Tumours of the oral cavity of dogs account for approximately 6% of all tumours (Hoyt & Withrow 1984). They commonly arise from the gingiva but can arise from buccal mucosa, mandible, maxilla, palate, dental structures, tongue and tonsils.

A number of histological subtypes are recognized and as the prognosis and treatment options depend upon the histological diagnosis, early biopsies are recommended:

For odontogenic tumours, the following tissues should be sampled:

In general these are tumours of older dogs, except FSA seen in large breed dogs (e.g. German Shepherds, Golden Retrievers and Labrador Retrievers), papillary SCC seen in dogs of less than 6 months of age (Ogilvie et al 1988), undifferentiated malignancy of young dogs (less than 2 years) and viral-induced papillomatosis (less than 1 year). MM and FSA are more common in male dogs (Dorn et al 1968).

The breeds most commonly affected with oral tumours are Cocker Spaniels, Poodles, Weimaraner, German Shepherds, Golden Retrievers, German Shorthaired Pointers and Boxers (Cohen et al 1964, Dorn & Priester 1976, Todoroff & Brodey 1979). Small breeds are at a higher risk of MM and tonsillar carcinomas. Dogs with heavily pigmented oral mucosa are predisposed to MM (Dorn & Priester 1976, Dorn et al 1968).

The clinical signs include decreased appetite, halitosis, loose/missing teeth, bloody saliva, exophthalmos, epistaxis, dysphagia, difficult/painful mastication and psychogenic polydipsia. Rostral tumours are most likely to by seen by the client and therefore diagnosed sooner than more caudally located tumours. Dyspnoea is common with large tonsillar, posterior pharyngeal tumours or lymphadenopathy.

Routine blood work, haematology and biochemistry are indicated, plus urinalysis.

Most dogs have stage III or IV disease at diagnosis.

Median survival times (MST) for dogs with oral melanoma treated with surgery are approximately 17–18, 5–6 and 3 months with stage I, II and III disease, respectively (Bergman 2007). Overall, 35% of patients with MM treated with mandibulectomy or maxillectomy will survive more than 1 year (Withrow & MacEwen 2001).

Prognostic indicators include:

Indicators that are not prognostic: age, breed, sex, degree of pigmentation, microscopic appearance (with routine histological assessment) and DNA ploidy.

The prognosis is poor to fair, as 95% of dogs will eventually develop metastatic disease. With no treatment, the MST is 65 days (Harvey et al 1981). Metastatic rate is site, size and grade dependent.

MM responds well to radiotherapy (83–100% response rate with up to 70% having a complete response) (Bateman et al, 1994a and Bateman et al, 1994b, Blackwood & Dobson 1996, Freeman et al 2003, Théon et al 1997a). Typically a coarse fractionation protocol is used (3–4 fractions of 8–9 Gy).

The use of radiation in the management of oral MM depends on the size, location and goals of treatment for each patient. Proulx et al (2003) compared a number of radiation protocols and showed no significant difference in survival times based on protocol (9 Gy × 4 fractions, 10 Gy × 3 fractions, or 2–4 Gy × 12–19 fractions).

A number of studies have also looked to evaluate the efficacy of platinum-based chemotherapy in conjunction with radiation, but unfortunately results to date have shown no survival advantage of these combinations (Freeman et al 2003, Proulx et al 2003). For patients that have undergone surgery to reduce bulky disease down to microscopic disease, radiotherapy is used in the adjuvant setting. Using a total of 6 fractions (once weekly for 6 weeks or twice weekly for 3 weeks) the 1-year survival rate is 48%, 2-year survival rate 21% and the median disease-free interval (MDFI) 235 days/8 months and the MST 240–363 days (Freeman et al 2003). Palliation of MM (coarse fractions, 4 treatments) gave 69% complete response and 31% partial response (Blackwood & Dobson 1996). Metastasis was the major cause of death, and 58% of patients were euthanized or died of metastatic disease.

For patients with large tumours, where surgery is declined or when there is already evidence of metastatic disease, radiotherapy is a palliative treatment that can reduce tumour burden and improve quality of life. Typically the authors recommend three treatments once weekly when minimal disease is present and four treatments once weekly for palliative treatment. Positive lymph nodes should also be either irradiated or excised as part of a palliative protocol.

MM is resistant to chemotherapy and there is no known effective chemotherapeutic agent for metastatic MM. A number of approaches using platinum-based drugs have attempted to improve long-term survival in patients with large primary tumours, e.g. intralesional cisplatin. In one study of 20 dogs, intralesional cisplatin gave 70% partial response with implant complications (mild local necrosis) in 85% of patients. MST was 51 weeks for responders (compared with 10 weeks for non-responders) (Kitchell et al 1994). Tumours located on the mandible had a better prognosis than maxillary tumours.

In another study on 27 dogs given carboplatin chemotherapy the response rate was 28%, with 4% complete response, 24% partial response and 36% each with stable disease and progressive disease. Median duration of partial response was 165 days; smaller dogs were more likely to respond but experienced a higher incidence of gastrointestinal side effects (Rassnick et al 2001).

