Introduction

Canine plasmacytomas are a solitary form of plasma cell tumor. They can arise from soft tissues (extramedullary plasmacytoma [EMP]) or bone (solitary osseous plasmacytoma). EMPs occur mainly in the skin, oral cavity, rectum, and colon but have been reported in the upper respiratory tract and lungs.^1–3

Oral EMPs represent 10–28% of all EMPs.^1,4 Canine EMPs carry a good prognosis after surgical excision, with surgery being curative in 90–95% of cases.^1,5 Without surgical excision, a median survival time of 138 days has been reported.⁴

Solitary osseous plasmacytomas are rare in dogs and are reported in the vertebrae, zygomatic arch, and ribs. They usually progress to multiple myeloma months to years after diagnosis, and optimal treatment consists of radiotherapy and/or surgical excision.¹

Overall, EMPs appear to respond favorably to complete surgical excision or radiotherapy, but spontaneous regression is very rarely reported in humans and has never been reported in dogs.^6–8 We report a case of a juvenile female dog with mandibular plasmacytoma, which spontaneously regressed over a period of 8 mo.

Case Report

A 4 mo old 10 kg female Finnish lapphund presented to a primary care center with a broken deciduous right lower third incisor tooth and associated mucopurulent discharge. The dog was fully vaccinated, had been in the owners’ possession for 2 mo, and had not traveled outside the United Kingdom. Amoxycillin-clavulanate^a 12.5 mg/kg per os (PO) q 12 hr was dispensed for 1 wk as an empirical treatment for a suspected abscess. One week later, the dog presented again with a gingival swelling around the broken tooth. The dog was admitted, and the broken tooth was extracted under general anesthesia along with the deciduous right lower canine and right lower second incisor. Deep pocketing was noted beneath the extraction sites, and there was purulent discharge and unexpected bleeding. Radiographs of the rostral mandible showed osteolysis in the area of the incisors’ roots. Grab biopsies of the tissues present at the extraction sites were performed and sent for histology and culture and sensitivity testing. The patient was discharged on meloxicam^b 0.1 mg/kg PO q 24 hr and clindamycin^c 11 mg/kg PO q 24 hr for 7 days.

The initial histopathology report described a poorly differentiated round cell tumor of the right mandible with >12 mitotic figures per high-power field^d, and microbiology only revealed growth of normal flora with no specific pathogens isolated. Immunohistochemistry was requested from these biopsies. The patient was then referred to our hospital for further management pending results. Physical examination revealed moderate swelling of the right rostral mandible, and hematology and biochemistry were within normal limits. Staging and treatment options were discussed with the owners and partial mandibulectomy was anticipated to be the mainstay of treatment. A head, neck, and thorax computed tomography (CT) scan was performed for tumor staging, which showed marked osteolysis of the rostral aspect of the right mandible caused by what appeared to be an expansile and lytic lesion within the mandible (Figures 1A, 2A). There were no signs of pulmonary metastasis. Given the uncertainty regarding the precise nature of the tumor reported by the referring veterinarian’s pathology laboratory, repeating biopsies to obtain larger samples was deemed appropriate. Five incisional biopsies of the mandibular lesion, measuring up to 6 × 6 × 2 mm, were obtained. Histopathological examination confirmed the presence of a moderately well-demarcated, unencapsulated, highly infiltrative, moderately pleomorphic neoplasm, variably arranged in sheets and loosely arranged streams, supported by scant fibrous connective tissue. Neoplastic cells were described as variably polygonal-to-round, with moderate eosinophilic cytoplasm that contained a single round-to-oval nucleus. Nuclei contained indistinct nucleoli and coarsely clumped chromatin (Figure 3A). Moderate anisocytosis and anisokaryosis were observed, with up to six mitoses in a single high-power field^e. Based on the second biopsies, differential diagnoses were varied and included osteosarcoma, a round cell tumor, or poorly differentiated carcinoma. Immunohistochemistry was not immediately requested on the second biopsies as it was pending from the first biopsies.

