Mammary gland tumours in the dog

Neoplasms of the mammary gland are the most common cancer in bitches, especially in Europe and in countries in which early sterilisation is not routinely practiced. For example, data from the Swedish Cancer Registry report that the incidence of malignant mammary tumours in bitches is 53.3%. Since these tumours are associated with the mammary gland, they can also develop in males, but in this case the incidence is about 1% and the tumours tend to be benign._³⁸ In bitches about 50% of tumours are malignant and of these approximately 50% can recur after surgical removal and metastasise, mainly to the lungs and lymph nodes. The average age of onset is 10-11 years; it is rare to find malignant mammary tumours in subjects under 4-5 years old. Inflammatory carcinoma, which should be discussed separately, is a highly malignant form and fortunately not very frequent. Mammary cancer in dogs is considered a good model for studying the analogous disease in humans because of some of the molecular characteristics and the behaviour of certain forms (e.g. inflammatory carcinoma). Breast cancer in cats will be discussed separately given the differences between the two species.

AETIOLOGY AND PATHOGENESIS

Although mammary cancer is found in all breeds of dog, there seems to be a predisposition for the Poodle, Spaniel, Dachshund, Yorkshire Terrier, German Shepherd and English Setter. Generally speaking, malignant mammary tumours occur 50% less often in breeds of small dogs than in breeds of large dogs._¹⁵

As often happens, the aetiology is multifactorial, but it is now clear that hormones play an important role in the pathogenesis of these tumours. In a prospective study_²² it was observed that 100% of normal or dysplastic mammary tissue and 95% of tissue with benign neoplasms was positive for oestrogen receptors (ER), while the expression of these receptors in carcinomas was significantly lower. Contrariwise, the expression of progesterone receptors (PR) was significantly lower in both benign and malignant tumours than in normal mammary tissue. Another study_³³ demonstrated negativity for both ER and PR in metastases; this could be explained as a loss of positivity for steroid hormone receptors as the aggressiveness of the tumour increases. The duration of exposure of the mammary gland to oestrogen affects the possibility of developing malignant mammary tumours. This is why early sterilisation has positive effects._⁴⁰Progesterone too, through its stimulating effect on growth hormone and, indirectly, on insulin-like growth factor (IGF-1), contributes to tumour development. A hormone whose expression seems to be linked to greater aggression is prolactin; the levels of prolactin were higher in the serum and mammary tissue of patients with malignant tumours.

In addition to being subject to hormonal influences, the development of canine mammary tumours is also linked to the expression of certain genes, as is the case in women (Table 1). New biomolecular investigation techniques have facilitated the detection of the presence and expression of molecules associated with tumour pathogenesis also in veterinary medicine, leading in recent years to a proliferation of literature in this field. A detailed discussion of these topics is beyond the scope of this author, who focuses here on a brief overview of the main factors studied. Among these, the most frequently investigated factors are c-erbB-2 (c-neu), p53, BRCA1 and 2. The first, c-erbB-2, which is identified in women using the Hercept test (now part of routine examinations and at the basis of human cancer therapy), has recently also been found with the same test in 34.5% of malignant mammary tumours in dogs._⁸However, the technique requires further testing before it can be introduced into veterinary practice. Mutations of the p53 tumour suppressor gene have been found in a small percentage of malignant canine mammary tumours,_¹² as have alterations of another tumour suppressor gene, BRCA1, which is the cause of certain hereditary malignant breast cancers in humans. However, also in this case the real role of these genes in the pathogenesis of canine mammary tumours is not yet known._^17,18,37,51

The expression of cyclo-oxygenase-2 (COX-2) has also been evaluated and found to be positive in both benign and malignant mammary tumours; an association between the degree of expression of cyclo-oxygenase and the histological subtype of the tumour has also been demonstrated._^7,13,34 The correlation between the expression of COX-2 and vascular endothelial growth factor (VEGF;the cause of angiogenesis in many tumours) has also recently been demonstrated to be an important factor in the development of metastatic malignant mammary tumours._^35,36,39 Finally, the expression of c-Kit, an oncogene that has attracted much attention in veterinary medicine, has been evaluated. Although the results need further validation, it seems possible that the expression of this gene can be used as a marker to discriminate between malignant and benign forms of canine mammary tumours._¹⁹

The action of mouse mammary tumour virus (MMTV) or its variants, which has sporadically been found in the mammary tumours of humans, has also been proposed to be among the factors capable of inducing breast cancer in this species. It is assumed that the same virus or its analogues may cause some mammary tumours even in dogs, although there are conflicting opinions among authors._^14,20

Table 1. Schematic summary of the clinical behaviour of mammary cancer in dogs.

