Original Article DOI: 10.1111/vco.12071 Canine oral mucosal mast cell tumours J. W. Elliott 1 , P. Cripps 1 , L. Blackwood 2 , D. Berlato 3 , S. Murphy 4 and I. A. Grant 5 1 Department of Veterinary Clinical Science, Liverpool University, Liverpool, UK 2 Small Animal Teaching Hospital, Liverpool University, Liverpool, UK 3 Oncology Unit, Animal Health Trust, Newmarket, UK 4 CSAS, Animal Health Trust, Newmarket, UK 5 Small Animal Hospital, University of Glasgow, Glasgow, UK Abstract Mast cell tumours (MCTs) are the most common cutaneous tumours of dogs, however rarely they can arise from the oral mucosa. This subset of MCT is reported to demonstrate a more aggressive clinical course than those tumours on the haired skin and the authors hypothesised that dogs with oral, mucosal MCT would have a high incidence of local lymph node metastasis at presentation and that this would be a negative prognostic factor. An additional hypothesis was that mitotic index (MI) would be prognostic. This retrospective study examines 33 dogs with MCTs arising from the oral mucosa. The results suggest that oral mucosal MCTs in the dog have a high incidence of lymph node metastasis at diagnosis (55%) which results in a poor prognosis. MI and nodal metastasis is highly prognostic. Loco-regional progression is common in these patients and dogs with adequate local control of their tumour had an improved outcome. Despite a more aggressive clinical course, treatment can result in protracted survivals, even when metastasis is present. Keywords chemotherapy, oncology, radiation oncology, small animal, surgical oncology, tumour biology, tyrosine kinase Introduction Mast cell tumours (MCTs) are the most common malignant skin tumours in dogs. 1,2 They are usually derived from haired skin, however, a proportion of the tumours are subcutaneous or arise from the mucocutaneous region or mucosal surfaces. 3 – 5 There is substantial variability in the biological behaviour of cutaneous MCTs in dogs, and a histologic grading scheme developed by Patnaik 6 for classification of MCTs affecting haired skin is one of the strongest prognostic indicators in affected patients. However, this grading scheme cannot be applied to MCTs that originate from the mucous membranes. 6 In the Patnaik scheme, extent of dermal invasion is one of the criteria used for determining tumour grade. 6 As the oral mucosa lacks a dermis, this subset of MCTs cannot be accurately graded using the Patnaik scheme, 6 and therefore histologic grade might not carry the same prognostic value as grading of cutaneous MCTs. A more recent grading scheme designed by Kiupel et al. 7 might be more applicable to this subset of tumours, because of the different criteria used. Anatomic location has also been described as a predictor of the biological behaviour of MCTs in dogs, with studies reporting more aggressive behaviour of MCTs involving gastro-intestinal, digital and visceral locations. 8 – 11 This aggressive behaviour is manifested by increased systemic involvement of the MCT, including lymph node and distant metastasis, compared with dogs having cutaneous MCTs. Patients with aggressive MCTs frequently die of their disease. 12 Little information exists on the behaviour of mucosal or mucocutaneous MCTs, with most reports discussing only a limited number of cases involving these locations. In one study, MCTs located on the muzzle of 24 dogs had a high regional metastatic rate. 13 Another reported a number of dogs with oral or peri-oral MCTs which revealed a high-regional lymph node metastatic rate and these dogs had a worse prognosis. 5 Only 14 of these cases originated from the oral mucosa Correspondence address: J. W. Elliott Department of Veterinary Clinical Science Liverpool University Liverpool CH64 7TE, UK e-mail: jameselliott44@ hotmail.com  2013 John Wiley & Sons Ltd 1