Sage Journals: Discover world-class research

Abstract

Feline mammary carcinomas are highly aggressive neoplasms. Several mechanisms are thought to be involved in their progression, including the loss of epithelial adhesion molecules. The present study was carried out on 21 adenomas and 139 mammary carcinomas. Of the carcinomas, 66 were not reported to have metastasized, while the remaining 73 had evidence of regional lymph node metastasis at the moment of diagnosis. The relationship was examined between the expression of the E-cadherin–β-catenin complex and basal (CK5/6, CK14) and luminal (CK8/18) cytokeratin expression. In the medical literature, carcinomas expressing basal cytokeratins are reported as having a poor prognosis in human breast cancer. Results revealed that preservation of the expression of E-cadherin and β-catenin is a significant feature of carcinomas without metastasis, whereas carcinomas with metastasis reveal the loss of one or both adhesion molecules. Additionally, basal cytokeratin expression was statistically associated with the presence of regional metastasis. Furthermore, the expression of E-cadherin–β-catenin was significantly correlated with the high expression of CK18 and low expression of CK5/6.

Keywords

feline mammary carcinoma immunohistochemistry E-cadherin β-catenin basal cytokeratins

Mammary neoplasia is the third-most-common tumor type affecting female cats, following hemopoietic neoplasms and skin tumors. Malignancy is frequently reported and presents with local and distant metastases and high mortality rates.⁴² Based on age, incidence, risk factors, histopathology, prognostic aspects, metastatic pattern, and response to therapy, feline mammary carcinomas have been proposed as a good model for human breast cancer.^2,42 Similar to reports in women, many veterinary studies describe regional lymph node involvement as one of the most important prognostic factors for feline mammary tumors.^11,16

Recent studies of human breast cancer by means of gene expression have defined different subgroups of breast tumors with distinct molecular signatures (luminal A, luminal B, Her-2+, or basal-like).²⁸ The same classification has also been made by immunohistochemistry.²⁵ Some of the subgroups, especially those described as basal-like, are correlated with an unfavorable prognosis.²⁷ This basal-like subtype includes carcinomas expressing markers that are characteristic of the myoepithelium of the normal mammary gland, such as cytokeratins CK14, CK5/6, and CK17 or epidermal growth factor receptor.¹⁴ Additionally, these carcinomas have been noted to display decreased levels of luminal cytokeratins (CK8/18).¹³ Several theories have been offered to explain the ability of epithelial tumoral cells to infiltrate regional circulation and establish new neoplastic colonies distant from the primary mass. Carcinomas may invade in the form of multicellular aggregates in a process known as collective cell migration, in which the carcinoma cells may retain their epithelial characteristics.⁸ Also, epithelial mesenchymal transition (EMT) may play a role in tumor invasion and metastasis.³⁹ EMT is an essential developmental process in which cells of epithelial origin lose their epithelial characteristics and polarity and acquire a mesenchymal phenotype with increased migratory behavior. This change is characterized by a loss of intercellular adhesion molecules (eg, E-cadherin) and the upregulation of mesenchymal markers, mainly vimentin.⁴⁰ Other changes, such as changes in location of β-catenin, have been associated with EMT.⁴ Coming full circle, a recent study has established that the basal-like phenotype (expressing CK5/6 or CK14 and with decreased expression of CK8/18) is especially prone to undergoing EMT in human breast cancer.³⁴

In the field of veterinary pathology, the loss of E-cadherin expression and the loss of its cytoplasmatic molecular anchorage, β-catenin, are well-established prognostic factors in canine mammary carcinomas.^1,5,10 However, few studies have been published on the expression of EMT markers (E-cadherin, β-catenin, or vimentin) in feline mammary carcinomas.^6,18,38 Those reports in the literature point to a high percentage of mammary carcinomas expressing vimentin as well as having a reduction of E-cadherin in the cell membranes of the neoplastic cells. Both findings could point to the existence of EMT phenomena in feline mammary carcinomas. However, no studies have been performed in feline carcinomas to establish the existence of molecular subtypes similar to those described in human breast cancer. The aim of the present work was to study a wide panel of feline mammary tumors, with and without metastasis at the moment of diagnosis, to ascertain the expression of basal and luminal cytokeratins as molecular subtype biomarkers. The relationship of the expression of these cytokeratins was analyzed in correlation with degree of expression of EMT markers, such as E-cadherin, β-catenin, and vimentin.

