions. The dog had no evidence of recurrence or metastasis of the hibernomas through a 4-month postoperative follow-up. This is apparently the first report of bilateral hibernomas in the femoral regions of a dog.

Key clinical message:

Bilateral hibernomas should be considered as a differential diagnosis for masses occurring in the femoral regions of dogs.

Hibernomas are rare, benign neoplasms that originate in brown adipose tissue, which is involved in thermogenesis in hibernating animals, human fetuses, and neonates (1). They were first described in humans by Merkel in 1906 and named by Gery in 1914 (2). Hibernomas have been reported in humans (2–5) and various animals, including dogs (6–11) and rats (12). In humans, hibernomas occur most frequently in the thighs but also occur on the scapular and head and neck regions (3). In rats, hibernomas occur mainly in the thoracic cavity (12). By contrast, in dogs, the periorbital region seems to be the most frequent site of hibernomas (9–11), whereas single instances of hibernomas in the nipple (6), the femoral muscles of the right hindlimb (7), and the omentum (8) have also been described in case reports. However, all these previously reported canine hibernomas occurred as single lesions, and canine hibernomas occurring as multiple lesions have apparently never been reported. Herein, we describe the clinical, histopathological, immunohistochemical, and ultrastructural features of bilateral hibernomas in the femoral regions of a dog.

Case description

A 14-year-old intact male Chihuahua dog with a mass located within the femoral muscles of the left hind limb was presented to a private veterinary clinic. Immediately before the surgery for this mass, palpation confirmed another mass located within the femoral muscles of the right hind limb. At the time of surgery, a solitary, spherical-to-oval mass was located between the biceps femoris and adductor muscles in the femoral region of each hind limb. Both masses could be fully surgically removed without leaving any adherent tissue because there were no firm attachments to the biceps femoris or adductor muscles. These resected masses were submitted to the Laboratory of Veterinary Histopathology at Kagoshima University (Kagoshima, Japan) for histopathological investigation.

The dog was discharged and had no evidence of recurrence or metastasis through a 4-month postoperative follow-up. As an unrelated finding, preoperative computed tomography imaging revealed an ileal mass ~2 cm in diameter that was resected and diagnosed as a gastrointestinal stromal tumor based on histopathological and immunohistochemical evaluation.

The femoral masses were fixed in 10% neutral-buffered formalin, embedded in paraffin, and sectioned at 3 μm. Sections were stained with hematoxylin and eosin (HE), periodic acid–Schiff (PAS), and phosphotungstic acid-hematoxylin (PTAH). Sudan black B staining of frozen tissues (8-micrometer sections) was also done.

Immunohistochemistry was done using a polymer method with the Histofine Simple Stain MAX PO (Multi) kit (Nichirei Biosciences, Tokyo, Japan). The primary antibodies used in this case are shown in Table S1 (available online from: www.canadianveterinarians.net). Endogenous peroxidase was inactivated with 3% H₂O₂-methanol. Detected antigens were visualized with 3, 3′-diaminobenzidine tetrahydrochloride reagent (Nichirei Biosciences). Sections were counterstained with Mayer’s hematoxylin. For ultrastructural examination, small pieces of formalin-fixed tissue were transferred to 0.5% glutaraldehyde and 1.5% paraformaldehyde, postfixed in 1% osmium tetroxide, dehydrated, and embedded in epoxy resin. Ultrathin sections were cut, stained with uranyl acetate and lead citrate, and examined using a transmission electron microscope (JEM-1010; JOEL, Tokyo, Japan).

Grossly, the left and right femoral masses were 2.0 × 1.0 cm and 1.0 × 1.0 cm in size, respectively. The gross cut surfaces of the formalin-fixed masses were firm, well-circumscribed, lobulated, and pale tan (Figure 1 A).

Figure 1.

A — Photographs of the gross cut surfaces of the formalin-fixed masses. These masses were firm, well-circumscribed, lobulated, and pale tan in color. R — Right; L — Left. B — The mass was composed of a mixture of neoplastic cells with a granular and eosinophilic cytoplasm, and multivacuolated and univacuolated cells resembling adipocytes. Hematoxylin and eosin stain. Scale bar: 50 μm. C — Neoplastic cells contained variably sized Sudan black B-positive lipid vacuoles. Sudan black B stain. Scale bar: 20 μm. D — Immunohistochemical labeling with uncoupling protein 1 (UCP1) revealed strong cytoplasmic immunoreactivity of neoplastic cells. Mayer’s hematoxylin counterstaining. Scale bar: 20 μm. E — Immunohistochemical labeling with myogenin sporadically showed nuclear immunoreactivity of neoplastic cells (arrows). Mayer’s hematoxylin counterstaining. Scale bar: 20 μm. F — Neoplastic cells contained abundant cytoplasmic mitochondria (arrows) and variably sized cytoplasmic lipid droplets. Transmission electron microscopy. Scale bar: 5 μm.

