Evaluation of Cytokeratin 7 Expression in Different Mammary Gland Neoplasms

Document Type : Pathology


1 Department of Pathology, Faculty of Veterinary Medicine, University of Tehran, Tehran, Iran

2 Mycology Research Center, Faculty of Veterinary Medicine, University of Tehran, Tehran, Iran

3 Department of Immunology, Faculty of Medicine, Shahid Sadoughi University of Medical Sciences, Yazd, Iran


Cytokeratins are non-contractile intermediate filaments engaged in anchoring and structural functions forming a network to support cytoplasm. Cytokeratin 7 (CK7) expression in human breast carcinomas has proved to be a useful differentiation marker, but its expression in canine mammary gland tumors is poorly understood.
Cytokeratin 7 (CK7) expression in human breast carcinomas has proved to be a useful differentiation marker, but its expression in canine mammary gland tumors is poorly understood.
This research was based on the immunohistochemical study of CK7 in 17 cases of canine mammary gland neoplasms obtained from the Department of Pathology, Faculty of Veterinary Medicine, University of Tehran. Masson’s trichrome staining was performed to differentiate between collagen fibers and smooth muscle.
CK7 protein was detected in both epithelial (1 benign mixed tumor, 1 fibroadenoma, 1 complex carcinoma, and 1 carcinoma mixed type) and myoepithelial (1 fibroadenoma, 1 benign mixed tumor, 3 complex carcinomas, 1 ductal carcinoma, and 1 carcinoma mixed type) cells. Fine and thick collagen fibers were observed in the sections stained by Masson’s trichrome.
Despite using CK7 as a differentiation marker in human breast cancer, CK7 had a controversial ex-pression in the epithelial and myoepithelial cells in canine mammary gland neoplasms. Based on the results, CK7 could not be considered as an independent marker for the canine mammary glands epithelial cell detection and a prognostic factor in canine mammary gland neoplasms


Article Title [فارسی]

ارزیابی بیان سیتوکراتین ۷ در نئوپلازی‌های مختلف غدد پستانی

Authors [فارسی]

  • الناز الهی راد 1
  • فرهنگ ساسانی 1
  • علیرضا خسروی 2
  • محمد جواد قراگزلو 1
  • فاطمه خان براری 3
1 گروه آسیب‌شناسی دانشکده دامپزشکی دانشگاه تهران، تهران، ایران
2 مرکز تحقیقات قارچ‌شناسی دانشکده دامپزشکی دانشگاه تهران، تهران، ایران
3 گروه ایمنولوژی، دانشکده پزشکی، دانشگاه علوم پزشکی شهید صدوقی، یزد، ایران
Abstract [فارسی]

زمینه مطالعه: سیتوکراتین‌ها فیلامان‌های بینابینی غیرانقباضی‌اند که در فعالیت‌های ساختمانی و اتصالی که در تشکیل شبکه‌ای برای حفاظت از سیتوپلاسم دخالت دارند. بیان سیتوکراتین ۷ در کارسینوم‌های پستان انسان مارکر مفیدی برای تمایز است اما بیان ان در تومورهای پستانی سگ به‌خوبی شناخته نشده است.
هدف: در بررسی تفکیک و تمایز تومور های پستانی در انسان بیومارکر CK7 نقش مفیدی دارد ولی نقش آن در تومور های پستانی سگ نا مشخص است.
روش کار: این مطالعه بر اساس مطالعه ایمنوهیستوشیمیایی سیتوکراتین ۷ در ۱۷ مورد نئوپلاسم پستان سگ اخذشده از گروه آسیب‌شناسی دانشکده دامپزشکی دانشگاه تهران و رنگ‌آمیزی تری‌کروم ماسون نمونه‌های مذکور برای نشان دادن رشته‌های کلاژن انجام شده است.
نتایج: پروتئین سیتوکراتین ۷ در هر دو سلول‌های اپیتلیالی (۱ تومور مختلط خوش‌خیم، ۱ فیبروآدنوما، ۱ کارسینوم کمپلکس، ۱1 کارسینوم مختلط) و میواپیتلیالی (۱ فیبروآدنوم، ۱ تومور مختلط خوش‌خیم، ۳ کارسینوم کمپلکس، ۱ کارسینوم مجرایی و ۱ کارسینوم مختلط) شناسایی شده است. رشته‌های کلاژن باریک و ضخیم در مقاطع رنگ‌آمیزی‌شده با تری‌کروم ماسون مشاهده شده است.
نتیجه‌گیری نهایی: با وجود استفاده از سیتوکراتین ۷ به‌عنوان مارکر تمایزی در سرطان پستان انسان، در بافت پستان سگ، سیتوکراتین ۷ بیان بحث‌برانگیز در سلول‌های اپیتلیالی و میواپیتلیالی داشت. بر اساس نتایج فوق، سیتوکراتین ۷ نمی‌تواند به‌عنوان مارکر مستقل برای شناسایی سلول‌های اپیتلیالی غدۀ پستانی سگ و فاکتور پیش‌آگهی‌دهنده در نئوپلاسم‌های پستان سگ در نظر گرفته شود.