Material and Methods

Tumors

In sum, 160 feline mammary tumors (21 adenomas and 139 carcinomas) were selected for the present study from among the 6113 feline mammary tumors (1054 adenomas and 5059 carcinomas) submitted to the Histovet diagnosis laboratory over a period of 6 years (2004–2009). Cases of multiple tumors from the same animal or recidivant neoplasms were discarded from the study. Diagnosis of adenoma or carcinoma was performed independently by 2 board certified veterinary pathologists (J.S. and M.V.). Carcinomas with diagnoses of metastasis were selected only where there was sufficient tissue in the lymph node to conclusively demonstrate the presence or absence of metastasis. Lymph node metastases were assessed by evaluation of hematoxylin and eosin–stained slides as well as immunohistochemistry using a pancytokeratin monoclonal antibody, as described below.²⁰ Among the selected carcinomas, 73 displayed regional metastasis at the moment of diagnosis, while in another 66 carcinomas, there was no regional metastasis. Carcinomas were classified according to their histologic pattern²⁴ as tubulopapillar (46 of 139), cribiform (62 of 139), or solid (31 of 139). Among the metastasizing carcinomas, the histologic pattern distribution consisted of the following: tubulopapillar, 31.5%; cribiform, 46.6%; and solid, 21.9%. Among the nonmetastasizing carcinomas, the breakdown of histologic patterns included tubulopapillar, 34.9%; cribiform, 42.4%; and solid, 22.7%. The gradation of the carcinomas was carried out following the modification of the Nottingham method for human breast cancer⁷ described by Castagnaro et al.³

Immunohistochemistry

Selected tissues were sectioned at 4 μm from formalin-fixed and paraffin-embedded samples from primary adenomas and carcinomas as well as from those tumors with lymph node metastases. All tissues were stained by the avidin–biotin–peroxidase complex method (ABC) as previously described³² with several primary antibodies, whose clone, source, and pretreatment for antigen retrieval are summarized in Table 1. Counterstaining involved Mayer’s hematoxylin. Normal feline mammary gland and skin served as positive control for each immunohistochemical series. Negative controls were carried out by replacing the primary antibody with phosphate buffered saline.

Table 1.

Characteristics of the Primary Antibodies

Antigen	Clone	Laboratory^a	Pretreatment
Pancytokeratin	AE1/AE3	Dako	Pronase 0.1%
Vimentin	V9	Santa Cruz Biotech	Heat (121°C in citrate buffer, pH 6, 5 min)
E-cadherin	NCH-38	Dako	Heat (121°C in citrate buffer, pH 6, 12 min)
β-catenin	β-catenin-1	Dako	Heat (121°C in citrate buffer, pH 6, 8 min)
Cytokeratin 5/6	D5/16 B4	Dako	Heat (121°C in citrate buffer, pH 6, 5 min)
Cytokeratin14	LL002	Thermo Scientific	Heat (121°C in citrate buffer, pH 6, 5 min)
Cytokeratin 8/18	5D3	Thermo Scientific	Heat (121°C in citrate buffer, pH 6, 5 min)

^aDako, Carpinteria, California. Santa Cruz Biotech, Santa Cruz, California. Thermo Scientific, Rockford, Illinois.

Quantification of Immunolabeling

E-cadherin and β-catenin immunoreactivity was classified as membranous (localized in the cell–cell boundaries, in cases in which the immunoreactivity was stronger than in the cytoplasm), cytoplasmatic (uniformly distributed through the cytoplasm, with no recognizable distinction between membrane and cytoplasm), or nuclear. Cases were grouped according to Brunetti et al¹ as preserved, when positivity was membranous and occurred in more than 75% of the neoplastic epithelial cells, and reduced in all the remaining samples, including negative tumors. A third variable was obtained by integrating the data from both proteins and so named E-cad/β-cat. The preserved expression of this variable included cases in which both proteins had a preserved status, whereas E-cad/β-cat was classified as reduced when the expression of at least one protein was reduced.

Basal (CK5/6 and CK14) cytokeratin expression was assessed semiquantitatively according to the percentage of immunoreactive cells: negative, <1% of positive cells; mild, 1%–10%; moderate, 10%–50%; and strong >50%. To obtain dichotomic variables to compare with other tumor features and following the definition of “basal-like” proposed by Rakha et al²⁹ in human breast cancer, carcinomas were classified as low expression (<10%; ie, negative or mild labeling) or high expression (>10%; ie, moderate or strong labeling).

