Germana Beha tesi

Germana Beha tesi
1
2
3
4
INDEX
1
1. INTRODUCTION
4
References
8
GENERAL SECTION
11
2. SOME INFORMATION ABOUT BREAST CANCER
12
Molecular taxonomy
12
Prognostic value
14
Target therapeutic and predictive approach
18
References
21
3. PHENOTYPIC CONCORDANCE AND DISCORDANCE BETWEEN
PRIMARY MAMMARY CARCINOMA AND ITS RELATED METASTASES
References
25
28
4. MOLECULAR PHENOTYPES IN CANINE MAMMARY TUMOURS
References
29
38
5. MOLECULAR PHENOTYPES IN FELINE MAMMARY TUMOURS
42
References
46
EXPERIMENTAL SECTION
49
1. MOLECULAR PORTRAIT-BASED CORRELATION BETWEEN
PRIMARY CANINE MAMMARY TUMOUR AND ITS LYMPH NODE
METASTASIS: POSSIBLE PROGNOSTIC-PREDICTIVE MODELS AND/OR
50
STRONGHOLD FOR SPECIFIC TREATMENTS?
Introduction
50
Methods
51
1
Results
54
Discussion
59
Conclusions
62
Figures
64
References
66
Publications and Proceedings
71
2. MOLECULAR PHENOTYPE IN MAMMARY TUMOURS OF QUEENS:
CORRELATION BETWEEN PRIMARY TUMOUR AND LYMPH NODE
72
METASTASIS
Introduction
72
Materials and Methods
73
Results
76
Discussion
79
Conclusions
82
Figures
83
References
85
Publications and Proceedings
88
3. MOLECULAR PHENOTYPE OF PRIMARY MAMMARY TUMOURS
AND DISTANT METASTASES IN FEMALE DOGS AND CATS
89
Introduction
89
Materials and Methods
90
Results
93
Discussion
95
Conclusions
99
Figures
100
References
102
Publications and Proceedings
106
FURTHER RESEARCH ON CANINE MAMMARY TUMOURS
4. MORPHOLOGY OF THE MYOEPITHELIAL CELL:
2
107
108
IMMUNOHISTOCHEMICAL CHARACTERIZATION FROM RESTING TO
MOTILE PHASE
Introduction
108
Aim
110
Materials and Methods
110
Results
112
Discussion
115
Conclusions
117
Figures
118
References
121
Publications and Proceedings
124
5. CD117 EXPRESSION INFLUENCES PROLIFERATION BUT NOT
SURVIVAL IN CANINE MAMMARY TUMOURS
125
Introduction
125
Materials and Methods
126
Results
129
Discussion
135
Conclusions
137
Figures
138
References
141
6. CONCLUSIONS
145
References
147
7. OTHER PUBLICATIONS OTHER PUBLICATIONS AND PROCEEDINGS
FROM JANUARY 2011 TO JANUARY 2014
3
148
1. INTRODUCTION
Breast cancer represents a heterogeneous group of tumours with varied morphologic and
biological features, behavior, and response to therapy (Rakha and Ellis, 2011). The above
mentioned heterogeneity is well recognized and raises the need for a revision of the World
Health Organization (WHO) classification, which was based on morphological features alone
(Eusebi, 2010). The beginning of a new era in cancer research is marked by the development
of methods to measure the expression of thousands of genes in tumour and in normal tissues.
The regulation of gene expression occurs through different molecular mechanisms such as the
transcriptional, transduction and post-transcriptional control. The first mechanism operates
at the frequency of transcription of messenger RNA from its own template DNA. The
transduction control regulates the speed with which a molecule of mRNA is translated into a
protein. The post-transcriptional control occurs in the moment when modulation of protein
synthesis is required, adjusting the speed or degrading a transcript that is already synthesized.
The functionality of these systems enables the processing of complex gene expression profiles
(Watson, 2008). Variation in transcriptional programs accounts for much of the biological
diversity of human cells and tumours. In each cell, signal transduction and regulatory systems
transduce information from the cell identity to its environmental status, thereby controlling
the level of expression of every gene in the genome (Perou et al, 2000; Sorlie et al. 2001).
The analysis of gene expression is the realization of the reform of tumour taxonomy, in
predicting the metastatic potential, prognosis and response to therapy. It is also used to reveal
patterns of gene expression that are dependent on the mutation of a single oncogene, and
finally, in analyzing the effects of hormones and environmental influences on carcinogenesis
(Robbins and Cotran, 2010).
4
Sorlie et al. (2001) laid the foundations for the new taxonomy demonstrating that changes in
patterns of gene expression, analyzed by cDNA microarray techniques and hierarchical
clustering, allowed a ‘molecular portrait’ to be defined for each tumour.
Recently, molecular characterization in human breast cancer has also been applied to the
metastasing lymph nodes and to the systemic metastases (Wu et al., 2008).
As in Human Medicine, mammary gland tumours commonly occur also in female dogs
(Gama et al., 2008) and cats (Morris et al., 2008), representing a remarkably heterogeneous
group in terms of morphology and biological behaviour (Nielsen et al., 2004). In the last
decades, veterinary attention has focused on the identification of reliable prognostic factors,
such as tumour size, histologic type, histologic grade, lymph node status (Misdorp et al.,
1999; Gama et al.2008; Sassi et al. 2010) and protein expression profile (Gama et al., 2008;
Sassi et al., 2010), essential in order to estimate the individual risk of unfavourable clinical
outcome (Misdorp, 2002; Zaidan, 2008). Gama et al. in 2008 and Sassi et al. in 2010 were
the first who applied, from human medicine, an immunohistochemical algorithm and
identified molecular phenotypes in canine mammary tumours.
Therefore, aims, developed as projects, of the past three years have been (1) to define the
molecular phenotype of feline mammary carcinomas and their lymph node metastases
according to a previous modified algorithm and to demonstrate the different relation between
the primary tumour and lymph node metastasis, (2) to analyze, in female dogs, the
relationship between the primary mammary tumor and its lymph node metastasis based on
immunohistochemical molecular characterization in order to develop the most specific
prognostic-predictive models and targeted therapeutic options, and (3) to evaluate the
molecular trend of cancer from its primary location to systemic metastases cats and dogs
with mammary tumors. Before the following research, no literature reports have addressed
the immunophenotyping of feline mammary carcinomas.
5
In order to make this study more complete, a parallel project has been carried out on the
myoepithelial cells of canine mammary tumours. In canine normal mamamary gland, the
lumina are delimitated by an inner layer of polarized epithelial cells resting on two outer or
basal layers of epithelial and myoepithelial cells (Ramalho et al., 2006). Both basal and
myoepithelial cells synthesize the basement membrane of ducts and alveoli and form a
structural barrier between the luminal epithelial cells and the surrounding stroma (Polyak and
Hu, 2006). The studies on mammary tumours, particularly in dogs, have drawn gradually
increasing attention not exclusively to the epithelial component, but also to the myoepithelial
cells. Myoepithelial cell proliferation is a frequent finding in the so-called complex and mixed
patterns (Misdorp et al., 1999) but it is an uncommon feature of breast cancer in women
(Sassi et al., 2010) and in cats. The lack of complete information on a valid panel of markers
for the identification of these cells in the normal and neoplastic mammary gland and the lack
of investigation of immunohistochemical changes from an epithelial to a mesenchymal
phenotype were the aims of a parallel research started in 2012, and which has been included
in the present report.
In addition, another study has been developed concerning CD117, a membrane-associated
tyrosine kinase growth factor receptor encoded by C-Kit gene (Yarden et al., 1987). CD117
is expressed in various cell types during embryonic development, it promotes different
functions of cells and has been shown to be expressed by neoplastic cells as well (Ronnstrand,
2004). Investigating mammary tumours, we noticed that only few studies had focused on the
expression of CD117, on the difference in expression between normal tissue and neoplastic
benign or malignant change in canine mammary tissue (Sailasuta et al., 2008). Since c-Kit is a
proto-oncogene that encodes a transmembrane tyrosine kinase growth factor receptor and
stimulates cell proliferation, it plays a crucial role in determining the existence of a correlation
between c-Kit expression and Ki67 index (Thompson et al., 2011). Therefore, it was decided
6
to further deepen the knowledge in order to to characterize the immunohistochemical staining
of CD117 in normal and neoplastic mammary tissue of the dog, and to correlate CD117
immunohistochemical results with mammary histotype, histological stage (invasiveness),
Ki67 index and patient survival time.
The reader of this paper will find the findings acquired in the three years 2011-2013, which
have been published or presented to national and international meetings.
7
References
Eusebi V (2010). Classifications and prognosis of breast cancer: from morphology to
molecular taxonomy. Breast Journal, 16, Suppl 1:S15-6.
Gama A, Alves A, Schmitt F (2008) Identification of molecular phenotypes in canine
mammary carcinomas with clinical implications: application of the human classification.
Virchows Archieve 453:123–132
Misdorp W (2002). Tumours of the mammary gland. In: Meuten DJ (ed) Tumour in domestic
animals. 4th edn. Iowa State Press, Iowa, pp 575–606
Misdorp W, Else RW, Hellmén E, Lipscomb TP (1999). Histological classification of
mammary tumour of the dog and the cat, Vol VII, 2nd series. Armed Forces Institute of
Pathology, American Registry of Pathology, Washington D.C., and the World Health
Organization Collaborating Center for Worldwide Reference on Comparative Oncology, pp
1–59
Morris JS, Nixon C, Bruck A, Nasir L, Morgan IM, Philbey AW (2008).
Immunohistochemical expression of TopBP1 in feline mammary neoplasia in relation to
histological grade, Ki67, ER alpha and p53. Veterinary Journal 175, 218-26.
Nielsen TO, Hsu FD, Jensen K, Cheang M, Karaca G, Hu Z, Hernandez-Boussard T, Livasy
C, Cowan D, Dressler L, Akslen LA, Ragaz J, Gown AM, Gilks CB, van de Rijn M, Perou
CM (2004). Immunohistochemical and clinical characterization of the basal-like subtype of
invasive breast carcinoma. Clinical Cancer Research 10(16):5367-5374.
Perou CM, Sorlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees CA, Pollack JR, Ross DT,
Johnsen H, Akslen LA, Fluge O, Pergamenschikov A, Williams C, Zhu SX, Lonning PE,
Borresen-Dale AL, Brown PO, Botstein D (2000). Molecular portraits of human breast
tumour. Nature 406:747–752.
8
Polyak K, Hu M Do Myoepithelial Cells Hold the Key for Breast Tumour Progression?
(2006). Journal of Mammary Gland Biology Neoplasia 3:pp. 231-246.
Rakha EA and Ellis IO (2011). Modern classification of breast cancer: should we stick with
morphology or convert to molecular profile characteristics. Advances in Anatomic Pathology
18:255-67
Ramalho LNZ, Ribeiro-Silva A, Cassali GD, Zucoloto S (2006). The Expression of p63 and
Cytokeratin 5 in Mixed Tumour of the Canine Mammary Gland Provides New Insights into
the Histogenesis of These Neoplasms. Veterinary Pathology 4:pp.424-429.
Robbins SL and Cotran RS (2010). Pathologic Basis of Disease, 8th edition, Saunders
Elsevier, Philadelphia.
Ronnstrand L (2004). Signal transduction via the stem cell factor/c-Kit. Cellular and
Molecular Life Sciences, 61, 2535-2548
Sailasuta A, Thomrongsuwannakij T, Romphotiyok K, Theeratammakom T, Wangnaitham S
(2008). Expression of c-Kit Oncogene Product in Mammary gland tumors in dogs. Proc 15th
FAVA congress-OIE Joint Symposium on Emerging Diseases, Bangkok, Thailand, October
27-30, pp 339
Sassi F, Benazzi C, Castellani G, Sarli G (2010). Molecular-based tumour subtypes of canine
mammary carcinomas assessed by immunohistochemistry. BMC Veterinary Research 6, 5.
Sorlie T, Perou CM, Tibshirani R, Aas T, Geisler S, Nobel A, Deng S, Johnsen H, Pesich R,
Geisler S, Demeter J, Perou CM, Lonning PE, Brown PO, Borresen-Dale AL, Botstein D
(2001). Gene expression patterns of breast carcinomas distinguish tumour subclasses with
clinical implications. Proceedings of the National Academy of Sciences of the United States
of America, 98, 10869-74.
9
Thompson JJ, Yager JA, Best SJ, Pearl DL, Coomber BL, Torres RN, Kiupel M, Foster RA
(2011). Canine subcutaneous mast cell tumors: cellular proliferation and KIT expression as
prognostic indices. Veterinary Pathology, 48, 169-181.
Watson JD, Caudy AA, Myers RM, Witkowski JA (2008). DNA ricombinante: Geni e
genomi; 2nd italian edition. Traslation by G. Forlani, N. Landsberger, M. Muzi Falconi.
Wu JM, Fackler MJ, Halushka MK, Molavi DW, Taylor ME, Teo WW, Griffin C, Fetting J,
Davidson NE, DeMarzo AM, Hicks JL, Chitale D, Marc L, Sukumar S, Argani P (2008).
Heterogeneity of Breast Cancer Metastases: Comparison of Therapeutic Target Expression
and Promoter Methylation Between Primary Tumour and Their Multifocal Metastases.
Clinical Cancer Research 10:1938–1946.
Yarden SY, Kuang WJ, Yang-Feng T, Coussens L, Munemitsu, Dull TJ, Chen E,
Schlessinger J, Francke U, Ullrich A (1987). Human proto-oncogene c-Kit: a new cell surface
receptor tyrosine kinase for an unidentified ligand. The EMBO Journal, 6, 3341-3351.
Zaidan Dagli ML (2008). The search for suitable prognostic markers for canine mammary
tumour: a promising outlook. Vet J 177:3–5
10
GENERAL SECTION
11
2. SOME INFORMATION ABOUT BREAST CANCER
Molecular taxonomy
The changes in gene expression patterns, using complementary DNA microarrays, provided a
distinctive molecular portrait of each tumour (Perou et al, 2000). Sets of co-expressed genes
were identified for which variation in messenger RNA levels could be related to specific
features of physiological variation. The tumours were classified into subtypes distinguished
by pervasive differences in their gene expression patterns of mammary epithelium (Perou et
al, 2000). Based on those differences four subtypes were identified:
-
Luminal like/ER+ (Estrogen Receprtor): so called because of the expression of many
genes is expressed by the luminal cells.
-
C-erb-B2: c-erbB-2 overexpression gene.
-
Basal like: expression of a clusters of gene proper of the basal epithelium.
-
Normal breast: overexpression of basal epithelium gene and low expression of genes
of the luminal epithelium.
Subsequently, Sorlie et al. (2001) modified the molecular classification, in search for
correlations between gene expression patterns and parameters of clinical relevance, such as
the survival rate and the likelihood of recidivism, thus demonstrating the prognostic value of
this subtyping. Two major branches of subtypes were obtained: ER- and ER+ (luminal). The
ER- cluster included the basal-like subtype (characterized by high expression of cytokeratin 5
and laminin 17 binding fatty acids), the c-erbB-2 subtype (characterized by a wide expression
of genes in the segment c-erbB-2), and the normal group breast-like (which showed the
expression of genes typical of adipose tissue and other non-epithelial cell types) (Sorlie et al.,
2001). These three clusters exhibited a remarkable expression of genes of the basal epithelium
and a reduced or no expression of genes of the luminal epithelium. The second main group,
12
ER+, obtained from the analysis included three hierarchical sub- expressing genes typical of
luminal epithelium, such as luminal A (high gene expression of ERα), luminal B and C
(moderate expression of specific genes of luminal epithelium). In particular, the luminal C
subtype was distinct from the other two luminal types for the expression of a set of genes
whose function was unknown. The gene expression pattern represent the tumour biology,
reflecting the biological diversity and correlate the different tumours with the clinical
genomics was thought to be the key to understand this diversity (Sorlie et al., 2001).
Further refinement for molecular phenotyping was given by Sorlie et al. (2003) in a
subsequent study, thanks to the increasing in number of tumour cases, from 115 to 534, and to
the use of intrinsic genes. The modified molecular phenotypes classification came to the
number of subtypes known at present: basal-like, c-erbB-2 overexpressing, two luminal-like
(A and B) and normal-like (Sorlie et al., 2003). The existence of different subtypes of breast
cancer was confirmed by protein expression patterns assessed by immunohistochemistry
(IHC) on tissue microarray (TMA), an efficient and reliable platform for subclassifying breast
cancers into relevant subtypes, using a limited number of markers (Matos et al., 2005). In
particular, the panels encompass at least anti- estrogen receptor (ER), anti-progesterone
receptor (PR), anti- c-erbB2 and antibasal cytokeratin antibodies (CK 5/6 and 14) (Cheang et
al, 2008; Kim et al., 2006). The obtained new classification was therefore, constituited by
two hormone (estrogen (ER) and/or progesterone (PR) receptor positive types [luminal A (ER
and/or PR+, c-erbB-2−), luminal B (ER and/or PR+, c-erbB-2+)], and three hormone receptor
negative [c-erbB-2 overexpressing (ER/PR−, c-erbB-2+), basal-like (ER/PR−, c-erbB-2−,
CK5/6 and/or CK14 and/or p63+), and normal-like (ER/PR−, c-erbB-2−, CK5/6 and/or CK14
and/or p63−)]. Even if much has been learned in the last few years about the molecular
taxonomy, it is still in evolution and likely to change over the coming years (Cummings et al.,
2011).
13
Prognostic value
Breast cancer outcome in women varies widely and prognosis is determined by the pathologic
examination of the primary carcinoma and the axillary lymph nodes. In 2002 the American
Joint Commitee on Cancer (AJCC) identified six major prognostic factors that were
recognised as strongest predictors of death. The major prognostic factors were:

Invasive carcinoma versus in situ disease. By definition, in situ carcinoma is confined
to the ductal system and cannot metastatize while invasive carcinoma usually locally
or distantly metastatize.

Distant metastases. Cure is unlike, especially in hormonally responsive tumours and
the type of tumour influences the timing and location of metastases.

Lymph node metastases. In absence of distant metastases, it is the most important
prognostic factor for invasive carcinoma. Approximally 10% to 20% of women
without axillary lymph node metastases have a recurrence outside the breast and
about the same number die from breast cancer. In these patients, metastasis may occur
via the internal mammary lymph nodes or hematogenously.

Tumour size. The size of a an invasive carcinoma is the second most important
prognostic factor. The risk of axillary lymph nodes metastases increases with the size
of the primary tumour, but both are indipendent of prognostic factors.

Locally advanced disease. Carcinomas invading into skin or skeletal muscle are
usually large and may be difficult to treat surgically.

Infllammatory carcinoma. Particularly poor prognosis is related to breast cancer
presenting
with breast swelling and skin thickening due to dermal lymphatic
involvement.
14
In addition to the six major factors used by the AJCC, a number of other factors have
become increasingly important as predictive of outcome.

Histologic Subtype. The survival rate of women with special type of invasive
carcinomas (tubular, mucinous, medullary, lobular, and papillary) is greater than
60%, compared with less than 20% for women with non special type.

Histologic grade. The commonly used grading system, the Nottingham Histologic
Score, combines nuclear grade, tubule formation and mitotic rate to classify
invasive carcinomas into three groups that are highly correlated with survival. A
higher survival for patients with well-differentiated grade 1 carcinoma, and, in
contrast, most deaths for poorly differentiated grade 3 carcinomas.

Estrogen and progesteron receptors. Immunohistochmistry assays are currently
used to detect nuclear hormone receptors. This latter finding is correlated with a
better outcome and is an important predictor of response to hormonal therapy.

C-erbB-2. Its overexpression is associated with poor survival, but its main
importance is as a predictor of response agents that target this transmembrane
protein (trastuzumab or lapatinib).

Lymphovascular invasion. The presence of tumour cells within vascular spaces is
strongly associated with the presence of lymph node metastases. It is a poor
prognostic factor for overall survival in women without lymph node metastases
and a risk factor for local recurrence.

Proliferative rate. Carcinomas with high proliferation rates have a poorer
prognosis but may respond better to chemotherapy. The rate can be measured by
mitotic counts or by immunohistochemical detection of cellular protein produced
during the in cell cycle (ki-67).
15

DNA content. Tumours with abnormal DNA indices have a slightly worse
prognosis.