It is in the field of immunotherapy that the next step forward is likely to occur with the development of a tumour vaccine.

It is known that administering bacillus Calmette–Guérin (BCG) or levamisole does not improve survival. There is a mild survival advantage with advanced local disease with surgery and Corynebacterium parvum compared with surgery alone (MacEwen et al 1986). Intralesional injection of granulocyte–macrophage colony-stimulating factor (GM-CSF) or interleukin-2 (IL-2) induced partial or complete response and prolonged survival time compared to historical controls (Dow et al 1998).

Surgery and liposome muramyl tripeptide immunotherapy (L-MTP) showed that 80% of patients with stage I disease were still alive at 2 years compared with 25% who underwent surgery alone. L-MTP did not influence survival in patients with stage II or III disease (MacEwen et al 1986). IL-2 and interferon (IFN) induce natural killer (NK) cells, lymphokine-activated killer (LAK) cells and augment antibody-dependent cellular cytotoxicity, and are promising avenues of research because MM is highly immunogenic. Immunotherapy and radiotherapy may be synergistic.

At the time of writing, a tumour vaccine has been given a conditional license by the US Food and Drug Administration for the treatment of canine patients with stage II–IV oral melanoma. The vaccine uses a DNA plasmid containing a human tyrosinase gene. Tyrosinase is present in canine melanoma cells and although similar to the canine tyrosinase the xenogeneic DNA stimulates an immune response against melanoma cells expressing this gene (Bergman et al 2006, Liao et al 2006). Currently the vaccine is recommended for patients that have received appropriate treatment, surgery or radiotherapy for local or regional disease that have an expected survival time of 60–150 days; those receiving the vaccine have been reported to have increased survival times up to 389 days (Bergman et al 2003).

SCC of the base of the tongue and the tonsil is highly metastatic (metastasis in up to 73%) and recurs locally or regionally (Beck et al 1986, Brooks et al 1988). Good prognostic factors include rostral location for SCC originating from the gum. Long-term survival and cure is possible for SCC treated with appropriate surgery (not including tonsillar or base of the tongue).

The 12-month survival rate is 50% for complete resection and 60–80% for complete resection with low-grade histology (Carpenter et al 1993). Due to the difficult location, early detection is imperative to allow complete excision.

Radiotherapy can be used in the adjuvant or neoadjuvant setting. Side effects from radiotherapy include fibrosis of the tongue.

Epulides are the most common benign oral tumours arising from the periodontal ligament. Fibromatous is the most common (57%), then ossifying (23%), acanthomatous (18%) and giant cell (2%) (Yoshida et al 1999). Another paper reported 40% acanthomatous, 32% ossifying and 25% fibromatous (Bjorling et al 1987). In this study, the median survival time post-surgery for all oral epulides was 49 months, with a 1-year survival rate of 92%.

These tumours arise from dental lamina and may arise from either dental epithelium or nests of epithelial cells. Odontogenic tumours should be classified according to whether they are of epithelial, mesenchymal or mixed epithelial and mesenchymal origin, rather than based on inductive changes (Gardner 1992).

These rare tumours occur in young cats (6–18 months), more frequently in males. They are usually located rostrally, including the canine teeth of the mandible or maxilla (more common). There is a variable degree of bone destruction and proliferation; expansion with deformity of the teeth is common. The treatment is either mandibulectomy/maxillectomy or radiotherapy. Metastasis is not reported (Stebbins et al 1989).

Both rare and benign, these tumours arise from the dental follicle during early stage tooth development. The diagnosis is made when there is evidence of induction of both enamel and dentin. These are intraosseous and locally invasive but do not metastasize. Treatment includes surgical debulking and cryosurgery or wide surgical excision (Figueiredo et al 1974).

The cause is horizontal transmission of papovavirus and usually occurs in young patients and more commonly in dogs (Norris et al 1985). The lesions are wart-like and multiple in the oral cavity, pharynx, tongue or lips. Spontaneous regression can occur in 4–8 weeks. Recommended treatment is surgery, cryosurgery or electrosurgery if severe or interfering with swallowing. Crushing lesions in situ may release antigen and result in immune-induced regression. The prognosis is excellent.

Seen in young dogs (1–7 years), there is a breed predisposition in Siberian Huskies and Cavalier King Charles Spaniels (Bredal et al 1996, Madewell et al 1980). These frequently raised, ulcerated lesions on the tongue often spontaneously regress, otherwise corticosteroids or surgical excision is required. Recurrence is rare.

These cysts arise within islands of odontogenic epithelium (Figure 13.7). They are benign, non-neoplastic circumscribed cystic lesions that may represent an early stage of a continuum to malignant epithelial tumours, e.g. basosquamous carcinoma (Poulet et al 1992). They are seen as closed cavities or cysts, with one or more teeth embedded in the cyst wall (Baxter 2004). Diagnosis is by radiographs with a characteristic radiolucent halo around a non-erupted tooth originating at the cemento-enamel junction and enveloping the crown of the tooth. Treatment of choice is surgical removal of the non-erupted tooth and cyst lining. Thorough curettage of the walls of the cyst is required to prevent recurrence.