Luminal (CK8/18) cytokeratin was assessed semiquantitatively according to the percentage of immunoreactive cells: negative, <10% of positive cells; mild, 10%–50%; moderate, 50%–75%; and strong, >50%. As with the basal cytokeratin expression assay, dichotomic variables were achieved to compare these with other tumor features, and in accordance with Sarrio et al,³⁴ carcinomas were classified as low expression (<50%; ie, negative or mild labeling) or high expression (>50%; ie, moderate or strong labeling).

Statistical Analysis

The association between the different markers was examined using the Pearson χ² test. For each association, the odds ratio was calculated with a confidence interval of 95%. In the particular case of tumoral grade (1–3), its association with the other markers was examined with a Mann-Whitney U test. Analyses were performed with SPSS 15.0. A P value < .05 was considered to be statistically significant.

Results

Immunohistochemical Analysis of the Adenomas

Immunolabeling with the markers used in the present study showed similar results for the 21 adenomas analyzed. The labeling pattern showed the preserved membranous expression of E-cadherin and β-catenin. Both basal CKs (CK5/6 and CK14) stained basally located cells as well as a limited number of luminal cells, while the luminal CK8/18 stained only luminal cells, predominantly in their apical cytoplasm. Vimentin expression was found in the epithelial cells in 1 of the adenomas studied.

Tumoral Grade

The 139 carcinomas studied were graded using the modified Nottingham method. This revealed 38 grade 1, 65 grade 2, and 36 grade 3 carcinomas. Among those that displayed metastases at diagnosis, 12 of 73 (16.4%) were grade 1, 40 of 73 (54.8%) were grade 2, and 21 of 73 (28.8%) were grade 3 carcinomas. Among those without metastasis, 26 of 66 (39.5%) were grade 1, 25 of 66 (37.8%) were grade 2, and 15 of 66 (22.7%) were grade 3 carcinomas. A statistically significant association (P = .019) was established between the existence of metastasis and grade of the carcinomas.

Immunohistochemical Analysis of E-cadherin and β-catenin Expression in Carcinomas

Analysis of the results obtained by labeling both molecules showed that E-cadherin expression was preserved in 51.8% of the carcinomas while β-catenin expression was preserved in 33.8% of the carcinomas. Only 26.6% of the carcinomas preserved the combined expression of E-cad/β-cat. When the data relating to adhesion molecules in the primary tumors were evaluated in tandem with the presence of regional metastasis, a significant association between the existence of metastasis in the carcinomas and the preserved/reduced expression of both molecules could be established using a χ² test (Table 2, Fig. 1). Analysis of the odds ratio demonstrated a negative correlation between the expression of E-cadherin, β-catenin, or their combined expression and the existence of metastasis, with this negative association being stronger for the combined expression of E-cad/β-cat (Table 2).

Figure 1. Mammary gland; cat. Reduced expression of E-cadherin in the primary focus of a metastasizing feline mammary carcinoma. ABC technique with Mayer’s hematoxylin. Figure 2. Mammary gland; cat. High expression of CK14 in the primary focus of a metastasizing feline mammary carcinoma. ABC technique with Mayer’s hematoxylin. Figure 3. Mammary gland; cat. Low expression of CK8/18 in the primary focus of a metastasizing a feline mammary carcinoma. ABC technique with Mayer’s hematoxylin. Figure 4. Mammary gland; cat. Association between high expression of CK5/6 and reduced expression of the E-cadherin–β-catenin complex in metastasizing feline carcinomas. (A) High expression of Figure 4. (continued) CK 5/6 in the primary focus of the carcinoma. (B) Reduced expression of E-cadherin. (C) Reduced expression of β-catenin. ABC technique with Mayer’s hematoxylin. Figure 5. Mammary gland; cat. Association between high expression of CK8/18 and preserved expression of the E-cadherin–β-catenin complex in nonmetastasizing feline carcinomas. (A) High expression of CK 8/18. (B) Preserved expression of E-cadherin. (C) Preserved expression of β-catenin. ABC technique with Mayer’s hematoxylin.

Table 2.