According to some authors, other risk factors are obesity at 1 year of age_³⁰and a diet with a high content of fats and red meat; this belief is supported by the fact that mammary tumours are very uncommon in herbivores._²⁶ Recently, it has been proposed that environmental pollution by pyrethroids may be a cause of malignant mammary tumours. These substances were in fact found in the peritumoural adipose tissue of dogs with mammary cancer._¹

BIOLOGY

In dogs, mammary tumours mostly affect the fourth and fifth mammary glands and in approximately 50% of cases multiple nodules of various sizes are present simultaneously in different glands, even in non-contiguous ones. It is important to remember that multiple tumours are often of different histotypes, which is why it is important to subject all excised tissue to histological examination. About 50% of the tumours will be benign, but even the remaining 50% could have a benign clinical evolution. Most mammary tumours originate from the epithelial component of the gland, giving rise to adenomas and adenocarcinomas; only a small part originates from the connective or myoepithelial component, although in these cases malignancy is often high. There are also mixed forms, called carcinosarcoma or malignant mixed tumour, characterized by the presence of malignant cells of both lines and a poor prognosis.

Metastases spread via blood and lymph vessels, and the most common locations for such metastases are, in decreasing order of frequency, the lungs, lymph nodes, liver, kidneys, bone (Figs. 1a, b, c) and other sites.

From a histological point of view, there are numerous subtypes of mammary tumours in dogs, unlike in women, and many attempts to classify the canine tumours have been made. Currently, the most widely used classification is the one  proposed to the World Health Organization (WHO) by Misdorp in 1999_²⁴ (Table 2). In general, malignancy increases from carcinoma in situ, a malignant tumour which has not yet invaded the basal membrane, to simple tubular (the most common) and solid (less differentiated) tumours, and to sarcomas (which probably originate from the malignant transformation of a benign mixed tumour). Osteosarcoma is the most common mesenchymal mammary tumour and its behaviour is comparable to that of appendicular osteosarcoma.

 Table 2. Histological classification of canine mammary tumours (Misdorp W. et al., 1999)

DIAGNOSIS

The clinical presentation of dogs with mammary tumours is variable and dependent on the attention of their owners. Since most dogs are asymptomatic, small nodules can go unnoticed for a long time or even be an occasional finding during a visit to the clinic for other reasons. The clinical history is very important, as the hormonal status can direct the diagnostic suspicion. Subjects with  pseudopregnancies often develop multiple benign nodules which are usually miliary, mobile, painless, and non-ulcerated. These tend to enlarge during the oestrous cycle and in the two subsequent months, whereas the malignant forms already appear ulcerated in the early stages (Figs. 2a, b and c), are poorly defined, inflamed and grow rapidly and independently of the hormonal status. An oophorectomized bitch should, therefore, always be evaluated critically, since a mammary lesion is much more likely to be malignant in these animals.

Very rarely pulmonary metastases are already evident at diagnosis (Figs. 3a and b).

Multiple mammary nodules (Figs. 4a and b) are found in approximately 60% of cases; this is not, however, considered a negative prognostic factor.

Following the clinical evaluation, the diagnosis and staging are completed with a blood-chemistry examination to fully assess the general condition of the animal; this examination should also include a coagulation profile in the case of suspected inflammatory carcinoma, which is often associated with disseminated intravascular coagulation. In rare cases dogs with mammary cancer can have hypercalcaemia because of the production of parathormone-like substances,  or increased resorption of calcium from bone metastases.

