Introduction

Oral malignant melanomas (OMMs) are the most common malignant oral tumor seen in dogs.¹ In addition to local invasion into bone and soft tissues, OMM has a high incidence of metastasis to regional lymph nodes and the lungs.^2,3 Treatment options and prognosis are highly dependent on the stage of OMM and typically involve both local and systemic therapies. The local treatment of choice is wide surgical excision, but when surgery is not feasible, hypofractionated radiation therapy may be used for tumor control and palliation of clinical signs. Response rates to radiation are 82–94%, and survival times after radiation therapy have been reported at 7–10.2 mo.^4–7 Although local tumor control can often be achieved with hypofractionated radiation therapy in cases of OMM, most dogs develop metastatic disease and succumb to disseminated cancer.

Mechanical transplantation of tumor cells has been reported in both the veterinary and human literature associated with fine-needle aspiration (FNA) and surgical procedures in patients with a variety of tumor types. In veterinary oncology, reports have described this complication in dogs with transitional cell carcinoma (TCC), pulmonary adenocarcinoma, thymoma, and carcinomatosis.^8–14 In human oncology, tumor seeding is a rare but recognized complication of percutaneous endoscopic gastrostomy (PEG) tube placement in patients with head and neck carcinomas, and mechanical transplantation of malignant melanoma cells has also been reported in patients undergoing FNA or biopsy.^15–20 This report describes the case of a dog with OMM who was treated with hypofractionated radiation therapy and later developed melanoma within the tracheal lumen. We hypothesize that endotracheal (ET) intubation during general anesthesia resulted in mechanical transplantation of melanoma cells from the oral cavity into the trachea. To the authors’ knowledge, this is the first clinical report of tumor seeding as a result of this etiology.

Case Report

A 10 yr old castrated male pug was referred to The University of Tennessee Veterinary Medical Center for further evaluation of an incompletely excised amelanotic OMM. Approximately 1 mo previously, the dog was presented to the primary care veterinarian for evaluation of a round, pedunculated, nonpigmented mass on the caudal hard palate measuring ∼3–4 cm in diameter. The dog was reported to have a good appetite but impaired mastication. Two weeks prior to referral, the primary care veterinarian had surgically debulked the mass and submitted tissue for histopathologic examination, which revealed an incompletely excised amelanotic OMM. Postoperatively, the owner reported that the dog was able to eat and drink normally. The dog was referred to The University of Tennessee Veterinary Medical Center for staging and further treatment.

Physical examination identified a red, raised, irregular mass on the right aspect of the hard palate measuring ∼5 cm in diameter. Other physical examination abnormalities included a large fatty tumor on the ventral abdomen, bilateral corneal pigmentation, clear nasal discharge, and full anal sacs. No other abnormalities were noted. A complete blood count, serum chemistry profile, and urinalysis showed no significant abnormalities. Thoracic radiographs revealed no evidence of pulmonary metastasis, and lateral thoracic views displayed no tracheal abnormalities. Abdominal radiographs and ultrasound identified bilateral hip dysplasia and focal mineralization of the prostate. Cytologic evaluation of an FNA from the right and left mandibular lymph nodes revealed no evidence of OMM metastasis.

Given the size and location of the dog’s OMM, complete surgical excision was not possible. For local tumor control, hypofractionated radiation therapy was recommended, and a radiation-planning computed tomography (CT) scan was performed under general anesthesia. The dog was administered 6 mg/kg propofol IV, intubated, and maintained on isoflurane and oxygen. The dog was placed in ventral recumbency on an indexed head frame^a. The dog’s body was positioned on a cushion^b, and his head was immobilized with a molded pillow^c and a mask^d. A roll of gauze was inserted in the mouth to achieve separation of the lower and upper jaws. Standard 0.9 helical precontrast images^e in a bone window were acquired and reconstructed in a soft-tissue window. Iodine contrast^f was administered at 2 mL/kg IV through an indwelling catheter at a cephalic vein. Standard postcontrast images in a soft-tissue window at 0.9 and 2 mm slice thickness were acquired. The CT scan revealed a rim-enhancing multilobulated mass surrounding the right maxillary bone at the level of the right fourth maxillary premolar which extended caudally into the rostral aspect of the nasopharynx. There were multifocal areas of bone lysis in the palatine and right maxillary bones.

