Assessment of Canine Mast Cell Tumor Mortality Risk Based on Clinical,Histologic,Immunohistochemical,and Molecular Features

Abstract

Mast cell tumor (MCT) is a frequent cutaneous neoplasm in dogs that is heterogeneous in clinical presentation and biological behavior, with a variable potential for recurrence and metastasis. Accurate prediction of clinical outcomes has been challenging. The study objective was to develop a system for classification of canine MCT according to the mortality risk based on individual assessment of clinical, histologic, immunohistochemical, and molecular features. The study included 149 dogs with a histologic diagnosis of cutaneous or subcutaneous MCT. By univariate analysis, MCT metastasis and related death was significantly associated with clinical stage (P < .0001, r _P = –0.610), history of tumor recurrence (P < .0001, r _P = –0.550), Patnaik (P < .0001, r _P = –0.380) and Kiupel grades (P < .0001, r _P = –0.500), predominant organization of neoplastic cells (P < .0001, r _P = –0.452), mitotic count (P < .0001, r _P = –0.325), Ki-67 labeling index (P < .0001, r _P = –0.414), KITr pattern (P = .02, r _P = 0.207), and c-KIT mutational status (P < .0001, r _P = –0.356). By multivariate analysis with Cox proportional hazard model, only 2 features were independent predictors of overall survival: an amendment of the World Health Organization clinical staging system (hazard ratio [95% CI]: 1.824 [1.210-4.481]; P = .01) and a history of tumor recurrence (hazard ratio [95% CI]: 9.250 [2.158-23.268]; P < .001]. From these results, we propose an amendment of the WHO staging system, a method of risk analysis, and a suggested approach to clinical and laboratory evaluation of dogs with cutaneous MCT.

Keywords

dog skin mast cell tumor mastocytoma prognosis clinical staging tumor grade mitotic count Ki-67

Mast cell tumor (MCT) is the second most frequently diagnosed malignancy in dogs,^9,18 representing 16% to 21% of all canine cutaneous neoplasms.²⁶ Nevertheless, canine MCT is highly heterogeneous in clinical presentation and biological behavior with a variable potential for recurrence and metastasis. Thus, there is a need to identify prognostic factors that predict outcome.^26,31,34

Current prognostic factors include: (1) clinical features, such as clinical stage,^4,26,46,47 breed,^26,28 anatomic site,^19,26 and history of tumor recurrence (ie, the dog presented with a recurrent tumor rather than a primary tumor)^6,29,33,44; (2) histological criteria considered for the 3-tier Patnaik grading system (grades 1, 2, and 3)³³ and the 2-tier Kiupel grading system (low and high grade),²⁰ although some histological features may represent isolated prognostic factors, as shown for the mitotic count^27,35; (3) immunohistochemical evaluation of Ki-67 labeling index⁴⁰ and KITr expression^14,20,48 and evaluation of metalloproteinases (MMP-2 and MMP-9)²³ and fibroblast activating protein¹⁵; (4) molecular features, emphasizing gain-of-functions mutations in c-KIT oncogene, such as internal tandem duplications in exon 11^2,49,50,53 as well as in exons 8 and 9^24,43 (which may also predict a favorable response to tyrosine kinase inhibitors currently licensed for use in dogs in United States and Europe, ie, masitinib and toceranib).^17,25 Despite all these prognostic factors, the decision-making process remains challenging for the small animal practioner.⁴⁷ It would be desirable to classify dogs with MCT as low, intermediate, or high risk for tumor recurrence or metastasis since this would be of value for clinical decisions as well as patient selection in clinical trials. The aim of this study was to develop a classification system for dogs with MCT based on the mortality risk, considering the assessment of clinical, histological, immunohistochemical, and molecular features.

Materials and Methods

Patients

This prospective-retrospective study was conducted from 2006 to 2016 and approved by the Ethics Committee on Animal Use of Universidade Federal de Minas Gerais, Brazil (protocol number 384/2013). Inclusion criteria was histological diagnosis of cutaneous or subcutaneous MCT attended in a primary case veterinary hospital, resulting in 316 cases. Exclusion criteria was lack of histological sample or staging information (ie, fine needle aspiration of regional lymph nodes and abdominal ultrasound) or patients who were lost to follow-up in less than 6 months after surgery, resulting in 149 dogs (84 prospectively enrolled). Patients who died due to MCT or unrelated causes were included, even if followed for less than 6 months. Follow-up consisted of phone calls and rechecks during treatment or every 4 to 6 weeks after treatment in the first 6 months and every 2 to 3 months after that.

Surgical and Medical Management

All dogs were evaluated by electrocardiogram, complete blood count, coagulation profile (time of activated partial thromboplastin and time of prothrombin), and serum biochemistry (total protein and albumin, blood urea nitrogen, creatinine, alkaline phosphatase, alanine transferase, aspartate transferase, and gamma-glutamyl transferase) prior to anesthesia. Patients were submitted to full staging, including physical exam, fine needle aspiration of lymph nodes, and abdominal ultrasound. Patients with suspected metastastes on lymph node evaluation had them removed for further staging.

Prednisone or prednisolone was used at the clinician’s discretion at a dose of 40 mg/m² daily for 7 to 10 days prior to surgery to promote cytoreduction or reduce peritumoral inflammation, as previously suggested.^9,39,47 According to staging, prognostic and clinical presentation after the cytoreductive treatment, the patients were treated by surgical resection (excisional biopsy) or incisional biopsy. Excional biopsy involved an elipitical incision with 2- to 3-cm lateral margins, followed by deep dissection aiming to include at least 1 muscular fascia.¹² The excised tissues were fixed in 10% neutral buffered formalin for subsequent histopathologic, immunohistochemical, and molecular evaluation. After surgery, treatments were chosen according to the oncologist's experience and prognostic factors evaluated by the laboratory findings.

Histological Analysis

Tumor tissue was fixed in formalin for 48 hours and 4-μm-thick sections were prepared routinely and stained with hematoxylin-eosin and toluidine blue. Histopahological examination was performed under light microscopy by a single veterinary pathologist (LM). Tumors were graded using the Patnaik³³ and Kiupel²⁰ systems. In addition, a combined Patnaik/Kiupel grade was assigned by combining these 2 grades. Neoplasms were also evaluated to characterize the tumor extent (subcutaneous, dermal, ulcerated, or within the muscle), architectural pattern (more than 50% in cords or in sheets), intratumoral necrosis, edema, hemorrhage, collagen hyalinization, cystic change of apocrine glands, desmoplasia, and eosinophilic infiltrate in a semi-quantitative analysis (absent, mild, moderate, or intense). Other assessments were the surgical margins (clean, narrow, or compromised), number of binucleated cells, and mitotic count. Multinucleated cells (more than 2 nuclei) were counted for grading according to the Kiupel system.²⁰ Mitotic figures, binucleation, and multinucleation were counted in fields (FN = 22 mm) with higher numbers of these events. Patients presenting with multiple lesions were classified according to the most undifferentiated pattern (ie, the highest grade, highest mitotic count and Ki-67 labeling index, abnormalities in KITr expression, and mutations in c-KIT oncogene, when present).