Response to nonadjuvant therapy. Pathologic response to chemotherapy can be
used as a short-term end point for clinical trials

Gene expression profiling. Expression profiling has been shown to predict survival
and recurrence-free interval, and also identifies patients who are most likely to
benefit from particular types of chemotherapy (Robbins and Cotran, 2010)
Standard clinical prognostic factors, such as patient age, tumour size, lymph node status,
tumour grade, hormone receptor status or c-erbB-2 provide valuable information on the risk
of recurrence; however, these estimates of clinical risk are coarse. On the contrary, multigene
classifiers have the potential to complement traditional methods through provision of
additional biological prognostic and predictive information in presently indeterminate risk
groups (Rakha and Ellis, 2011). Indeed, the different molecular subtypes have been associated
with different prognoses (Sorlie et al., 2003), treatment planning (Carey, 2011), metastatic
sites (Fountzilas et al. 2012; Gabos et al. 2006) and survival rates (Blows et al., 2010). Sorlie
et al. (2001), based on a study of 49 patients with locally advanced disease status but without
systemic metastases, conducted a survival analysis correlating ther subtypes with the overall
survival rate and recurrence-free interval . The recurrence-free interval was defined as the
interval between the date of breast surgery and the date of subsequent diagnosis of breast
cancer. The overall survival rate is defined as the interval between the date of breast surgery
and the date of death, related to breast cancer (Sotiriou et al., 2003). The added value
identified by Sorlie et al. (2001) gave a clinical significance to these tumour subtypes,
demonstrating a low survival rate and a high relapse rate for the Basal-like and c-erB-2+
subtypes, accompanied by a poor prognosis, with a high frequency of mutation in the TP53
gene and amplification of the c-erbB-2 gene, respectively. The luminal presented a more
16
favorable prognosis than the previous but with different outcomes between luminal A and B.
The latter showed a worse outcome than the first (Sorlie et al., 2001).
The molecular processes at the basis of tumour progression have not been fully characterized
as well as the factors that determine the metastatic potential of a tumour, both to the
respective regional lymph node or to distant sites. A common hypothesis argues that the
metastatic process is affected by the molecular changes that evolve within few clones of the
primary tumour that metastasize. If this were true, the metastasizing ability of a small
proportion of the primary tumour clones could not be predicted by the analysis of the primary
tumour. An alternative hypothesis argues that the metastatic capacity is largely determined by
the sum of the molecular changes that characterize the majority of cells in the primary
tumour. In this case, the metastases could be predictable. The status of the regional lymph
node, determined at the time of diagnosis, may represent a functional, even if imperfect
surrogate of the process of metastasis. Lymph node metastases reflect the biological
properties of the tumour (eg, motility, invasiveness) (Lu et al., 2008).
The concept that the molecular information of the primary tumour may help
define the
metastatic potential was investigated by Lu et al. (2008). Their formulated models,
constructed on the basis of gene expression data, demonstrated predictive accuracy to
biological characteristics, such as the status of ER, c-erbB-2, and tumour grade. Instead, the
model failed to demonstrate the predictive accuracy to anatomical features such as lymph
node status, tumour size and lymph-vascular invasion. The authors concluded by saying that
the regional metastases develop in relation to time and to a "propensity" that could be due to
inherent biological differences, best reflected by molecular subtypes, and that their prediction
is of limited value (Lu et al., 2008).
A further merit that was given to the molecular analysis methods was the possibility to
establish the preferential sites of distant metastases in association with different molecular
17
subtypes. It was found that bone metastases were more consistently associated with the
luminal subtypes, less frequently with the basal phenotypes. The opposite was true for
metastases to the lung and brain, with no specific correlation between lung metastases and
luminal A type. The pleural metastases were found almost exclusively in the luminal subtypes
(Smid et al., 2008). The ability of the tumour to head, survive and proliferate in certain
distant organs requires a different set of genes, as well as the ability of the tumour to
metastasize. A lot of genes differentially expressed were identified, many of which were
common among the subtype and the site in which the subtype
had preferentially
metastasized; wingless-type MMTV integration site family (WNT) signalling was upregulated in the basal subtype as well as in its specific brain metastasis. The five molecular
subtypes are therefore clearly different with regard to their ability to metastasize to different
body sites (Smid et al., 2008).
The profile of gene expression is thus a most powerful predictor of the outcome of the
disease, than traditional systems based on clinical and histological criteria (van de Vijver et
al., 2002).
Target therapeutic and predictive approach
While prognostic factors intend to objectively predict the clinical outcome independently
from treatment, the predictive approaches are intended to predict the response of patiens to
specific therapeutic therapies and are also associated with tumour sensitivity or resistance to
therapy (Weigel et al., 2010).
The discovery of molecular subtypes of breast cancer gave a further proof of the fact that
biological diversity consequently denies a unique therapeutic approach (Peppercorn et al.,
18
2008). The therapeutic approach to breast cancer varies according to the different molecular
phenotypes.
The luminal subtypes are hormone receptor-positive, therefore appropriate for endocrine
therapy with tamoxifen combined with chemotherapy (Peppercorn et al., 2008). This is
especially true for the luminal A, in which conversely only a small amount of those tumours
respond to classical chemotherapy. The luminal B, also referred to as triple positive for the
expression of c-erbB-2 as well as hormone receptor, is often responsive to chemotherapy
(Robbins and Cotran, 2010).
The expression of ERα is lately undoubtedly considered the most important biomarkers in
breast cancer, because it provides an index of sensitivity to endocrine treatment (Weigel and
Dowsett, 2010).
The therapeutic approach to c-erbB-2-overexpressing phenotypes is by treating patients with
chemotherapy, and c-erbB-2 targeted therapy such as trastuzumab (Carey, 2011).
The c–erbB-2 overexpressing phenotypes, besides being a predictor factor of resistance to
endocrine therapy or selective resistance to tamoxifen predicts resistance to some
chemotherapies (Harris et al., 2007).
The therapeutic approach to basal-like and normal-like phenotypes represent today a huge
challange since they are not responsive to hormone therapy, but they are treated with
chemotherapy alone (Carey, 2011; Matos et al., 2005). Treatment of these phenotypes only
with chemiotherapy implies an estimate of the residual risk of 30-40% (Fig. 1) (Carey, 2011).
19
Fig. 1: Estimate residual risk
Adjuvant therapy for breast cancer in the initial state, with the initial risk of recurrence of approximately 60%. If
the patient present a tumour hormone receptor positive phenotype, she will be treated with endocrine therapy and
chemotherapy, with a subsequent residual risk of recurrence <25%. In case of c-erbB-2 overexpressing
phenotype (c-erbB-2+), the patient will be treated with anti- c-erbB-2 therapy in addition to chemotherapy, with
estimates residual risk of <25%. If the cancer present a triple negative phenotype the treatment is chemotherapy
only, with residual risk> 30-40% (Modified from Carey, 2011).
20
References
Blows FM, Driver KE, Schmidt MK, Broeks A, van Leeuwen FE, Wesseling J, Cheang MC,
Gelmon K, Nielsen TO, Blomqvist C, Heikkilä P, Heikkinen T, Nevanlinna H, Akslen LA,
Bégin LR, Foulkes WD, Couch FJ, Wang X, Cafourek V, Olson JE, Baglietto L, Giles GG,
Severi G, McLean CA, Southey MC, Rakha E, Green AR, Ellis IO, Sherman ME, Lissowska
J, Anderson WF, Cox A, Cross SS, Reed MW, Provenzano E, Dawson SJ, Dunning AM,
Humphreys M, Easton DF, García-Closas M, Caldas C, Pharoah PD, Huntsman D (2010)
Subtyping of breast cancer by immunohistochemistry to investigate a relationship between
subtype and short and long term survival: a collaborative analysis of data for 10,159 cases
from 12 studies. PLoS Medicine 7:1–11
Carey LA (2011) Directed therapy of subtypes of triple-negative breast cancer. Oncologist,
15:49–56.
Cheang MC, Voduc D, Bajdik C, Leung S, McKinney S, Chia SK, Perou CM, Nielsen TO
(2008) Basal-like breast cancer defined by five biomarkers has superior prognostic value than
triple-negative phenotype. Clinical Cancer Research, 14(5):1368-1376.
Cummings MC, Chambers R, Simpson PT, Lakhani SR (2011) Molecular classification of
breast cancer: is it time to pack up our microscopes? Pathology, 43(1):1-8.
Early Breast Cancer Trialists' Collaborative Group (EBCTCG) (2005). Effects of
chemotherapy and hormonal therapy for early breast cancer on recurrence and 15-year
survival: an overview of the randomised trials. Lancet 365:1687-717.
Farmer P, Bonnefoi H, Becette V, Tubiana-Hulin M, Fumoleau P, Larsimont D, Macgrogan
G, Bergh J, Cameron D, Goldstein D, Duss S, Nicoulaz AL, Brisken C, Fiche M, Delorenzi
M, Iggo R (2005). Identification of molecular apocrine breast tumours by microarray analysis.
Oncogene 24(29):4660-4671
21
Fountzilas G, Dafni U, Bobos M, Batistatou A, Kotoula V, Trihia H, Malamou-Mitsi V,
Miliaras S, Chrisafi S, Papadopoulos S, Sotiropoulou M, Filippidis T, Gogas H, Koletsa T,
Bafaloukos D, Televantou D, Kalogeras KT, Pectasides D, Skarlos DV, Koutras A,
Dimopoulos MA (2012). Differential response of immunohistochemically defined breast
cancer subtypes to anthracycline-based adjuvant chemotherapy with or without paclitaxel.
PLoS One. 7(6).
Gabos Z, Sinha R, Hanson J, Chauhan N, Hugh J, Mackey JR, Abdulkarim B (2006)
Prognostic significance of human epidermal growth factor receptor positivity for the
development of brain metastasis after newly diagnosed breast cancer. J Clin Oncol 24:56585663.
Harris L, Fritsche H, Mennel R, Norton L, Ravdin P, Taube S, Somerfield MR, Hayes DF,