Keywords [فارسی]

  • تری‌کروم ماسون
  • سگ
  • سلول میواپیتلیال
  • سیتوکراتین ۷
  • نئوپلاسم پستان


One of the most frequently occurring tumors in dogs is the mammary gland tumors (Salas et al., 2015; Pastor et al., 2018; Sharma et al., 2018). Histologically and clinically, there are similarities between canine mammary glands and human breast tissue. Hence, dogs can be considered as in vivo models for the human breast cancer (Cassali, 2013; Abdelmegeed and Mohammed, 2018). The architecture of mammary tissue is composed of three cell populations: luminal epithelial cells, basal/myoepithelial cells, and mesenchymal cells (Sorenmo et al., 2011). The wide heterogeneity of these tumors has increasingly made the classification scheme complex (Schlafer, 2016). In 6 out of 34 histological types of neoplasms, myoepithelial cell proliferation was demonstrated; three of them showed malignant transformation of myoepithelial (Goldschmidt et al., 2011).

Myoepithelial cells are well-developed in canine sweat and mammary glands (Beha et al., 2012; Ingthorsson et al., 2015). In mammary tissue, they form a continuous layer around ducts and teat sinuses (spindle-shaped) and a discontinuous layer around alveoli (star-shaped). They are located between luminal epithelial cells and stromal fibroblasts, which makes them suitable to link with these two cell populations and they have characteristics of muscle and epithelial cells (cytokeratins) (Beha et al., 2012).

Different types of epithelial cells synthesize filaments that are products of co-polymerization of at least one type I and one type II cytokeratin. These filaments which are abundant in the cytoplasm of epithelial cells are called soft keratin. They form intra-cytoplasmic cytoskeleton which causes resistance against mechanical stress (Cooper, 2000). Intermediate filaments are the most resistant filaments against mechanical stress among all filaments that form the cytoskeleton. Cytokeratins are important intermediate filaments that are categorized into six types according to the protein structure. Type I and II are acidic (CK-1, CK- 2, CK-3, CK-4, CK-5, CK-6, CK-7, CK-8) and basic (CK-9, CK-10, CK-11, CK-12, CK-13, CK-14, CK-15, CK-16, CK-17, CK-18, CK-19, CK-20) cytokeratins, respectively. Also, according to the molecular weight they are categorized into high molecular weight cytokeratins (squamous keratins) and low molecular weight cytokeratins (simple keratins) (Rekhtman, 2011). They are widely used as epithelial cell markers to assess the tissue origin of cancer (Jacob, 2018). Cytokeratin 7 (CK7) is a type II keratin (basic keratin, low molecular weight) which improves the integrity of cellular structures (Moll et al., 2008). It has been shown that this intermediate filament is expressed in normal canine apocrine glands (Espinosa De Los Monteros et al., 1999). Previously, CK7 expression was subjected to several studies on various canine tissues including cutaneous lesions and mammary glands (Pieper et al., 2015; Eivani and Mortazavi 2016). CK7 is used as a differential diagnosis marker for urothelial tumors, adenocarcinomas, bile duct and lung epithelia (Painter et al., 2010; Pieper et al., 2015; Luo et al., 2017). 

