Tonsillar plasmacytoma in a dog

Abstract

A 10-year-old greyhound dog was presented because of an incidental finding of a tonsillar mass. Excisional surgical biopsy was performed and the dog was diagnosed with an incompletely resected plasma cell tumor. Adjuvant therapy was declined. One year later there was no local recurrence or distant metastasis of the mass or clinical signs associated with the tonsillar plasmacytoma.

Extramedullary plasmacytomas (EMPs) are solitary plasmacytic tumors of soft tissues. The most common locations in the dog are cutaneous and mucous membranes of the oral cavity, colon, and rectum (1). Other locations that have been reported include eye, genitalia, intracranial sites, larynx, liver, spleen, sinonasal cavity, stomach, third eyelid, trachea, and uterus (2–8). Typically, cutaneous and oral EMPs in dogs are benign and metastasis is rare (1). However, EMPs that arise from other locations (i.e., gastrointestinal tract) have been reported to have a different behavior with a tendency towards metastasis. Extramedullary plasmacytomas that arise from the tonsils are rare in dogs and their clinical behavior is unknown. In this report, we describe the clinical signs, examination findings, and outcome in a dog with EMP in the tonsil.

Case description

A 10-year-old, spayed female retired racing greyhound dog was initially presented to the referring DVM with an acute history of left-sided sialorrhea which was presumed to be related to dental disease. As the dog was undergoing tracheal intubation for anesthesia in preparation for a dental cleaning, a mass was discovered originating from the left tonsillar fossa, and the dog was referred to Michigan State University Veterinary Teaching Hospital (MSU-VTH) for further diagnostic work-up 2 wk later. On presentation at the MSU-VTH, the dog was bright, alert, and responsive. Peripheral lymph nodes were small and symmetrical, aside from the left mandibular lymph node which was subjectively slightly enlarged. Blood analysis consisting of complete blood (cell) count (CBC) and serum chemistry profile (i.e., total protein, albumin, globulin, alkaline phosphatase, alanine aminotransferase, aspartate aminotransferase, total bilirubin, urea nitrogen, creatinine, glucose, and electrolytes) were unremarkable. There was no evidence of abnormalities on 3-view thoracic radiographs. A fine-needle aspiration of the left mandibular lymph node was done. Cytopathologic evaluation of the left submandibular lymph node revealed reactive lymphoid hyperplasia.

The dog was anesthetized with midazolam (Novaplus, Eatontown, New Jersey, USA), 0.2 mg/kg body weight (BW), IM, and butorphanol (Akorn, Lake Forest, Illinois, USA) 0.2 mg/kg BW, IM as premedication, and propofol (Novaplus, Lake Zurich, Illinois, USA) IV as induction. Under general anesthesia, oral examination revealed a 6 cm × 4 cm × 3.7 cm soft tissue mass originating from the left tonsillar fossa and protruding into the oral cavity. Normal tonsillar tissue was not present within the fossa. The right tonsil was within normal limits, and there were no other abnormalities of the soft tissues of the oral cavity. The dog was intubated using an endotracheal tube. After the intubation, isoflurane and oxygen were used to maintain general anesthesia. Excisional surgical biopsy of the left tonsillar mass was completed using a vessel sealing device (Ligasure Precise; Valleylab, Boulder, Colorado, USA) and the tonsillar fossa epithelium was closed with absorbable suture (Monocryl Ethicon; Johnson & Johnson, Somerville, New Jersey, USA). The dog recovered uneventfully after the procedure. Codeine (West-Ward Pharmaceuticals, Eatontown, New Jersey, USA), 1.25 mg/kg BW, PO, q8h, and Clavamox (Zoetis, Parsippany, New Jersey, USA), 15.6 mg/kg BW, PO, q12h, were prescribed for pain control and septic prophylaxis due to severe dental disease, respectively.

The tissue was fixed in 10% neutral buffered formalin for approximately 8 h. Tissue was trimmed and routinely embedded. Paraffin-embedded tissues were sectioned at 5 μm, mounted on frosted glass slides, routinely stained with hematoxylin and eosin (H&E), and examined by light microscopy. Sections were also immunolabeled using a mouse monoclonal anti-MUM-1 antibody (Dako Cytomation, Carpinteria, California, USA) and a rabbit polyclonal anti-CD20 antibody (Bioz Thermo Fisher Scientific Neomarkers, Palo Alto, California, USA).