Radiography of the chest, at least in the two lateral projections, and ultrasonography of the abdomen, to evaluate the iliac and sublumbar lymph nodes (which are sometimes palpable transrectally, if their volume is increased), are needed to exclude metastases. Recently, it has been shown that chest assessment by computed tomography, if performed routinely, can pick up early lung metastases that would otherwise not be visible (Fig. 5); the use of this imaging technique is, therefore, indicated._²⁷

Cytological examination by fine needle aspiration of the mammary mass is not used routinely. In fact, until a few years ago it was believed, in part based on the literature, that this investigation could not enable "active" breast tissue to be differentiated from low-grade malignant neoplasms. It was, therefore, considered useful only in cases of very aggressive tumours or to differentiate mammary tumours from inflammatory breast disease or other tumours (e.g., mast cell tumour). However, a paper published in 2009,_⁴² reporting data from 50 dogs, showed that a cytological examination, if done properly and taking multiple samples from the same lesion, has an accuracy, sensitivity and specificity suitable for the diagnosis of malignant mammary tumours (Fig. 6).

There is no doubt about the usefulness of cytology for the evaluation of enlarged or palpable regional lymph nodes, especially axillary ones. While inguinal lymph nodes are always removed during surgery (and hence subject to histological examination) since they are closely connected to the last mammary gland, axillary lymph nodes require a specific access, and previous knowledge about their status is important when planning surgery. In addition, recent studies have shown that while in healthy mammary glands the draining  lymph nodes are the axillary nodes for the two cranial mammary glands and the superficial inguinal nodes for the two caudal mammary glands and, in most cases, for the third mammary gland (Fig. 7), in the case of neoplastic mammary glands the lymph nodes to be evaluated are the sternal nodes for the first four glands, the medial iliac nodes for the third, fourth and fifth glands, and the inguinal and popliteal nodes for the fifth gland_⁴⁶.

This must also be considered when planning surgery. A recent study_¹¹ proposed an alternative method to synthetic dyes for the detection of sentinel lymph node(s): haemosiderin, obtained by haemolysis of the animal’s blood, thereby avoiding the risk of anaphlyaxis, can be injected into a peripheral regional vein (cephalic or saphenous vein) instead of a synthetic dye. Brownish  traces of the substance are then searched for in the regional lymph nodes. If confirmed, this method could facilitate the identification of lymph nodes to be removed.

In any case, the final diagnosis is made only from the histological examination of the lesion (Figs. 8, 8a and 8b), often performed directly during the therapeutic surgery, as the biopsy is generally excisional.

Based on the evaluation of the data obtained, it is then possible to complete the staging of the tumour according to the Tumour, Node, Metastasis (TNM) system established by the WHO_²⁸and subsequently modified (Table 3), which considers the size of the tumour and the involvement of lymph nodes as significant prognostic factors.

Finally, the possibility of using certain molecular markers in dogs is currently being studied for the early detection of tumour cells in the bloodstream. In particular, polymerase chain reaction markers have been used for CK19, ERBB2, EGFR, CLDN7, and ELF3, all substances correlated in humans with the presence of circulating tumour cells and early indicators of metastasis. The positive results obtained in dogs are promising and it is to be hoped that in the future this technique could be applied routinely also in dogs._⁶

Table 3. Modified staging system of mammary tumours in the dog. (Owen LN, 1980)²⁸

TREATMENT

With the exception of inflammatory carcinoma, the treatment of choice for mammary tumour is surgery, the extension of which must be sufficient to remove the entire neoplastic mass with a margin of healthy tissue; more invasive surgery does not improve the prognosis and increases costs and morbidity. This implies a more or less extensive excision depending on the number of mammary glands involved, in contrast to what is done in cats. In detail, the types of surgery are:

• Lumpectomy: this is almost exclusively considered for biopsy purposes and recommended for nodules with a diameter not exceeding 0.5 cm, not adherent to deep tissue planes, not ulcerated and which are thought to have a benign origin (Fig. 9). If the subsequent histological examination reveals the presence of a malignancy, a further broader intervention is indicated.

• Mastectomy: this is removal of a single mammary gland, possibly together with the underlying muscle or fascia if the tumour is adherent to deep tissue planes. It is indicated for single lesions located in the centre of the mammary parenchyma (Fig. 10). Given he connection of the mammary tissue between the fourth and fifth mammary glands and between the first and second, it is often easier to remove two glands rather than a single gland. However, there are no contraindications to removing a single gland providing that the histological examination confirms the completeness of excision.