The postcontrast sequence with 2 mm slice thickness was used for radiation-planning. Organs at risk were contoured including the eyes (left and right individually), inner ears (left and right individually), and brain. Gross tumor volume was defined as the visible contrast-enhancing tumor on CT images, and a 0.5 cm clinical target volume expansion was done in the rostral and caudal directions. The planning target volume was a symmetrical 0.5 cm expansion of the clinical target volume. The dog was prescribed four fractions of 8 Gray each for a total dose of 32 Gray administered via 6 MV photons. Inverse planning^g was used and was prescribed to ensure that 100% of the radiation dose covered a minimum of 90% of the planning target volume. Radiation treatments were delivered once weekly with a linear accelerator^h, and an intensity-modulated radiation therapy technique was used. General anesthesia was induced for each treatment using the same anesthetic protocol used during the radiation-planning CT scan. Port films were taken before each treatment to verify positioning and adjustments were made accordingly. Three weeks after starting radiation therapy, the tumor was decreased in size by 50%. Radiation treatments were well tolerated; the side effects consisted of alopecia and self-limiting mild dry desquamation.

Systemic treatment options for OMM, including immunotherapy and chemotherapy, were also discussed with the owner, but these treatments were declined. Regular rechecks were recommended including periodic thoracic radiographs to evaluate for pulmonary metastasis.

Seven months later, the dog was presented to the primary care veterinarian with a 3 day history of intermittent dyspnea, cough, inappetence, and inability to breathe while sleeping. Thoracic radiographs were performed that revealed an angular soft-tissue opacity within the trachea just distal to the thoracic inlet (Figure 1). Differentials included a tracheal foreign body, granuloma, or neoplasia. No evidence of pulmonary metastasis was seen. The dog was referred to The University of Tennessee Veterinary Medical Center for further evaluation.

View Figure

• Download as Powerpoint
• Download Figure

On physical examination, the dog was gasping and dyspneic. Oral examination identified the OMM on the right aspect of the hard palate, measuring ∼2 cm in diameter. because of the patient’s unstable condition, emergency tracheoscopy was recommended.

Premedication was performed with administration of 0.01 mg/kg acepromazine and 0.01 mg/kg glycopyrrolate IM. The patient was then given 6 mg/kg propofol and 2.5 mg/kg ketamine IV to facilitate placement of an ET tube for general anesthesia. Initial attempts to intubate using a laryngoscope and ET tube were unsuccessful, and therefore, a 4 mm introducer catheter was used to guide placement of a 7 mm ET tube. The dog was positioned in sternal recumbency, and a 5.9 mm bronchoscopeⁱ was inserted through the ET tube and advanced into the trachea. The appearance and texture of the proximal tracheal mucosa was grossly normal. At the level of the distal trachea, just proximal to the carina, a pedunculated, nonpigmented mass arising from the left tracheal wall was visualized. The mass occupied ∼65–75% of the tracheal lumen and measured ∼2 cm in length. No gross mucus, hemorrhage, or other lesions were noted. The bronchoscope was advanced past the mass to the level of the carina, and no additional abnormalities were identified. Following retraction of the scope cranial to the mass, the pedunculated lesion had shifted slightly in positioning so that ∼80–90% of the tracheal lumen was obstructed.

While the patient remained under general anesthesia, treatment options were presented to the owners including tracheal stenting or tracheal resection and anastomosis. Because of the poor prognosis, the owners elected humane euthanasia.

Postmortem examination identified a locally extensive, raised and ulcerated pink mass measuring 2 × 1 × 0.2 cm on the palatine surface medial to the right fourth premolar, first molar, and second molar (Figure 2A). At the level of the thoracic inlet, there was a 2 × 1 × 1 cm soft, tan mass within the lumen of the trachea (Figure 2B). Histologically, the oral mass was an infiltrative neoplasm composed of polygonal to spindle cells in organized packets and nests separated by fine fibrovascular stroma (Figure 2C). The neoplastic cells had moderate amounts of eosinophilic cytoplasm with occasional variably sized brown (melanin) granules and distinct cell borders. The nuclei were round to oval and vesicular with 1–3 prominent nucleoli. Anisokaryosis and anisocytosis were moderate. There were 13 mitotic figures in 10 (400×) high-power fields. Moderate multifocal lymphocytic infiltrates were interspersed within the neoplasm. There was moderate focally extensive ulceration of the overlying mucosa with an overlying serocellular crust, underlying granulation tissue, and neutrophil infiltration. The tracheal mass had a similar microscopic appearance (Figure 2D). Immunohistochemistry for Melan-A confirmed melanoma in both the oral cavity (Figure 2E) and the trachea (Figure 2F). As a result of strong Melan-A positivity at both sites, additional staining for PNL2, TRP-1, and TRP-2 was not performed. There was no gross or histologic evidence of neoplastic cells in any other location. All other gross and microscopic findings were incidental.

