Background: The human epidermal growth factor type 2 (HER2) receptor is a membrane glycoprotein tyrosine kinase. In woman, HER2 expression is diagnosed in 30% of breast carcinomas and it is associated with a worse prognosis, higher rate of recurrence and mortality. In the bitch, the HER2 overexpression in canine mammary tumors is still controversial and the prognostic value remains uncertain. Thus, we aimed to verify the HER2 expression in canine mammary carcinomas and relate it to the type and histological grade, lymph node metastasis and clinical staging.Materials, Methods & Results: Ninety bitches diagnosed with mammary carcinoma were included in this study. The inclusion criteria were bitches with complete clinical examination, thoracic radiographic examination and submitted unilateral or bilateral mastectomy. Ninety-nine samples of mammary carcinoma were used and the fragments of tumor and regional lymph nodes were fixed in 10% neutral formalin for histopathological and immunohistochemistry analysis. The lesions were evaluated by two pathologists and classified according to the type and histological grade. HER2 expression was performed by semi-quantitative analysis of the slides according to the HerceptTestTM (Dako) recommended score. Simple carcinomas were the most frequent (51.51%) followed by complex carcinomas (46.47%) and in situ carcinoma (2.02%). The histological grade of 97 carcinoma samples was attributed, except in situ carcinoma, 37 (38.14%) of the neoplasms were grade I, 50 (51.55%) grade II and only 10 (10.31%) tumors were classified as grade III. Forty bitches were submitted to clinical staging (TNM) and 42.50% of the bitches received staging in grade I and, 25% of the bitches staged in grade IV and V, with metastases. The HER2 expression, 13/99 samples (13.13%) received score +2, 19/99 (19.19%) score +1 and absence of marking (score 0) was identified in 67 samples (67.80 %). Immunostaining in hyperplastic or normal epithelial cells was evidenced, often in association with weak or moderate cytoplasmic labeling. Of the samples expressing +2 score for HER2 (n = 13), eight samples (17.39%) were complex carcinoma and five (9.80%) simple carcinomas. There was no relationship between HER2 immunostaining with age, tumor size, TNM, histological type, histological gradation, lymph node metastasis and distance. Animals with lymph node metastasis, as well as those diagnosed with distant metastasis, did not present HER2 expression in the tumors.Discussion: The simple carcinoma seems to be the most frequent type histological diagnosed in canine mammary carcinomas, followed by carcinoma in mixed tumor and complex carcinoma. Tubulopapillary carcinomas are more invasive in the female dogs as well as in the woman. Carcinomas grade I and II are more frequent and present a better prognosis for the dog. However, bitches with grade III carcinoma survived for a shorter time when compared to dogs with grade I or II tumors. A factor that may have contributed to the lower number of bitches at worst prognostic stage (EC IV and V) is the current owners’ awareness that they have sought veterinary help earlier, as soon as they detect small nodules in mammary gland. Overexpression of HER2 in women breast cancer is diagnosed in 20-30% of cases, whereas in bitches, this expression is variable. Also the different percentages of canine HER2 immunostaining are due to the lack of standardization for the analysis of the immunostaining, the immunohistochemical techniques employed and the non-specificity of the HER2 antibody. In canine mammary carcinomas the HER2 expression in low and this immunostaining is not related to other established prognostic factors. This study reinforces the hypothesis put forward by other authors that in the bitch the expression of HER2 may not be related to malignancy and tumor progression.