• Regional Mastectomy: this is the most commonly used operation in dogs (Fig. 11a, b and c). It is planned based on the lymphatic and venous connections between the mammary glands, although exceptions may occur, especially in cases of very aggressive cancers. In general, the first, second and third glands should be removed together, as should the fourth and fifth glands together with the inguinal lymph nodes. The axillary lymph nodes are removed only if enlarged or cytologically metastatic.

• Unilateral or bilateral total mastectomy: the removal of the entire row of mammary glands of one or both sides is performed only when there are multiple lumps on non-contiguous mammary glands. As bilateral mastectomy involves significant tension on the skin, it is usually preferable to perform the surgery in two separate sessions, about a month apart, starting from the row that appears most clinically involved (Figs. 12a, b, c, d, e).

If the surgery has to be performed in one session, an attempt can be made to preserve healthy tissue between the two rows, in the cranial and caudal portion of the abdomen (Fig. 13), or to implement tension-relieving techniques.

The choice of the type of procedure can also be influenced by the age of the subject: a recent study showed that 58% of dogs treated with regional mastectomy developed a new cancer in the same mammary row,_⁴⁷ regardless of whether the initial diagnosis was a benign or malignant tumour. Total mastectomy may, therefore, be indicated as prophylaxis, especially in young subjects. Another study carried out on 90 dogs with 236 tumours found that the malignant forms are generally larger (> 3 cm), affect older subjects and can develop from neglected benign forms. A similar behaviour has been described for breast cancer in women. This is demonstrated by the fact that in many of the malignant tumours diagnosed, the same histological section or even sections in adjacent regions contained  benign areas and/or carcinoma in situ._⁴⁵The choice of the type of surgery might be changed, partly in view of this new piece of information.

Chemotherapy, used widely in women, is rarely used in dogs. 5-fluorouracil, one of the first drugs used in humans, has proven effective even in the most aggressive forms of canine mammary tumours; in a study of eight dogs, treatment with surgery and adjuvant chemotherapy, started 1 week after surgery and consisting of the combined intravenous administration of 5-fluorouracil (150 mg/m_²) and cyclophosphamide (100 mg/m_²) once a week for 4 weeks, significantly improved the survival of the subjects compared to a similar group treated with only surgery._¹⁶In contrast, a more recent study, conducted on 31 dogs treated with doxorubicin[33] or docetaxel (taxanes are commonly used in human medicine for breast cancer), did not show that this management produced statistically significant differences in terms of survival, relapse-free survival or metastases compared to treatment with surgery alone._⁴¹ The decision to use chemotherapy is, therefore, based on the degree of malignancy, the presence of neoplastic lymphatic or blood vessel emboli and the completeness of surgical excision. In patients at high risk of recurrence or metastases, the medical management can be justified, even though the results are not impressive.

Another widely used treatment in human medicine is radiotherapy, which is not used for mammary tumours in the dog._⁴⁴

The same applies to hormone therapy, another cornerstone of human medicine. In dogs, however, it has been observed that hormone-dependency is lost as the malignancy of the tumour increases; thus this type of treatment is effective exclusively in benign tumours. In fact, even in women the use of hormone therapy is limited to ER-positive cases, because if ER are not expressed, the treatment would be useless. It is not yet possible to determine receptor positivity in daily veterinary practice. In addition to blocking the oestrogen stimulus with receptor antagonists (tamoxifen), the same result can be achieved with aromatase inhibitors, agonists of luteinising hormone (LHRH) or surgical removal of the ovaries. Oophorectomy concomitant with mastectomy may, therefore, be useful in patients with ER-positive tumours._^43,44 A recent study showed that this approach could also be beneficial in patients under treatment with tamoxifen, in view of the high incidence of pyometra resulting from the use of this drug. In addition, tamoxifen induces vulvar oedema, a purulent vulvar discharge and retinitis in 50% of subjects treated._⁴⁹

The new frontier for the treatment of the most malignant mammary tumours could be vaccines, which are currently being studied for several types of cancer. Vaccines administered with the aid of a viral vector appear to be active in vitro against experimental models of canine mammary carcinoma,_¹⁰ while a adenoviral vector vaccine against telomerase reverse transcriptase (TERT) and HER-2/neu, both expressed in some malignant canine mammary tumours, has given good results in vivo in terms of inducing immunity against these two tumour-associated antigens._³²

Finally, as for other malignancies, in selected cases of malignant mammary tumours in dogs, isolated metastatic pulmonary nodules can be removed. In order for this to be possible, however, certain conditions must be met:

• the primary tumour should have been under control for at least 300 days;
• there must be no more than two visible pulmonary nodules (preferably by computed tomography studies);
• the metastases must have a doubling time of at least 30 days;
• there must be no other metastases evident in other sites.