The routine staining method of hematoxylin and eosin (H&E) is incapable of distinguishing various eosinophilic tissue components, though special staining techniques such as Masson’s trichrome have proved to be cost-effective methods to demonstrate the presence and pattern of collagenous tissue. Despite being old-fashioned, this multi-colored method is still a popular stain for the connective tissue assessment in modern histology (Alturkistani et al., 2016). In this study, this method was used to investigate the increase in collagenous tissue, either as a desmoplasia (stromal cell proliferation) or probable changes in myoepithelial cell nature in the neoplastic process.

The current study aimed to investigate the expression of CK7 in canine normal and neoplastic mammary gland epithelial and myoepithelial cells and to evaluate its association with malignancy grade and histologic characteristics of neoplasms. Also, Masson’s trichrome was chosen to differentiate between smooth muscle fibers and collagen fibers.

Materials and Methods

 Samples and Histopathology

This study was based on 17 cases of canine mammary gland neoplasms (n=17) obtained from Department of Pathology, Faculty of Veterinary Medicine, University of Tehran. The samples were fixed in 10% neutral buffered formalin, then cut into 5 µm sections and stained with routine H&E and Masson’s trichrome methods. Histopathological types, subtypes, and grades were evaluated according to the Peña method (Peña, 2013) (Table 1). Histopathological variables included histological type and grade of tumors, intra-tumoral necrosis, and lymphovascular invasion.



Table 1. Method for grading. Modification of Pena method for histologic grading of canine mammary cancer

Type of   tumor

Criteria   for Histological Malignancy Grade

Total   scoring (A+B+C)

Grade of   Malignancy

A.Tubule   formation (of the specimen)

B.Nuclear   pleomorphism

C.Mitoses   per 10 HPF

Benign   tumors






Malignant   tumors

> 75%   (1)*

Uniform   nucleus and occasional nucleoli (1)*

0-9 (1)*



10-75% (2)*

Moderate variation in nuclear size   and shape, hyperchromatic nucleus and presence of nucleoli (2)*

10-19 (2)*



< 10%   (3)*

Marked   variation in nuclear size, hyperchromatic nucleus and prominent nucleoli (3)*

>20 (3)*



a * Points

bHPF, high-power field.

Source: Peña, L., De Andres, P.J., et al. (2013) Prognostic value of histological grading in noninflammatory canine mammary carcinomas in a prospective study with two‐year follow‐up: Relationship with clinical and histological characteristics. Vet Pathol 50:94–105. Reproduced with permission of SAGE Publications


Masson’s Trichrome Staining

After deparaffinization and rehydration, the slides were stained with hematoxylin for 10 min, then Biebrich scarlet-acid fuchsin solution for 2 min. After that they were placed in the phosphomolybdic-phosphotungstic acid solution for 10-15 min and aniline blue solution for 5 min. Finally, the slides were placed in 1% acetic acid for 3-5 min, and then dehydration and clearing were performed. 

Antibodies and Immunohistochemistry

Immunohistochemistry for CK7 was performed on formalin-fixed, paraffin-embedded (FFPE) tissue sections at 4 µm diameter. The procedure was performed using ready-to-use mouse monoclonal antibody (clone OV-TL 12/30 Novocastra Leica Biosystems), which recognizes human CK7 protein. A horseradish peroxidase (HRP) polymer detection method was used for the CK7 antibody. After deparaffinization and rehydration, antigen retrieval was carried out by microwave oven at 750 W and 180 W (citrate buffer, 15 min, pH=6, 0.01 M). To neutralize the endogenous peroxidase, all slides were incubated in hydrogen peroxide 3% for 5 min. To reduce the non-specific bindings, all slides were incubated in protein blocker reagents for 5 min. The slides were incubated with the primary antibody CK7, the post-primary antibody, and Novolink polymer (HRP) Each step for 1 h. Next, the diaminobenzidine (DAB) working solution was used as the chromogen for 10 min. Finally, the slides were counterstained in hematoxylin.

Positive and negative controls were set by adjacent normal mammary tissue and replacing the primary antibody by PBS. The semi-quantitative assessment was done based on the Ramalho et al. study (2006) by two observers through randomly chosen 30 fields at high power (×400). Tumors exhibiting brown cytoplasmic staining at >10% neoplastic cells were regarded as positive.