Immunohistochemistry

For this technique, 4-μm-thick sections were obtained from representative paraffin blocks of each case and placed into gelatin-coated slides. Slides were deparaffinized and rehydrated in an alcohol series. Heat-induced antigen retrieval was performed (Target Retrieval Solution Citrate pH 6, Dako Cytomation) under pressurized heat (20-25 mmHg, 125°C, 2 minutes). Endogenous peroxidase was blocked by immersion in 3% hydrogen peroxide, and protein blockage was performed with a Protein Block solution (Thermo Scientific Ultravision). Primary antibodies CD117 (polyclonal, 1:800, Dako Cytomation) and Ki-67^d (MIB-1, 1:25, Dako Cytomation) were incubated at 4°C for 16 hours. Secondary antibody (Advance HRP Link, Dako Cytomation) was incubated in a humid chamber for 30 minutes, and the reaction was amplified by the polymer (Advance HRP Enzyme, DakoCytomation). The immunoreaction was revealed with the chromogen 3,3–diaminobenzidine tetrahydrochloride (Liquid DAB + Substrat Chromogen System, Dako Cytomation) and counterstained with Harris hematoxylin. Sections from canine MCT known to express Ki-67 and CD117 were used as positive controls. Negative controls were assessed using normal serum instead of the primary antibodies.

The immunolabeling pattern for KITr was evaluated by counting membrane, focal, or diffuse cytoplasmic immunoexpression of CD117 (KIT patterns I, II, or III, respectively) in 10% of the tumor fragment at 400× magnification. Each MCT was assigned the highest staining pattern present in at least 10% of the neoplastic cell population or present in large clusters of neoplastic cells within the tumor.²¹ Ki-67 labeling index was determined as the percentage of positive nuclei in at least 500 neoplastic cells in 3 to 5 hot-spots (ie, 3-5 400× fields with higher number of labeled nuclei). Each nucleus with evidence of immunolabeling was considered positive for Ki-67.^1,37

Screening of Mutations in the c-KIT Oncogene

The PCR reaction for amplification of the fragment of interest in the c-KIT oncogene (exon 11 – nucleotides 1661 to 1787 of chromosome 13) was performed by Progen (Vetpat Laboratory) using DNA extracted from paraffin-embedded tumor by the proteinase K method. The primers (Invitrogen) were: forward 5’-ATCTGTCTCTCTTTTCTCCCCC-3’ and reverse 5’-TGGGGTTCCCTAAAGTCATTGT-3’. The product was 225 bp in the absence of mutations (native c-KIT). Reactions were carried out in a GenPro thermocycler (BIOER Technology), with a maintenance at 95°C for 5 minutes, then 30 cycles of 94°C for 45 seconds for denaturation, 63°C for 45 seconds for annealing, 72°C for 1 minute for extension, then maintained at 72°C for 10 minutes for molecular stabilization. The amplified material was separated by electrophoresis at 100 V with free amperage. The positive and negative controls were canine healthy skin samples and water, respectively.

Assessment of Response

Clinical outcome was defined as a categorical variable in 3 levels: (1) no local recurrence, metastasis, or tumor-related death; (2) MCT local recurrence or metastasis but no tumor-related death; (3) MCT local recurrence or metastasis and tumor-related death. The disease-free interval (DFI) was considered from the beginning of therapy for patients that achieved complete remission or those for whom surgery was amenable. The overall survival (OS) was determined for all patients and defined as the time from the beginning of therapy until tumor-related death. Patients lost to follow-up or with unrelated deaths were censored. Cytology was used to confirm the diagnosis in case of progressive disease and appearance of new lesions.

Statistical Analysis

Statistical analysis was performed using the Stata Statistical Software (v. 12.0, Stata Corp). Differences were considered significant when P < .05.

A matrix correlation considering prognostic factors and clinical outcome was conducted with Pearson test. Significant correlations were considered strong when they occurred in over 49% of the studied population (r > 0.7), moderate when occurring in 9% to 49% (0.3 < r <0.7), and weak when occurring in less than 9% of the population (r < 0.3).

A ROC curve was used to set a cut-point for quantitative variables that were significantly correlated to clinical outcome. The cut-point was considered with maximum specificity without compromising sensitivity for predicting the clinical outcome.

The overall survival of patients in this study was estimated by the Kaplan-Meier curve, and the logrank test of Cox-Mantel was used to compare the curves within each prognostic factor. Cox proportional hazard model was performed along with factor analysis.

Results

This study included 149 dogs, aging from 3 to 16 years old (9.6 ± 3.1; mean ± SD). Aproximately 19.5% (29/149) were mixed-breed dogs, and the remaining were of 34 breeds, including Labrador Retriever (n = 30), Poodle (n = 10), Golden Retriever (n = 7), Pinscher (n = 7), Pit-bull (n = 6), Boxer (n = 5), Dachshund (n = 5), and Cocker Spaniel (n = 5).

MCT were considered solitary in 95.3% of cases (142/149). The anatomic sites included the limbs (70/142, 49.3%) and trunk (39/142, 27.5%) as well as scrotum (9.9%), vulva (2.8%), perineum (2.8%), face (2.8%), prepuce (2.1%), oral mucosa (2.1%), and pinna (0.7%).

Clinical Outcomes and Associated Parameters

Of the 149 cases, 97 (65.1%) had no local recurrence, metastasis, or tumor-related death. Local tumor recurrence was not observed in this study, even in patients with narrow (n = 14) or incomplete surgical margins (n = 84). Progressive disease with concomitant metastasis to the skin and/or lymph node occurred in 52 cases. Of these, 6 of 52 (11.5%) remained alive at the end of the study, and 46 were considered tumor-related deaths (ie, 46/149, 30.8%, tumor-related deaths in the full population). Tumor-related death resulted from euthanasia in 42 cases due to disease progression and clinical deterioration. Four cases died because the owner refused euthanasia; of these, 1 had metastasis to iliac lymph nodes and paraneoplastic syndromes including anaemia and gastritis in addition to paralytic ileus, and 3 developed marked anaemia related to bone marrow invasion (with no metastases to the liver or spleen, detected through ultrasound or autopsy). Threrefore, clinical outcome was defined as (1) no local recurrence, metastasis, or tumor-related death; (2) MCT metastasis but no related death; (3) MCT metastasis and related death. Significant correlations between parameters are shown in Table 1 and as Supplemental Material. Regarding clinical features, the clinical outcome was not related to breed, sex, age, or location of the tumor (multicentric, trunk, limb, face/muzzle, mucosa, scrotum, prepuce/vulva/perineum), but it was moderately correlated to WHO clinical stage (P < .0001, r _P = –0.610) and history of tumor recurrence (whether it was a primary tumor or a recurrence) (P < .0001, r _P = –0.550).

Table 1.

Statistically Significant Correlations Among Clinical and Pathologic Parameters in 149 Dogs With Mast Cell Tumors.