These are rare in dogs. Mean age of affected dogs is greater than 10 years. No consistent breed predisposition has been reported (Hammer et al 2001). Mandibular and parotid glands are the most frequently involved (Carberry et al 1987). Although most are malignant (adenocarcinomas), benign lesions do occur (adenoma and lipoma) (Brown et al 1997, Stubbs et al 1996). Other malignant histological types include SCC, MCT, etc. Sialadenitis, sialoceles and salivary gland abscesses are differential diagnoses. Metastasis to RLN, lung, bone, eyes and kidneys has been reported (Grevel et al 1978, Habin & Else 1995).

Fine needle aspiration (FNA) cytology and biopsy are important for diagnosis (incisional biopsy prior to definitive surgery). The RLN should be palpated and assessed by FNA, or biopsy if enlarged. Thoracic radiographs/CT should be taken as part of staging. CT/MRI scans of local tumour may be of use to plan definitive surgical resection ± radiotherapy. Removal of the tumour with wide margins may be difficult due to location; however, a unilateral vagosympathetic trunk, jugular vein and carotid artery may be sacrificed with minimal morbidity in the dog. The surgeon should be familiar with normal regional anatomy of this area.

Radiation and chemotherapy may be used adjuvantly to surgery. This may improve survival times over surgery alone; however, detailed reports of survival outcomes comparing these groups are not available due to the small number of cases.

The authors suggest that dogs with non-metastatic disease have a better prognosis when treated with surgery than cats, as wider margins are easier to achieve in dogs. For large fixed tumours, radiotherapy can be utilized to shrink the tumour.

Histopathological features were not found to be prognostic in one study (Hammer et al 2001), although clinical stage was predictive.

Invasive SCC is usually preceded by a protracted course of disease (months to years) that progresses through stages. First there is crusting and erythema, then superficial erosions and ulcers (typically carcinoma in situ or early SCC) and, finally, deeply invasive and erosive lesions (Figure 13.8). Because SCC has a low rate of metastasis, which, if it occurs, does so late in the course of disease, dogs can have prolonged survival even if left untreated, although there is unacceptable morbidity associated with an ulcerated, invasive, deforming and secondarily infected cancer.

Cytology and superficial biopsies are not useful. Lymph nodes are usually negative except with advanced disease and thoracic radiographs are invariably negative for metastases. Regional radiographs are generally unrewarding. CT/MRI are valuable for staging dogs with SCC of the nostril to help define the posterior extent of disease and hence level of resection. Rhinoscopy can also assist in determining the extent of caudal disease and to look for multiple lesions in dogs (Banks & Straw 2004).

It may be possible to prevent or arrest the course of preneoplastic disease by limiting exposure to sun or tattooing to add pigment protection. Topical sunscreens are readily licked off and rarely help.

When inflammation and ulceration are present it is very difficult to maintain the tattoo as it is rapidly removed by macrophages. At best, tattoos would need to be repeated regularly.

Attempts to increase epithelial differentiation with synthetic derivatives of vitamin A are generally unsuccessful for advanced disease, but may be helpful in reversing or limiting the growth of preneoplastic lesions.

SCC can be divided into two categories depending on the degree of invasiveness:

In the authors’ experience, Golden Retrievers are over-represented with infiltrative SCC. If large, these tumours are difficult to control. The nasal planum should be completely removed with the tumour, if possible with margins of 1 cm of grossly normal tissue. If the margins are incomplete, adjuvant radiotherapy has been successful, resulting in a cure in one of four dogs (25%); however, the other three dogs had local recurrence at a median of 9 weeks (Lascelles et al 2000). Combined removal of premaxilla and nasal planum has been reported in the dog, with acceptable cosmetic appearance, good function and tumour control (Kirpensteijn et al 1994). An improved cosmetic appearance is reported with the use of bilateral labial mucocutaneous rotation-advancement flaps after nasal planectomy and premaxillectomy for SCC in the dog (Gallegos et al 2007). Another paper describes aggressive rostral maxillectomy and nasal planectomy. Complete margins were obtained in all cases, but one dog had local recurrence 10 months after surgery. All others had no signs of local recurrence during an 11–66 month follow-up period (Lascelles et al 2004) (Figure 13.9A,B).

Combined systemic chemotherapy and then surgery have been used in one dog to reduce tumour size, allowing it to be more amenable to wide surgical resection (Banks & Straw 2004); however, radiotherapy is more likely to be successful in the neoadjuvant setting. Surgical resection of the nasal planum with clean margins correlates to prolonged survival and potential cure (Gallegos et al 2007, Kirpensteijn et al 1994, Lascelles et al 2000). One potential problem in the months following surgery is stricture of the combined nasal orifice, and attempts at managing this in dogs include wide skin excision and removal of the nasal septum rostrally, laser ablation, rubber stents or permanent placement of a stainless steel intraluminal expansile stent.

Radiotherapy is applicable in the management of deeply infiltrative tumours as palliative therapy in patients where surgery is declined or when disease is very extensive; survival times range from 8 weeks to greater than 1 year. For early disease, excisional biopsy and radiotherapy is an option with good cosmesis and long-term control (Figure 13.9C).