Two out of 17 tumors (11.76%) were benign, and 15 out of 17 were malignant. Among the malignant tumors (n=15), 10 were classified as grade I (66.66%) and 5 were classified as grade II (33.33%). In Masson’s trichrome staining, epithelial cells, myoepithelial cells stained red, and the nucleus of fibroblasts, chondroblasts, and osteoblasts were stained dark blue, and collagen fibers, bone, and cartilage matrices were stained blue. No muscular tissue was observed in the tumor sections. Summarized results of the histopathology and immunohistochemical assessments of the examined sections of neoplastic mammary tissues were presented in Table 2.



Table 2. Classification, grading, and immunohistochemical results of canine mammary gland tumors evaluation


Tumor type


Histological type








Microscopic features


CK7 expression


Tumor parenchyma

Tumor stroma



Benign (11.76%)


1 (5.88%)


Ductal, tubular pattern

Myxoid matrix



Benign mixed tumor

1 (5.88%)


Tubuloalveola, solid   sheet

Desmoplasia,   lymphoplasmacytic infiltration





Complex carcinoma

7 (41.17%)


Ductal, intraductal papillary   pattern,

Desmoplasia, lymphoplasmacytic   infiltration

- (41.17%)

- (n=5, 29.41%)

± (n=2, 11.76%)

1 (5.88%)


chondromucinous appearance,   tubuloalveolar with tubulopapillary pattern

± (5.88%)

+ (n=1, 5.88%)

Ductal carcinoma*

2 (11.76%)


Ductal, papillary   pattern, chondromucinous appearance

Desmoplasia, mixed   inflammatory infiltration

-  (n=2)

- (n=1, 5.88%)

+ (n=1, 5.88%)

Carcinoma mixed type

1 (5.88%)


Ductal, tubular pattern

Myxoid matrix,   lymphoplasmacytic infiltration, chondroid metaplasia



1 (5.88%)


Tubuloalveolar, tubular pattern




2 (11.76%)


Tubuloalveola,   tubulopapillary pattern*

Hemangiopericytoma-like   pattern, chondroid metaplasia, bone metaplasia



Carcinoma and malignant   myoepithelioma

1 (5.88%)


Tubuloalveola, tubular pattern

Desmoplasia, myxoid matrix




a *Local invasion

b numbers in the brackets are the relative frequencies.


H&E and Connective Tissue Staining Findings

Benign Neoplasms

The proliferation of ducts was observed in a hypocellular stroma in fibroadenoma and desmoplasia in the benign mixed tumor. In Masson’s trichrome staining, the nucleus of spindle and stellate stromal fibroblasts were demonstrated as dark blue, and the thick collagen fibers of fibroadenoma and benign mixed tumor were demonstrated as blue (Figure 1).

Malignant Neoplasms

The malignant neoplasms are divided into malignant epithelial neoplasms and mixed neoplasms. In complex carcinomas (n=8) (malignant neoplasms), the proliferation of both epithelial and myoepithelial cells, areas of myxoid matrix, cholesterol cleft, and lipofuscin laden macrophages were observed. In one ductal carcinoma, areas of squamous differentiation, foci of necrosis, myxoid matrix, cholesterol cleft, and mixed inflammatory infiltration were observed.In carcinoma and malignant myoepithelioma, nests and trabeculae of epithelial cells and malignant proliferation of myoepithelial cells with scant myxoid matrix, necrosis, cholesterol cleft, desmoplasia, and lipofuscin pigments were observed. Periductal, interlobular, and intralobular collagen fibers were demonstrated as blue fibers in Masson’s trichrome stain (Figure 1. A & D). This regular pattern was observed in 8 complex carcinomas, 2 ductal carcinomas, and 1 carcinoma and malignant myoepithelioma. In some tumors, intratumoral collagen fibers were recognized as fine distinct fibers interspersed between proliferative normal and malignant epithelial and myoepithelial cells (1 benign mixed tumor, 8 complex carcinomas, 1 carcinoma and malignant myoepithelioma) (Figure 1).





Figure 1. Canine mammary gland neoplasia. Intratumoral fine (arrows) and thick (pointed arrows) collagen fibers. Masson’s Trichrome. (A & B. Complex carcinoma, 100Î, C. Carcinoma and malignant myoepithelioma, 200Î & D. Fibroadenoma, 100Î).


