Introduction

Extramedullary plasmacytoma (EMP) is an uncommon neoplasm of plasma cells that arises from soft tissue, independent of bone or bone marrow. EMPs constitute ∼2.4% of all neoplasms in dogs.¹ Most commonly, EMPs affect the skin, mucous membranes of the oral cavity and lips, or the gastrointestinal tract.^1–6 Other less commonly reported locations include the eye, third eyelid, esophagus, trachea, larynx, stomach, liver, spleen, genitals, and intracranial sites.^7–13 Cutaneous and oral mucocutaneous EMPs are generally considered benign, whereas EMPs originating in other locations may behave more aggressively.^1,4–6,14

Parotid salivary gland EMPs are a rare presentation of head and neck EMPs in humans, with the literature consisting predominantly of individual case reports.^15,16 There is only one report of parotid salivary gland EMPs in nonhuman species, described in two Syrian hamsters.¹⁷ The present study describes the first reported case of parotid salivary gland EMP with local lymph node metastasis in a dog.

Case Report

A 12 yr old spayed female mixed-breed dog presented to The Ohio State University Veterinary Medical Center for evaluation of a recurrent grade II soft-tissue sarcoma on the left medial elbow. The dog had a previous diagnosis of idiopathic epilepsy that was reportedly well controlled with levetiracetam^a (32 mg/kg per os q 12 hr) and potassium bromide^b (39 mg/kg per os q 24 hr). On physical examination, a 5 × 3cm firm, round, and slightly movable subcutaneous mass was palpated ventral to the left ramus of the mandible. There was also a firm lobulated mass (2 × 2 cm) over the medial aspect of the left elbow, and multiple (∼2 × 1 cm) raised soft subcutaneous masses that were freely movable over the right scapula, ventral midline, and left inguinal region. There were no other pertinent abnormalities on physical examination.

A fine-needle aspirate of the mandibular mass was performed. Cytology revealed individual and loosely adherent clusters of atypical round to polygonal cells with a moderate amount of deeply basophilic cytoplasm, with occasional perinuclear clearing. Eccentrically placed round to irregularly shaped nuclei had condensed to coarsely stippled chromatin with small, faint nucleoli. (Figures 1A, B). There was mild to moderate anisocytosis and anisokaryosis with rare binucleation and trinucleation. The cytologic diagnosis was a neoplastic process with concern for a round-cell tumor or basal-cell tumor. Results of a complete blood count and serum biochemical profile were unremarkable, including a serum globulin concentration that was within normal limits (2.3 g/dL, reference range: 2.2–2.9 g/dL).

View Figure

• Download as Powerpoint
• Download Figure

Because of the concern for a round-cell neoplasm, a sedated abdominal ultrasound was performed (dexmedetomidine^c 3 μg/kg intravenously; butorphanol,^d 0.2 mg/kg intravenously) to look for evidence of systemic disease. No visceral architectural abnormalities were observed. Ultrasound-guided fine-needle aspirates of the liver and spleen were performed, and there was no cytologic evidence of neoplastic plasma cells.

The following day, the dog was placed under general anesthesia for a computed tomography^e scan to determine the origin of the mandibular mass. Images were obtained pre- and postcontrast (iohexol,^f 480 mg/kg intravenously) of the head and neck. An ovoid soft-tissue mass (2.9 cm in diameter) was noted within the left parotid salivary gland causing rightward displacement of the hyoid apparatus (Figure 2). The computed tomography scan also revealed an enlarged left medial retropharyngeal lymph node (3.4 × 2.2 × 3.9 cm), with multiple regions of hypoattenuation that did not enhance with contrast, suggestive of necrosis. Incidental dystrophic mineralization of the falx and dura mater was also observed. No other bony or soft-tissue abnormalities were observed.

View Figure

• Download as Powerpoint
• Download Figure

An ultrasound-guided fine-needle aspirate and cytology of the left medial retropharyngeal lymph node revealed a similar uniform population of round cells to those seen in the parotid salivary gland with moderate amounts of basophilic cytoplasm with occasional perinuclear clearing (Figure 1B). Nuclei were round with moderately condensed chromatin and mild to moderate anisocytosis and anisokaryosis and small or inconspicuous nucleoli. There were occasional binucleated and trinucleated cells. The diagnosis was round-cell neoplasia consistent with a plasma-cell tumor.

An incisional biopsy of the left parotid salivary gland was performed under the same anesthesia. Histopathologic examination and immunohistochemistry of the biopsy sample were overseen by a board-certified anatomic pathologist and an anatomic pathology resident (CP and MJD). The tissue specimen was preserved in 10% neutral buffered formalin and processed for histopathologic evaluation by routine methods. Examination of hematoxylin and eosin–stained tissue sections revealed sheets and cords of poorly differentiated neoplastic round to polygonal cells. The cells contained ovoid to reniform nuclei with one to two nucleoli and a moderate amount of pale eosinophilic to amphophilic cytoplasm. There was mild anisocytosis and moderate anisokaryosis and a mitotic index of four mitotic figures per ten 400× fields. Immunohistochemistry demonstrated that the neoplastic population demonstrated strong positive cytoplasmic to membranous staining with CD45 in 90–100% of neoplastic cells, weak granular positive nuclear staining with MUM-1 in 90–100% of neoplastic cells (Figure 1F), and strong positive membranous staining with CD18 in 20% of neoplastic cells. The neoplastic cells were predominantly negative for CD20, with <20% of neoplastic cells demonstrating strong positive membranous staining. These results supported a diagnosis of round-cell tumor of plasma cell origin. Surgical excision was recommended.