Bast RC Jr (2007). American Society of Clinical Oncology. Update of recommendations for
the use of tumour markers in breast cancer. Journal of Clinical Oncology 25(33):5287-5312.
Hu Z, Fan C, Oh DS, Marron JS, He X, Qaqish BF, Livasy C, Carey LA, Reynolds E,
Dressler L, Nobel A, Parker J, Ewend MG, Sawyer LR, Wu J, Liu Y, Nanda R, Tretiakova M,
Ruiz Orrico A, Dreher D, Palazzo JP, Perreard L, Nelson E, Mone M, Hansen H, Mullins M,
Quackenbush JF, Ellis MJ, Olopade OI, Bernard PS, Perou CM (2006). The molecular
portraits of breast tumour are conserved across microarray platforms. BMC Genomics 7:96.
Kim MJ, Ro JY, Ahn SH, Kim HH, Kim SB, Gong G (2006). Clinicopathologic significance
of the basal-like subtype of breast cancer: a comparison with hormone receptor and Her2/neuoverexpressing phenotypes. Human Pathology, 37, 1217–1226
Lu X, Lu X, Wang ZC, Iglehart JD, Zhang X, Richardson AL (2008). Predicting features of
breast cancer with gene expression patterns. Breast Cancer Research Treatment 108:191-201.
22
Matos I, Dufloth R, Alvarenga M, Zeferino LC, Schmitt F (2005). p63, cytokeratin 5, and Pcadherin: three molecular markers to distinguish basal phenotype in breast carcinomas.
Virchows Archive, 447, 688-94.
Peppercorn J, Perou CM, Carey LA (2008). Molecular subtypes in breast cancer evaluation
and management: divide and conquer. Cancer Investigation 26:1-10.
Perou CM, Sorlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees CA, Pollack JR, Ross DT,
Johnsen H, Akslen LA, Fluge O, Pergamenschikov A, Williams C, Zhu SX, Lonning PE,
Borresen-Dale AL, Brown PO, Botstein D (2000) Molecular portraits of human breast
tumour. Nature 406:747–752.
Prat A, Parker JS, Karginova O, Fan C, Livasy C, Herschkowitz JI, He X, Perou CM (2010)
Phenotypic and molecular characterization of the claudin-low intrinsic subtype of breast
cancer. Breast Cancer Research 12(5):R68
Rakha EA, Ellis IO (2011) Modern classification of breast cancer: should we stick with
morphology or convert to molecular profile characteristics. Advances in Anatomic Pathology
18:255-267.
Robbins SL, Cotran RS (2010) Pathologic Basis of Disease, 8th edition, Saunders Elsevier,
Philadelphia.
Smid M, Wang Y, Zhang Y, Sieuwerts AM, Yu J, Klijn JG, Foekens JA, Martens JW (2008).
Subtypes of breast cancer show preferential site of relapse. Cancer Research 68:3108-3114.
Sorlie T, Perou CM, Tibshirani R, Aas T, Geisler S, Nobel A, Deng S, Johnsen H, Pesich R,
Geisler S, Demeter J, Perou CM, Lonning PE, Brown PO, Borresen-Dale AL, Botstein D
(2001) Gene expression patterns of breast carcinomas distinguish tumour subclasses with
clinical implications. Proceedings of the National Academy of Sciences of the United States
of America 98, 10869-10874.
23
Sorlie T, Tibshirani R, Parker J, Hastie T, Marron JS, Nobel A, Deng S, Johnsen H, Pesich R,
Geisler S, Demeter J, Perou CM, Lønning PE, Brown PO, Børresen-Dale AL, Botstein D
(2003). Repeated observation of breast tumour subtypes in independent gene expression data
sets. Proceedings of the National Academy of Sciences of the United States of America
100:8418-8423.
Sotiriou C, Neo SY, McShane LM, Korn EL, Long PM, Jazaeri A, Martiat P, Fox SB, Harris
AL, Liu ET (2003). Breast cancer classification and prognosis based on gene expression
profiles from a population-based study. Proceedings of the National Academy of Sciences of
the United States of America 100:10393-10398.
van de Vijver MJ, He YD, van't Veer LJ, Dai H, Hart AA, Voskuil DW, Schreiber GJ, Peterse
JL, Roberts C, Marton MJ, Parrish M, Atsma D, Witteveen A, Glas A, Delahaye L, van der
Velde T, Bartelink H, Rodenhuis S, Rutgers ET, Friend SH, Bernards R (2002). A geneexpression signature as a predictor of survival in breast cancer. New England Journal of
Medicine 347:1999-2009.
Weigel MT, Dowsett M. Current and emerging biomarkers in breast cancer: prognosis and
prediction (2010). Endocrine Related Cancer 17:R245-262.
24
3. PHENOTYPIC CONCORDANCE AND DISCORDANCE BETWEEN
PRIMARY MAMMARY CARCINOMA AND ITS RELATED
METASTASES
The study by Weigelt et al. (2003) supported the concept that the ability to metastasize to
distant sites is a genetic properties of breast carcinomas. Shown that the gene expression
profile of the primary tumour is closely similar in distant metastases of the same patient,
indicative of the hypothesis that the ability of the metastatic breast cancer is an innate
characteristic and it is not based on a clonal selection (Weigelt et al., 2003).
Previously, Fidler et al. (1977) found that clones, obtained in vitro from a parental malignant
melanoma murine cell culture, varied greatly in their ability to produce metastatic colonies.
This involved the striking conclusion that the sub-cell population having highly metastatic
power pre-existed in parental population (Fidler et al., 1977). This model implies that
metastases arise from a sub-clone, which could be molecularly distinct from their primary
tumour. The Fidler clonal concept is widely accepted, although the metastatic process has also
been described as a stochastic event, having the primary tumour cells with equal metastatic
potential (Weigelt et al., 2003; Milas et al., 1983).
Many human tumours develop through a succession of genetic and epigenetic changes that
confer to the cells neoplastic characteristic. This process can be likened to a Darwinian
evolution within the microcosm of living tissues, in which individual cells represent the
selection unit. Single cells that possesses advantageous characteristics (such as survival and
proliferation) are selected to become the progenitor of a successor cell population that
eventually will dominate the tumour mass. The initiator of the next round of successors clones
represent a rare variant that arises between the many successor cells. About 6-10 cycles of
clonal sequences are needed to generate highly malignant tumour cells. A different model of
25
metastatic mechanism relies on the fact that the tendency to metastasize is largely determined
by the identity of the mutant alleles, which are acquired relative at the beginning of the
various carcinogenesis stages (Bernards et al., 2002).
In order to investigate the molecular differences between the primary tumour and metastases
Ramaswamy et al. (2003) compared the gene expression profile of metastatic adenocarcinoma
form multiple tumour phenotypes with not corresponding primary adenocarcinomas. They
proved that a part of primary tumours resembled the metastatic ones regarding the gene
expression signature (Ramaswamy et al., 2003).
Weigelt et al. (2005), studying the gene expression profile of a set of 70, already proved to be
efficacy in predicting the metastatic potential, found that distant metastases reflect the
molecular subtypes and prognostic signature (70 genes) of their primary tumour. These results
support the hypothesis that the molecular subtypes originate from different cell types within
the mammary gland and thus reflect the different biological entities, which are kept in the
metastatic process (Weigelt et al. 2005).
Divergent results were deduced from a study of Aitken et al. (2010), which carried out
changes in expression through quantitative analysis of ER, PR, c-erbB-2 in primary mammary
site tumour and its related lymph node metastasis. Given that nodes are excised in standard
surgical practice, as sentinel node biopsy, axillary node sample, or node clearance, and
assessed by a pathologist for routine staging, the authors showed that there may be added
benefit to molecular testing on nodal metastases as well as the primary tumour in order to
guide adjuvant therapy. While this might incur additional financial costs for specimen
processing and molecular analysis, savings could be made by avoiding overtreatment. The
authors demonstrated that a significant number of patients showed a quantitative difference in
expression of the molecular markers between primary tumour and lymph node metastasis and
this data may confer increase therapeutic sensitivity or resistance to targeted therapy.
26
Therefore, the difference in receptor expression levels may be one of the causes of treatment
failures, found in some clinical cases and the phenotyping of both primary and lymph node
would reduce morbidity for patients and ultimately has the potential to produce more
favourable clinical outcomes (Aitken et al., 2010).
In conclusion the different studies regard the comparison between the phenotypes of primary
site and lymph node metastasis have highlighted the existence of a concordance in which the
primary tumour and the lymph node metastasis have the same phenotype, and a discordance
with differences in phenotype between primary and metastases. On the basis of this
knowledge, the future treatment decisions should be based on the gene expression profile of
both the primary tumour and its related lymph node metastasis (Aitken et al., 2010).
27
References
Aitken SJ, Thomas JS, Langdon SP, Harrison DJ, Faratian D (2010). Quantitative analysis of
changes in ER, PR and HER2 expression in primary breast cancer and paired nodal
metastases. Annals of Oncology 21(6):1254-61.
Bernards R, Weinberg RA (2002). A progression puzzle. Nature 418(6900):823.
Fidler IJ, Kripke ML (1977). Metastasis results from preexisting variant cells within a
malignant tumour. Science 197(4306):893-895.
Milas L, Peters LJ, Ito H (1983). Spontaneous metastasis: random or selective? Clinical
Experimental Metastasis 1(4):309-315.
Ramaswamy S, Ross KN, Lander ES, Golub TR (2003). A molecular signature of metastasis
in primary solid tumour. Nature Genetics 33(1):49-54.
Weigelt B, Glas AM, Wessels LF, Witteveen AT, Peterse JL, van't Veer LJ (2003). Gene
expression profiles of primary breast tumour maintained in distant metastases. Proceedings of
the National Academy of Sciences of the United States of America 100(26):15901-15905.
Weigelt B, Hu Z, He X, Livasy C, Carey LA, Ewend MG, Glas AM, Perou CM, Van't Veer
LJ (2005). Molecular portraits and 70-gene prognosis signature are preserved throughout the
metastatic process of breast cancer. Cancer Research 65(20):9155-9158.
28
4. MOLECULAR PHENOTYPES IN CANINE MAMMARY TUMOURS
Mammary tumours are among the most common tumour in dogs, in fact, they represent 25–
50% of all neoplasms in this species, with an average age of occurrence between 6 and 10
years (Millanta et al., 2005a).
The high homology between the canine and the human genomic sequence, as the many
similarities regarding the morphology, biological behavior, and clinical course of mammary
tumours, is the starting point for the study of comparative pathology that is proposing the dog