In carcinoma mixed type (mixed neoplasms), foci of chondroid metaplasia with the proliferation of epithelial cells were observed. In one carcinosarcoma proliferation of epithelial cells, the malignant hemangiopericytoma-like proliferation pattern of stromal fibroblasts and areas of chondroid metaplasia and bone metaplasia were observed. In Masson’s trichrome staining, collagen fibers were recognized as fine distinct fibers in carcinoma mixed type and interlobular and intralobular fibers in carcinosarcomas.

CK7 Immunoreactivity

CK7-labeled epithelial cells were observed in a normal case and four cases of tumors [23.53%] (1 benign mixed tumor, 1 fibroadenoma, 1 complex carcinoma, and 1 carcinoma mixed type) (Figure 2. A & E) and myoepithelial cells in seven cases [41.17%] (1 fibroadenoma, 1 benign mixed tumor, 3 complex carcinomas, 1 ductal carcinoma, and 1 carcinoma mixed type) (Figure 2. C & F). Negative labeling of epithelial (Figure 2. B, D, & F) and myoepithelial (Figure 2. A, B, D, & E) cells were observed in 13 (76.47%) and 10 (58.82%) tumors, respectively (Table 2).




Figure 2. Canine mammary gland carcinoma. Note positive immunolabeling of CK7 in epithelial cells (A, 200Î & E, 100Î) and myoepithelial cells (C, 200Î & F, 400Î) and negative immunolabeling of CK7 in epithelial cells (B, 200Î & F) and myoepithelial cells (A, B & D, 400Î). Proliferative myoepithelial cells stained positive. (Epithelial cells: Pointed arrows, myoepithelial cells: Arrows, A, B & E. Complex carcinoma, C & F. Carcinoma mixed type, D. Carcinoma and malignant myoepithelioma). C & F are from the same samples.




Tumor studies could provide advantages in terms of the etiology, development, and treatment of cancer. Among all domesticated species, dog is more susceptible to tumor progression because of serving as a companion animal to humans and exposure to the same predisposing factors as humans (Salas et al., 2015; Baioni et al., 2017; Pastor et al., 2018). In veterinary medicine, several studies have focused on normal canine mammary gland histology and its morphological changes through the estrous cycle. There is some evidence to suggest that the mammary glands undergo various stages of growth (development and regression), depending on its hormonal environment (Santos et al., 2010; Sorenmo et al., 2011). Hence, it is not surprising that mammary neoplasms have a higher incidence in intact female dogs than in male dogs (Salas et al., 2015; Baioni et al., 2017).

Epithelial and myoepithelial tumors are dominant tumors of the mammary glands (Schlafer, 2016). The origin of these cell populations has not been elucidated, but a stem/progenitor cell and bipotent mammary precursors have been suggested (Visvader and Stingl, 2014). In mammary glands, epithelial and myoepithelial cells may have a polygonal morphology (Beha et al., 2012).

In the present study, the ratio of benign-to-malignant neoplasms was obtained 2:15. Various studies have reported different ratios (Baioni et al., 2017; Gabli et al., 2017). Conversely, in women, most breast masses detected on clinical examinations are categorized as benign breast disease. In this research, complex carcinomas had the highest number (8:17) among all histological types (Table 2). Peña et al. (2013) and Santos et al. (2013) reported complex carcinoma as the most frequent neoplasm of their studies (Peña et al., 2013; Santos et al., 2013).

In different neoplasms, epithelial cells reacted differently to CK7 antibody. In benign neoplasms and hyperplastic glands, epithelial cells mainly reacted with the CK7 antibody. Proliferative myoepithelial cells reacted positively to the CK7 antibody. This different immunoreactivity to CK7 antibody suggests that these cells have followed different paths in their keratin maturation or expression. Another possible explanation is related to the CK7 content in the epithelial and myoepithelial cells. Since CK7 content is low, it is not detectable until after proliferation or neoplastic transformation. The positive reactivity for CK7 in elongated myoepithelial cells could be indicative of gaining CK7 expression during progression toward chondromucinous tissue by these cells. 

It has been postulated that stellate motile myoepithelial cells change into fibroblasts due to the discontinuous labeling of Alpha-SMA, loss of CK14, CK5/6, and p63 expression (myoepithelial suprabasal markers) and retention of an affinity for vimentin (Ramalho et al., 2006).

