Current Management Recommendations for Canine Mast Cell Tumors

Margaret C. McEntee, DVM, DACVIM (oncology), DACVR (radiation oncology)
Cornell University


Canine mast cell tumors are frequently encountered in clinical practice. Information available on the treatment of mast cell tumors has focused on the utility of surgery and radiation therapy. New information is becoming available on the use of chemotherapy. There has also been a recent revision of the recommendations for staging of dogs with mast cell tumors. The following will provide a review and update on canine mast cell tumors.

Biological Behavior

A number of prognostic indicators (Table 1) have been identified. The histologic grading system provides useful information for predicting behavior. However, grade has to be considered in light of tumor location and other factors to more accurately prognosticate. Tumors located in the preputial, inguinal, and perineal regions behave more aggressively and site can be more important than grade. A clinical history of rapid or infiltrative growth typically indicates a more malignant tumor, whereas tumors that have been present for a long time have been associated with a better prognosis. Other prognostic indicators include DNA ploidy, AgNOR's (agyrophilic nucleolar organizing region), and PCNA (proliferating cell nuclear antigen). The role of c-kit mutations in the development of mast cell tumors is being investigated. The proto-oncogene c-kit encodes a transmembrane receptor with tyrosine kinase activity (kit) and plays an important role in development and function of normal mast cells. C-kit is a receptor for stem cell factor (SCF). Point mutations and deletions can lead to constitutive activation of the receptor in the absence of the ligand (SCF). Mutations of c-kit have been identified in canine mast cell tumors, and may contribute to the development and/or progression of mast cell tumors. Mutations appear to occur in intermediate- or high-grade tumors. Metastasis typically occurs to the regional lymph node(s) first and then can involve the liver, spleen, and bone marrow. Mast cell tumors rarely metastasize to the lung.

TABLE 1 : Prognostic Factors for Canine Mast Cell Tumors

Prognostic Factor

Relationship to behavior

Histopathologic grade

Higher grade associated with more aggressive behavior.

Clinical stage

Stage 0 or 1 with localized disease has a better prognosis.


Preputial, inguinal, perineal, subungual, and mucocutaneous sites associated with a poorer prognosis. Visceral/bone marrow disease associated with grave prognosis.

Conjunctival site is associated with a good prognosis.

Growth rate

Present > 7 months without any significant growth associated with a better prognosis; rapid growth is associated with a poor prognosis and increased likelihood of metastasis.


Boxers tend to have well differentiated mast cells and have a better prognosis.

Systemic signs

Systemic signs are associated with more aggressive disease and a poorer prognosis.

DNA ploidy

Aneuploid tumors show trend towards shorter survival and higher stage

AgNOR count

Higher AgNOR counts are associated with a poorer prognosis


PCNA counts are higher in tumors that metastasize


Recurrence after surgery may be associated with a poorer prognosis

c-KIT mutations

Appear to occur more commonly in higher grade tumors

Clinical Presentation

The most common presentation in the dog is a solitary dermal or subcutaneous mass. Dermal masses are usually well-defined, raised, the majority are < 3 cm in diameter, can be hairless, erythematous, may be ulcerated, although they can occasionally present as diffuse, edematous thickening of the dermis/epidermis. Identification of subcutaneous mast cell tumors can be more problematic as they are typically soft, ill-defined masses that feel like lipomas. It is important to perform aspiration cytology on any mass that palpates like a lipoma, as it is possible that it may be a mast cell tumor. Approximately 50% of the dermal/subcutaneous mast cell tumors are located on the trunk and perineum, and 40% are located on the extremities. Multiple mast cell tumors are identified at initial presentation in 14% of dogs. Other forms of mast cell tumors are much less common. If there is visceral involvement (spleen, liver, kidney) it is usually metastatic disease. Other primary sites of involvement that have been sporadically reported include: oral cavity, nasopharynx, larynx, trachea, mediastinum, intestine, and hepatopancreatic lymph nodes. Mast cell leukemia is rare.

Possible systemic signs of disease at presentation include: anorexia, vomiting, diarrhea, melena, and even hypotensive shock due to mast cell degranulation.


