Mast Cell Tumours in the Dog. What Are the Facts?
Mast cells originate from pluripotent hematopoietic stem cells, and perhaps also from undifferentiated mesenchymal cells. The dermis, submucosa of the gastrointestinal and respiratory tract, serous membranes and bone marrow is rich in mast cells, often residing in small clusters in perivenous areas. Mast cells contain granules that can be made visible with specials stains, such as May-Grunwald Giemsa used in cytology, or Toluidine-blue, used in histology. These granules are filled with bio-active factors such as histamine, heparin, or leukotrienes, playing a role in the inflammatory response in which mast cells may take part. Release of these factors may occur if appropriate stimuli occur (such as antigen binding to IgE at the surface of mast cells, e.g., in hypersensitivity reactions) but also at cell death (e.g., in necrotic tumor areas). These latter conditions may lead to paraneoplastic signs related to the presence of mastocytoma (MCT). A possible link between chronic inflammation of skin and MCT has been suggested. A viral origin has been hypothesized but not been confirmed. A deregulated expression of stem cell factor (SCF) receptor--mediating the SCF growth signal to mast cells, and encoded by the c-kit gene--has been observed in some MCTs.
MCTs rank first in incidence of skin tumors in the dog, and second in the cat. In the dog, peak incidence is at about 9 years, but the age at which this tumor may occur ranges widely. Occasionally multiple regressing MCTs are seen in young animals, which more likely are of hyperplastic rather than neoplastic nature. Boxers, bulldogs, Boston terriers, bullterriers, and retrievers are more or less at increased risk. In the majority of boxers the disease is of the less aggressive, histologically well differentiated type. The majority of canine MCT originate in the skin, but they may arise in spleen, liver, gastro-intestinal or (cranial) respiratory tract (including palatum, pharynx, larynx) and mediastinum. Multiplicity of skin MCTs is seen in roughly 25% of dogs (either as lesions manifested simultaneously or subsequently) and likely in a even greater percentage of cats. Multiplicity may represent growth of metastases, but also independent primary lesions.
Diagnostic Workup for Clinical Staging and Histological Grading
Diagnosis by cytology of FNABs (often successful) or by histology of incisional/excisional biopsy (if cytology inconclusive) preferably should take place before complete treatment planning. The latter should be considered in relation with assessment of prognosis. Thus, extension (and number) of primary tumor(s) should be determined, location recorded, and regional and distant lymph nodes should be examined. If any suspicion of lymph node involvement exists, FNABs should be taken (something to be considered in all dogs with MCT). Blood examination includes CBC, and measures of kidney and liver function. Abdominal ultrasound should be followed by guided FNABs of any suspected lesion. Still, the presence of mast cells in lymph nodes, bone marrow, spleen, liver or blood is not proof of neoplastic involvement, although a large increase in number, as well as signs of cellular de-differentiation may increase the likely-hood of tumor involvement proportionally. Radiography of the thorax may be preserved for animals with tumor extension in the cranial part of the body, in particular of regional nodes are involved. In the latter instance, increased mass of intra-thoracic nodes may be suggestive of tumor involvement, whereas lung involvement is a rare event. Based on this workup the animal can be clinically staged. A new staging system will be discussed.
In the dog, histological classification of malignancy grade bears prognostic value, and discriminates between grades 1 (well-differentiated), 2 (moderately differentiated) and 3 (poorly differentiated type). Increased grade is associated with of increased aggressiveness. Grading, however, is based on a subjective view, and results may vary widely between pathologists (and between studies). Absolute predictions on the relation between grade and prognosis cannot be made--and grade alone should not be considered a reliable basis to for treatment planning in individual cases. Still, relative differences in prognosis for dogs with MCTs of different grades exist: metastasis is very rare in grade 1 tumors, if treated properly, and may develop in 5-30% of grade 2 tumors, while being more common in grade 3 tumors.
No factor precisely will predict tumor behavior and treatment outcome. Discrimination should be made between true local recurrence, regional (nodal/satellite) metastasis, distant metastasis, and development of new primary lesions, but this is not always feasible. Still, clinical stage (which is related to factors such as the histological malignancy grade) to some extent has shown to relate to prognosis, but no consensus exist on the precise staging system. As an example, number of skin lesions may inversely relate to disease free period based more on the likely presence of yet occult lesions than development of metastases. Size, extent of infiltrative growth and ulceration mark important features of the primary lesion. Regional lymph node involvement clearly heralds high risk of further metastasis, while presence of distant macro-metastases will be related to increased risk of a more rapid death.
Localization also has shown to be of prognostic weight. Firstly, because it may influence the ease of radical surgery (greater on trunk than at distal extremities). Secondly, there are some sites of primary manifestation, that are recorded in some studies to be related with greater aggressiveness: These include the inguinal, preputial, perianal, scrotal area of skin, and mucocutaneous sites elsewhere. However, the consistency of these findings is low. In other studies no relationship between localization and prognosis was found. The visceral form of the disease carries a grave prognosis.
In addition to the histological malignancy grade, other factors can be determined, that may be related to the biological aggressiveness. Histological staining for argyrophilic nucleolar organizer regions (AgNORs), as well as for other markers of proliferation, may have prognostic value, as can be said of the presence of paraneoplastic signs (ill sign). Recently, increased cytoplasmic KIT staining was significantly associated with an increased rate of local recurrence and a decreased survival rate.
