|Home | About | Journals | Submit | Contact Us | Français|
A 3-year-old female gerbil developed a non-healing skin wound due to a malignant neoplasm. Histology, immunohistochemistry (cytokeratin 19 positive; vimentin, estrogen, and progesterone receptor negative), and electron microscopy (no desmosomes or melanosomes) revealed an undifferentiated carcinoma with pulmonary metastasis. Unlike in previous reports, it did not arise from the abdominal pad’s sebaceous gland.
Carcinome cutané d’origine non sébacée peu différencié chez une gerbille de Mongolie âgée de 3 ans (Meriones unguiculatus). Une gerbille femelle âgée de 3 ans a développé une plaie cutanée qui ne guérissait pas en raison d’un néoplasme malin. Des examens histologiques, par immunohistochimie (positif pour la cytokératine 19; négatif pour les récepteurs de vimentine, d’œstrogène et de progestérone) et par microscopie électronique (pas de desmosomes ni de mélanosomes) ont révélé un carcinome indifférencié avec métastase pulmonaire. Contrairement aux rapports antérieurs, il n’était pas causé par la glande sébacée du coussinet abdominal.
(Traduit par Isabelle Vallières)
A 64-g, 3-year-old, female gerbil (Meriones unguiculatus) was presented with a 0.5-cm diameter and 0.2-cm deep draining wound in the left axilla and alopecia on the left front limb (Figure 1A). The lesion had first been noticed by the owners approximately 3 wk earlier. The wound was flushed with sterile saline and cleaned with 4% chlorhexidine gluconate solution (Germi-Stat 4%; Germiphere Corporation, Brantford, Ontario). Radiographs of the left forelimb revealed a soft tissue swelling around the humerus and elbow, as well as a radiolucent region within the proximal radius and ulna (Figure 1B). Based on differential diagnoses of severe ulcerative dermatitis, traumatic dermatitis, or underlying neoplasia, prescribed therapy consisted of meloxicam (Metacam; Boehringer Ingelheim, Burlington, Ontario), 0.5 mg/kg body weight (BW) PO, q12h for 10 d, oral suspension of enrofloxacin (Baytril; Bayer, Toronto, Ontario), 10 mg/kg BW, PO, q12h for 10 d, and silver sulfadiazine cream 1% (Flamazine; Smith and Nephew, St. Laurent, Quebec) topically twice daily for 10 d.
There was poor response to therapy and the lesion worsened into ventral swelling and bruising of the left forelimb, alopecia, and deep ulceration with serous discharge leaking from the caudoventral surface of the wound. The gerbil was unable to bear weight on the front left limb and was bradypneic. A 3 × 1-cm mass was noted on radiographs; it extended from the left submandibular region along the left side of the neck to the left axilla and was multilobulated with many firm regions. Cytological examination of a fine-needle aspirate of the mass revealed primarily clusters of neoplastic cells and fewer inflammatory cells, such as neutrophils and macrophages (Figure 1C). The neoplastic cells were polygonal with little-to-moderate blue mottled cytoplasm occasionally containing small (0.5 to 1 μm in diameter) clear round vacuoles; nuclei were ovoid, with coarse chromatin and distinct, often very large, nucleoli. Anisocytosis and anisokaryosis were moderate to marked, with occasional nuclear molding and bi- or multi-nucleation. Due to poor prognosis, euthanasia by intracardiac injection of pentobarbital sodium (Euthansol; Shering-Plough, Pointe-Claire, Quebec) was carried out under general anesthesia with isoflurane (Isoflurane; Pharmaceutical Partners of Canada, Richmond Hill, Ontario).
On macroscopic postmortem examination, the gerbil was in good body condition with adequate subcutaneous and visceral fat stores and had a 1 × 0.5-cm area of alopecia on the left front limb above the radius/ulna. Beneath the left mandible at the cranioventral thorax and attached to the underlying muscle tissue was a well-circumscribed, firm, tan, multilobulated L-shaped subcutaneous neoplasm. The short portion of the mass (1 × 1.5 cm) was present in the neck, while its long portion (3 × 1 cm) extended ventrally and caudally along the external surface of the rib cage; the mass did not reach the umbilical region and was not associated with the abdominal pad. Additionally, the right ovary was grossly enlarged by a pale tan mass 1.5 × 2 cm, or approximately 10 times the diameter of the left ovary (unidentified ovarian tumor). Tissue samples were fixed in 10% neutral buffered formalin (NBF) and processed for histologic examination and electron microscopy (Figure 1D).