as a valuable comparative and predictive model for human breast cancer. Recent studies have
shown that the general basic biology of cancer in dogs such as tumour establishment and
metastatic progression is similar to what happens in human cancer (Pinho et al., 2012). Most
of the cancer-associated genetic alterations (Fig. 2) that are known to play a role in mammary
tumour development and progression are similar in both species (Rivera et al., 2011).
A recent study demonstrated that many genes deregulated in human breast cancer were also
found deregulated in canine mammary tumours compared with normal mammary tissue. This
first genome-wide comparative analysis demonstrated that the pathways showing
upregulation in tumour from both species are those related to increased proliferation activity,
whereas the pathways related to cell development, cell matrix adhesion, and cell
communication are downregulated. Furthermore, it has also been shown a great degree of
homology between human and canine mammary tumours in the perturbation of many cancer
related pathways (Uva et al., 2009). The genome similarities of dogs and humans support
strongly the genetic homology between both species (Pinho et al., 2012).
29
Fig. 2: Comparative analysis of the role of the critical genes and signaling pathways involved in the
carcinogenesis of human breast cancer and canine mammary tumour
The first study regarding the gene expression of mammary tumours was accomplished by Rao
et al. (2008) who characterized, with the use of the cDNA microarray, the expression profile
of three different cell lines of canine mammary carcinoma originating from histologically
distinct primary tumours:
- CMT (Canine Mammary Tumour) - U335 (histological type osteosarcoma)
- CMT- U229 (atypical benign in mixed tumour histological type)
- P114 (anaplastic histological type carcinoma)
The evaluation of the gene expression profile of the cell lines showed correspondence with
the tumour of origin and the differential regulation of several pathways such as the WNT,
integrins, cell cycle, alternative complement cascade, cytokine/Rho-GTPase. These pathways
showed overlaps with those found in humans, therefore the expression profile of spontaneous
canine mammary carcinomas was supposed to act as a biological sieve for the identification
of pathways or of gene expression profiles that are involved in carcinogenesis (Rao et al.,
2008). The highlighting of molecular subtypes of canine mammary tumours by using gene
expression patterns of a intrinsic gene set, similarly to what was done in humans, has not been
30
performed in veterinary medicine because of a reduced representation of similar genes on the
canine microarray (Rao et al., 2008).
The new classification of distinct molecular phenotypes was developed by Sarli et al. (2007)
using an immunohistochemical panel with anti-cytokeratin 19 (CK19), -CK14, -CK 5/6, -ER,
-PR, -vimentin, -c-erbB-2, in order to identify the following phenotypes:
• luminal -like (CK19+, ER+/-, PR+/-, CK14-, CK5/6-) type A (c-erbB-2-) and type B (cerbB -2+);
• basal- like (CK19-, ER-, PR-, CK14+, CK5/6+, c-erbB-2-);
• c-erbB-2+ (CK19-, ER-, PR-, CK14-, CK5/6-, c-erbB-2+).
The subtypes luminal A and luminal B showed the same phenotype but differed in the
expression of c-erbB-2, which is considered an important prognostic index
of tumour
progression in canine mammary tumours as well as in woman breast cancer. The
myoepithelial proliferation, in complex and in mixed tumour, resulted negative for
myoepithelial cells markers (CK14 and CK5/6), which characterize the basal- like phenotype,
but were positive for vimentin. The positivity for cytokeratins was expressed only when the
cells were surrounded by the myoepithelial luminal counterpart. Sarli et al. (2007) suggested
that the above mentioned data could be indicative of lack of specific phenotypes in canine
pathology or of the need to investigate further for myoepithelial markers.
On the basis of the immunohistochemical panel proposed by Matos et al. (2005), Gama et al.
(2008) characterized 102 canine mammary carcinomas based on the immunohistochemical
panel which involved the evaluation of five molecular markers (ER, c-erbB-2, CK5, p63 and
P-cadherin). ER and p63 positive cases showed the characteristic nuclear staining, whereas
those positive for CK5 showed a cytoplasmic pattern. The c-erbB-2 positive tumours to have
presented a membranous staining, P-cadherin positive tumours showed both a cytoplasmic or
a membranous staining. The tumour classification was effected on the basis of ER and c31
erbB-2 status, in accordance with the algorithm formulated by Nielsen et al. (2004); positive
tumours to ER were classified as luminal, distinguishing them further into A and B on the
basis of negativity or positivity to c-erbB-2, respectively. Tumour that did not express ER, but
were positive for c-erbB-2 were defined as c-erbB-2 overexpressing phenotype, or those cases
that continued to be negative but positive for basal markers CK5, p63, and P-cadherin were
classified as basal. Tumours that did not express any markers were identified as “none
phenotype”. CK5, p63 and P-cadherin are proteins that are expressed early in epithelial
differentiation and may contribute to a committed stem cell and/or progenitor phenotype
(Boecker and Buerger, 2003; Boecker et al., 2002). In the study by Gama et al. (2008), the
authors demonstrated that these markers were upregulated in the basal subtype, similarly to
the previous results by Matos et al. (2005). In fact, the basal subtype rarely expressed just
one basal marker but frequently expressed them simultaneously, which suggested a more
undifferentiated profile. C-erbB-2 overexpressing subtype was also characterised by an upregulation of basal markers, confirming some human breast studies, which suggested that cerbB-2-overexpressing tumours should be included in a bona fide basal-like subclass (Gama
et al., 2008; Matos et al. 2005). In contrast, the majority of luminal tumours in Gama et al.
(2008) series were simultaneously negative for basal cell markers, with some cases showing
basal marker expression, which was also described by some authors who reported coexpressing basal CK and hormone receptors or c-erbB-2.
The correlation between subtypes obtained and histology showed that ER positive luminal A
tumours more frequently associated with complex tumour type, low histological grade, less
invasive and low proliferative tumours (Gama et al., 2008; Sorlie et al., 2003), whereas basallike and c-erbB-2 overexpressing subtypes were associated with simple and carcinosarcoma
tumour types, high histological grade, lymphovascular invasion and high proliferation,
32
features that are in accordance to the ones described in human literature for basal-like cancers
(Gama et al., 2008; Kim et al., 2006; Matos et al., 2005; Sorlie et al., 2001).
In terms of prognosis, it is already well known that molecular phenotypes are associated with
different clinical outcomes in canine mammary tumours as well as in human tumours (Gama
et al., 2008; Sorlie et al., 2003). Basal subtype is associated with lower survival rates and
appearence of recurence, similarly to human breast cancer studies. In contrast to basal
subgroup, luminal and c-erbB-2 overexpressing subtypes show increased survival rates.
The fact that luminal tumours were associated with a better prognosis is not surprising since
ER positive human breast carcinomas are usually associated with a more favourable clinical
outcome. In veterinary pathology, however, the prognostic value of ER in canine mammary
cancer is still a matter of debate. C-erbB-2 overexpressing tumours were found usually
associated with established indicators of poor prognosis such as large tumour size, high
histologic grade, invasion, simple histologic type and high proliferative indices (Gama et al.,
2008; Rakha et al., 2006; Sorlie et al., 2008). However, Kaplan–Meier analysis in the study
by Gama et al. (2008) revealed that C-erbB-2 subtype was related with a more favourable
clinical outcome, findings that were in contrast with human studies, which describe similar
survival rates for c-erbB-2 overexpressing and basal-like subtypes (Gama et al., 2008; Rakha
et al., 2006: Sorlie et al., 2008).
In order to evaluate the prognostic potential of the hormone receptors expression in canine
malignant mammary tumours, Chang et al. (2009) conducted a study on benign and malignant
tumours. The expression of ER and PR was significantly more frequent in benign tumours
compared to the malignant counterpart. Moreover, among the malignant samples only those
expressing ER and PR had a high survival rate, compared with malignant tumours expressing
ER but not PR suggesting, therefore, that the PR could be considered as an important
prognostic factor (Chang et al., 2009).
33
It is known that malignant tumours negative to PR receptors have a higher proliferation index
compared to those PR+, suggestive of the fact that the progression to malignancy in
spontaneous mammary tumours was accompanied by a decrease in the dependence hormonal
steroid (Geraldes et al., 2000).
In the study by Millanta et al. (2005), the expression pattern of canine steroid receptors
showed that ER expression is significantly high in the normal tissue, in hyperplastic and
dysplastic lesions, and in benign tumours, but significantly lower in carcinomas. There were
not apparent significant changes in the reactivity of PR in normal tissue, dysplastic and
benign neoplastic lesions except for a significant reduction of expression in carcinomas. The
absence of ER and PR only in the female dog is associated with a high mortality rate and
these data confirm previous observations that demonstrated that canine mammary tumours
have features in common with hormone-dependent human breast cancer (Millanta et al.,
2005a)
The immunohistochemical investigation of ER in canine mammary tumours was found to be a
simple application technique with prognostic value that could be useful for appropriate
hormonal therapies selection (Nieto et al., 2000).
With regard to the prognostic value of c-erbB-2, studies present in literature reported
controversial data, in spite of what is known in the woman. According to Hsu et al. (2009),
the relationship between the clinical course and the protein expression of c-erbB-2 in dogs
with malignant mammary tumour indicated a greater overall survival rate in c-erbB-2
overexpressing tumours compared to those with normal levels of the antigen. The reason for
such a difference between c-erbB-2 overexpressing phenotype in canine and in women breast