View Figure

• Download as Powerpoint
• Download Figure

Discussion

In the case described here, the only additional tumor site identified at necropsy was located within the trachea. At the time of death, there was no evidence of metastasis to the regional lymph nodes or lungs on gross or histologic examination. Thus, we hypothesize that the development of tracheal melanoma may have been a direct result of mechanical implantation of neoplastic cells from the OMM that were dislodged and transplanted during ET intubation. The dog in this case was anesthetized and intubated on six separate occasions prior to development of tracheal melanoma: for debulking of the OMM by the primary care veterinarian, for a radiation-planning CT scan, and for four weekly radiation treatments. Because the OMM in this case was located on the hard palate, it is likely that the ET tube made contact with the mass during intubation. Additionally, because of the dog’s brachycephalic conformation, an elongated soft palate may have contributed to challenges in intubation and resulted in a greater likelihood of traumatic dislodging of tumor cells from the hard palate with transplantation into the tracheal lumen.

This sequela has not been previously reported in dogs with OMM, but direct implantation of tumor cells has been reported to occur after a variety of diagnostic and therapeutic procedures in both dogs and humans. The most common tumor type associated with iatrogenic seeding in dogs is TCC.^8–11 In one study, seeding of neoplastic cells to the abdominal wall as a result of medical procedures, including cystotomy and ultrasound-guided FNA of the primary tumor was reported in 24 of 544 dogs with urinary tract TCC.⁸ Another case series of dogs with TCC of the urinary bladder or urethra reported the metastasis of neoplastic transitional cells to the dermis or subcutis in 12 dogs.⁹ Additionally, another reported cause of iatrogenic TCC seeding is through the placement of percutaneous ureteral stents to relieve malignant obstruction; a case report describes the development of bilateral subcutaneous nodules at the sites of pyelocentesis in a patient with TCC.¹⁰ Biopsy of these nodules confirmed tumor metastasis. Although the overall incidence of tumor seeding appears to be rare in canine TCC, it is an important potential complication which influences the choice of diagnostic and therapeutic procedures, requires careful surgical techniques, and necessitates client education on risk. Other canine tumor types that have been associated with iatrogenic tumor cell seeding include pulmonary adenocarcinoma, thymoma, and carcinomatosis.^12–14

In humans, other iatrogenic procedure-related tumor development has also been described. Spread of head and neck cancers to an abdominal PEG tube site is a rare but recognized complication in humans.^15,16 During placement of a PEG tube, a guidewire is directed through the mouth, esophagus, and into the stomach, and the pull-through technique is commonly used for placement. In patients with head and neck cancers, direct contact of the PEG tube with the tumor during placement can lead to mechanical translocation of malignant cells to the abdominal wall. In patients with oropharyngeal and esophageal malignancies, one study showed that malignant cells were present at the abdominal transcutaneous incision site in 22.5% of patients immediately after pull-through PEG placement, and local tumor development was confirmed in 9.4% of patients at follow-up.¹⁵

Mechanical transplantation of malignant melanoma cells has also been reported in humans. In one case report, tumor cell seeding occurred following a transretinal biopsy in a patient with retinal melanoma.¹⁷ During the procedure, tumors cells were likely seeded into the choroid, as melanoma developed in the conjunctiva 14 mo after the biopsy was performed. Seeding of tumor cells was also reported after a diagnostic punch biopsy in a patient with dermal melanoma.¹⁸ During the biopsy procedure, tumor cells were directly inoculated deeper within the subcutaneous fat, causing tumor depth to increase from 1.8 to 9.0 mm. Several other case reports also describe iatrogenic melanoma seeding, including cutaneous tumor development secondary to percutaneous FNA of a metastatic peripheral lymph node and endoscopic ultrasound-guided FNA of a metastatic perigastric lymph node leading to gastric wall melanoma growth.^19,20

Conclusion

In conclusion, we hypothesize that the cause of tracheal melanoma development in this case was related to mechanical tumor cell seeding during ET tube placement for general anesthesia. Because of the highly metastatic nature of OMM, the possibility of hematogenous metastasis cannot be excluded. However, the presence of a solitary tracheal mass and absence of other sites of tumor metastasis to more common locations such as the lymph nodes and lungs makes this possibility less likely. Thus, the risk of iatrogenic tumor seeding should be considered when placing ET tubes in dogs with OMM.