Statistical Association Between Expression of E-cadherin, β-catenin, or Their Combination and the Existence Metastasis at the Time of Diagnosis

	Carcinomas, n (%)
Expression	Metastasizing	Nonmetastasizing	n	P	Odds Ratio (95% Confidence Interval)
E-cadherin				.003	0.610 (0.43–0.85)
Preserved	29 (40.2)	43 (59.8)	72
Reduced	44 (65.7)	23 (34.3)	67
β-catenin				.001	0.424 (0.23–0.70)
Preserved	15 (31.9)	32 (69.4)	47
Reduced	58 (63.1)	34 (36.9)	92
E-cad/β-cat				< .001	0.335 (0.17–0.63)
Preserved	10 (27.0)	27 (73.0)	37
Reduced	63 (61.8)	39 (38.2)	102
Total	73	66	139

Additionally, when data about molecular adhesion expression were related with the grade of carcinomas using the Mann-Whitney U test, a statistically significant correlation was found between the preservation of β-catenin and E-cad/β-cat and tumoral grade (P = .023 and P = .026, respectively). Grade 1 carcinomas displayed significantly higher preserved expression of both β-catenin and E-cad/β-cat than grade 2 and grade 3 carcinomas. No statistically significant association was found between E-cadherin and tumoral grade (P = .243).

Immunohistochemical Analysis of CK8/18, CK14, and CK5/6 Expression in Carcinomas

Analysis of the results obtained for CK8/18 pointed to negative labeling in 11.5% of the carcinomas, mild labeling in 48.2%, moderate labeling in 22.2%, and strong labeling in 18.1%. Using the terms of low and high expression, the percentage of low-expression carcinomas was 59.7% and high-expression carcinomas, 40.3%. When the labeling of basal cytokeratins was analyzed, the results for CK14 were as follows: negative labeling in 7.2% of carcinomas, mild labeling in 15.8%, moderate labeling in 54.7%, and strong labeling in 22.3% (ie, 23% low-expression carcinomas and 77% high-expression carcinomas). Results for CK5/6 were 38.1% for negative labeling, 27.3% for mild labeling, 23.1% for moderate labeling, and 11.5% for strong labeling so that 65.4% of the carcinomas displayed low expression of CK5/6 while 34.6% of the carcinomas displayed high expression.

Data for basal cytokeratin displayed a significant association between the existence of metastasis in the carcinomas and the expression of both CK5/6 and CK14 using the χ² test (Table 3, Fig. 2). Analysis of the odds ratio demonstrated a positive correlation between the high expression of CK5/6 or CK14 and the existence of metastasis (Table 3). A significant association between the existence of metastasis in the carcinomas and the expression of the CK8/18 luminal cytokeratin was established using a χ² test (Table 3, Fig. 3). Analysis of the odds ratio demonstrated a negative correlation between the high expression of CK8/18 and the existence of metastasis (Table 3).

Table 3.

Statistical Association Between Expression of Basal and Luminal Cytokeratins and the Existence of Metastasis at the Time of Diagnosis

	Carcinomas, n (%)
Expression	Metastasizing	Nonmetastasizing	n	P	Odds Ratio (95% Confidence Interval)
CK-5/6				.039	1.649 (1.01–2.68)
High	31 (64.6)	17 (35.4)	48
Low	42 (46.1)	49 (53.9)	91
CK-14				.001	1.336 (1.10–1.61)
High	65 (59.6)	44 (40.4)	109
Low	8 (26.7)	22 (77.3)	30
CK-8/18				.004	0.542 (0.35–0.83)
High	21 (37.5)	35 (62.5)	56
Low	52 (62.7)	31 (37.3)	83
Total	73	66	139

Additionally, when data for CKs expression were related with the grade of carcinomas using the Mann-Whitney U test, an association was found between the CK5/6 and CK14 positive labeling and tumoral grade (P = .005 and P = .037, respectively). Grade 1 carcinomas displayed significantly lower labeling of both basal CKs than grade 2 and grade 3 carcinomas. No statistically significant association was found between CK8/18 labeling and tumoral grade (P = .237).