cancer remained unclear for the authors. Certainly c-erbB-2 seemed to play an important role
in tumour formation, but did not seem to be directly correlated with the progression to
malignancy (Hsu et al., 2009). Given the high expectation of suraival rate in canine mammary
34
c-erbB-2+ tumours, the role of this protein in oncogenesis may be different from that played
in breast cancer (Hsu et al., 2009).
Opposite results results exerted that c-erbB-2+ tumours were correlated with indicators of
poor prognosis, such as tumour size >3cm, histological grade III [presence of neoplastic
emboli (Gilbertson et al., 1983)], type of invasive growth, absence of hormone receptors and
a period of <6 months without relapse after surgery. Those results were in agreement with
what has been observed in the human, having characteristics indicative of a worse prognosis
(Martin de las Mulas et al., 2003; Sorlie et al., 2003).
Mammary tumours in the dog are characterized by significant high molecular heterogeneity,
thereby not surprising that individual markers could not accurately estimate the heterogeneity
of breast cancer (Rakha et al., 2009) and that, on the contrary, could benefit from a
classification based on their molecular differences.
Sassi et al. (2010) applied an immunohistochemical panel (anti –ER, -PR, c-erbB-2, -CK5/6
and -CK14) to a series of canine mammary carcinomas in order to: identify molecular
phenotypes based on a modified molecular classification, find possible correlation between
the phenotypes and stage and histological grade, and use the phenotypes as a prognostic aid
in veterinary practice. The molecular phenotypes have been identified in accordance with the
flowchart classification proposed by Conforti et al. (2007) (Fig.3).
35
Fig. 3: Flowchart classification based on Sassi et al. (2010)
Out of 45 samples, the panel of antibodies has identified only three tumour groups (luminal
A, luminal B and Basal-like) out of the five groups known in Human Medicine (Sassi et al.,
2010) and still different from the four identified (luminal A and B, Basal-like, c-erbB-2
overexpressing) found by Gama et al. (2008). The uses of different panels and criteria to
define markers positivity was probably the cause of the differences found between Gama et
al. (2008) and Sassi et al. (2010) studies. Sassi et al. (2010) demonstrated a correlation
between the molecular based classification and the histological grade, but they did not find a
relation between the stage and the morphological classification of tumours. In addition,
luminal A phenotype included a high percentage of grade I tumour, compared to luminal B in
which prevailed grade II and III tumours. On the contrary, Gama et al. (2008) showed that
only the basal-like phenotype was correlated with grade and vascular invasion. Sassi et al.
36
(2010) did not detect any similarity between the molecular classification system and survival
rates, while Gama et al. (2008) found only for the basal phenotype the association with short
survival.
In multivariate analysis, according to Sassi et al. (2010), staging and histological grade
showed an independent association with survival rate, while the phenotypes and the
histological types did not show any association. These data suggested that caution should be
used when applying the new classification system for canine mammary tumours, in which the
fundamental prognostic information derived from the staging and histological grade (Sassi et
al., 2010).
Canine mammary tumours have been treated in different ways with surgery as the first choice
of therapy, either alone or in combination with chemotherapy (Lana et al., 2007) even though
no standard therapeutic protocols are available (Lavalle et al., 2012). Receptor evaluation
has been introduced to use an anti-estrogen therapy, whose side-effects include endometritis
in female dogs with ER negative tumours (Morris et al., 1993)
37
References
Boecker W, Buerger H (2003). Evidence of progenitor cells of glandular and myoepithelial
cell lineages in the human adult female breast epithelium: a new progenitor (adult stem) cell
concept. Cell Proliferation 36(Suppl 1):73–84
BoeckerW,Moll R, Poremba C, Holland R, Van Diest PJ, Dervan P, Burger H, Wai D, Ina
Diallo R, Brandt B, Herbst H, Schmidt A, Lerch MM, Buchwallow IB (2002). Common adult
stem cells in the human breast give rise to glandular and myoepithelial cell lineages: a new
cell biological concept. Laboratory Investigation 82:737–746
Chang CC, Tsai MH, Liao JW, Chan JP, Wong ML, Chang SC (2009). Evaluation of
hormone receptor expression for use in predicting survival of female dogs with malignant
mammary gland tumour. Journal of American Veterinary Medical Association 235(4):391396.
Conforti R, Boulet T, Tomasic G, Taranchon E, Arriagada R, Spielmann M, Ducourtieux M,
Soria JC, Tursz T, Delaloge S, Michiels S, Andre F (2007). Breast cancer molecular
subclassification and estrogen receptor expression to predict efficacy of adjuvant
anthracyclines-based chemotherapy: a biomarker study from two randomized trials. Annual
Oncology 18(9):1477-1483.
Gama A, Alves A, Schmitt F (2008). Identification of molecular phenotypes in canine
mammary carcinomas with clinical implications: application of the human classification.
Virchows Archiv 453(2):123-132.
Geraldes M, Gärtner F, Schmitt F (2000). Immunohistochemical study of hormonal receptors
and cell proliferation in normal canine mammary glands and spontaneous mammary tumours.
Veterinary Record 146(14):403-406.
38
Gilbertson SR, Kurzman ID, Zachrau RE, Hurvitz AI, Black MM (1983). Canine Mammary
Epithelial Neoplasms: Biologic Implications of Mophologic Characteristics Assessed in 232
Dogs. Veterinary Pathology 20:127–142.
Hsu WL, Huang HM, Liao JW, Wong ML, Chang SC (2009). Increased survival in dogs with
malignant mammary tumours overexpressing HER-2 protein and detection of a silent single
nucleotide polymorphism in the canine HER-2 gene. Veterinary Journal 180(1):116-123.
Kim MJ, Ro JY, Ahn SH, Kim HH, Kim SB, Gong G (2006) Clinicopathologic significance
of the basal-like subtype of breast cancer: a comparison with hormone receptor and
Her2/neuoverexpressing phenotypes. Human Pathology 37:1217–1226.
Lana SE, Rutteman GR, Withrow SJ (2007) Tumour of the mammary gland. In: Withrow &
MacEwen’s Small Animal Clinical Oncology, ed. Withrow SJ and Vail DM, 4th ed., pp. 631.
Saunders Elsevier Press, Philadelphia.
Lavalle GE, De Campos CB, Bertagnolli AC, Cassali GD (2012) Canine malignant mammary
gland neoplasms with advanced clinical staging treated with carboplatin and cyclooxygenase
inhibitors. In Vivo, 26, 375–379.
Matos AJ, Baptista CS, Gärtner MF, Rutteman GR (2012) Prognostic studies of canine and
feline mammary tumour: The need for standardized procedures. Veterinary Journal 193, 24–
31
Matos I, Dufloth R, Alvarenga M, Zeferino LC, Schmitt F (2005) P63, cytokeratin 5, and Pcadherin: three molecular markers to distinguish basal phenotype in breast carcinomas.
Virchows Archiv 447:688–694.
Morris JS, Dobson JM, Bostock DE (1993) Use of tamoxifen in the control of canine
mammary neoplasia. Veterinary Records 27:539–542.
Nielsen TO, Hsu FD, Jensen K, Cheang M, Karaca G, Hu Z, Hernandez-Boussard T, Livasy
C, Cowan D, Dressler L, Akslen LA, Ragaz J, Gown AM, Gilks CB, van de Rijn M, Perou
39
CM (2004). Immunohistochemical and clinical characterization of the basal-like subtype of
invasive breast carcinoma. Clinical Cancer Research 10(16):5367-5374.
Nieto A, Peña L, Pérez-Alenza MD, Sánchez MA, Flores JM, Castaño M (2000).
Immunohistologic detection of estrogen receptor alpha in canine mammary tumour: clinical
and pathologic associations and prognostic significance. Veterinary Pathology 37(3):239-247.
Pinho SS, Carvalho S, Cabral J, Reis CA, Gärtner F (2012). Canine tumour: a spontaneous
animal model of human carcinogenesis. Translational Research 159(3):165-172.
Rakha EA, El-Rehim DA, Paish C, Green AR, Lee AH,Robertson JF, Blamey RW,
Macmillan D, Ellis IO (2006) Basal phenotype identifies a poor prognostic subgroup of breast
cancer of clinical importance. Journal of European Cancer 42:3149–3156.
Rakha EA, El-Sayed ME, Reis-Filho J, Ellis IO (2009). Patho-biological aspects of basal-like
breast cancer. Breast Cancer Research Treatment 113(3):411-22.
Rao NA, van Wolferen ME, van den Ham R, van Leenen D, Groot Koerkamp MJ, Holstege
FC, Mol JA (2008). cDNA microarray profiles of canine mammary tumour cell lines reveal
deregulated pathways pertaining to their phenotype. Animals Genetic 39(4):333-145.
Rivera P, von Euler H (2011). Molecular biological aspects on canine and human mammary
tumour. Veterinary Pathol 48(1):132-146.
Sarli G, Sassi F, Brunetti B, Benazzi C (2007). Luminal-like A and B types in canine
mammary carcinomas in: Proceedings of the 25th Annual Meeting of the European Society of
Veterinary Pathology, s.l, s.n, 2007 pp. 187 (25th Annual Meeting of the European Society of
Veterinary Pathology).
Sassi F, Benazzi C, Castellani G, Sarli G (2010). Molecular-based tumour subtypes of canine
mammary carcinomas assessed by immunohistochemistry. BMC Veterinary Research 28:6:5.
Sorlie T, Tibshirani R, Parker J, Hastie T, Marron JS, Nobel A, Deng S, Johnsen H, Pesich R,
Geisler S, Demeter J, Perou CM, Lønning PE, Brown PO, Børresen-Dale AL, Botstein D
40
(2003). Repeated observation of breast tumour subtypes in independent gene expression data
sets. Proceedings of the National Academy of Sciences of the United States of America
100:8418-8423.
Uva P, Aurisicchio L, Watters J, Loboda A, Kulkarni A, Castle J, Palombo F, Viti V, Mesiti
G, Zappulli V, Marconato L, Abramo F, Ciliberto G, Lahm A, La Monica N, de Rinaldis E
(2009). Comparative expression pathway analysis of human and canine mammary tumour.
BMC Genomics 10:135.
41
5. MOLECULAR PHENOTYPES IN FELINE MAMMARY TUMOURS
Mammary tumours are common in female cats, especially in older animals, and the
majority of them are malignant and very aggressive (MacEwen and Withrow 1996;
Multon, 1990). Feline malignant tumours grow rapidly, and metastases are reported to occur
in 50–90% of affected animals (Hayden and Nielsen, 1971). Metastasis to regional lymph
nodes (83%), lungs (83%), pleura (22%) and liver (25%) are most common. However,
various studies have also documented widespread metastases to the adrenal glands,
diaphragm and kidneys (Giménez et al., 2010).
Nowadays, the most important prognostic factor in cats with mammary gland neoplasia is
tumour size, which significantly affects both disease-free interval and survival time (Lana
et al., 2001). Other important prognostic factors are:

histopathologic grade: An association between survival time and histopathologic
grade has been demonstrated. The rate of death 1 year after surgery was 0% in cats
with well-differentiated carcinoma, and 100% in those with poorly differentiated
carcinoma. However, there was not a good correlation between moderate
differentiation and survival time (Jeglum et al., 1985);

mitotic rate: The number of mitotic figures found in tumour tissue has been shown
to be of prognostic value. Longer survival times were seen in animals with
tumours exhibiting fewer than two mitotic figures per high power field (Weijer et
al., 1983);

disease stage: Clinical stage at presentation has been shown to be significantly
associated with survival time (Ito et al. 1996);

surgical approach: In a study of cats
adenocarcinoma,
bearing malignant mammary
which had undergone radical mastectomy,
42
a significantly
reduced rate of local recurrence was compared with cats that had undergone
conservative surgery (MacEwen et al., 1984);

molecular markers: The expression of some genes, receptors or proteins can be
altered during the malignant process in feline mammary tumours. These so-called
molecular markers can be detected by immunohistochemistry and yield useful
prognostic information in some cases (Giménez et al., 2010). Most studied markers
are: hormone receptors ER and PR (Millanta et al., 2006), c-erbB-2 (Millanta et
al., 2005b), cyclin A, protein p53, macrophage-stimulating protein receptor
(RON), vascular endothelial growth factor (VEGF), cyclo-oxygenase (COX) and
topoisomerase IIβ binding protein 1 (TopBP1) (Giménez et al., 2010).
Regarding the hormone receptors state, Martín de las Mulas et al. (2002) demonstrated
through the analysis of the immunohistochemical expression of PR receptors in normal,
dysplastic and neoplastic mammary glands of cats that benign mammary lesions, like those in
dogs, appeared to have higher PR receptor profiles than mammary carcinomas, with 66.7% of
benign tumours and dysplasias and 37.5% of malignant mammary lesions being progesterone
receptor positive. The elevate percentage of feline PR+ tumours carcinomas suggested the
possible role of PR as an early promoter of tumour growth in the cat (Millanta et al., 2006). In
2000, Martín de las Mulas et al. consided the expression of ER, detecting significant decrease
in ER positivity between normal and dysplastic tissues and invasive carcinomas. It has been
suggested that steroid hormones may play a role in promoting cellular proliferation in the
early stages of development of canine mammary tumours (Rutteman, 1990). Cellular
proliferation has been recognised as a prognostic indicator in ER−human breast carcinomas,
whereas benign mammary lesions in both cats and dogs (Rutteman et al., 1993) had higher
ER profiles. Estrogens might act by binding ER normally present in epithelial and stromal
cells of mammary gland inducing in these cells an increased expression of PR receptors. In
43
this hypothesis the un-reactivity of epithelial and stromal cells for ER could be explained with
a previous presence of ER in these tissues in the initial stage of the diseases and a progressive
loss during the evolution of the disease (Millanta et al, 2005a). The high negative rates of ER
expression seem to be a characteristic feature of feline mammary carcinomas when compared
with human and canine tumours and suggest a lack of estrogen dependency as previously
proposed. Also, the reduced percentage of ER+ invasive carcinomas demonstrate the behavior
and phenotype of typically aggressive mammary tumour in cats (Martín de las Mulas et al.,
2000). The evaluation of ER receptors state in normal, dysplastic and neoplastic feline
mammary tissue assumed a prognostic value as ER expression was significantly higher in
normal and hyperplastic tissue more than in neoplastic tissue. In light of the correlation
between the hormone receptors expression and the proliferation index obtained by
immunolabeling with MIB-1 (Mouse Monoclonal IgG1), ER+ phenotypes had a proliferation
index significantly lower than ER- phentypes, having therefore a poorer prognosis (Millanta
et al., 2006). Considering both estrogen and progesterone receptors,
feline mammary
carcinomas showed features in common with hormone-independent human breast cancer
(Millanta et al., 2005a)
The feline c-erbB-2 was found to be barely detectable in normal mammary gland, increased in
mammary benign tumours, and elevated in a high percentage in carcinomas. The correlation
between c-erbB-2 overexpression and the overall survival rate showed that cats affected by cerbB-2 overexpressing phenotype tumours had shorter overall survival time. These findings
suggested that c-erbB-2 status could provide valuable prognostic and predictive information
(Giménez et al., 2010). The effect of c-erbB-2 overexpression in feline mammary tumours
seemed to be more similar to human tumours than in those of the dog (Millanta et al., 2005b).
44
Treatment options that have been studied for feline mammary neoplasia are surgical excision, chemotherapy, immunotherapy and radiation therapy. These differ in terms of clinical
outcome (Giménez et al., 2010; Lana et al., 2007). Regarding the chemiotherapy, a recent
retrospective study on the use of adjuvant doxorubicin therapy after surgical excision in 67
cats reported a Kaplan Meier median survival time of 448 days and called for randomized
trials to prove the true efficacy of chemotherapy (Novosad et al., 2006).
As for canine mammary carcinomas, also for queens no standard therapeutic protocols are
available (Lavalle et al., 2012)
45
References
Giménez F, Hecht S, Craig LE, Legendre AM (2010). Early detection, aggressive therapy:
optimizing the management of feline mammary masses. Journal of Feline Medicine and
Surgery 12(3):214-224.
Hayden DW, Nielsen SW (1971). Feline mammary tumours. Journal of Small Animal
Practice 12:687–698.
Ito T, Kadosawa T, Mochizuki M, Matsunaga S, Nishimura R, Sasaki N (1996). Prognosis of
malignant mammary tumour in 53 cats. Journal of Veterinary Medical Science 58:723–726.
Jeglum KA, de Guzman E, Young KM (1985). Chemotherapy of advanced mammary
adenocarcinoma in 14 cats. Journal of the American Veterinary Medical Association 187:
157–60.
Lana SE, Rutteman GR, Withrow SJ (2001). Tumours of the mammary gland. In: Withrow
SJ, Vail DM, eds. Small animal clinical oncology. 4th edn. Canada: Saunders Elsevier, pp.
628–636.
Lavalle GE, De Campos CB, Bertagnolli AC, Cassali GD (2012) Canine malignant mammary
gland neoplasms with advanced clinical staging treated with carboplatin and cyclooxygenase
inhibitors. In Vivo, 26, 375–379.
MacEwen EG, Hayes AA, Harvey HJ, Patnaik AK, Mooney S, Passe S (1984). Prognostic
factors for feline mammary tumours. Journal of American Veterinary Medicine Association
185: 201–204.
Macewen EG and Withrow SJ (1996). Tumours of the mammary gland. In Small Animal
Clinical Oncology. 2nd edn. Eds E. G. MacEwen, S.J. Withrow. London, W.B. Saunders. pp
356-379
46
Martín De Las Mulas J, Van Niel M, Millán Y, Blankenstein MA, Van Mil F, Misdorp W
(2000). Immunohistochemical Analysis of Estrogen Receptors in feline mammary gland
benign and malignant lesions: comparison with biochemical assay. Domestic Animal
Endocrinology 18:111-125.
Martín De Las Mulas J, Van Niel M, Millán Y, Ordás J, Blankenstein MA, Van Mil F,
Misdorp, W (2002). Progesterone receptors in normal, dysplastic and tumourous feline
mammary glands. Comparison with oestrogen receptors status. Research in Veterinary
Science 72:153-161.
Millanta F, Calandrella M, Bari G, Niccolini M, Vannozzi I, Poli A (2005a). Comparison of
steroid receptor expression in normal, dysplastic, and neoplastic canine and feline mammary
tissues. Research Veterinary Science 79(3):225-232.
Millanta F, Calandrella M, Citi S, Della Santa D, Poli A (2005b). Overexpression of HER-2
in feline invasive mammary carcinomas: an immunohistochemical survey and evaluation of
its prognostic potential. Veterinary Pathology 42(1):30-34.
Millanta F, Calandrella M, Vannozzi I, Poli A (2006). Steroid hormone receptors in normal,
dysplastic and neoplastic feline mammary tissues and their prognostic significance.
Veterinary Records 158(24):821-824.
Moulton JE (1990). Mammary tumours of the cat. In: Moulton JE, ed. Tumours in domestic
animals. 3rd edn. Berkeley: University of California Press, pp. 547–552.
Novosad CA, Bergman PJ, O’Brien MG, McKnight JA, Charney SC, Selting KA, Graham
JC, Correa SS, Rosenberg MP, Gieger TL. (2006) Retrospective evaluation of adjunctive
doxorubicin for the treatment of feline mammary gland adenocarcinoma: 67 cases. Journal of
the American Animal Hospital Association 42:110-120.
Rutteman, GR (1990). Hormones and mammary tumours disease in the female dog: an
update. In Vivo 4, 33–40.
47
Rutteman GR, Misdorp W, Blankenstein MA, van den Brom WE (1988). Oestrogen (ER) and
progestin receptors (PR) in mammary tissue of the female dog: different receptor profile in
nonmalignant and malignant states. British Journal of Cancer 58, 594–599.
Weijer K, Hart AA (1983). Prognostic factors in feline mammary carcinoma. Journal of the
National Cancer Institute 70: 709–716.
48
EXPERIMENTAL SECTION
49
1. MOLECULAR PORTRAIT-BASED CORRELATION BETWEEN
PRIMARY CANINE MAMMARY TUMOUR AND ITS LYMPH NODE
METASTASIS: POSSIBLE PROGNOSTIC-PREDICTIVE MODELS
AND/OR STRONGHOLD FOR SPECIFIC TREATMENTS?
Introduction
Canine mammary tumour and human breast cancer are heterogeneous diseases commonly
occurring in female dogs [1,2] and in women [3,4].
Traditionally, breast cancer has been classified by morphological criteria in both human [5]
and veterinary [6,7] medicine. Recent veterinary attention, as widely explained in the general
part regard canine mammary tumours, has focused on the protein expression profile and four
main carcinoma subtypes have been identified [8,9]: luminal A, luminal B, c-erbB-2
overexpressing and basal-like. Recently, molecular characterization in human breast cancer
has also been applied to the metastasing lymph nodes [15]. The metastatic process is in fact
the most urgent, important and difficult issue to approach in human [16] and animal cancer
medicine. The relationship between the primary tumour and the lymph node metastasis from
50
the same patient was studied by Wu et al. [15] to determine if satellite tumour are uniform or
divergent in molecular properties and to provide new information of diagnostic and
therapeutic significance [16].
Recent publications emphasized several similarities between human breast cancer and canine
mammary tumour, such as the relative age at onset, incidence, risk factors, biological
behaviour, metastatic pattern, histopathological and molecular features, metastases-associated
expression profile [17], and response to therapy [18,19].
The aim of this study was to analyze the relationship between the primary mammary tumour
and its lymph node metastasis based on immunohistochemical molecular characterization to
develop the most specific prognostic-predictive models and targeted therapeutic options.
Methods
Samples
Specimens of mammary carcinomas from 20 female dogs were collected from the database of
the Pathology Service of the Department of Veterinary Medical Science of Bologna
University and from the Department of Animal Pathology of Pisa University.
The 20 dogs comprised 11 mixed breed, two German Shepherd, one Yorkshire, two English
Setter, two Doberman, one Maremma Shepherd and one Poodle. Dog ages ranged from six to
14 years with a mean age of 10.3 and a median of 10.5. Two-year follow-up survival data
were available for 11 out of the 20 animals with invasive carcinomas included in the study.
Overall survival time was defined as the time from the day of diagnosis until the day of death
or last follow-up. All the latest data are summarized in Table 1.
Cases were selected based on both the primary mammary neoplasia and histological grade II
(grade II: invasive carcinoma and metastases to regional lymph nodes) according to
51
Gilbertson et al. [25]. No cases displayed systemic metastases. Samples were available as
sections stained with hematoxylin and eosin and obtained from formalin-fixed and paraffinembedded tissue block.
Table 1: Individual data (Beha et al., 2012)
Histological diagnosis and immunohistochemistry