Recent studies have focused on finding a trusted marker or panel of markers for myoepithelial cells, but none of them enjoyed 100% sensitivity and specificity, though p63 and CK14 have been recommended as suitable markers for myoepithelial cells (Gama et al., 2003; Beha et al., 2012; Moritani et al., 2015).

There is a growing body of evidence that is changing our point of view toward carcinogenesis to a more complicated process consisting of interactions between epithelial, myoepithelial, and mesenchymal (mainly extracellular matrix) elements (Sirka et al., 2018). Myoepithelial cells in the earliest stage of proliferation, form spherical masses in which cells with elongated to stellate nuclei are trapped in a slightly basophilic ground substance (chondromucinous appearance) (Moulton, 1978). In this stage, alkaline phosphatase reactivity is diminished. This process suggests that this phenomenon is caused by the collapse of the morphostat gradient and disruption of the tissue architecture (extracellular matrix) in a prolonged process (Baker et al., 2010). Similarly, the negative results of CK7 labeling in some tumors could be the result of this process which changes the immunoprofile of myoepithelial cells – «converted myoepithelial cells» (Baker et al., 2010). Regarding different benign to malignant tumor ratios in humans and dogs, authors suggest that different stages of neoplastic cell differentiation could affect differentiation marker expression and its expression pattern. 

Masson’s trichrome special stain was used to differentiate between smooth muscle cells and fibrous connective tissue. We did not observe smooth muscle cells in any of the samples.

The most notable observation in this study was the negative reaction of some epithelial cells to CK7 antibody in canine mammary gland neoplasms. Considering only the expression or lack of expression, CK7 expression is not significantly associated with tumor type and grade. Asimilar observation  was reported by Eivani and Mortazavi (2016). Despite using cytokeratins as epithelial cell marker in human breast cancer, their role in canine mammary gland tumor prognosis is not clear.



In conclusion, CK7 could not be considered as an independent marker for the canine mammary glands epithelial cell detection and a prognostic factor in canine mammary gland neoplasms, contrary to human.



The authors wish to thank Amol’s pathobiological lab for their staining technical support and Dr. Mohammad Taghi Sheibani for his support in preparing photomicrographs.


Conflict of Interest

Authors declared no conflicts of interest.