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

View Figure

• Download as Powerpoint
• Download Figure

The patient was sent home pending results, and receipt of the immunohistochemistry was delayed. The patient did not receive any additional antibiotic or anti-inflammatory therapy beyond what had been prescribed by the referring veterinarian. The final report from the referring practice’s pathology laboratory revealed that the neoplastic cells were negative for PAX5 (B-cell marker) and CD3 (T-cell marker) but positive for skeletal muscle actin and desmin. This supported a diagnosis of embryonal rhabdomyosarcoma, and partial mandibulectomy was recommended to the dog’s owners. The patient was 6 mo old when she was seen again for planned surgery 5 wk after initial presentation to our hospital. On physical examination the previous gingival swelling had resolved. Given the delay since the initial consultation and the improvement in clinical signs, repeating the CT scan of the head, neck, and thorax was recommended prior to embarking on surgery. The repeat CT studies showed marked improvement in the osteolytic lesion of the rostral right mandible (Figures 1B, 2B). These findings were discussed with the owners, who were advised that the apparent improvement rendered close monitoring an acceptable option. The owners opted for monitoring and regular repeat staging rather than mandibulectomy.

The patient was re-examined 1 mo later (at 7 mo of age), and physical examination was unremarkable. A CT scan was performed again and showed further improvement of the mandibular lesion, with only minimal lytic areas remaining in the rostral aspect of the right mandible (Figures 1C, 2C).

Four months later, at 11.5 mo of age, the patient returned and was reported to be doing well at home. Physical examination was unremarkable. A CT scan was repeated and showed further resolution of the mandibular lesion (Figures 1D, 2D) and no sign of metastatic disease in the pulmonary parenchyma.

At the time of writing this report, the patient is 15 mo old and is reported to be in good health, with no sign of recurrence of the mandibular lesion.

Given the exceptional progression of the lesion, requesting an extended immunohistochemistry profile from all biopsies to our pathology laboratory was deemed appropriate, in the hope that the results would help understand the tumor behavior and establish a prognosis.

On repeat immunohistochemistry, the neoplastic population within both biopsies was positive for vimentin (mesenchymal marker) and interferon regulatory factor 4 (MUM-1; plasma cell marker; Figures 3B, D). Very low numbers were positive for CD18 (leukocyte marker). The neoplastic population was negative for actin (Figure 3C) and desmin (muscle marker), osterix (osteoblast marker), e-cadherin (intercellular adhesion molecule), cytokeratin (epithelial marker), CD20 (B-lymphocyte marker), and CD3 (T-lymphocyte marker). This immunohistochemistry profile supported a diagnosis of plasmacytoma.⁹

Discussion

Plasmacytomas are uncommon tumors in dogs, and both EMPs and solitary osseous plasmacytomas are generally treated with surgical excision or radiotherapy. Solitary osseous plasmacytomas may also progress to multiple myeloma in dogs.¹⁰

We report a case of a 4 mo old Finnish lapphund with spontaneously regressing mandibular plasmacytoma. The diagnosis was made based on histopathology and immunohistochemistry. Based on histopathology, diagnosis of a poorly differentiated round cell neoplasm was made by two different pathologists on two independent biopsies. On immunohistochemistry the neoplastic population was found to be positive for MUM-1 and vimentin. Vimentin positivity is compatible with a mesenchymal origin, as would be expected in a plasma cell tumor. MUM-1 is reported to be highly specific for plasma cells in dogs and therefore indicative of the presence of a plasmacytoma.⁹

Given that full excision of the lesion and subsequent histopathology was not performed, it is uncertain whether the mandibular lesion represented an oral EMP that extended to the mandibular alveolar bone or a solitary osseous plasmacytoma of the mandible. Multiple myeloma was ruled out given the absence of other lesions on CT and the clinical course of the disease.