Relationship Between E-cadherin and β-catenin Expression and Basal and Luminal Cytokeratins

The relationship between the E-cadherin–β-catenin complex and luminal and basal cytokeratin expression was analyzed using the χ² test. A significant association was observed between the existence of positive labeling for CK5/6 in the carcinomas and the preserved/reduced expression of both E-cadherin and β-catenin and their combination using a χ² test (Table 4, Fig. 4). Analysis of the odds ratio demonstrated a negative correlation between the expression of E-cadherin and the combined expression and the existence of positive labeling for CK5/6 (Table 4). Analysis of labeling for CK8/18 in the carcinomas also demonstrated significant association with the preserved/reduced expression of both E-cadherin and β-catenin and their combination (Table 5, Fig. 5). Analysis of the odds ratio demonstrated a positive correlation between the expression of E-cadherin and β-catenin and the existence of positive labeling for CK8/18 (Table 5).

Table 4.

Statistical Association Between Preserved/Reduced Expression of E-cadherin, β-catenin, or Their Combination and Cytokeratin 5/6 Expression

	CK5/6, n (%)
Expression	High Expression	Low Expression	n	P	Odds Ratio (95% Confidence Interval)
E-cadherin				.014	0.558 (0.34–0.90)
Preserved	18 (25.0)	54 (75.0)	72
Reduced	30 (44.8)	37 (55.2)	67
β-catenin				.049	0.582 (0.32–1.03)
Preserved	11 (23.4)	36 (76.6)	47
Reduced	37 (40.2)	55 (59.8)	92
E-cad/β-cat				.020	0.471 (0.23–0.95)
Preserved	7 (18.9)	30 (81.1)	37
Reduced	41 (40.2)	61 (59.8)	102
Total	73	66	139

Table 5.

Statistical Association Between Preserved/Reduced Expression of E-cadherin, β-catenin, or Their Combination and Cytokeratin 8/18 Expression

	CK8/18, n (%)
Expression	High Expression	Low Expression	n	P	Odds Ratio (95% Confidence Interval)
E-cadherin				.016	1.675 (1.08–2.58)
Preserved	36 (50.0)	36 (50.0)	72
Reduced	20 (29.8)	47 (70.1)	67
β-catenin				.001	1.957 (1.32–2.88)
Preserved	28 (59.6)	19 (40.4)	47
Reduced	28 (30.4)	64 (69.6)	92
E-cad/β-cat				.017	1.654 (1.12–2.44)
Preserved	21 (18.9)	16 (81.1)	37
Reduced	35 (40.2)	67 (59.8)	102
Total	73	66	139

Immunohistochemical Analysis of Vimentin Expression in Carcinomas

Vimentin expression was found in the epithelial cells of 70.5% of the studied carcinomas (98 of 139). Coexpression of cytokeratin and vimentin could be observed in most of tumors (Fig. 6). No statistically significant association could be established between the expression of vimentin and metastasis or between the expression of vimentin and the other biomarkers analyzed in the present study.

Figure 6.

Mammary gland; cat. Presence of coexpression of cytokeratin and vimentin in nonmetastasizing feline carcinomas. (A) Labeling with anti-pancytokeratin antibody. (B) Labeling with anti-vimentin antibody. ABC technique with Mayer's hematoxylin.

Discussion

The present study points to the correlation of 2 groups of biomarkers, E-cadherin–β-catenin complex and basal and luminal cytokeratins, with the presence of regional metastasis, a well-established feature of malignancy in feline mammary carcinomas. Obviously, every carcinoma has the potential to metastasize, and the absence of metastasis described in this study should be understood as a phase in the process of tumor progression, which does not preclude possible future changes in antigen expression or in metastasis status. Notwithstanding, expression of the studied antigens and their association with the presence of metastases provides valuable knowledge for understanding the evolution of mammary neoplasms. Furthermore, expression of the antigens studied in the present work was associated not only with the presence of metastasis at the moment of diagnosis but also with tumoral grade, recently described as an independent prognostic factor associated with overall survival and disease-free survival in feline mammary carcinomas.³⁶ In fact, an association between the expression of basal cytokeratins and tumoral grade was found in which grade 2 and 3 carcinomas showed higher expression of these cytokeratins. An association was also found between combined E-cadherin and β-catenin preservation and tumoral grade, with grade 1 carcinomas having shown the preservation of both molecules.