Abdelmegeed, S. M. and S. Mohammed (2018). Canine mammary tumors as a model for human disease. Oncol Lett. 15(6), 8195-8205.  [DOI:10.3892/ol.2018.8411] [PMID] [PMCID]
Alturkistani, H. A., F. M. Tashkandi and Z. M. Mohammedsaleh (2016). Histological stains: a literature review and case study. Glob J Health Sci. 8(3), 72.  [DOI:10.5539/gjhs.v8n3p72] [PMID] [PMCID]
Baioni, E., E. Scanziani, M. C. Vincenti, M. Leschiera, E. Bozzetta, M. Pezzolato, et al. (2017). Estimating canine cancer incidence: findings from a population-based tumour registry in northwestern Italy. BMC Vet Res. 13(1), 203. [DOI:10.1186/s12917-017-1126-0] [PMID] [PMCID]
Baker, S. G., A. Cappuccio and J. D. Potter (2010). Research on early-stage carcinogenesis: Are we approaching paradigm instability? J Clin Oncol. 28(20), 3215. [DOI:10.1200/JCO.2010.28.5460] [PMID] [PMCID]
Beha, G., Sarli, G., Brunetti, B., Sassi, F., Ferrara, D., & Benazzi, C. (2012). Morphology of the myoepithelial cell: immunohistochemical characterization from resting to motile phase. The Scientific World Journal, 2012.  [DOI:10.1100/2012/252034] [PMID] [PMCID]
Cassali, G. D. (2013, April). Comparative mammary oncology: canine model. In BMC proceedings (Vol. 7, No. 2, pp. 1-2). BioMed Central. [DOI:10.1186/1753-6561-7-S2-K6] [PMID] [PMCID]
Costa, G. M., Araujo, S. L., Xavier Júnior, F. A. F., Morais, G. B. D., Silveira, J. A. D. M., Viana, D. D. A., & Evangelista, J. S. A. M. (2019). picrosirius red and masson’s trichrome staining techniques as tools for detection of collagen fibers in the skin of dogs with endocrine dermatopathologies. Ciência Animal Brasileira20. [DOI:10.1590/1089-6891v20e-55398]
Eivani, D. and P. Mortazavi (2016). The relationship between basal and luminal cytokeratins with histopathologic characteristics of canine mammary gland cancer. Pol J Vet Sci. 19(2):261-9. [DOI:10.1515/pjvs-2016-0033] [PMID]
Espinosa De Los Monteros, A., A. Fernandez, M. Millan, F. Rodriguez, P. Herraez and J. Martín De Las Mulas (1999). Coordinate expression of cytokeratins 7 and 20 in feline and canine carcinomas. Vet Pathol. 36(3): 179-190. [DOI:10.1354/vp.36-3-179] [PMID]
Gabli, Z., L. Beddar, Z. Djerrou and E. Gomez (2017). Prevalence and histopathologic analyses of mammary tumors in female dogs in the Northeast of Algeria. J Biol Sci. 17, 166-177. [DOI:10.3844/ojbsci.2017.166.177]
Gama, A., A. Alves, F. Gartner and F. Schmitt (2003). p63: a novel myoepithelial cell marker in canine mammary tissues. Vet Pathol. 40(4), 412-420. [DOI:10.1354/vp.40-4-412] [PMID]
Goldschmidt, M., L. Peña, R. Rasotto and V. Zappulli (2011). Classification and grading of canine mammary tumors. Vet Pathol. 48(1), 117-131.  [DOI:10.1177/0300985810393258] [PMID]
Ingthorsson, S., B. Hilmarsdottir, J. Kricker, M. K. Magnusson and T. Gudjonsson (2015). "Context-Dependent Function of Myoepithelial Cells in Breast Morphogenesis and Neoplasia. Curr Mol Biol Rep. 1(4), 168-174. [DOI:10.1007/s40610-015-0027-x] [PMID] [PMCID]  
Lo, P.-K., Y. Zhang, Y. Yao, B. Wolfson, J. Yu, S.-Y. Han, et al. (2017). Tumor-associated myoepithelial cells promote the invasive progression of ductal carcinoma in situ through activation of TGFβ signaling. J Biol Chem. 292(27), 11466-11484.  [DOI:10.1074/jbc.M117.775080] [PMID] [PMCID]
Luo, H.-T., C.-X. Liang, R.-C. Luo and W.-G. Gu (2017). Identification of relevant prognostic values of cytokeratin 20 and cytokeratin 7 expressions in lung cancer. Biosci Rep. 37(6), [DOI:10.1042/BSR20171086] [PMID] [PMCID]  
Makki, J., O. Myint, A. A. Wynn, A. T. Samsudin and J. Daisy Vanitha (2015). Expression distribution of cancer stem cells, epithelial to mesenchymal transition, and telomerase activity in breast cancer and their association with clinicopathologic characteristics. Clin Med Insights Pathol. 8, CPath. S19615.  [DOI:10.4137/CPath.S19615]
Moll, R., M. Divo and L. Langbein (2008). The human keratins: biology and pathology. Histochem Cell Biol. 129(6): 705-33.    [DOI:10.1007/s00418-008-0435-6] [PMID] [PMCID]