	Correlated Parameters	P Value	r _P Value	Correlated Parameters	P Value	r _P Value
Clinical outcome	Clinical staging (WHO)	<.0001	–0.610	Architectural pattern	<.0001	–0.452
	Clinical staging (WHO amendment)	<.0001	–0.602	Mitotic count	<.0001	–0.325
	Recurrence history	<.0001	–0.550	Ki-67 index	<.0001	–0.414
	Patnaik grade	<.0001	–0.380	KITr	.01	–0.207
	Kiupel grade	<.0001	–0.500	c-KIT status	<.0001	–0.356
	Patnaik-Kiupel grade	<.0001	–0.375	—	—	—
Clinical staging (WHO)	Clinical staging (WHO amendment)	<.0001	0.974	Architectural pattern	<.0001	0.470
	Recurrence history	<.0001	0.538	Eosinophilic infiltrate	.04	–0.237
	Tumor location	.001	0.271	Binucleated cells	.009	0.292
	Patnaik grade	<.0001	0.291	Mitotic count	<.0001	0.326
	Kiupel grade	<.0001	0.457	Ki-67 index	<.0001	0.374
	Patnaik-Kiupel grade	<.0001	0.345	c-KIT status	<.0001	0.368
Clinical staging (WHO amendment)	Recurrence history	<.0001	0.538	Cystomatosis	.04	–0.273
	Tumor location	.004	0.242	Eosinophilic infiltrate	.05	–0.226
	Patnaik grade	.001	0.277	Binucleated cells	.009	0.283
	Kiupel grade	<.0001	0.454	Mitotic count	<.0001	0.352
	Patnaik-Kiupel grade	<.0001	0.393	Ki-67 index	<.0001	0.417
	Architectural pattern	<.0001	0.450	c-KIT status	<.0001	0.397
Recurrence history	Age	.03	–0.200	Cystomatosis	.03	–0.301
	Patnaik grade	.002	0.247	Mitotic count	<.0001	0.333
	Kiupel grade	<.0001	0.325	Ki-67 index	.001	0.267
	Patnaik-Kiupel grade	<.0001	0.323	KITr	.04	0.168
	Architectural pattern	.02	0.245	c-KIT status	<.0001	0.299
Tumor location	Kiupel grade	.05	0.165	Cystomatosis	.02	–0.327
Tumor location	Architectural pattern	.02	0.248	Binucleated cells	.04	0.230
Patnaik grade	Kiupel grade	<.0001	0.467	Mitotic count	<.0001	0.495
	Patnaik-Kiupel grade	<.0001	0.426	Ki-67 index	<.0001	0.413
	Architectural pattern	.001	0.355	KITr pattern	<.0001	0.379
	Binucleated cells	<.0001	0.513	c-KIT status	<.0001	0.295
Kiupel grade	Patnaik-Kiupel grade	<.0001	0.626	Mitotic count	<.0001	0.507
	Tumor extension	.009	0.244	Ki-67 index	<.0001	0.454
	Architectural pattern	<.0001	0.453	KITr pattern	<.0001	0.312
	Desmoplasia	.02	0.304	c-KIT status	<.0001	0.384
Patnaik-Kiupel grade	Architectural pattern	<.0001	0.385	Ki-67 index	<.0001	0.482
	Cystomatosis	.03	–0.286	KITr pattern	.008	0.215
	Mitotic count	<.0001	0.454	c-KIT status	<.0001	0.345
Tumor extent	Architectural pattern	.006	0.293	c-KIT status	.02	0.212
Architectural pattern	Necrosis	.03	0.270	Mitotic count	.02	0.239
	Desmoplasia	.03	0.271	Ki-67 index	.007	0.284
	Cystomatosis	.01	–0.334	KITr pattern	<.0001	0.368
	Binucleated cells	.007	0.311	—	—	—
Necrosis	Mitotic count	.006	0.322	KITr pattern	.02	0.266
Collagenolysis	c-KIT status	.01	–0.116	—	—	—
Desmoplasia	Binucleated cells	.05	0.244	—	—	—
Cystomatosis	Ki-67 index	.006	–0.372	KITr pattern	.03	–0.302
Binucleated cells	Mitotic count	<.0001	0.484	KITr pattern	.001	0.363
Binucleated cells	Ki-67 index	.04	0.238	—	—	—
Mitotic count	Ki-67 index	<.0001	0.434	c-KIT status	<.0001	0.370
Mitotic count	KITr pattern	<.0001	0.328	—	—	—
Ki-67 index	KITr pattern	.002	0.253	c-KIT status	<.0001	0.385
KITr pattern	c-KIT status	.001	0.275	—	—	—

Significant correlations were considered strong when they occurred in more than 49% of the study population (r _P > 0.7), moderate when they occurred in 9% to 49% (0.3 < r _P < 0.7), and weak when they were less than 9% of the population (r _P < 0.3). Clinical outcome: no recurrence or metastasis versus metastasis versus metastasis with tumor-related death. Recurrence history: primary tumor versus recurrence at site of prior excision. Extent: dermal versus subcutaneous versus ulcerated versus muscle invasion. Tumor location: trunk versus limbs versus head versus scrotum versus perineal versus mucosa. Architectural pattern: microscopic arrangement in cords versus sheets. P, significance level; r _P, Pearson’s coefficient.

Among histological criteria, clinical outcome (ie, MCT metastasis or related death) was correlated to Patnaik grade (P < .0001, r _P = –0.380) and Kiupel grade (P < .0001, r _P = 0.500), predominant architectural pattern (ie, cords vs sheets) of neoplastic cells (P < .0001, r _P = –0.452), and mitotic count (P < .0001, r _P = –0.325). However, in this study, there was no correlation of clinical outcome with presence of neoplastic cells at the surgical margin, tumor extent (ie, subcutaneous, dermal, ulcerated, or within the muscle), or the presence or intensity of edema, necrosis, hemorrhage, collagenolysis, cystic change of apocrine glands, desmoplasia, eosinophilic infiltrate, and number of binucleated cells. Histological features are shown in Figs. 1 and 2.

Figures 1–2. Mast cell tumor (MCT), dog, skin. Hematoxylin and eosin. Figure 1. Kiupel low grade MCT with evident cell granulation and desmoplasia, moderate anisocytosis, and anisokaryosis. Figure 2. Kiupel high grade MCT with a binucleated cell (arrow). Figures. 3–4. Mast cell tumor (MCT), dog, skin, IHC for Ki-67 (MIB-1). Figure 3. Ki-67 labeling in 3% of nuclei. Figure 4. Ki-67 labeling in 45% of nuclei.

On immunohistochemical evaluation, a correlation with clinical outcome was obtained for Ki-67 labeling index (P < .0001, r _P = –0.414) and KITr pattern (P = .01, r _P = –0.207) (Figs. 3–7).

Figures. 5–7. Mast cell tumor, dog, skin. Immunohistochemistry for KITr. Figure 5. KIT pattern I, with membranous staining. Figure 6. KIT pattern II, with focal cytoplasmic staining. Figure 7. KIT pattern III, with diffuse cytoplasmic staining in more than 10% of the cells.