The loss of E-cadherin, which permits the detachment of the epithelial tumoral cells and their movement, has been associated with tumor invasiveness and metastasis as part of the EMT process.³⁹ Several studies of canine mammary tumors have reported that loss of E-cadherin, alone or in combination with loss of β-catenin, is associated with the presence of metastasis or with a shorter overall survival period.^1,5,10 Studies on E-cadherin expression in feline mammary tumors are scarce. An earlier study examining normal mammary tissue, hyperplastic lesions, and a number of neoplasms (4 adenomas and 14 carcinomas) reported that the loss of E-cadherin and abnormalities in the pattern of membranous location were associated with some of the carcinomas but not with adenomas.⁶ Another study, using 8 cell lines from feline mammary tumors, showed the reduced expression of E-cadherin in cell lines obtained from a regional lymph node metastasis.³⁸ Our findings for E-cadherin confirm these previous reports: The highest percentage of reduction was observed in carcinomas with metastasis, while no reduction was observed in adenomas. The present study shows that a high percentage of carcinomas, mainly in those with regional metastasis, had a reduced expression of β-catenin, which suggests the presence of nonfunctional E-cadherin in some of the tumors that preserve the expression of this adhesion molecule. Loss of β-catenin has been described in human invasive lobular carcinomas.³⁰ Indeed, the association observed between the combined preservation of E-cad/β-cat and metastasis was stronger than in each of the individual biomarkers, suggesting that this variable would be a better prognostic factor that the study of E-cadherin preservation alone.

Basal cytokeratins are high molecular weight cytokeratins associated with the basal layers of the stratified epithelium. In the mammary gland, these cytokeratins are also expressed in the basally located myoepithelial cell layer, in a small proportion of luminal gland cells and in the so-called mammary stem or a progenitor cells.¹³ Thus, “basalness” has a dual meaning and could be related with a myoepithelial or progenitor/stem cell origin. The discovery in human breast cancer of in situ carcinomas expressing basal cytokeratins seems to rule out the myoepithelial origin of basal-like tumors.¹⁵ Mammary progenitor or stem cells are also characterized by a low expression of CK8/18.³⁷ The possible association between EMT and the existence of a basal-like phenotype has recently been demonstrated in a study of human breast cancer, suggesting that EMT may not be a sign of overall tumor dedifferentiation but, rather, the manifestation of a specific phenotype of stem cell origin showing more aggressive behavior.³⁴ In fact, cells with a basal phenotype probably have a higher proclivity to develop EMT changes and subsequently undergo invasion. This feature has not been studied in feline mammary tumors, and our results point to another interesting similarity between human and feline mammary tumors.

The results related with the immunolabeling of vimentin demonstrated that a high percentage of carcinomas expressed this intermediate filament in contrast with the negative labeling found in most of the adenomas. The percentage of vimentin-positive carcinomas is similar to that found in other studies carried out in feline mammary carcinomas.^18,33 However, no association was found between the expression of vimentin and well-established features of malignancy, such as regional metastasis or tumoral grade, unlike other EMT markers, such as the reduced expression of E-cadherin or β-catenin. The results described in the present study suggest that vimentin expression could be considered as a hallmark of carcinoma but not as a marker of invasiveness. In reports studying human breast cancer, some authors claim that vimentin is one of the first steps in carcinoma progression and that these vimentin positive cells are prone to develop subsequent EMT changes, such as loss of cadherins.¹⁷

In the field of veterinary pathology, the carcinogenesis of canine mammary tumors has been widely studied,¹² and several attempts have been made to establish a model of a molecular-based subgrouping of malignant tumors similar to that developed for human breast cancer.^9,35 To our knowledge, no studies on molecular subgrouping have been performed in feline mammary tumors, although the loss of hormone receptors in tumor progression is well known,^19,21,23 and several reports have evaluated the role of HER-2 overexpression in mammary carcinomas of cats.^22,26,31,41 A more profound study of feline carcinoma molecular subtypes would be valuable, since these carcinomas present a number of similarities with the disease in humans, such as the highly malignant behavior or the almost complete absence of myoepithelial proliferations (a common feature of mammary tumors in dogs, the so-called complex and mixed tumors).

Although this work is not a prognostic study, the relatively high number of cases studied, the findings regarding the association between different markers, and the presence of metastasis or tumoral grade and the similarities found with some aspects of breast cancer in women open an interesting field for future prognosis studies.

Footnotes

Acknowledgement

C. Peñafiel-Verdu is the recipient of a predoctoral grant from the Fundacion Seneca (Regional Govern of Murcia).

Declaration of Conflicting Interests

The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The authors received no financial support for the research, authorship, and/or publication of this article.