Microscopically, the 2 masses had similar histopathological features. The masses were well-circumscribed and composed of lobules of round-to-polygonal neoplastic cells separated by thin, fibrovascular stroma. The neoplastic cells comprised a mixture of cells with eosinophilic granular cytoplasm and multivacuolated and univacuolated cells resembling adipocytes (Figure 1 B). Nuclei were round to oval and centrally to eccentrically arranged with finely stippled chromatin and 1 to 2 nucleoli. Anisocytosis and anisokaryosis were mild, and the number of mitotic figures varied from 0 to 1 per 10 high-power fields (400×, 2.37 mm²). No evidence of vascular invasion was observed.

The cytoplasm of the neoplastic cells contained variably sized Sudan black B-positive lipid vacuoles (Figure 1 C) but did not contain PAS-positive granules or PTAH-positive cross-striation. Immunohistochemically, the neoplastic cells diffusely expressed vimentin, uncoupling protein 1 (UCP1), and S100 protein, and sporadically expressed myogenin (Figure 1 D and E). The neoplastic cells were negative for keratin/cytokeratin (panCK), ionized calcium-binding adapter molecule 1 (Iba-1), and melanoma marker (PNL2).

Ultrastructurally, the neoplastic cells with the eosinophilic granular cytoplasm contained large numbers of electron-dense mitochondria and lipid droplets (Figure 1 F). A few mitochondria were also present between lipid droplets in the vacuolated cells. Based on the gross, histopathological, immunohistochemical, and ultrastructural findings, a diagnosis of bilateral hibernomas in the femoral regions was made.

Discussion

Brown adipose tissue is distributed mainly in the thoracic and abdominal cavities and in interscapular subcutaneous tissues in neonates, rodents, and humans (1). However, no information about the distribution of the brown adipose tissue in dogs has apparently been published. In addition, hibernomas may arise from the brown adipose tissue remnants, ectopic growth or migration of brown adipose cells, or aberrant differentiation of mesenchymal cells (5). Considering this, perhaps aberrant differentiation of mesenchymal cells was involved in tumorigenesis in the case described here. However, the exact mechanism of tumorigenesis in this case is uncertain and further studies are needed.

Histopathologically, human and canine hibernomas typically appear as well-circumscribed lesions composed of lobules of round-to-polygonal neoplastic cells with granular eosinophilic, multivacuolated, and univacuolated cytoplasm with mild nuclear atypia (4,9). Transmission electron microscopy revealed abundant, electron-dense mitochondria and lipid droplets within the cytoplasm of granular cells and a few mitochondria between the abundant lipid droplets within vacuolated cells (4,9).

Immunohistochemical labeling with UCP1 has been considered a useful tool for diagnosing canine hibernomas (6,9–11). Note that the UCP1 antibody used in this case has been reported to react in rodent brown adipose tissues (13), and canine tumor tissue in this case showed a similar staining behavior (Figure 1 D; and see Figure S1, available online from: www.canadianveterinarians.net). However, the lack of validation for the cross-reactivity of the UCP1 antibody in canine brown adipose tissues is an important limitation of this case. Immunohistochemical evaluations of UCP1 antibody using brown adipose tissues obtained from canine fetuses or neonates should be completed to confirm specificity of this antibody in dogs.

In some cases, hibernomas also have sporadically expressed myogenic markers (MYOD1, myogenin, etc.), which is attributed to the common embryological origins of brown adipocytes and myocytes (9,14). These characteristic features are consistent with those in this case and support a diagnosis of hibernoma. In addition, rhabdomyoma, granular cell tumor, oncocytoma, xanthogranuloma, balloon cell melanoma, and liposarcoma should be considered in the differential diagnosis of hibernomas (9). However, in this case, these diseases could be ruled out based on the histopathological, special staining, immunohistochemical, and electron microscopic findings (see Table S2, available online from: www.canadianveterinarians.net).

In human medicine, complete surgical resection is the standard treatment for hibernomas (3). The prognosis is generally favorable and no distant metastases have been reported; however, hibernomas have recurred after surgical resection in 6 patients (2). By contrast, information on the prognosis for canine hibernomas is limited to 3 reports, with no recurrence or metastasis observed through 12-, 14-, and 18-month postoperative follow-ups (6,10,11). Although no reports about multiple canine hibernomas have been published, the absence of any recurrence or metastasis through the 4-month postoperative period in this case suggests that surgical resection was likely curative.

In conclusion, to the best of our knowledge, this is the first report of bilateral hibernomas in the femoral regions of a dog. Bilateral hibernomas should be considered as a differential diagnosis for masses occurring in the femoral regions of dogs.