As in women, it has also been proven in dogs that survival time and quality of life are improved following removal of metastases.

PROGNOSIS

Some of the risk factors considered important in the treatment of canine mammary tumours are:_{^2,9,31,45,46}

• hormonal factors: oestrogen and progesterone play important roles in inducing the development of mammary tumours (hence the positive effect of early sterilisation);
• growth factors: epidermal growth factor and transforming growth factor are associated with the presence of ER and PR in malignant mammary tumours; the excess of growth hormone induced by progestins is involved in the development of mammary tumours, while receptors for prolactin have been found in neoplastic tissue;
• obesity at a young age, as well as a diet rich in fats and red meat, promotes the development of mammary tumours.

The prognostic factors include:

• advanced age at diagnosis: shorter survival and disease-free survival, though not all authors agree on this point;
• rapid and infiltrative growth;
• fixation to deep planes;
• large size (> 3 cm);
• early skin ulceration;
• clinical stage;
• lymph node involvement: 80% of dogs with lymph node metastases have a recurrence within 2 years after surgery;
• sarcomas: generally more malignant than carcinomas (except inflammatory carcinoma);
• malignant mixed tumour, squamous cell carcinoma, solid carcinoma: worse prognosis;
• ER-negative: indicator of less differentiation;
• density of the microcirculation in the tumour: an indicator of neo-angiogenesis, therefore a negative factor;
• aneuploidy;
• positivity for Ki-67
• presence of other genetic alterations (p53 mutations)
• previous removal of a malignant mammary tumour: increased risk of developing other tumours in the course of an animal’s lifetime.

PREVENTION

Does spaying prevent tumours?

As mentioned above, mammary tumours are under hormonal influence. It has been shown that early sterilisation can drastically reduce the risk of developing tumours, even malignant ones: if the oophorectomy is performed before the first heat, the risk of developing malignant mammary tumours is 0.05%; after the first heat it rises to 8%, after the second heat 26%._⁴⁰ Thereafter the preventive effect becomes insignificant. There is proof of this in Anglo-Saxon countries, in which most bitches are spayed before puberty: the incidence of malignant mammary tumours is very low compared to that in other European countries, in which this is the most common type of tumour among bitches.

What about pregnancy?

Pregnancy, the number of pregnancies, the number of pups born and age at first pregnancy do not appear to influence the development of malignant mammary tumours. Although the risk appears to be higher in nulliparous bitches, no study has ever proven this with certainty.

- What is the effect of oestrogens and progestins?

The use of oestrogens or progestins in dogs favours the development of benign tumours, while the combination of high-dose oestrogens and progestins can induce the formation of malignant tumours.

- Is sterilisation advisable when removing the mammary tumour?

There is a diversity of opinions regarding this subject. While Yamagami_⁵⁰ and Morris_²⁵ argue that survival for 2 years after mastectomy for malignant mammary tumours is not affected by concomitant sterilisation, Sorenmo_⁴³believes that sterilisation can be useful if it is performed concomitantly with or very soon after onset of the tumour. In any case, eliminating the hormonal stimulus makes the breast parenchyma hypotrophic. For this reason, even though the protective effect against the development of new malignancies has not been demonstrated, the reduction in volume of the mammary gland itself can facilitate an early diagnosis of new nodules. It is incumbent on the veterinarian to recommend the type of surgery to be performed on a case-by-case basis, since the new theories argue that animals that have developed a malignant mammary tumour are more likely to develop other malignant tumours,_⁴⁷ especially if they are young (there being more time for further malignant transformation). Early diagnosis is, therefore, essential in these cases.