Microscopic examination of the mass revealed a proliferation of neoplastic cells. Neoplastic cells were round, had variably distinct cell borders, and had small to moderate amounts of eosinophilic cytoplasm. Nuclei were ovoid to reniform, coarsely stippled, and occasionally peripheralized. There were occasional bi- and tri-nucleated cells (Figure 1a, b), and there were 10 mitotic figures in 10 high power fields. Neoplastic cells extended to the margins of the sections examined; therefore, complete excision could not be confirmed.

Figure 1.

Histopathology of tonsillar mass. The tonsillar stroma is markedly expanded and effaced by a poorly demarcated, highly cellular proliferation of neoplastic plasma cells arranged in dense sheets supported by fine fibrous stroma (bar = 50 μm) (a). Neoplastic cells are round with variably distinct cell borders, small to moderate amounts of eosinophilic cytoplasm, and contain coarsely stippled ovoid to reniform nuclei that are often peripheralized. There are occasional bi- and tri-nucleated cells (bar = 20 μm) (b). Hematoxylin and eosin (H&E). Immunohistochemically, neoplastic cells exhibit moderate to strong (arrows) nuclear labelling for MUM-1 (c) and are non-immunoreactive for CD20 (d). Rare individual B-cells show cytoplasmic labeling for CD20 (d). Diaminobenzidine chromogen (bar = 50 μm).

Immunohistochemistry was done to define the origin of these cells and showed that cells were immunoreactive for MUM-1 and non-immunoreactive for CD20, which was most supportive of a neoplasm of plasma cell histogenesis (9) (Figure 1c, d).

The owner did not want to pursue possible treatments but agreed to monitor the disease with frequent computed tomography (CT) scans and radiographs. A CT scan of the head and thorax was performed 10 wk after surgical biopsy. Mild hyperattenuation in the region of the right tonsil was noted with differential diagnoses of metastatic disease or reactive scar tissue from surgery.

After 10 mo, the dog was presented for a follow-up examination. She had been doing well at home between the recheck times, and had unchanged activity level and normal appetite. The sialorrhea had completely resolved. The tonsils were evaluated by oropharyngeal scoping and there was no local recurrence on the left tonsil (Figure 2). The right tonsil was normal in size and shape. On a repeat CT scan of the head and thorax, there was no evidence of metastasis to the regional lymph nodes or lungs.

Figure 2.

Oropharyngeal scope image 10 months after surgery. There was no tumor recurrence at the left tonsillar site (arrow). The right tonsil was normal in size and shape (arrowhead).

Discussion

Myeloma-related disorders arise when plasma cells or immunoglobulin-producing B-lymphocyte precursor lineage cells transform into a neoplastic population. Myeloma-related disorders include multiple myeloma (MM), EMP, IgM macro-globulinemia, solitary osseous plasmacytoma, and Ig-secreting lymphomas and leukemias (1). Extramedullary plasmacytomas in dogs are most commonly found in the skin, oral cavity, and other parts of the alimentary tract (1).

Dogs with EMP of the alimentary tract and other abdominal organs treated with surgical excision alone or in combination with systemic chemotherapy can achieve long-term survival times (1). In the veterinary literature, 5.2% to 22% of EMPs occur in the oral cavity and lips in dogs (1,10–13). In the oral cavity, gingival and lingual EMPs are the most common, composing 54.8% and 37.8%, respectively (11). Oral EMPs are locally aggressive, often involving bony destruction, but rarely metastasize. Dogs with single and multicentric oral EMPs treated with surgical excision can achieve long-term survival times (1,13–15). However, dogs treated with only cytoreductive surgery have regrowth of the tumor at the median recurrence time of 50 d and the median survival time of dogs that had no treatment or incomplete surgical margins without any adjuvant therapy is 90 d (13). Similar to what has been observed in humans, progression from a cutaneous plasmacytoma to a plasma cell leukemia in dogs has rarely been reported and is associated with a poor prognosis (16). Oral EMPs can progress to infiltrative plasmacytosis in multiple organs (13).