References

Brunetti

Sarli

Preziosi

. E-cadherin and beta-catenin reduction influence invasion but not proliferation and survival in canine malignant mammary tumors. Vet Pathol. 2005;42:781–787.

Burrai

Mohammed

, Miller MA, et al.Spontaneous feline mammary intraepithelial lesions as a model for human estrogen receptor- and progesterone receptor-negative breast lesions. BMC Cancer. 2010;10:156.

Castagnaro

Casalone

, Bozzetta E, et al.Tumour grading and the one-year post-surgical prognosis in feline mammary carcinomas. J Comp Pathol. 1998;119:263–275.

Cowin

Rowlands

Hatsell

. Cadherins and catenins in breast cancer. Curr Opin Cell Biol. 2005;17:499–508.

De Matos

Lopes

Faustino

. E-cadherin, beta-catenin, invasion and lymph node metastases in canine malignant mammary tumours. APMIS. 2007;115:327–334.

Dias Pereira

Gartner

. Expression of E-cadherin in normal, hyperplastic and neoplastic feline mammary tissue. Vet Rec. 2003;153:297–302.

Elston

Ellis

. Pathological prognostic factors in breast cancer: I. The value of histological grade in breast cancer: experience from a large study with long-term follow-up. Histopathology. 1991;19:403–410.

Friedl

Wolf

. Tumour-cell invasion and migration: diversity and escape mechanism. Nat Rev Cancer. 2003;3:362–374.

Gama

Alves

Schmitt

. Identification of molecular phenotypes in canine mammary carcinomas with clinical implications: application of the human classification. Virchows Arch. 2008;453:123–132.

10.

Gama

Paredes

Gärtner

. Expression of E-cadherin, P-cadherin and beta-catenin in canine malignant mammary tumours in relation to clinicopathological parameters, proliferation and survival. Vet J. 2008;177:45–53.

11.

Hahn

Bravo

Avenell

. Feline breast carcinoma as a pathologic and therapeutic model for human breast cancer. In Vivo. 1994;8:825–828.

12.

Klopfleisch

von Euler

Sarli

. Molecular carcinogenesis of canine mammary tumors: news from an old disease. Vet Pathol. 2011;48:98–116.

13.

Korsching

Jeffrey

Meinerz

. Basal carcinoma of the breast revisited: an old entity with new interpretations. J Clin Pathol. 2008;61:553–560.

14.

Korsching

Packeisen

Liedtke

. The origin of vimentin expression in invasive breast cancer: epithelial-mesenchymal transition, myoepithelial histogenesis or histogenesis from progenitor cells with bilineal differentiation potential? J Pathol. 2005;206:451–457.

15.

Livasy

Perou

Karaca

. Identification of a basal-like subtype of breast ductal carcinoma in situ. Hum Pathol. 2007;38:197–204.

16.

MacEwen

Hayes

Harvey

. Prognostic factors for feline mammary tumors. J Am Vet Med Assoc. 1984;2:201–204.

17.

Maeda

Johnson

Wheelock

. Cadherin switching: essential for behavioral but not morphological changes during a epithelium-to-mesenchyme transition. J Cell Sci. 2005;118:873–997.

18.

Martin de

las

Mulas

Espinosa de

los

Monteros

Bautista

. Immunohistochemical distribution pattern of intermediate filament proteins and muscle actin in feline and human mammary carcinomas. J Comp Pathol. 1994;111:365–381.

19.

Martin de

las

Mulas

Van Niel

Millan

. Progesterone receptors in normal, dysplastic and tumorous feline mammary glands. Comparison with oestrogen receptors status. Res Vet Sci. 2002;72:153–161.

20.

Matos

Faustino

Lopes

. Detection of lymph node micrometastasis in malignant mammary tumours in dogs by cytokeratin immunostaining. Vet Rec. 2006;158:626–630.

21.

Millanta

Callandrela

Bari

. Comparison of steroid receptor expression in normal, dysplastic, and neoplastic canine and feline mammary tissues. Res Vet Sci. 2005;79:225–232.

22.

Millanta

Calandrella

Citi

. Overexpression of HER-2 in feline invasive mammary carcinomas: an immunohistochemical survey and evaluation of its prognostic potential. Vet Pathol. 2005;42:30–34.

23.