Moritani, S., S. Ichihara, Y. Yatabe, M. Hasegawa, A. Iwakoshi, W. Hosoda, et al. (2015). Immunohistochemical expression of myoepithelial markers in adenomyoepithelioma of the breast: a unique paradoxical staining pattern of high-molecular weight cytokeratins. Virchows Arch. 466(2), 191-198.   [DOI:10.1007/s00428-014-1687-2] [PMID]
Moulton, J. E. and J. E. Moulton (1978). Tumors in Domestic Animals, University of California Press.
Painter, J., N. Clayton and R. Herbert (2010). Useful immunohistochemical markers of tumor differentiation. Toxicol Pathol. 38(1), 131-141. [DOI:10.1177/0192623309356449] [PMID] [PMCID]
Pandey, P. R., J. Saidou and K. Watabe (2010). Role of myoepithelial cells in breast tumor progression. Front Biosci. 15, 226. [DOI:10.2741/3617] [PMID] [PMCID]
Pastor, N., N. C. Caballé, M. Santella, L. J. Ezquerra, R. Tarazona and E. Duran (2018). Epidemiological study of canine mammary tumors: age, breed, size and malignancy. Austral J Vet Sci. 50(3), 143-147.   [DOI:10.4067/S0719-81322018000300143]
Peña, L., P. D. Andrés, M. Clemente, P. Cuesta and M. Perez-Alenza (2013). Prognostic value of histological grading in noninflammatory canine mammary carcinomas in a prospective study with two-year follow-up: relationship with clinical and histological characteristics. Vet Pathol. 50(1), 94-105.  [DOI:10.1177/0300985812447830] [PMID]  
Pieper, J. B., A. W. Stern, S. M. LeClerc and K. L. Campbell (2015). Coordinate expression of cytokeratins 7 and 14, vimentin, and Bcl-2 in canine cutaneous epithelial tumors and cysts. J Vet Diagn Invest. 27(4), 497-503. [DOI:10.1177/1040638715594115] [PMID]
Ramalho, L. N. Z., A. Ribeiro-Silva, G. Cassali and S. Zucoloto (2006). The expression of p63 and cytokeratin 5 in mixed tumors of the canine mammary gland provides new insights into the histogenesis of these neoplasms. Vet Pathol. 43(4), 424-429. [DOI:10.1354/vp.43-4-424] [PMID
Rekhtman, N., & Bishop, J. A. (2011). Quick reference handbook for surgical pathologists (p. 180). Berlin: springer. ISBN-13: 978-3642200854. [DOI:10.1007/978-3-642-20086-1]
Røsland, G. V., S. E. Dyrstad, D. Tusubira, R. Helwa, T. Z. Tan, M. L. Lotsberg, et al. (2019). Epithelial to mesenchymal transition (EMT) is associated with attenuation of succinate dehydrogenase (SDH) in breast cancer through reduced expression of SDHC. Cancer Metabol. 7(1), 6.  [DOI:10.1186/s40170-019-0197-8] [PMID] [PMCID]
Salas, Y., A. Márquez, D. Diaz and L. Romero (2015). Epidemiological study of mammary tumors in female dogs diagnosed during the period 2002-2012: A growing animal health problem. PLoS One 10(5), e0127381.  [DOI:10.1371/journal.pone.0127381] [PMID] [PMCID]
Santos, A. A., C. C. Lopes, J. R. Ribeiro, L. R. Martins, J. C. Santos, I. F. Amorim, et al. (2013). Identification of prognostic factors in canine mammary malignant tumours: a multivariable survival study. BMC Vet Res. 9(1), 1.   [DOI:10.1746.1186-6148-9-1] [PMID] [PMCID]
Santos, M., R. Marcos and A. Faustino (2010). Histological study of canine mammary gland during the oestrous cycle. Reprod Domest Anim. 45(5), e146-e154.  [DOI:10.1111/j.1439-0531.2009.01536.x]
Schlafer, D. H., Foster, R. A. (2016). Pathology of Female genital system. Pathology of domestic animals. M. G. Maxie. USA, Elsevier. 3,912. [DOI:10.1016/B978-0-7020-5319-1.00015-3] [PMCID]
Sharma, N., A. Gupta, R. Bhat, M. Yatoo and O. Parray (2018). Epidemiology and Treatment of Canine Mammary Tumours in Jammu Region of India. J Dairy Vet Anim Res, 7(2), 59-62. [DOI:10.15406/jdvar.2018.07.00190]
Sirka, O. K., E. R. Shamir and A. J. Ewald (2018). Myoepithelial cells are a dynamic barrier to epithelial dissemination. Int J Cell Biol. 217(10), 3368-3381.  [DOI:10.1083/jcb.201802144] [PMID] [PMCID]
Sorenmo, K., R. Rasotto, V. Zappulli and M. Goldschmidt (2011). Development, anatomy, histology, lymphatic drainage, clinical features, and cell differentiation markers of canine mammary gland neoplasms. Vet. Pathol. 48(1), 85-97.  [DOI:10.1177/0300985810389480] [PMID]
Visvader, J. E. and J. Stingl (2014). Mammary stem cells and the differentiation hierarchy: current status and perspectives. Genes Dev. 28: 1143-1158.  [DOI:10.1101/gad.242511.114] [PMID] [PMCID]