Internal tandem duplications in exon 11 of the c-KIT oncogene was found by PCR in 10.7% (16/149) of dogs in this study, and it was associated with a higher risk of MCTs metastasis or related-death (P < .0001, r _P = –0.356). Only 2 patients (1.3%) presented with deletions in the PCR amplicon, and both of these had confirmed lymph node metastasis and died within 30 and 134 days of treatment.

The ROC curve established a cut-point for the mitotic count of 2 mitotic figures in 10 HPF (AUC = 0.8163; 95% CI, 0.741-0.891; OR = 1.944; sensitivity = 84%; specificity = 56%) and for the Ki-67 labeling index of 5.6% Ki-67-positive cells in 500 neoplastic cells (AUC = 0.786; 95% CI, 0.706-0.867; OR = 1.777; sensitivity = 92%; specificity = 42%).

Survival Time and Tumor-Related Death

The mean follow-up time for the 149 patients was 358 days, and the median overall survival was 1020 days.

Clinical staging, as proposed by the World Health Organization,³² resulted in 64.4% (96/149), 12.8% (19/149), 21.5% (32/149), and 1.3% (2/149) of patients in stages I, II, III, and IV, respectively. Median survival was not reached for stage I; median survival was 203, 164, and 15 days for stages II, III, and IV, respectively, and these were significantly different (P < .0001), although there was no difference between stages II and III (Fig. 8). An amendment to the WHO staging system (Table 2) was done to divide stage III into III.1 (new stage II; multiple tumors with no regional lymph node involvement; n = 18; OS = 381 days) and III.2 (new stage IV; large/infiltrative or multiple tumors with regional lymph node involvement; n = 14; OS = 90 days). The OS was significantly different between all stages (P < .0001) and was significantly higher in stage III.1 compared to stages II (new stage III) or III.2 (Fig. 9).

Figures. 8–13. Kaplan-Meier survival curves for dogs with mast cell tumors. Figure 8. Clinical staging according to the World Health Organization system (n = 149). The median survival time was not reached in stage I but was 203, 164, and 15 days for stages II, III, and IV, respectively (P < .0001, although there was no significant difference among stages II and III). Figure. 9. Amendment to the World Health Organization system of clinical staging (n = 149). The median survival time was not reached in stage I but was 381, 203, 90, and 15 days for stages II (III.1: division of the previous stage III compromising dogs with multiple mast cell tumors but without lymph node involvement), III (previous stage II), IV (III.2: division of the previous stage III compromising dogs with large and infiltrative mast cell tumors or multiples tumors with regional lymph node involvement), and V (previous stage IV), respectively (P < .0001). Figure. 10. Primary or recurrent mast cell tumor (MCT) (n = 149). The median survival time was not reached in patients with primary MCT (ie, the initial clinical presentation), but it was 131 days in dogs with a history of tumor recurrence (P < .0001). Figure. 11. Kiupel grade (n = 149). The median survival time was not reached for low-grade tumors but was 213 days for high-grade tumors (P < .0001). Figure. 12. Mutational status of exon 11 of the c-KIT oncogene (n = 147). The median survival time was reached only for cases with internal tandem duplications on exon 11 (n = 16; median of 164 days; P < .0001). Patients with deletions (n = 2) were not included. Figure. 13. Risk of tumor recurrence or metastasis defined through multivariate analysis of prognostic factors (see Table 3). Low risk (n = 27), intermediate risk (n = 64), and high risk (n = 58). The median survival time was reached only for the high-risk group (213 days; P < .0001).

Table 2.

Proposed Amendment to the WHO System of Clinical Staging of Cutaneous and Subcutaneous Mast Cell Tumors in Dogs.

Stage	Description
I	Single tumor, without regional lymph node involvement
II	Multiple tumors (≥3), without regional lymph node involvement
III	Single tumor, with regional lymph node involvement
IV	Large and infiltrative tumors, without delimitation, or multiple tumors (≥3), with regional lymph node involvement
V	Any tumor with distant metastasis, including bone marrow invasion and the presence of mast cells in the peripheral blood

Previous WHO stage II became stage III. Previous WHO stage III was divided in III.1 (current stage II) and III.2 (current stage IV). Previous WHO stage IV became stage V.

A history of tumor recurrence (ie, the dog presented with a recurrent tumor rather than a primary tumor) was present in 30 of 149 (20.1%) cases. MCT-related death occurred in 24 of 30 (80.0%) dogs with recurrent MCT (median survival of 131 days), compared to 22 of 119 (18.5%) of dogs with initial (nonrecurrent) MCT (median survival was not reached for this group) (P < .0001; Fig. 10).

For the Patnaik grading system, grade 1, 2, and 3 tumors were found in 4.0% (6/149), 79.9% (119/149), and 16.1% (24/149) of cases. Tumor-related mortality occurred in 0% (0/6) of grade 1, 25% (30/119) of grade 2, and 67% (16/24) of grade 3 cases. Median survival of 203 days was reached only for grade 3 cases, differing from low and intermediate grade tumors (P < .0001).

For the Kiupel grading system, high-grade MCT were found in 32.9% (49/149) and low-grade found in 67.1% (100/149) of cases. Tumor-related mortality occurred in 61% (30/49) of high grade and 16% (16/100) of low grade MCT. A median survival of 213 days was reached only for high-grade cases (P < .0001; Fig. 11).

Using the Patnaik/Kiupel grading systems together resulted in 6 grade 1/low grade, 94 grade 2/low grade, 25 grade 2/high grade, and 24 grade 3/high grade, and tumor-related mortality occurred in 0%, 17% (16/94), 56% (14/25), and 67% (16/24), respectively. Median OS was not reached for grade 1/low grade and grade 2/low grade, and these were not significantly different. There was also no difference in OS between grade 2/high grade (OS = 226 days) and grade 3/high grade (OS = 203 days). Nevertheless, low grade differed from high grade (P < .0001).

There were no MCT-related deaths in cases with a diffuse cord-like achitectural pattern, and the median survival was therefore not reached. In contrast, cases with sheet-like architectural pattern reached the median survival at 337 days (P < .0001).

Using the cut-points defined by the ROC curves, the median survival was reached only for patients whose tumors had values that exceeded the cut-point: 260 days for cases with more than 2 mitotic figures in 10 hpf (P < .0001) and 524 days for Ki-67 labeling index of more than 5.6% (P < .0001). Two samples were excluded from the survival analysis because they lacked immunolabeling for Ki-67 even in the internal positive control (epithelial cells).

Regarding KITr expression, the median survival was not reached in the group with a membranous KIT pattern (KIT I), but it was 1020 and 224 days for cytoplasmic focal (KIT II) or diffuse (KIT III) expression, respectively. However, statistical significance was found only between KIT I and KIT III (P = .008).

In a survival analysis, excluding those with deletions in the c-KIT amplicon (n = 2), the median survival was not reached for patients without mutations (n = 131), but it was 164 days for those with internal tandem duplications in exon 11 of the c-KIT oncogene (n = 16; P < .0001; Fig. 12). MCT related-deaths occurred in 69% (11/16) of patients with such mutations.