Millanta

Callandrela

Vannozzi

. Steroid hormone receptors in normal dysplastic and neoplastic feline mammary tissues and their prognostic significance. Vet Rec. 2006;158:821–824.

24.

Misdorp

. Tumours of the mammary gland. In: Meuten

, ed. Tumors in Domestic Animals. 4th ed. Ames, IA: Iowa Press; 2002:575–605.

25.

Nielsen

Hsu

Jensen

. Immunohistochemical and clinical characterization of the basal-like subtype of invasive breast carcinoma. Clin Cancer Res. 2004;10:5367–5374.

26.

Ordás

Millán

Dios

. Proto-oncogene HER-2 in normal, dysplastic and tumorous feline mammary glands: an immunohistochemical and chromogenic in situ hybridization study. BMC Cancer. 2007;7:179.

27.

Peppercorn

Perou

Carey

. Molecular subtypes in breast cancer evaluation and management: divide and conquer. Cancer Invest. 2008;26:1–10.

28.

Perou

Sørlie

Eisen

. Molecular portraits of human breast tumours. Nature. 2000;406:747–752.

29.

Rakha

El-Sayed

Green

. Breast carcinoma basal differentiation: a proposal for pathology definition based on basal cytokeratin expression. Histopathology. 2007;50:434–438.

30.

Rakha

Patel

Powe

. Clinical and biological significance of E-cadherin protein expression in invasive lobular carcinoma of the breast. Am J Surg Pathol. 2010;34:1472–1479.

31.

Rasotto

Caliari

Castagnaro

. An immunohistochemical study of HER-2 expression in feline mammary tumours. J Comp Pathol. 2011;144:170–179.

32.

Sanchez

Ramirez

Buendia

. Immunohistochemical characterization and evaluation of prognostic factors in canine oral melanomas with osteocartilaginous differentiation. Vet Pathol. 2007;44:676–682.

33.

Sarli

Brunetti

Benazzi C

. Mammary mucinous carcinoma in the cat. Vet Pathol. 2006;43:667–673.

34.

Sarrio

Rodriguez-Pinilla

Hardisson

. Epithelial-mesenchymal transition in breast cancer relates to the basal-like phenotype. Cancer Res. 2008;68:989–997.

35.

Sassi

Benazzi

Castellani

. Molecular-based tumour subtypes of canine mammary carcinomas assessed by immunohistochemistry. BMC Vet Res. 2010;6:5.

36.

Seixas

Palmeira

Pires

. Grade is an independent prognostic factor for feline mammary carcinomas: a clinicopathological and survival analysis. Vet J. 2011;187:65–71.

37.

Shackleton

Vaillant

Simpson

. Generation of a functional mammary gland from a single stem cell. Nature. 2006;439:84–88.

38.

Takauji

Watanabe

Uyama

. Expression and subcellular localization of E-cadherin, alpha-catenin, and beta-catenin in 8 feline mammary tumor cell lines. J Vet Med Sci. 2007;69:831–834.

39.

Thiery

Acloque

Huang

. Epithelial-mesenchymal transitions in development and disease. Cell. 2009;139:871–890.

40.

Tomaskovic-Crook

Thompson

Thiery

. Epithelial-mesenchymal transition and breast cancer. Breast Cancer Res. 2009;11:213.

41.

Winston

Craft

Scase

. Immunohistochemical detection of HER-2/neu expression in spontaneous feline mammary tumours. Vet Comp Oncol. 2005;3:8–15.

42.

Zappulli

De Zan

Cardazzo

. Feline mammary tumours in comparative oncology. J Dairy Res. 2005;72:98–106.

Reduced Expression of E-cadherin and β-catenin and High Expression of Basal Cytokeratins in Feline Mammary Carcinomas With Regional Metastasis

Abstract

Keywords

Material and Methods

Tumors

Immunohistochemistry

Quantification of Immunolabeling

Statistical Analysis

Results

Immunohistochemical Analysis of the Adenomas

Tumoral Grade

Immunohistochemical Analysis of E-cadherin and β-catenin Expression in Carcinomas

Immunohistochemical Analysis of CK8/18, CK14, and CK5/6 Expression in Carcinomas

Relationship Between E-cadherin and β-catenin Expression and Basal and Luminal Cytokeratins

Immunohistochemical Analysis of Vimentin Expression in Carcinomas

Discussion

Footnotes

Acknowledgement

Declaration of Conflicting Interests

Funding

References