Inflammatory carcinoma

This type of tumour, described in a dog for the first time in 1983 by Susaneck and colleagues,_⁴⁸ is a highly malignant form of cancer that leads to the death of the animal within a few weeks. Clinically, the animal presents with oedematous mammary glands that are hot, painful and reddened. Often the oedema, caused by the cancer cells invading cutaneous lymph vessels (important in the diagnosis of inflammatory carcinoma), extends to the limbs, making movement painful (Fig. 14a and b).

Two forms of inflammatory carcinoma have been described in the dog: a primary, idiopathic form and a secondary form. The primary form has a more aggressive behaviour, is very painful, the neoplastic cells are negative for the expression of PR, there is increased expression of p53 and a sarcoma-like dermal infiltration._²⁹ The secondary form occurs after mastectomy and has a slightly less malignant behaviour.

Both forms must be differentiated from mastitis, which does not have such a rapid and drastic course. In most cases, inflammatory carcinomas are characterized by significant neoplastic embolisation of blood and lymphatic vessels (Fig. 15). Lymph node metastases can even develop retrograde to the popliteal lymph node, or reach the iliac lymph nodes. Clement and colleagues_⁴ also demonstrated that inflammatory carcinoma metastasises to sites different from those of the other malignant mammary tumours, as it more frequently involves the bladder and genitalia and less frequently the lungs and bones.

A recent study_³ has shown that inflammatory carcinoma, apart from not expressing ER and PR, expresses VEGF more strongly than do other mammary malignancies; this factor is important for tumour angiogenesis and also explains the high metastasising capacity of this histological type of carcinoma. Moreover, newly formed microvascular structures without endothelial cells_²³ have been found in this tumour, something also found in the analogous malignancy in humans. These findings help to demonstrate the highly invasive potential of inflammatory carcinoma._⁴

From a clinical point of view, once the diagnosis has been made (based mainly on the course of the disease and the findings of the clinical and cytological examination), treatment is almost totally ineffective. Indeed, despite the name inflammatory carcinoma, cytological examination of this tumour in dogs does not show an inflammatory infiltration but rather highly malignant tumour cells. In most cases surgery is not recommended, although it may be associated with medical therapy in selected cases,^₂₁ while palliative treatment may be attempted using the positivity of many inflammatory carcinomas for COX-2 receptors. In one study the median survival of dogs treated with piroxicam was 185 days, compared to 30 days in dogs treated with doxorubicin, with or without cyclophosphamide._⁵ Other chemotherapeutic agents, such as carboplatin, cisplatin and capecitabine have not produced encouraging results._²¹