To the authors’ knowledge, this is the first report of a spontaneously regressing plasmacytoma in a dog. Spontaneous tumor regression is frequently encountered in the dog associated with cutaneous histiocytoma and viral papilloma. Each is associated with induction of an immune response that has been reported to variably include CD8 and CD4/CD8-positive lymphocytes, respectively.^11,12 Large numbers of CD3-positive lymphocytes were not identified in these samples, but regression occurred following sampling, so CD3-positive lymphocytes would not necessarily be expected to be present if involved. Reports of spontaneously regressing aggressive neoplasms in dogs are sparse. One case report described a suspected osteosarcoma in an 11 yr old spayed female boxer, which spontaneously resolved within 18 mo, although the diagnosis was made on radiographs alone.¹³ Mehl et al. reported four cases of spontaneously regressing osteosarcoma diagnosed on histopathology.¹⁴ Two of the four dogs received nonsteroidal anti-inflammatory treatment. The authors advanced the hypothesis of an immune-mediated response to explain tumor regression.¹⁴

In humans, solitary plasmacytomas also occur in the bone or as EMPs (in soft tissue). EMPs are mainly reported in the upper respiratory tract and can also progress to multiple myeloma.¹⁵ Spontaneous regression of solitary plasmacytoma in humans is rare and has only been reported in a gastric plasmacytoma following Helicobacter pylori eradication, in a chest wall plasmacytoma following incisional biopsy, and in a nasosinusal plasmacytoma following biopsy.^6–8 Of these three cases, two spontaneous regressions occurred after the lesion was biopsied, similarly to our case.

Some human benign neoplasms are known to sometimes spontaneously regress; 24 cases of spontaneously regressing osteochondromas have been reported, interestingly, all in patients younger than 19 yr.^16,17 Cerebellar astrocytomas in children regress in 14–45% of cases after incomplete resection.¹⁸ Fifty percent of cardiac rhabdomyomas diagnosed at birth, or in utero, spontaneously regress within the first year of life.¹⁹ Neuroblastomas sometimes undergo spontaneous regression or differentiation in children under 18 mo of age, but they are more aggressive in older children.²⁰

Proposed mechanisms for tumor regression in humans include genetic predisposition, induction of cell apoptosis or differentiation through intervention of neurotrophin receptors, immune-mediated response, or loss of telomerase activity.²⁰ Arunabh et al. proposed that biopsy of the lesion could induce an increase in antitumor antibodies, leading to tumor regression.⁷ Meziane et al. also reported that spontaneous regression of tumors often follows events such as biopsy, incomplete excision, local infection, or autoimmune illness.⁸

In this case of spontaneously regressing plasmacytoma, the mechanisms leading to regression remain unclear. It is possible that mechanisms similar to those proposed in humans were involved, such as an immune-mediated response. It is also possible that the incisional biopsy was an initiating event for the spontaneous regression, as has been reported in humans.^7,8 There was also an inflammatory process at the tumor site prior to biopsy, and in humans, infection (viral or bacterial) and local inflammation have been reported to stimulate an immune response that can sometimes result in tumor regression.²¹

Local inflammation and infection could also have contributed to the CT appearance of the lesion in this case. Osteomyelitis was part of the differential diagnoses based on the initial CT appearance but was subsequently considered unlikely based on the results of histological examination of the incisional biopsies, which did not show any inflammatory cells or infectious agents. Osteomyelitis would also be considered less likely given the spontaneous improvement of such an extensive lesion without additional antibiotic therapy.

Conclusion

This case shows the unexpected spontaneous regression of a mandibular plasmacytoma in a juvenile dog. Solitary plasmacytomas are normally treated by surgical excision or radiotherapy, and in this case, wide surgical excision was discussed with the owners at the time of diagnosis. However, when the dog presented again for planned surgery, spontaneous improvement could be observed, and the patient was managed with close monitoring until full resolution of the mandibular lesion was observed. This case suggests that biological behavior of primary neoplasms in juvenile dogs might be more unpredictable than usually described.