In the factor analysis, 11 variables failed to show an association with clinical outcome (P values above .6), including tumor extent (.67), necrosis (086), colagenolysis (.72), desmoplasia (.74), cystomatosis (.75), inflammatory infiltrate (.77), surgical margins (.89), and patient age (.83), Patnaik grade (.62), Ki-67 labeling index (.75), and KITr pattern (.66) although the last three were correlated to clinical outcome in an univariate analysis. The multivariate analysis with Cox proportional hazard model demonstrated that only 2 of the variables were independent predictors of OS: both WHO stage and the proposed amendment of the WHO clinical staging system (hazard ratio [95% CI]: 1.824 [1.210-4.481]; P = .01) and the history of tumor recurrence (hazard ratio [95% CI]: 9.250 [2.158-23.268]; P < .001).

From the analysis of correlations, OS, and multivariate analysis, a model of the risk of recurrence or metastasis was developed for canine MCT (Table 3). In this model, dogs diagnosed with MCT were assigned to the high-risk group if they presented with advanced stage disease (stage II, III.2, and IV or III, IV and V, according to the proposed amended of the WHO staging system) and/or a history of tumor recurrence. Cases were assigned to the intermediate risk group if they had no history of tumor recurrence, WHO stage I or II (according to the proposed amended of the WHO staging system shown in Table 2), plus any one of: combined Patnaik/Kiupel grade 2/high or grade 3/high, sheet-like architectural pattern, mitotic count higher than 2 in 10 HPF, Ki-67 labeling index above 5.6%, KIT pattern III, or internal tandem duplications in exon 11 of the c-KIT oncogene. In this study, none of these parameters were independent even if patients with recurrent tumors and advanced stage disease were withdrawn. Cases were assigned to the low-risk group if they had none of the prognostic factors listed in Table 3. Of the 149 cases, 58 were in the high-risk group; these had a median survival time of 213 days, and 36 of 58 (62%) had MCT-related deaths. The intermediate risk group included 64 cases, and 9 of 64 (14%) had MCT-related deaths. The low-risk group included 27 cases, and 1 of 27 (4%) had MCT-related death. Median survival was not reached in low- or intermediate-risk groups but differed from that of the high-risk group (P < .0001, Fig. 13).

Table 3.

Negative Prognostic Features Used to Establish the Risk of Metastasis and Tumor-Related Death in Dogs With Cutaneous or Subcutaneous Mast Cell Tumor.

Clinical Features	Histological Features
Clinical stage III, IV. and V (amendment of the World Health Organization clinical staging system)	Architectural pattern (sheets rather than cords)
History of tumor recurrence (ie, the dog presented with a recurrent tumor rather than a primary tumor)	Mitotic count (cut-point ≥2 mitotic figures in 10 high-power fields) Patnaik/Kiupel grade 2/high or grade III/high
Immunohistochemical features	Molecular features
Ki-67 index (cut-point ≥5.6%) KITr pattern III	Internal tandem duplications in exon 11 of the c-KIT oncogene

High-risk patient: stage III disease or higher (using the amendment to the WHO system shown in Table 2) and/or recurrent mast cell tumors. Intermediate-risk patient: stage I or II disease (using the amendment to the WHO system shown in Table 2) and primary tumor (no history of recurrence) but with at least 1 other negative prognostic features listed. Low-risk patient: none of the negative prognostic features listed.

Discussion

An extensive list of prognostic factors was used in this study, allowing the definition of the risk of recurrence or metastasis in canine MCT. Numerous correlations were present, but only a few of them were related to clinical outcome. Other correlations were also identified. Mitotic count was correlated to Ki-67 labeling index, as expected,^3,36 but also to the number of binucleated cells and presence/intensity of intratumoral necrosis. Mutations in the c-KIT oncogene were positively correlated to clinical stage, history of tumor recurrence, Patnaik and Kiupel grading systems, mitotic count, and Ki-67 labeling index. After a multivariate analysis, only the WHO staging and proposed amended system and history of tumor recurrence were independent predictors of poor outcome in canine MCT.

Clinical staging, as proposed by the World Health Organization,³² allows evaluation of the extent of the disease, and it has been recommended for canine MCT once diagnosis is established.^4,26 Clinical staging is directly related to prognosis of canine MCT^10,26,52 as observed in this study. However, the true influence of stages II and III is long debated in the literature, and it is hard to define. This is related to: (1) difficulty in the diagnosis of lymph node metastases and (2) uncertainty of the importance of multiple lesions and multicentric presentations, especially when there is no concomitant involvement of regional lymph nodes.^10,26,52 Here we propose that stage III is divided into III.1 in the absence of lymph node involvement and III.2 when there is a confirmed metastasis in the lymph node. In this study, a better outcome was observed in dogs in stage III.1 compared to stage II. Therefore, we propose an amendment of the canine MCT staging system, based on 5 tiers, which may improve the definition of clinical stages and their progression. Although the WHO staging was also a significant independent variable associated with clinical outcome, the amended staging system better predicted the outcome of dogs with MCT.

The relevance of canine MCT staging is often questioned, but its importance was clearly demonstrated in this study. Nonetheless, the probability of metastasis may be extremely low in some patients,⁴⁶ and a complete stating is rarely necessary.⁴ Considering this, a protocol for staging is proposed in 3 levels:

Simple/basic staging: detailed physical examination searching for new cutaneous lesions but also for regional lymph node involvement, following the lymphosome territory proposed for dogs.⁴¹ This should include cytological evaluation and/or histological assessment through an incisional or excisional biopsy. This may be performed during an incisional biopsy of the primary tumor itself or during the definitive excision. Cytological evaluation of lymph nodes for dogs presenting with MCT is limited as up to 24% of healthy dogs may have an increased population of normal mast cells in the lymph node,⁵ but there are cytological criteria available that can be used.²² A histological classification of the level of involvement of the regional lymph node in dogs with MCT may further improve clinical staging and refine clinical and therapeutic decisions.⁵¹ If a lymph node is confirmed to have metastasis (high-risk patient) or if the primary tumor is located in a site with potential drainage to internal lymph nodes, complete staging is recommended.

Complete staging: besides the basic staging, complete staging should include imaging, especially abdominal ultrasound (for perineal MCT) and thoracic radiographs (for neck and trunk MCT),^4,46 but also computed tomography if the previous images suggest possible metastatic lesions. In these cases, an extended staging should be performed.⁴

Extended staging: cytological evaluation of suspected metastatic lesions, performed through a fine needle aspiration (FNA), guided by ultrasound or computed tomography. Additionally, a bone marrow assessment may be recommended, although there is also a lack of criteria for diagnosing bone marrow metastasis, and such involvement is rare in canine MCT and usually associated with severe and terminal disease.^10,26,47,52

Recurrent canine MCT are associated with a poor prognosis, with related death rates dramatically reaching 86% to 100% of cases, as reported in previous studies.^6,33 Although recent studies in dogs with recurrent MCT have demonstrated effective disease control with surgery and adjuvant therapies,⁴⁴ mortality rates remain high. Tumor-related mortality occurred in 80% of patients with recurrent MCT in this study, even though adjuvant treatments were recommended and performed in most cases.^6,33

In this study, in contrast to the literature, differences in the clinical outcome were not associated with breed and tumor site. Probably due to epidemiological differences in the distribution of canine breeds, we did not observe a greater frequency of MCT in breeds presenting the Bulldog as a common ancestral.^26,28 The wide variation of breeds in this study compromised evaluation of overall survival according to many breeds, and additional studies are necessary for defining such association.