Histologically, the neoplastic mass had invaded the dermis and subcutis, was unencapsulated, multilobular, and expansile, and was composed of nests, cords, and trabeculae of polygonal to oval cells with marked anisocytosis and anisokaryosis (Figure 2A). Neoplastic cells had scant, pale, faintly granular eosinophilic cytoplasm and large nuclei often with a single, prominent nucleolus. Tumoral nests and cords were separated by a thick collagenous stroma, as well as multiple small foci of central necrosis infiltrated in some areas by small numbers of lymphocytes, plasma cells, and macrophages. The mitotic rate was 3 per high power (40 ×) field. There was no evidence of myoepithelial differentiation. In the lungs, intravascular metastatatic emboli with similar appearance to the subcutaneous mass occasionally spilled into alveolar spaces (Figure 2B). As no histopathological examination of the regional bones was done, it is uncertain whether the radiolucency observed in the proximal radius and ulna was due to neoplastic metastasis.
When immunohistochemical stains were applied to sections of skin infiltrated with tumoral nests (Figure 2C), approximately 90% of neoplastic cells had intracytoplasmic immunopositivity for Cytokeratin 19 (CK19). The healthy overlying epidermis served as a positive internal control and follicular epithelium, sebaceous glands, and surrounding dermal fibrous connective tissue served as negative internal controls (Figure 2D). Neoplastic cells were diffusely negative for melanin pigment (Fontana Masson stain) and for Vimentin C (Figure 2E). The regional dermis was diffusely positive for Vimentin, serving as the internal positive control and the overlying epidermis was diffusely negative, serving as an internal negative control. Neoplastic cells were also diffusely negative for estrogen and progesterone receptors (immunohistochemical staining conducted at the Michigan State Diagnostic Laboratory). Electron microscopy (EM) revealed the occasional presence of multiple small nucleoli and several variably sized mitochrondria, free and bound ribosomes, and occasional membrane-bound clear vacuoles of varying sizes in the cytoplasm (Figure 1D). Unfortunately, fixation for EM was poor so that intact intercellular junctions were rare and no distinct desmosomes were noted. Based on cytologic, histologic, immunohistochemical, and electron microscopic characteristics, the mass was diagnosed as a poorly differentiated subcutaneous carcinoma of non-sebaceous origin.
The origin of the ovarian tumor was not determined. This tumor was composed of sheets of tightly packed round-to-polygonal cells with scant, slightly pale, granular eosinophilic cytoplasm, and prominent nuclei with 1 to 2 nucleoli. Ovarian tumors, particularly granulosa cell tumors, are not uncommon in female gerbils (1). Differentiation between granulosa cell tumors and luteomas in gerbils is complicated by the common luteinization of granulosa tumor cells and the inconsistent presence of Call-Exner bodies, which are characteristic of human granulosa cell tumors (2). Ovarian tumors in gerbils rarely metastasize and, when they do, metastases seem limited to invasion of adjacent peritoneal viscera (2). Because the histologic morphology of the neoplasm in the ovary was distinct from that of the tumors found in the skin and metastatic foci in the lung, and considering the low probability of an ovarian neoplasm metastasizing to both skin and thoracic viscera, the ovarian tumor in this gerbil was considered distinct from the other tumors and an incidental finding with little to no clinical significance.
The finding of neoplasia in a gerbil is common, as approximately 25% to 40% of gerbils over 2 y of age will develop a tumor (3). Cutaneous tumors, among the most common in gerbils, are reported to arise almost exclusively from the sebaceous glands of the abdominal pad (4–6). Tumors of the abdominal pad are more frequently found in males and may have a predominantly sebaceous or squamous cell differentiation. The function of the abdominal pad is mostly related to marking of territory and, in males, it involutes after castration (5). Little is known of the behavior of the tumors. The few cases for which such information exists suggest that metastasis does not occur (4,5) and that resection, particularly with concurrent castration, may be curative (5).