References

1. Andrade F.H.E., Figueiroa F.C., Bersano P.R.O. et al. Malignant mammary tumor in female dogs: environmental contaminants. Diagn Pathol 5:45-49, 2010
2. Chang S.C., Chang C.C., Chang T.J. et al. Prognostic factors associated with survival two years after surgery in dogs with malignant mammary tumors: 79 cases (1998-2002). J Am Vet Med Assoc 227:1625-29, 2005
3. Clemente M., Pérez-Alenza M.D., Illera J.C. et al. Histological, immunohistological, and ultrastructural description of vasculogenic mimicry in canine mammary cancer. Vet Pathol 47:265-74, 2010a
4. Clemente M., Pérez-Alenza M.D., Peña L. Metastasis of canine inflammatory versus non-inflammatory mammary tumours. J Comp Pathol 143:157-63, 2010b
5. de M. Souza C.H., Toledo-Piza E., Amorin R. et al. Inflammatory mammary carcinoma in 12 dogs: Clinical features,cyclooxygenase-2 expression, and response to piroxicam treatment Can Vet J 50:506–10, 2009
6. Da Costa A., Oliveira J.T., Gärtner F. et al. Potential markers for detection of circulating canine mammary tumor cells in the peripheral blood. The Vet J, e-pub novembre 2010
7. Dore M., Lanthier I., Sirois J. Cyclooxygenase-2 expression in canine mammary tumors. Vet Pathol 40:207-12, 2003
8. Dutra A.P., Granja N.V.M., Schmitt F.A. et al. C-erb-B-2 expression and nuclear pleomorphism in canine mammary tumors. Braz J Med Biol Res 37:1673-81, 2004
9. Ferreira E., Bertagnolli A.C., Cavalcanti M.F. et al. The relationship between tumour size and expression of prognostic markers in benign and malignant canine mammary tumours. Vet Comp Oncol, 7:230-35, 2009
10. Gentschev I., Ehrig K., Donat U. et al. Significant growth inhibition of canine mammarycarcinoma xenografts following treatment with oncolytic vaccinia virus GLV-1h68. J Oncol e-pub, 2010
11. Gonzaga Porto Pinheiro L., Santos de Oliveira Filho R., Vasques P.H.D. et al. Hemosiderin. A new marker for sentinel lymph node identification. Acta Cirúrgica Brasileira 24:432-36, 2009
12. Haga S., Nakayama M., Tatsumi K. et al. Overexpression of the p53 gene product in canine mammary tumors. Oncology Reports 8:1215-19, 2001
13. Heller D.A., Clifford C.A., Goldschmidt M.H. et al. Cyclooxygenase-2 expression in associated with histologic tumor type in canine mammary carcinoma. Vet Pathol 42:776-80, 2005
14. Hsu W.L., Lin H.Y., Chiou S.S. et al. Mouse mammary tumor virus-like nucleotide sequences in canine and feline mammary tumors. J Clin Microbiol 48:4354-62, 2010
15. Itoh T., Uchida K., Ishikawa K. et al. Clinicopathlogical survey of 101 canine mammary gland tumors: differences between small-breed dogs and others. J Vet Med Sci 67:345-47, 2005
16. Karayannopoulou M., Kaldrimidou E., Costantinidis T.C. et al. Adjuvant post-operative chemotherapy in bitches with mammary cancer. J Vet Med A 48:85-96, 2001
17. Klopfleisch R., Gruber A.D. Increased expression of BRCA2 and RAD51 in lymph node metastases of canine mammary adenocarcinomas. Vet Pathol 46:416-22, 2009
18. Klopfleisch R., von Euler H., Sarli G. et al. Molecular carcinogenesis of canine mammary tumors: news from an old disease. Vet Pathol online first, dicembre 2010
19. Kubo K., Matsuyama S., Katayama K. et al. Frequent expression of the c-kit proto-oncogene in canine malignant mammary tumor. J Vet Med Sci 60:1335-40, 1998
20. Laumbacher B., Fellerhoff B., Herzgerger B. et al. Do dogs harbour risk factors for human breast cancer? Med Hypotheses  67:21-6, 2006
21. Marconato L., Romanelli G., Stefanello D. et al. Prognostic factors for dogs with mammary inflammatory carcinoma: 43 cases (2003-2008). J Am Vet Med Assoc 235:967-72, 2009
22. Millanta F.,Calandrella M.,Bari G. et al. Comparison of steroid receptor expression in normal, dysplastic, and neoplastic canine and feline mammary tissues. Res Vet Sci.79:225-32, 2005
23. Millanta F., Caneschi V., Ressel L. et al. Expression of Vascular Endothelial Growth Factorin canine inflammatory and non-inflammatory mammary carcinoma. J Comp Path 142:36-42, 2010
24. Misdorp W., Else R., Hellmén E. et al. Histological classification of mammary tumors of the dog and the cat. Vol. VII. Armed Forces Institute of Pathology & American Registry of Pathology & the World Health Organization Collaborating Center for Worldwide Reference on Comparative Oncology, Washington D.C., USA, 1999