Although tumor localization was not associated with clinical outcome in this study, it may affect the prognosis of some patients.^7,38,42 A poor prognosis is reported for canine MCT located in the mucosa, muco-cutaneous junctions, prepuce, scrotum, vulva, and perineum; however, clinical outcomes similar to other locations may be obtained for cases with tumors at these sites if prognosis and treatment are properly evaluated and applied.^7,38 It is reported that canine MCT located on the muzzle may also have high probability of metastasis, but they may also have a good outcome under proper treatment.¹⁹ The only muzzle MCT included in this study presented with an inoperable lesion and confirmed lymph node metastasis. It was a high-grade tumor with an internal tandem duplicaiton in exon 11 of the c-KIT oncogene, evaluated from an incisional biopsy. Treatment with TKI in this patient resulted in complete remission and prolonged disease-free interval (240 days), and the patient was still alive at the end of this study.

In this study, MCT metastasis and/or related deaths occurred in 40.7% of patients with compromised or exiguous surgical margins and 16% of those with tumor-free margins, regardless of treatment; however, no significant correlation was found. It is suggested that evidence of tumor-free histologic margins was not correlated with DFI or OS for canine MCT, and such information should not indicate a worse outcome.^29,30 However, additional studies are needed to verify the true impact of histological margins on the prognosis of canine MCT.

The Patnaik grading system³³ has been criticized for giving too much importance to extent of the tumor and including subjective criteria that result in the inclusion of more than 40% of MCT in an intermediate grade (grade 2), gathering a wide and diverse group of lesions with distinct biological behavior.¹⁶ In fact, 80.5% of MCT in this study were classified as a grade 2. Furthermore, tumoral extent (ie, restricted to dermis or subcutaneous tissue, skin ulceration within deeper layers including muscle) was not correlated with the clinical outcome for patients in this study. The Kiupel 2-tier grading system²⁰ reduces the subjectivity of the Patnaik grading system and has shown to be highly reproducible by different pathologists.^33,36 A study of 162 cases demonstrated superiority of the Kiupel grading system. Furthermore, in the present study, the group with combined Patnaik/Kiupel grade 2/low grade MCT did not reach the median survival, while the group with grade 2/high grade MCT reached a median OS of 698 days (P = .001). However, there was also a difference between grade 2/high grade and grade 3 MCT/high grade (P = .027).³⁶ Prognostic superiority of this system was confirmed in this study, and Patnaik grade failed as a relevant prognostic tool in the factor analysis. A Kiupel high grade is sufficient for inclusion of dogs with MCT in an intermediate-risk group, whatever the Patnaik grade. Nevertheless, 19 of 49 (39%) of patients with high-grade tumors were still alive at the end of this study, and there were 16 of 100 (16%) patients with low-grade tumors that sucumbed to the disease, suggesting that other prognostic factors must be used along with the Kiupel grading system.

Among histological features, the architectural pattern (microscopic arrangement as cords vs sheets) and the mitotic count were also associated with the clinical outcome in this study. These data are included (with variable importance) in the Patnaik³³ and Kiupel grading systems,²⁰ but they should be evaluated separately. The mitotic count evaluated in 10 HPF is probably one of the best prognostic factors for grade I and II MCT in early stages of disease at first diagnosis (nonrecurrent MCT).^11,35,45 Longer survival rates are reported for patients with MCT that have less than 5 mitotic figures in 10 HPF.³⁵ Nevertheless, a cut-point of 5 was associated with a high specificity but low sensitivity, with many patients succumbing to MCT despite of a mitotic count lower than 5.^11,45 A system with 2 cut-points of mitotic counts was proposed: 0, associated with low risk; 1–7, associated with intermediate risk; and >7, associated with high risk.¹¹ In fact, in this study, a low sensitivity for the cut-point of 5 was observed (aproximately 50%; OR = 6045), and therefore a new cut-point of 2 is proposed to have a higher sensitivity, as reported by other authors.^11,45 We propose that MCT with a mitotic count ≥2 should be considered at greater risk of adverse outcome.

Ki-67 labeling index was moderately related to the mitotic count for canine MCT, and both have a similar association with tumor behavior.³ Ki-67 protein (recognized by the monoclonal antibody MIB-1) is expressed in every phase of the cell cycle except G0.³⁷ Different cut-points were proposed in the literature, but the 1.8% value is the most accepted one for Patnaik grade 2 MCT.²⁷ In the current study, a cut-point of 5.6% was associated with the risk of recurrence or metastasis, with a sensitivity of 92%. This cut-point approaches that established by another study, which reported a mortality rate of 14.3% and 55.6% in dogs with a KI-67 labeling index lower or higher than 7%, respectively.⁴⁰ It is the authors’ experience that patients with MCT and Ki-67 labeling indices above 1.8% often have long-term control of the disease, even in the absence of adjuvant therapies. In this study, a 1.8% cut-off had high sensitivity (98%) but poor specificity (10%), with an OR of 1.088. It must be noted that the cut-points used in this study have high specificity but low sensitivity and must not be used as a separate prognostic factor.

A decreased OS in patients with KIT pattern III was also observed, confirming its relationship to the prognosis of canine MCT.^14,21,48,49 However, among relevant prognostic factors, it might be considered a less informative one since in this study, differences were found only between patients with a membranous pattern (KIT I) and those with a cytoplasmic diffuse pattern (KIT III).^8,13 Recent studies^8,13 also demonstrated a weak association between expression of KITr and prognosis. In this study, Ki-67 and KITr failed as relevant prognostic features in the factor analysis. Immunohistochemistry is a complementary technique to histology and a solid partnership between the clinical oncologist and pathologist is necessary for better results.