Twenty-eight days after the incisional biopsy of the left parotid salivary gland, the dog was placed under general anesthesia, and surgical excision of the left parotid salivary gland and the left medial retropharyngeal lymph node was performed. Both samples were submitted for histopathologic examination. A marginal excision of the left elbow soft-tissue sarcoma was also performed because of recurrence of the mass 2 mo before. The dog recovered from surgery without complication and was discharged from the hospital the following day.

Histopathologic examination of the left parotid salivary gland and left medial retropharyngeal lymph node exhibited similar findings. Both neoplasms were unencapsulated and well circumscribed and contained sheets, cords, and nests of round cells. The neoplastic cells contained moderate amounts of pale eosinophilic cytoplasm and ovoid, reniform, or elongated nuclei with clumped chromatin. There was mild anisocytosis and moderate anisokaryosis with occasional multinucleation, and a mitotic index of three mitotic figures per ten 400× fields. Scant aggregates of pale, eosinophilic matrix were present throughout the neoplasm that stained positive for Congo red and exhibited apple green birefringence under cross-polarized light, consistent with amyloid. These findings confirmed a diagnosis of metastatic, moderately differentiated malignant parotid salivary gland EMP. Follow-up was recommended in 5–7 days with The Ohio State University Veterinary Medical Center to remove surgical sutures and to discuss future treatment options.

Follow-up care was pursued with the primary care veterinarian. Approximately 7 mo after surgical excision of the left parotid salivary gland, a complete blood count and serum biochemical profile were performed revealing no clinically significant abnormalities. At that time, the serum globulin concentration measured within normal limits (3.0 g/dL, reference range 2.3–4.0 g/dL). The dog was euthanized because of hindlimb paresis, pain, and decreased quality of life 685 days after surgical removal of the left parotid salivary gland and left medial retropharyngeal lymph node. No diagnostics were performed at that time.

Discussion

The present case report describes the diagnosis and treatment of the first reported parotid salivary gland EMP in a dog. The most common cause of salivary gland disease in the dog is neoplasia, followed by sialadenitis, sialocele, and salivary gland infarction.^18,19 The most common salivary gland affected by salivary disease is the parotid.¹⁸ EMPs have been reported in salivary glands in people but occur rarely.^15,16 The present report indicates that EMPs should be considered as a rare differential diagnosis for salivary gland disease in dogs.

The biologic behavior of EMPs in dogs may be linked to tumor location. In dogs, tumors originating from the skin or mucous membranes of the oral cavity commonly have a benign biologic behavior with a low likelihood of metastasis or conversion to multiple myeloma.^4–6,14 In contrast, tumors originating outside of these locations, such as the gastrointestinal tract, have been shown to metastasize to localor regional lymph nodes.^1,3,7

The only report of parotid gland EMP in nonhuman species described the disease in two Syrian hamsters.¹⁷ Metastasis to the mandibular lymph node was noted at necropsy in one hamster; the other hamster did not have evidence of metastatic disease. There are few reports of lymph node metastasis in human EMPs. One study in humans found that only 5% of EMP tumors of the head and neck metastasized to regional lymph nodes.²⁰ However, a literature review of parotid salivary gland EMPs in humans found that ∼25% of the cases exhibited metastasis to regional lymph nodes.¹⁶ Although rare in humans, parotid salivary gland EMPs typically occur in older patients and are usually treated with surgery with or without adjunctive radiation therapy and chemotherapy.¹⁶ With treatment, parotid salivary gland EMPs tend to have a favorable prognosis in humans, although it remains a very uncommon clinical entity.¹⁶ Findings from these human reports, the two cases in Syrian hamsters, and the present report suggest that parotid salivary gland EMPs can have aggressive behavior and should be evaluated for regional lymph node metastasis.

EMPs of the head and neck in people progress to multiple myeloma in ∼20% of cases.^16,20 In dogs, local invasion and metastasis varies by location and occurs more frequently in EMPs involving the gastrointestinal tract.^1,3,6,7 Even in more aggressive EMPs, there is minimal evidence of progression to multiple myeloma. No hyperglobulinemia, visceral disease, bony lesions of the head and neck, or other evidence of multiple myeloma were detected in this dog before surgical excision of the tumor and the metastatic lymph node. Given the dog was euthanized ∼2 yr after surgery, the patient’s death may have been unrelated to the parotid salivary gland EMP. Unfortunately, diagnostics were not performed before euthanasia to determine the cause of paresis or to detect evidence of systemic plasma cell disease; thus, we cannot rule out progression of salivary EMP to multiple myeloma as a cause of the paresis and euthanasia.

Conclusion

Parotid salivary gland EMPs have not been reported in dogs. This case report describes a dog with a parotid salivary gland EMP with local lymph node metastasis treated with surgical excision of locoregional disease. Although rare, EMPs should be considered as a differential diagnosis for an enlarged salivary gland in dogs, and surgical removal should be considered for treatment.