The first step in diagnosing a mast cell tumor is usually as simple as performing aspiration cytology. Diff-Quik may at times fail to stain mast cell granules, and giemsa stain is recommended. The next step is to perform an incisional or excisional biopsy for histopathology and tumor grading. Determination of tumor grade prior to definitive therapy may aid in the decision process regarding treatment and prognosis. The Patnaik system is used to grade mast cell tumors. Grade I (low grade) tumors have clearly defined cytoplasmic boundaries with regular, spherical, or ovoid nuclei, mitotic figures rare; cytoplasmic granules large, deep staining, and plentiful. Grade II (intermediate grade) tumors have cells closely packed with indistinct cytoplasmic boundaries; nucleus-to-cytoplasm ratio lower than that of anaplastic; mitotic figures infrequent; more granules than anaplastic. Grade III (high grade) tumors have highly cellular, undifferentiated plasmic boundaries, irregular size and shape of nuclei; frequent mitotic figures; low number of cytoplasmic granules. Tumor grading has classically required histopathologic examination of a biopsy sample. Another option is a silver staining technique to assess agyrophilic nuclear organizer regions (AgNOR's; loops of DNA in the nucleoli; sites of ribosomal RNA transcription), an indirect marker of proliferation. The assessment of AgNOR's can be done on either paraffin-embedded tissues or cytologic preparations and the counts have been shown to correlate with both histologic grade and post-surgical outcome in dogs with mast cell tumors. AgNOR frequency increases with increasing tumor grade.

Mast Cells in Normal Tissue

Aspiration cytology performed on a normal lymph node will reveal a few to several mast cells. Also, aspiration of a normal spleen will reveal a few mast cells. Mast cells are rare in a normal bone marrow aspirate and it has classically been stated that typically no mast cells are identified on a buffy coat examination. There are inflammatory diseases including parvovirus enteritis that have been associated with positive buffy coat smears. In a study by McManus on the frequency and severity of mastocytemia in dogs with and without mast cell tumors it was determined that the severity of mastocytemia in dogs without mast cell tumors was often greater than that with, and that random detection of mast cells in blood smears is usually not secondary to mast cell tumors. In normal tissue there would be only 0-1 mast cells/HPF (< 1% of cells). This increases to 1-5 mast cells/HPF (10% of cells) in hypersensitized tissue. Aspiration cytology of a mast cell tumor will typically reveal that at least 50% of the cells are mast cells.

Staging of Canine Mast Cell Tumors

Staging of mast cells should most importantly include aspiration cytology of the regional lymph node, and if necessary an incisional or excisional biopsy of the regional lymph node. If the aspirate is equivocal, i.e., a few mast cells are seen but they appear normal, then a lymph node biopsy should be performed. If not prior to definitive surgery on the primary tumor then at the time of surgery the regional lymph node should be evaluated. The vast majority of the time the first site of spread of mast cell tumors is to the draining lymph node. If the regional lymph node is negative then it is much less likely that positive results will be obtained on the remaining staging tests. Mast cell tumors can spread to the liver, spleen, bone marrow and blood as well as other sites. Comprehensive staging may include abdominal radiographs, abdominal ultrasound, ultrasound-guided aspiration of the spleen and/or liver, buffy coat smear, and bone marrow aspiration cytology. Specific recommendations for staging will be discussed. Thoracic radiographs are usually not indicated but may be done as part of the evaluation of a geriatric patient. Rarely pulmonary involvement is identified and it is more likely to present as a diffuse infiltrate and more rarely a nodular pattern. Thoracic radiographs may be useful for identifying intrathoracic lymph node involvement.



Surgery is the treatment of choice. It has been recommended that 3 cm surgical margins are important for complete resection, although this has been questioned as to the necessity of such a wide margin for all mast cell tumors. Pretreatment with an antihistamine is recommended prior to surgical manipulation. Surgery is typically curative for Grade I tumors and may be curative for dogs with Grade II tumors with localized disease that have a complete resection. Again, other prognostic factors such as tumor location and growth rate need to be factored into this equation to accurately prognosticate for an individual patient. If a tumor is nonresectable at presentation then neoadjuvant chemotherapy may effectively reduce the tumor in size such that surgery can then be performed. If it is determined that it was an incomplete resection of a Grade I or II tumor, then local radiation therapy is recommended. At times a second surgery may be attempted to effect local control, but particularly with extremity tumors this may not be a viable option. For dogs with metastatic, systemic or Grade III tumors chemotherapy is indicated.