Furthermore, gain of function mutations in the c-KIT oncogene, especially the internal tandem duplications in exon 11, had a significant impact on the outcome of dogs with MCT and may even drive treatment toward the use of TKIs. These are in fact the most frequent mutations in this oncogene in canine MCT and lead to constitutive activation of KITr.^49,50 However, the overall incidence of such mutations is low in canine MCT (approximately 10% in this study), especially if low-risk MCT are also considered.^{2,43,49,50,53} Although mutations may occur in patients with aberrant KITr expression, those are separate events, and 80% of patients with abnormalities in KITr expression did not have mutations in this locus.⁴⁹ In a study performed with dogs with internal tandem duplications (n = 8), a DFI of 130 days was reached, while those without mutations (n = 39) had a DFI of 345 days.⁴³ Nonetheless, screening for mutations may not be done in all cases of canine MCT, particularly if resources are limited and there are no prognostic features to indicate an intermediate or high risk of recurrence or metastasis. Although it was not included in this study, screening for exons 8 and 9 mutations has been recommended for high- to intermediate-risk MCT as those mutations may confer gain of function by increasing the affinity of the extracellular domain to the stem cell factor. Nevertheless, the prevalence of such mutations in exons 8 and 9 are around 4.7% and 6.4% for exon 8 and 14.9% for exon 9, and their impact on prognosis and treatment is still not well established.^24,43

Despite the wide range of treatments, comparison of different treatments was not a goal of this study. Considering the clinical experience and information about treatment and prognosis of canine MCT,²⁶ the best available treatment was ethically offered for each case.

This study allowed a classification of dogs with MCT according to the risk of recurrence or metastasis. An amendment of the WHO clinical staging system can be used, considering that patients with multiple lesions without regional lymph node involvement (III.1, new stage II) usually have a better prognosis than those within the previous stage II, meaning single lesion with lymph node involvement (now stage III). Patients with large and infiltrative MCT or multiple MCT with involvement of the lymph node (III.2) remain with a worse prognosis and therefore are included in stage IV. Therefore, previous stage IV becomes now stage V. Clinical staging and history of tumor recurrence proved to be independent factors in this study, and a high risk may be assumed for patients presenting with advanced stage disease (II, III.2, and IV) or a history of tumor recurence, while an intermediate risk may be assigned for patients whose tumors present a primary tumor (no recurrence) in stage I or III.1 and at least 1 of a Patnaik/Kiupel grade 2/high grade or grade 3/high grade, mitotic count above 2, Ki-67 labeling index above 5.6%, KIT pattern III, or internal tandem duplications in c-KIT exon 11. Although the presence of exiguous or compromised surgical margins was not related to clinical outcome, controlled studies with precise evaluation of margins are needed to verify its true impact on the prognosis of MCT in dogs. Other studies have suggested that a worse prognosis may be associated with certain breeds (ie, Shar-pei) and tumor sites (mucosa, scrotum, prepuce, perineum, and vulva), but these require further study. These findings may be of value in preventing tumor recurrence by directing patients with intermediate to high risk to receive adjuvant therapies according to the oncologist’s experience and clinical condition of each patient.

Footnotes

Supplementary material for this article is available online.

Authors’ Note

Institution where the work was conducted: Universidade Federal de Minas Gerais (RH, GL, LM, GC, RA)

Acknowledegments

We thank FEPMVZ (Fundação de Estudo e Pesquisa em Medicina Veterinária e Zootecnia; Belo Horizonte, MG – Brazil), Vetpat Laboratory and Progen Biotecnologia (São Paulo, SP – Brazil).

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Abadie

Amardeilh

Delverdier

. Immunohistochemical detection of proliferating cell nuclear antigen and Ki-67 in mast cell tumors from dogs. J Am Vet Med Assoc. 1999;215(11):1629–1634.

Avery

. Molecular diagnostics of hematologic malignancies in small animals. Vet Clin North Am Small Anim Pract. 2012;42(1):97–110.

Berlato

Murphy

Monti

. Comparison of mitotic index and ki67 index in the prognostication of canine cutaneous mast cell tumours. Vet Comp Oncol. 2015;13(2):143–150.

Blackwood

Murphy

Buracco

. European consensus document on mast cell tumoursin dogs and cats. Vet Comp Oncol. 2012;10(3):e1–e29.

Bookbinder

Butt

Harvey

. Determination of the number of mast cells in lymph node, bone marrow, and buffy coat cytologic specimens from dogs. J Am Vet Med Assoc. 1992;200(11):1648–1650.

Bostock

. The prognosis following surgical removal of mastocytomas in dogs. J Small Anim Pract. 1973;14(1):27–41.

Cahalane

Payne

Barber

. Prognostic factors for survival of dogs with inguinal and perineal mast cell tumors treated surgically with or without adjunctive treatment: 68 cases (1994–2002). J Am Vet Med Assoc. 2004;225(3):401–408.

Costa Casagrande

de Oliveira Barros

Fukumasu

. The value of molecular expression of KIT and KIT ligand analysed using real-time polymerase chain reaction and immunohistochemistry as a prognostic indicator for canine cutaneous mast cell tumours. Vet Comp Oncol. 2015;13(1):1–10.

Dobson

Samuel

Milstein

. Canine neoplasia in the UK: estimates of incidence rates from a population of insured dogs. J Small Anim Pract. 2002;43(6):240–246.

10.

Dobson

Scase

. Advances in the diagnosis and management of cutaneous mast cell tumours in dogs. J Small Anim Pract. 2007;48(8):424–431.

11.

Elston

Sueiro

FAR

Cavalcanti

. Letter to the editor: the importance of the mitotic índex as a prognostic factor for survival of canine cutaneous mast cell tumors: a validation study. Vet Pathol. 2009;46(2):362–365.

12.

Fulcher

Ludwing

Bergman

. Evaluation of a two-centimeter lateral surgical margin for excision of grade I and grade II cutaneous mast cell tumors in dogs. J Am Vet Med Assoc. 2006;228(2):210–215.

13.

Giantin

Vascellari

, Morello EM, et al. c-KIT messenger RNA and protein expression and mutations in canine cutaneous mast cell tumors: correlations with post-surgical prognosis. J Vet Diagn Invest. 2012;24(1):116–126.

14.

Gil da Costa

. C-kit as a prognostic and therapeutic marker in canine cutaneous mast cell tumours: from laboratory to clinic. Vet J. 2015;205(1):5–10.

15.

Giuliano

Horta

Hoather

. Expression of fibroblast activating protein (FAP) and correlation with histological grade, mitotic index and Ki67 expression in canine mast cell tumours. J Comp Pathol. 2017;156(1):14–20.

16.

Gross

Ihrke

Walder

. Mast cell tumors. In: Gross

Ihrke

Walder

. Skin disease of the clinical and histopathologic diagnosis. 2nd ed. Oxford, UK: Blackwell Publishing Company; 2005:853–865.

17.

Hahn

Ogilvie

Rusk

. Masitinib is safe and effective for the treatment of canine mast cell tumors. J Vet Intern Med. 2008;22(6):1301–1309.

18.

Hauck

. Tumors of the skin and subcutaneous tissues. In: Withrow

Vail

Page

, eds. Withrow and Macewen’s Small Animal Clinical Oncology. 5th ed. Philadelphia, PA: Saunders; 2013:305–320.

19.

Hillman

Garrett

De Lorimier

. Biological behavior of oral and perioral mast cell tumors in dogs: 44 cases (1996–2006). J Am Vet Med Assoc. 2010;237(8):936–942.

20.

Kiupel

Webster

Bailey

. Proposal of 2-tier histologic gradind system for canine cutaneous mast cell tumors to more accurately predict biological behavior. Vet Pathol. 2011;48(1):147–55.

21.