The carcinoma found in the skin of this female gerbil was highly pleomorphic, but morphologically and immunohistochemically consistent with an epithelial tumor. Unlike the most common cutaneous tumors described in previous reports, the tumor did not arise from the abdominal pad and had no evidence of sebaceous or squamous differentiation, but appeared instead to be of basal cell origin. Immunohistochemical staining of the tumor and adjacent skin with CK19 resulted in positive staining of the neoplastic cells and adjacent epidermis, but negative staining of the follicular epithelium and sebaceous glands, further supporting the non-sebaceous origin of the neoplasm. The lack of positive staining in the healthy sebaceous glands in this case matches what little is known about cutaneous immunohistochemical characteristics in gerbils. Although sebaceous glands in humans tend to stain positively with CK19, gerbil sebaceous glands seem to be negative to pan-cytokeratin staining, which includes CK19 (6).
Alternative origins for the cutaneous neoplasm, given its location and histologic morphology, include a carcinoma of the apocrine, salivary, or mammary glands. Mammary gland tumors of gerbils are rarely reported in the literature, despite the common occurrence in other small rodent species. Salivary tumors have been reported in mice, sometimes in association with polyoma-virus infection (7,8), but have not been reported in gerbils. The mammary markers used in this case (estrogen and progesterone receptors) were negative, but this does not completely rule out the possibility of a mammary origin for the subcutaneous mass as it is unknown whether healthy mammary gland tissue in gerbils is immunopositive for these markers. Unfortunately, staining for prolactin, to which mammary tumors in other rodents are often positive (7), was not attempted in this case.
Less likely possibilities include a poorly differentiated squamous cell carcinoma and an amelanotic melanoma. Squamous cell carcinomas tend to include keratinous differentiation and desmosomes formation, neither of which was detected on histopathology or EM (9). Melanoma in gerbils tends to arise in the ear, foot, or base of the tail, and often contains melanin granules visible by Fontanna Masson staining and melanosomes visible on EM (7,9); neither feature was present in this case. A trichoblastic carcinoma was considered, but it is not supported by the immunohistochemical staining: the normal hair follicular epithelium was negative for CK19 staining, whereas the tumor cells were strongly positive.
Another possibility is an oncocytoma, a tumor that characteristically has cytoplasmic eosinophilic granules corresponding to numerous mitochondria. Such a tumor of the parotid gland is rare, but has been previously reported in humans and cats (10,11). Oncocytomas have also been reported in the pinnae of rats, as well as in many locations in humans, often associated with glands (thyroid, parathyroid, salivary, and lacrimal glands) (10,12). The exact cellular origin of oncocytes is not understood, although it is thought to be epithelial. The tumors tend to be benign in humans, although rare malignancies have been reported (10). The final consideration of cellular origin, albeit unlikely given the lack of specific evidence, was a salivary gland tumor, which has not to our knowledge, been previously reported in a pet gerbil. Salivary gland tumors are rare in veterinary species, with an overall incidence of 0.17% in dogs and cats (13). Although salivary-specific markers are available for human tumors, none has been validated for use in veterinary species (14).
This report describes an undifferentiated carcinoma in a female gerbil that did not arise from the sebaceous glands of the abdominal pad, as is common in the species, and illustrates the value of using immunohistochemical staining as part of the diagnostic protocols for non-domestic species.
The authors extend special thanks to the diagnosticians, veterinary students, and intern who helped us with this case, including Dr. Matti Kiupel, Aimee Elson, Dr. Letitia Chow, and Sarah Mouri, and to Kathy Jones who conducted some of the immunohistochemical staining. The authors also thank the anonymous reviewer whose comments helped to improve this manuscript. CVJ
Use of this article is limited to a single copy for personal study. Anyone interested in obtaining reprints should contact the CVMA office (gro.vmca-amvc@nothguorbh) for additional copies or permission to use this material elsewhere.