In humans, 82.8% of cases of EMP occur in the upper aerodigestive tract. The most common subsites are nasopharynx and oropharynx, nasal cavity, and paranasal sinuses (17). Approximately, 10.5% involve the tonsil or soft palate. Nodal involvement of EMP in the upper aerodigestive tract was reported in 7.6% of cases (18). In addition, bilateral EMP of the palatine tonsils has been reported (19,20). After radiation and/or surgery as therapy for EMP in the upper aerodigestive tract, 61.1% of all patients had no recurrence or conversion to multiple myeloma. However, 22.0% of these patients had local recurrence of EMP and 16.1% were reported to have conversion to multiple myeloma (18). Surgery alone gives the best prognosis in cases of EMP in which complete resection is achieved. However, if complete margins are not obtained or if regional or distant metastasis has occurred, adjunctive therapy is recommended (18). Due to the risk of a poor cosmetic result with surgery, radiation therapy at a dose of 40 to 60 Gy over 4 to 6 wk has been recently used as a primary treatment. After radiation therapy, the estimated 10-year overall survival, disease-free survival, and multiple myeloma-free survival are reported as 68.4%, 49.3%, and 55%, respectively (17). The current National Comprehensive Cancer Network guidelines for MM recommend that follow-up imaging should be used in cases of solitary plasmacytomas only if clarification is required for changes on the imaging studies noted at baseline or if there is suspicion of disease progression in the form of new onset bone pain or positive laboratory test results (21).

Immunohistochemistry for CD20 and MUM-1 was used to confirm plasma cell histiogenesis. CD 20 is a transmembrane phosphoprotein predominantly expressed from the pre-B cell stage to the activated B-cell stage (22). Multiple myeloma oncogene 1 (MUM-1) is a family of transcription factors and is required for immunoglobulin light-chain rearrangement at the pre-B stage of lymphocyte maturation (23), and high levels of MUM-1 expression lead to plasma cell differentiation (24).

MUM-1 is a very specific marker for plasma cell tumors. However, some B-cell lymphomas will label for MUM-1 depending on their direction of differentiation. Histologic features must always be used in conjunction with immunohistochemical results to make a diagnosis. In a study assessing the immunohistochemical features of plasma cell tumors in dogs, 101 of 109 plasma cell tumors (92.7%) were immunoreactive for MUM-1, while only 21 (19.3%) were immunoreactive for CD20 (9). In this study, 10 B-cell lymphomas out of 139 nonplasmacytic tumors were immunoreactive for MUM-1, but these neoplasms should also be immunoreactive for other B-cell markers, including CD20 and CD79a (9). Based on the histomorphologic features combined with the immunohistochemistry results, the tonsillar mass in this case is consistent with a plasma cell tumor.

Prognostic factors are unknown for cutaneous plasmacytomas in dogs. Grades and proliferation rates did not correlate with prognosis (12,25). In 1 report, all dogs with local recurrence or metastasis had polymorphous-blastic types (12). In humans, age, solitary plasmacytoma of bone, larger size (> 5 cm), and the Ig-G1 monoclonal antibody against Ki-67 antigen (MIB1) scores are reported as potential prognostic factors (26–28).

In this case, the right tonsil showed mild hyperattenuation on CT scan 10 wk after the surgery. However, 1 report revealed that there is no difference for attenuation between normal tonsils and tumors (29). Thus it is difficult to evaluate tonsils without histology. In human medicine, Fludeoxyglucose (¹⁸F) positron emission tomography (FDG-PET) permits a whole-body investigation with an overall sensitivity of 90% and specificity of 75% for the detection of myeloma lesions, which is superior to CT-scan and MRI (30). In addition, the prognostic value of FDG-PET has been studied. One study comparing FDG-PET and MRI in 33 patients with MM at diagnosis concluded that FDG-PET had a prognostic value above MRI (31). We understand the suitability of using FDG-PET-CT scan in this case, as recommended initially by supervising clinicians. Unfortunately, the usage of FDG-PET scan is still limited in veterinary medicine and only explored in a limited number of neoplastic malignancies (i.e., hemangiosarcoma, osteosarcoma, lymphoma, mast cell tumor, feline squamous cell carcinoma, and fibrosarcoma) (32–34).

To our knowledge, canine tonsillar EMP is rare and its clinical behavior is not fully understood, especially after incomplete resection (35). After 1 1/2 y from diagnosis and resection, there is no clinical evidence of recurrence or distant metastasis. Extramedullary plasmacytomas may be considered as a differential diagnosis for dogs with a tonsillar mass. CVJ