In 1973 Bostock reported on the survival of 114 dogs with mast cell tumors treated with surgery alone. This original paper utilized a grading system wherein a low-grade tumor was designated as a Grade 3 tumor, and high-grade as a Grade 1 tumor. The grading scheme was modified by Dr. Patnaik in 1984 to the system that is in use today. In Bostock's report the mean survival by Grade was as follows: Grade 3 (n=39 dogs) = 51 weeks, Grade 2 (n=30 dogs) = 28 weeks, and Grade 1 (n=45 dogs) = 18 weeks. In this study dogs alive > 7 months post-operatively were apparently cured. Also, if the tumor had been present > 7 months pre-operatively the patient had a better prognosis.

Patnaik reported on 83 dogs with mast cell tumors treated surgically, and he reported his results in terms of the percentage of dogs alive greater than 1500 days. The results were: Grade I = 93% (28/30), Grade II = 47% (16/36), and for high-grade tumors Grade III = 6% (1/17). In 2000 at the Veterinary Cancer Society mid-year meeting an abstract was presented by Sequin et al on the results of 54 dogs with 59 Grade II mast cell tumors treated with surgery alone. Of the group 21 dogs had previous surgery and had a second surgical procedure performed due to incomplete resection at the time of the first surgery. The median tumor diameter was 2 cm with a range of 1-4 cm. The local recurrence rate was 5% with a median time to recurrence of 62 days (range = 51-252 days). Of interest, mast cell tumors developed at other locations in 11% of the dogs; at a median time of 240 days (range = 51-1480 days). It is not uncommon for a dog that has had one mast cell tumor to develop another mast cell tumor at another unrelated site. It is always important to inform clients of this possibility and have them check their dog on a regular basis. Any new masses that develop should be assessed due to the possibility of a new mast cell tumor.

At the same meeting Henderson et al reported on a contiguous series of canine mast cell tumors that were excised using the tumor diameter as the minimum surgical margin, as opposed to the standard 3 cm margin for all mast cell tumors regardless of tumor size. For example, a 0.5 cm tumor was excised using a measured 0.5 cm margin circumferentially and deep. To date the study has included 10 dogs (Grade I tumors in 8 dogs, Grade II tumors in 2 dogs) and clean margins have been obtained in all dogs. At this time this practice of margin determination has been adopted for all mast cell tumors at Auburn University, and with additional cases a 95% success rate has been realized. Additional study will be necessary to determine the appropriate extent of surgery for mast cell tumors.


Despite the fact that mast cell tumors are encountered frequently and chemotherapy has been used widely in the treatment of mast cell tumors, little information has found its way into the literature. L-asparaginase is the only drug that has been shown experimentally to have activity against mast cell tumors. Doxorubicin and cytoxan have been considered to be ineffective based on data from treatment of humans with mast cell disease. The following table (Table 2) outlines chemotherapy protocols that have been used or are being investigated.

TABLE 2: Chemotherapy Protocols





1) Prednisone 1 mg/kg PO SID

2) Prednisone 40 mg/m2 PO SID x 7 days, then 20 mg/m2 PO SID x 14 days, then 20 mg/m2 PO EOD x 14 days, then 10 mg/m2 PO EOD (6 months total*)

McCaw (1994)


Vincristine 0.75 mg/m2 IV weekly for 4 treatments

McCaw (1997)

Prednisone, vinblastine

Prednisone 2 mg/kg PO SID, with dosage tapered and discontinued over 3-6 months

Vinblastine 2 mg/m2 IV every 1-2 weeks; usually weekly for 4 treatments, then every other week for an additional 4 treatments

Thamm (1999)


CCNU (lomustine) 90 mg/m2 PO every three weeks; may need to reduce to 70 mg/m2

Rassnick (1999)

Vinblastine, CCNU

Vinblastine 2 mg/m2 IV on days 1 and 14

CCNU 60 mg/m2 PO on day 28

Cycle repeated starting on day 49

Current protocol in use at CUHA

A report by McCaw, et. al., on 25 dogs (23% Grade II, 77% Grade III) treated with prednisone alone showed a 20% overall response rate with one complete response (CR) and four partial responses (PR). The survival times for the 5 responders were 3, 5, 6, 7.5 and greater than 28 months. The same author reported on the use of vincristine alone in 27 dogs. The conclusion was that vincristine alone is ineffective in the treatment of measurable disease, and it resulted in an undesirable number of adverse reactions.