25. Morris J.S., Dobson J.M., Bostock D.E. et al. Effect of ovariohysterctomy in bitches with mammary neoplasms. Vet Rec 142:656-58, 1998
26. Munson L., Moresco A. Comparative pathology of mammary gland cancers in domestic and wild animals. Breast Disease 28:7-21, 2007
27. Otoni C.C., Rahal S.C., Vulcano L.C. et al. Survey radiography and computerized tomography imaging of the thorax in female dogs with mammary tumors. Acta Veterinaria Scandinavica 52:20-9, 2010
28. Owen L.N. Classification of tumors in domestic animals. Ginevra, 1980 World Health Organization
29. Peña L., Pérez-Alenza M.D., Rodriguez-Bertos A. et al.Canine inflammatory mammary carcinoma: histopathology, immunohistochemistry and clinical implications of 21 cases Breast Cancer ResTreat78:141-8, 2003
30. Pérez-Alenza M.D., Rutteman G. R., Peña L. et al. Relation between habitual diet and canine mammary tumours in a case-control study. J Vet Intern Med, 12:132-39, 1998
31. Pérez-Alenza M.D., Peña L., del Castillo N. et al. Factors influencing the incidence and prognosis of canine mammary tumours. J Small Anim Pract 41:287-91, 2000
32. Peruzzi D., Mesiti G., Ciliberto G. et al. Telomerase and HER-2/neu as targets of genetic cancer vaccines in dogs. Vaccine 28:1201-8, 2010
33. Queiroga F.L., Pérez-Alenza M.D., Silvan G. et al. Role of steroids hormones and prolactin in canine mammary cancer. J Steroid Biochem Mol Biol 94:181-87, 2005
34. Queiroga F.L., Pires I., Lobo L. et al. The role of Cox-2 expression in the prognosis of dogs with malignant mammary tumours. Res in Vet Sci 88:441-45, 2010
35. Queiroga F.L., Pires I., Parente M. et al. COX-2 over-expression correlates with VEGF and tumour angiogenesis in canine mammary cancer. The Vet J e-pub luglio 2010
36. Qiu C.W., Lin D.G., Wang J.Q. et al. Expression and significance of PTEN and VEGFin canine mammary gland tumours. Vet Res Commun 32:463-72, 2008
37. Rivera P., Melin M., Biagi T. et al. Mammary tumor development in dogs is associated with BRCA1 and BRCA2. Cancer Res 69:8770-74, 2009
38. Saba C.F., Rogers K.S., Newman S.J. et al. Mammary gland tumors in male dogs J Vet Intern Med 21:1056-59, 2007
39. Santos A.A.F., Oliveira J.T., Lopes C.C.C. et al. Immunohistochemical expression of VascularEndothelial Growth Factor in canine mammary tumours. J Comp Path 143:268-75, 2010
40. Schneider R., Dorn C.R., Taylor D.O.N. Factors influencing canine mammary cancer development and postsurgical survival. J Natl Cancer Inst 43:1249-61, 1969
41. Simon D., Schoenrock D., Baumgärtner W. et al. Postoperative adjuvant treatment of invasive malignant mammary gland tumors in dogs with doxorubicin and docetaxel. J Vet Intern Med 20:1184-90, 2006
42. Simon D., Schoenrock D., Nolte I. et al. Cytologic examination of fine-needle aspirates from mammary glands tumors in the dog: diagnostic accuracy with comparison to histopathology and association with postoperative outcome. Vet Clin Pathol 38:521-28, 2009
43. Soremno K.U., Shofer F.S., Goldschmidt M.H. Effect of spaying and timing of spaying on survival of dogs with mammary carcinoma. J Vet Intern Med 14:266-70, 2000
44. Soremno K.U. Canine mammary gland tumors. Vet Clin Small Anim Pract 33:573-96, 2003
45. Soremno K.U., Kristiansen V.M., Cofone M.A. et al. Canine mammary gland tumours; a histological continuum from benign to malignant; clinical and histopathological evidence. Vet Comp Oncol 7:162-72, 2009
46. Soremno K.U., Rasotto R., Zappulli V. et al. Development, anatomy, histology, lymphatic drainage, clinical features, and cell differentiation markers of canine mammary gland neoplasms. Vet Pathol online first, dicembre 2010
47. Stratmann N., Failing K., Richter A. et al. Mammary tumor recurrence in bitches after regional mastectomy. Vet Surg 37:82-6, 2008
48. Susaneck S.J., Allen T.A., Hoopes J. et al. Inflammatory mammary carcinoma in the dog. J Am Anim Hosp Assoc 19:971-76, 1983
49. Tavares W.L.F., Lavalle G.E., Figueiredo M.S. et al. Evaluation of adverse effects in tamoxifen exposed healthy female dogs. Acta Veterinaria Scandinavica52:67, 2010
50. Yamagami T., Kobayashi T., Takahashi K. et al. Influence of ovariectomy at the time of mastectomy on the prognosis for canine malignant mammary tumours. J Small Anim Pract 37:462-64, 1996
51. Yoshikawa Y., Morimatsu M., Ochiai K., et al. Novel variations and loss of heterozygosity of BRCA2 identified in a dog with mammary tumors. Am J Vet Res 69:1323-28, 2008