Kiupel

Webster

Kaneene

. The use of KIT and Tryptase expression patterns as prognostic tools for canine cutaneous mast cell tumors. Vet Pathol. 2004;41(4):371–377.

22.

Krick

Billings

Shofer

. Cytological lymph node evaluation in dogs with mast cell tumours: association with grade and survival. Vet Comp Oncol. 2009;7(2):130–138.

23.

Leibman

Lana

Hansen

. Identification of matrix metalloproteinases in canine cutaneous mast cell tumors. J Vet Intern Med. 2000;14(6):583–586.

24.

Letard

Yang

Hanssens

. Gain-of-function mutations in the extracellular domain of kit are common in canine mast cell tumors. Mol Cancer Res. 2008;6(7):1137–1145.

25.

London

Malpas

Wood-Follis

. Multi-center, placebo-controlled, double-blind, randomized study of oral toceranib phosphate (SU11654), a receptor tirosine kinase inhibitor, for the treatment of dogs with recurrent (either local or distant) mast cell tumor following surgical excision. Clin Cancer Res. 2009;15(11):3856–3865.

26.

London

Thamm

. Mast cell tumors. In: Withrow

Vail

Page

, eds. Withrow and Macewen’s Small Animal Clinical Oncology. 5th ed. Philadelphia, PA: Saunders; 2013:335–355.

27.

Maglennon

Murphy

Adams

. Association of Ki67 index with prognosis for intermediate-grade canine cutaneous mast cell tumours. Vet Comp Oncol. 2008;6(4):268–274.

28.

Mcniel

Prink

O’Brien

. Evaluation of risk and clinicaloutcome of mast cell tumours in pug dogs. Vet Comp Oncol. 2004;4(1):2–8.

29.

Michels

Knapp

Denicola

. Prognosis following surgical excision of canine cutaneous mast cell tumors with histopathologically tumor-free versus nontumor-free margins: a retrospective study of 31 cases. J Am Anim Hosp Assoc. 2002;38(5):458–466.

30.

Miller

Van Lelyveld

Warlan

. A retrospective review of treatment and response of high-risk mast cell tumours in dogs. Vet Comp Oncol. 2016;14(4):361–370.

31.

O’connell

Thomson

. Evaluation of prognostic indicators in dogs with multiple, simultaneously occurring cutaneous mast cell tumours, 63 cases. Vet Comp Oncol. 2013;11(1):51–62.

32.

Owen

. TNM Classification of Tumours in Domestic Animals. Geneva, Switzerland: World Health Organization; 1980:53.

33.

Patnaik

Ehler

Macewen

. Canine cutaneous mast cell tumor: morphologic grading and survival time in 83 dogs. Vet Pathol. 1984;21(5):268–274.

34.

Rogers

. Mast cell tumors. Vet Clin North Am Small Anim Pract. 1996;6(1):87–101.

35.

Romansik

Reilly

Kass

. Mitotic index is predictive for survival for canine cutaneous mast cell tumors. Vet Pathol. 2007;44(3):335–341.

36.

Sbattini

Scarpa

Berlato

. Histologic grading of canine mast cell tumor: is 2 better than 3? Vet Pathol. 2015;52(1):70–73.

37.

Scase

Edwards

Miller

. Canine mast cell tumors: correlation of apoptosis and proliferation markers with prognosis. J Vet Int Med. 2006;20(1):151–158.

38.

Sfiligoi

Rassnick

Scarlett

. Outcome of dogs with mast cell tumors in the inguinal or perineal region versus other cutaneous locations: 124 cases (1990–2001). J Am Vet Med Assoc. 2005;226(8):1368–1374.

39.

Stanclift

Gilson

. Evaluation of neoadjuvant prednisolone administration and surgical excision in treatment of cutaneous mast cell tumors in dogs. J Am Vet Med Assoc. 2008;232(1):53–62.

40.

Strefezzi

Kleeb

Xavier

. Avaliação da proliferação celular como indicador prognóstico para mastocitomas cutâneos caninos. Pesquisa Vet Brasil. 2010;30(7):559–565.

41.

Suami

Yamashita

Soto-Miranda

. Lymphatic territories (lymphosomes) in a canine: an animal model for investigation of postoperative lymphatic alterations. PLoS One. 2013;8(7):e69222.

42.

Takahashi

Kadosawa

Nagase

. Inhibitory effects of glucocorticoids on proliferation of canine mast cell tumor. J Vet Med Sci. 1997;59(11):995–1001.

43.

Takeuchi

Fujino

Watanabe

. Validation of the prognostic value of histopathological grading or c-kit mutation in canine cutaneous mast cell tumours: a retrospective study. Vet J. 2013;196(3):492–498.

44.

Thamm

Mauldin

Vail

. Prednisone and vinblastinechemotherapy for canine mast cell tumor—41 cases (1992–1997). J Vet Intern Med. 1999;13(5):491–497.

45.

Van Lelyveld

Warlan

Miller

. Comparison between ki-67 index and mitotic index for predicting outcome in canine mast cell tumours. J Small Anim Pract. 2015;56(5):312–319.

46.

Warland

Amores-Fuster

Newbury

. The utility of staging in canine mast cell tumours. Vet Comp Oncol. 2014;12(4):287–298.

47.

Warland

Brioschi

Owen

. Canine mast cell tumours: decision-making and treatment. In Pract. 2015;37(7):315–332.

48.

Webster

Kiupel

Yuzbasiyan-Gurkan

. Evaluation of the kinase domain of c-KIT in canine cutaneous mast cell tumors. BMC Cancer. 2006;6(85):1–8.

49.

Webster

Yuzbasiyan-Gurkan

Kaneene

. The role of c-KIT in tumorigenesis: evaluation in canine cutaneous mast cell tumours. Neoplasia. 2006;8(2):104–111.

50.

Webster

Yuzbasiyan-Gurkan

Thamm

. Evaluation of prognostic markers for canine mast cell tumors treated with vinblastine and prednisone. BMC Vet Res. 2008;4(32):1–8.

51.

Weishaar

Thamm

Worley

. Correlation of nodal mast cells with clinical outcome in dogs with mast cell tumour and a proposed classification system for the evaluation of node metastasis. J Comp Pathol. 2014;151(4):329–338.

52.

Welle

Bley

Howard

. Canine mast cell tumours: a review of the pathogenesis, clinical features, pathology and treatment. Vet Dermatol. 2008;19(6):321–339.

53.

Zemke

Yamini

Yuzbasiyan-Gurkan

. Mutation in the juxtamembrane domain of c-KIT are associated with higher grade mast cell tumors in dogs. Vet Pathol. 2002;39(5):529–535.

Supplementary Material

Please find the following supplemental material available below.

For Open Access articles published under a Creative Commons License, all supplemental material carries the same license as the article it is associated with.

For non-Open Access articles published, all supplemental material carries a non-exclusive license, and permission requests for re-use of supplemental material or any part of supplemental material shall be sent directly to the copyright owner as specified in the copyright notice associated with the article.

0.00 MB

0.49 MB