Canine cutaneous mast cell tumours (CMCTs) account for 11.2 to 17.6% of all cutaneous tumours in dogs. These tumours show variable biological behaviour, with some being cured by surgical excision while others recur locally or at distant sites. A three-tiered grading system devised by Patnaik et al (1984) is in currently used to predict the behaviour and prognosis of canine CMCTs. However, this system is based on a specific set of histological criteria, and CMCTs with characteristics of more than one grade are often placed in the grade two category by default. This reduces the prognostic significance of this system. Also, having rigid categories for each grade has led to variation between veterinary pathologists when assigning grades to CMCTs. Hence, the main aim of this study was to develop a new grading system for canine CMCTs; one that is based on tumour characteristics that are shown to be most closely linked with time to death.
The patient data and the histological slides of 324 CMCTs surgically removed from 246 dogs were analysed. Tumours most commonly arose as a single mass in the caudal half of the animal and the majority were graded as a two using the current grading system.
The risk for developing a CMCT was highest in dogs aged 3.5 to 12 years of age.
The risk decreased once animals were older than 12.5 years of age. Boxers and Staffordshire Bull Terriers were found to be at greatest risk of CMCT development.
Tumour characteristics with the greatest impact on survival time in dogs with CMCTs were tumour diameter, degree of granulation, completeness of tumour excision and mitotic rate. Dogs with CMCTs > 50mm in diameter, with a mitotic rate of >3 per HPF, and that were incompletely excised had the highest risk of death from their tumour.
Utilising this data, two versions of a new CMCT grading system were developed. The first system, designed as a flow chart gave tumours a grade of 1 to 4, with 1 being the most benign and 4 the most malignant. Median times to death were established for each grade to give an indication of likelihood of survival. The second grading system, based on the same characteristics, gave an indication of how much more likely an animal was at risk of dying from their CMCT compared to a dog with the most benign mass. These two versions may provide clinicians with improved information on the prognosis for dogs with CMCTs as they are only rely on a small number of prognostically significant tumour characteristics, and time to death has been estimated for most combinations of these variables. However, validation of these new grading systems is still needed using a much larger population of dogs with CMCTs.
The second aim of this study was to investigate whether mutations in the c-kit gene were associated with the histological appearance of CMCTs and prognosis. This gene encodes a protein receptor, KIT, which is responsible for the growth and differentiation of mast cells. As yet no one has carried out a study to determine if there are any associations between the gross and histological appearance of CMCTs and the presence of mutations in a particular area of the gene. In doing so, it could highlight a particular gross or histological tumour characteristic that could be used to identify animals at higher risk of death. The entire transcript sequence of c-kit mRNA from 16 CMCTs was isolated and sequenced.
The CMCTs were selected using the tumour characteristics determined earlier in the study to have prognostic significance; nine had characteristics associated with the best prognosis (Group A), and seven exhibited one or more characteristics associated with the worst prognosis (Group B).
Two significant types of mutations were found in the c-kit gene; nonsynonomous single nucleotide polymorphisms (SNPs) and internal tandem duplications (ITDs). In Group A, 17 nonsynonomous SNPs were identified in six CMCTs. Two ITDs which affected the fifth Ig-like domain and the juxtamembrane domain (JMD) of the KIT protein were also detected. The Group B tumours contained only a total 4 SNPs and 2 ITDs. The ITDs in this group were both located in the JMD.
The nonsynonomous SNPs affected only one to two clones per tumour and none were identical. Therefore the presence of c-kit mutations and the gross and histological characteristics of CMCTs were not able to be compared. Also, the prognostic significance of SNPs in c-kit was not able to be determined in the study population. However, three dogs with CMCTs with ITDs in the JMD of the c-kit had tumour recurrence, and two from the Group B also had metastases to the regional lymph node. These same two dogs had also died by the conclusion of the study which suggests these duplications may be a negative prognostic finding.