Radical surgery is the first treatment option. As always in oncologic surgery safe margins should be considered, but this is especially true for mast cell tumors, as they have often infiltration of the surrounding tissues. Margins of 3 cm with one extra plane in the depth are advocated. When these margins cannot be obtained this should never be an excuse NOT to operate. There are two reasons for this. First of all a 3 cm margin is not always necessary for complete removal of the tumor. In a recent study with 23 grade 1 and 2 mast cell tumors a 2 cm margin resulted in 100% complete removal of the tumor and a 1 cm margin in 75%. Secondly, when there is a 'dirty margin' (tumor cells present on the edge of the cut) this is not always related with a recurrence of the tumor. Several studies have demonstrated that only 15-35% of these cases will lead to a recurrence. Radical surgery (with >3 cm margins in width, with excision of fascia/muscle in continuity of the primary MCT) of a solitary MCT confined to the skin, at sites linked with a more favorable biological behavior, may have a good prognosis if lesions are grade 1, less so if grade 2, and will carry a poor prognosis if it concerns grade 3 lesions. Prognosis in this respect concerns absence of local or distant recurrence. The risk of manifestation of new primary tumors elsewhere may more directly be related with the number of MCTs presented so-far. In view of the increased likely-hood of recurrent disease in dogs with MCTs where clinico-pathological features deviate from the above mentioned more favorable condition, radiotherapy--if the disease is likely local--and/or chemotherapy should be considered in such cases. The benefit of adjuvant de-ionised water has not successfully been proven.
Mast cell tumors are in general responsive to radiotherapy. External beam fractionated radiotherapy (45-57 Gray) of local disease alone may lead to about 50% good clinical response-rate for > 1 year, while > 80% response-rates have been achieved if RT was used as an adjunct following non-radical surgery for grade 1 or 2 MCTs in the dog.
Major involvement of regional lymph nodes is highly suggestive of presence of distant (micro)metastasis, and thus may ask for systemic chemotherapy. Similarly, grade 3 tumors, and MCTs in perianal/inguinal areas (as well as other unfavorable sites or prognostic conditions, mentioned above) may benefit from (adjunctive) chemotherapy. Still, surprisingly little information exists with respect to chemotherapy response rates. Prednisolone oral medication was found to induce a complete response in only 1/21 dogs, while combination chemotherapy did so in 23-33% of dogs with MCT. The presence of distant macro-metastasis appears to preclude a good response to chemotherapy. On the other hand, if a complete response of clinical disease to chemotherapy occurs in dogs with MCT within the first month, very long disease free survival may follow in some. Efficacy of combination chemotherapy protocols (including vinblastine or vincristine, cyclophosphamide, prednisolone) appeared better than the sole use of lomustine, an oral medication with significant risk of hepatotoxicity.
Receptor tyrosine kinases are excellent candidates for molecular targeted therapy, because they play key roles in controlling cell proliferation and survival and are frequently dysregulated in a variety of malignancies. Point mutations in the kinase domain of c-kit leading to tyrosine phosphorylation in the absence of ligand binding have been identified internal tandem duplications (ITDs) in exons 11 and 12 of c-kit in mast cell tumors (MCTs) of dogs. When related to histological grade ITDs were found in 1 of 12 (8%) grade-1, 42 of 119 (35%) grade-2, and 9 of 26 (35%) grade-2 tumors (overall prevalence, 52 of 157 [33%]). SU1165411, a small molecule receptor tyrosine kinase inhibitor, was used in a phase I clinical trial including canine MCTs and showed durable objective responses (partial responses and complete responses); 9 of these MCTs possessed ITDs in the JM domain of c-kit. Recently, a commercial available canine selective inhibitor of KIT (Masitinib) was tested in a double-blind, randomized, placebo-controlled phase III clinical trial. Masitinib increased overall TTP compared with placebo from 75 to 118 days (P = 0.038). This effect was more pronounced when masitinib was used as first-line therapy, with an increase in the median TTP from 75 to 253 days (P = 0.001) and regardless of whether the tumors expressed mutant (83 versus not reached [P = 0.009]) or wild-type KIT (66 versus 253 [P = 0.008]). Another veterinary tyrosine kinase inhibitor (Tocenarib) was able to achieve an overall response rate of 37% (7 CR, 25 PR in 86 dogs; compared to 5 PR in 63 placebo treated dogs), with a median response duration of 3 months.
Apart from therapeutic measures, dogs and cats at threat of secondary (local/inflammatory, distant/paraneoplastic) MCT-related symptoms may benefit from a symptomatic approach. Thus, corticosteroids may alleviate inflammatory complications of MCT, while histamine-antagonists may help prevent gastro-intestinal or respiratory complaints. In the management of MCTs, however, one should bear in mind their great variation in biological behavior. Thus, presence of tumor cells in surgical margins may be without development of local recurrence in 30% of cases. Some dogs with large MCTs, or with positive lymph nodes, or occasionally even distant metastasis, may live long without complicating clinical symptoms. On the other hand, dogs with aggressive progression of the disease, often had limited tumors that remained indolent for a long time and manageable all the same time.