Thamm, et. al., reported on the response of 41 dogs to a combination of oral prednisone and intravenous vinblastine. Treatment was administered in an adjuvant setting in 23 dogs and 18 dogs had gross disease. Adverse effects were observed in 20% (8/41) usually after the first dose of vinblastine; side effects were mild in 6 dogs, and resulted in discontinuation of therapy in 2 dogs. The overall response rate in dogs with gross disease was 47% (7/15) with 5 CR's and 2 PR's with a median response duration of 154 days (range = 24 to >645 days). In the adjuvant setting there was a 57% 1- and 2-year disease-free rate. The median survival time for all dogs had not been reached by the time the report was done, with a median follow up of 573 days. Of interest, the median survival for dogs with Grade III tumors was 331 days.

Rassnick, et. al., reported on the use of CCNU (lomustine) in 23 dogs; cutaneous tumors (n=22), nasal cavity tumor (n=1). Tumor Grade was I (1), II (2), or III (8). Response was evaluable in 19 dogs with 8/19 (42%) having a measurable response. There was one CR; euthanized at 440 days for an unrelated problem. There were 7 PR's with a median duration of 77 days, and a range of 21-254 days. Stable disease was documented in 6 dogs; median duration of 78 days; range = 42-347 days. Dose reduction was necessary in 4 dogs due to neutropenia and fever (reduced from 90 to 70 mg/m2). It is important to remember that there is a potential for delayed and cumulative neutropenia and thrombocytopenia with CCNU as well as hepatotoxicity, and close monitoring with repeat CBC's and periodic chemistry panels is imperative.

Based on the apparent efficacy of vinblastine and CCNU in the treatment of canine mast cell tumors a protocol has been initiated at Cornell University that combines the two drugs (Table 2). Weekly complete blood cell counts should be done to monitor for bone marrow suppression. The current protocol entails reassessment of response after 2 cycles; with potentially a total of 4 cycles depending on response to therapy. Two cycles may be adequate in the adjuvant setting with Stage 0 disease. Weekly CBC's are done during the first 2 cycles to closely monitor for bone marrow suppression. Chemistry panels are repeated prior to each cycle.

Radiation Therapy

Radiation therapy is most commonly used postoperatively for incompletely resected Grade II tumors on the extremity patients with localized disease. Definitive and palliative radiation protocols have been used depending on a combination of factors including the tumor grade and stage, and the goals of therapy.

Turrel reported on 85 dogs that were treated with orthovoltage or Cobalt 60 teletherapy with a variable total dose of radiation ranging from 36-48 Gy; 24 of the dogs also received corticosteroids. The histopathologic grade was available for 52 of the dogs (Grade I in 1 dog; Grade II in 33 dogs; Grade III in 18 dogs). The overall median survival was 19 months, and the mean survival was 61.2 months. The 1- and 2-year survival rates were 76.2% and 73.2%, respectively. Prognostic factors identified included stage, grade, and location (poorer prognosis associated with tumors in the inguinal and perineal region).

Al-Sarraf, et. al., reported on a homogeneous population of dogs and response to radiation therapy. Thirty-two dogs were irradiated post-operatively for intermediate grade tumors that had been incompletely resected, and in this population there was no evidence of regional or distant metastasis. Cobalt 60 teletherapy was used on an alternate day basis (Monday, Wednesday, Friday) to deliver 3 Gy/fraction in 18 fractions for a total dose of 54 Gy. The 1-year survival rate was 100%, and the 2-5 year survival rate was 96%.

In a 1998 publication by LaDue, et. al., the overall median disease free interval (DFI) was 32.7 months, range of 19-70 months (n =56 dogs). The median DFI was 12 months for dogs with measurable disease vs 54 months for dogs with microscopic disease suggesting the importance of cytoreductive surgery.