Histological diagnosis was made according to the WHO classification system [6]. Seven
consecutive 4μm-thick sections were cut from the paraffin blocks containing representative
tumour samples and labeled by immunohistochemistry with the following antibodies: antiER, -PR, -c-erbB-2, -CK5/6, -CK14, -CK19, -p63. Data on the primary antibodies are
summarized in Table 2.
52
Table 2: Primary antibodies, resources and dilutions used in immunohistochemistry (Beha et al., 2012)
ANTIBODY (−ANTI)
CLONE
MANUFACTURER
DILUTION
ER
B-10
Abcam, Cambridge, UK
1: 300
PR
PR4-12
Oncogene TM, Boston, MA, USA
1: 100
c-erbB-2
Polyclonal
Dako, Glostrup, Denmark
1: 250
Cytokeratins 5/6
D5/16B4
Zymed (South San Francisco, CA, USA)
1: 100
Cytokeratin 14
Ab-1 (LL002)
NeoMarkers (Fremont, CA, USA)
1: 300
Cytokeratin 19
BA17
Dako (Glostrup, Denmark)
1: 50
p63
4A4
Dako (Glostrup, Denmark)
1: 50
Sections were dewaxed in toluene and rehydrated. Endogenous peroxidase was blocked by
immersion in H2O2 0.3% in methanol for 20 min. Sections were then rinsed in Tris Buffer
and antigen was retrieved with citrate buffer (2.1 g citric acid monohydrate/liter distilled
water), pH 6.0 (except for CK 5/6 which use EDTA, pH 8.0), and heating for two 5-min
periods in a microwave oven at 750 W, followed by cooling at room temperature for 20 min.
All antibodies were incubated with the tissue sections overnight at 4°C, and their binding was
revealed by a commercial streptavidin-biotin-peroxidase technique (LSAB Kit, Dako,
Amsterdam, The Netherlands). Diaminobenzidine (0.05% for 10 min at room temperature)
was used as chromogen. Slides were counterstained with Papanicolaou's hematoxylin.
As a negative control, the primary antibody was replaced with an irrelevant, isotype-matched
antibody to control for non-specific binding of the secondary antibody. As positive controls to
assess the cross-reactivity with canine tissues and the specificity of the immunohistochemical
stain, sections of normal canine uterus (for anti-ER and -PR antibodies), canine skin (for antiCK5/6, -CK14, -CK19 and -p63 antibodies) were used following the same protocols. A
53
human poorly differentiated invasive ductal mammary carcinoma (kindly provided by P.
Viacava, Department of Oncology, University of Pisa, Italy) known to react with c-erbB-2
antibody was used as positive control.
The staining result was classified semi-quantitatively with a dichotomous evaluation: positive
or negative. The sample was considered positive:
•
when presenting cytoplasmic stain in more than 1% of the invasive tumour cells for
anti-CK-19, CK-5/6 and anti-CK14 antibodies [26];
•
when presenting complete membranous stain in more than 10% of tumour cells for
anti- c-erbB-2 antibody according to the Hercept-test [27];
•
when presenting nuclear stain in more than 5% of tumour cells for anti-ER and anti-
PR antibodies [28];
•
when presenting nuclear stain in more than 10% of tumour cells for anti-p63 antibody
[29].
Molecular taxonomy
The application of the panel allowed cases to be grouped into five molecular subtypes
according to an algorithm by Sassi et al. [9] modified as follows:
•
Luminal-A: ER+ and/or PR+, c-erbB-2–, regardless of CK5/6, CK14, p63 staining.
•
Luminal-B: ER+ and/or PR+, c-erbB-2+, regardless of CK5/6, CK14, p63 staining.
•
c-erbB-2 overexpressing: ER–, PR–, c-erbB-2+ regardless of CK5/6, CK14, p63
staining.
•
Basal-like: ER–, PR–, c-erbB-2–, CK5/6+ and/or CK14+ and/or p63+.
•
Normal-like: Negative to all markers.
Results
54
Diagnosis
Eight of the 20 primary tumour were classified as simple tubulopapillary carcinomas, eight as
solid carcinomas, two as complex carcinomas and two as anaplastic carcinomas.
Immunohistochemistry and molecular phenotypes
Immunohistochemistry for ER, PR, c-erbB-2, CK5/6, CK14, p63 in the primary tumour and
in the respective lymph node metastasis is summarized in Table 3. In each case the epithelial
origin of cancer was confirmed by CK19 staining. Based on the applied algorithm, molecular
phenotypes were obtained in the primary mammary tumour and in their lymph node
metastases (Table 4). Four phenotypes (luminal A (Figure 1A-B, line 1-2-3), luminal B
(Figure 2A-B, line 1-2-3-4), c-erbB-2 overexpressing (Figure 3A-B), basal-like (Figure 4A-BC)) were diagnosed in primary tumour (eight (40 %), seven (35 %), two (10%), three (15%)
respectively) and five (luminal A (Figure 1C-D-E, line 1), luminal B (Figure 2C-D-E, line 1),
c-erbB-2 overexpressing (Figure 2C-D-E, line 2 and Figure 3c-d-e), basal-like (Figure 2c-d-e,
line 3 and Figure 4c-d-e), normal-like (Figure 1c-d-e, line 3 and line 4 of Figure 2c-d-e)) in
the lymph node metastases (five (25%), three (15%), four (20%), six (30%), two (10%)
respectively).
Relationship between molecular phenotype in the primary mammary tumour
and its related lymph node metastasis
Phenotypic concordance was found in 13 of the 20 cases (65%) (five luminal A (Figure 1,
line 1), three luminal B (Figure 2, line 1), two c-erbB-2 overexpressing (Figure 3) and three
basal-like (Figure 4)). Seven cases (35%) showed discordance with lymph node phenotypic
profile differing from the primary tumour (two luminal A became basal-like (Figure 1, line 2),
one luminal A became normal-like (Figure 1, line 3), two luminal B became c-erbB-2
55
overexpressing (Figure 2, line 2), one luminal B became basal-like (Figure 2, line 3), one
luminal B became normal-like (Figure 2, line 4)) (Table 3). C-erbB-2 overexpressing and
basal-like primary tumour were 100% concordant with the molecular phenotype of the
respective lymph node metastases (Figure 3 and Figure 4). Luminal A and luminal B were
concordant in 65.2% and 42.9% respectively of primary tumour for the same comparison.
Table 3: Summary of immunohistochemical staining
56
Table 4: Molecular phenotypes
Histological diagnosis in primary mammary tumour compared to molecular
phenotypes and to concordance/discordance
In the primary tumour the luminal A phenotype displayed a different histological tumour
pattern, i.e. simple tubulopapillary (one case) (Figure 1A-B, line 1), solid (six cases) (Figure
1A-B, line 2), anaplastic (one case) (Figure 1A-B, line 3). Luminal B phenotype displayed a
57
different histological tumour pattern, i.e. simple tubulopapillary (four cases) (Figure 2A-B,
line 1 and 4), solid (two cases) (Figure 2A-B, line 2) and complex (one case) (Figure 2A-B,
line 3). The c-erbB-2 overexpressing phenotype showed two patterns, i.e. tubulopapillary
(one case), and anaplastic (one case) (Figure 3A-B). The basal-like phenotype displayed
different histological tumour patterns, i.e. simple tubulopapillary (two cases) (Figure 4A-B-C)
and complex (one case). No association was found between histological diagnosis and
phenotype in the primary tumour (Pearson Chi-square, P=0.065; for statistics only the
tubulopapillary and the solid carcinomas were considered because the other two types had
only one case for phenotype).
Six cases with tubulopapillary pattern displayed concordance between the primary tumour
and its related lymph node metastasis whereas the other two cases showed discordance. In the
solid pattern concordance was found in five cases and discordance in three. In the anaplastic
and complex patterns both concordance and discordance were present with one case for each
type. Comparing the four histological patterns, no differences in the percentages of
concordance/discordance were evident (Pearson Chi square P=0.857, for statistics only the
tubulopapillary and the solid carcinomas were considered because the other two types had
only one case for each phenotype).
Influence of the molecular phenotype and concordance/discordance on dog’s
survival rate
Table 1 reports the survival data of the 11 animals. Few data are available to group the
subjects according to molecular phenotypes and concordance/discordance to perform survival
analysis even though it appears that all five animals alive at 24 months post-surgery were
concordant luminal A or luminal B cases. The other six dead animals bore primary
tumour/lymph node concordant (2 basal-like and 2 c-erbB-2 overexpressing) or discordant (1
58
luminal A/normal like; 1 luminal B/normal-like) cases in which less differentiated molecular
types were present in both sites or only in the lymph node compared to the live animals.
Discussion
Canine mammary carcinomas can become fatal due to the development of distant metastases.
One of the most important prognostic factors in the diagnosis is the acknowledgment of
metastases to the regional lymph node that represents an early step in metastatic spread [30].
Klopfleisch et al. [17] and Lu et al. [31] demonstrated that metastatic spread of canine
mammary tumour to the lymph nodes is associated with a gene expression profile of increased
cell cycle progression, altered cell differentiation and decreased growth factor signaling.
Metastasis development is a complex process involving invasion, intravasation, survival in
the bloodstream, extravasation and homing and proliferation at the site of metastasis [32].
Although some phenotypes showed greater aggressiveness and metastatic capability, only a
selected subpopulation was able to metastatize in the multiple and heterogeneous tumour cell
population. In this case the phenotype may have been transient and these selected cells have
had an intrinsic program to transition to a phenotype enhancing their ability for heterotypic
interaction and survival proliferation in distant organs [32] as Darwinian clonal evolution.
Conversely, the metastatic process has also been described as a stochastic event, the primary
tumour cells having equal metastatic capability, characterized by a phenotypic overlap
between the primary tumour and its metastases [33,34]. Thus the identification of molecular
phenotypes in primary tumour and metastases can provide predictive information on the most
likely metastatic profile, not the condition in the primary tumour.
Sassi et al. [9] identified three phenotypes out of the four detected by Gama et al. [8],
demonstrating that basal-like subtypes were associated with a better outcome than luminal A
and luminal B tumour, in contrast with the findings of Gama et al. [8]. The prognostic role of
59
c-erbB-2 overexpression remains controversial despite what is known in human medicine.
According to a study by Hsu et al. [35], the relationship between the clinical course and
protein expression of c-erbB-2 in dogs with malignant mammary neoplasia indicated a
greater survival rate in tumour overexpressing c-erbB-2 compared to those having
nonoverexpressed levels of antigen. Certainly, c-erbB-2 plays an important role in
carcinogenesis, but does not seem to be directly correlated with progression to malignancy
[35]. In the present investigation it seems that luminal A or B concordance should be
considered a positive prognostic factor, whereas concordance for the other molecular types or
discordance should not, even if these results await confirmation in a larger number of cases
and proper statistical analysis.
This study revealed four out of the five protein expression phenotypes of breast cancer in
primary tumour (20 cases): luminal A (40%), luminal B (35%), c-erbB-2 overexpressing
(10%) and basal-like (15%). The prevalence of luminal phenotypes (75%) over the others
(25%) is in accordance with findings both in human [11,36,37] and veterinary [8;9]
medicine.
Based on the present study and in agreement with Brunetti et al. [38], labeling for CK and
p63 would only appear necessary when a tumour is negative for ER, PR and c-erbB-2.
With regard to luminal A and B phenotypes, the expression profiles of ER and PR are
essential to decide on the application of endocrine therapy [39] in breast cancer and canine
mammary neoplasia, and also seem to play a minor role in predicting tumour biological
behaviour [40,41]. Wu et al.’s study [15] in breast cancer confirmed the observation that ER
and/or PR could be lost when carcinomas metastasizes, thereby resisting endocrine therapy.
The present study shows almost overlapping results, losing hormone receptors by moving
from luminal A to basal-like (2 cases) and/or to normal-like (1 case) phenotypes and from
60
luminal B to c-erbB-2 overexpressing (2 case) and/or to basal-like (1 case), and/or to normal
like (1case), confirming that the gene expression profile in canine mammary tumour may
prove a helpful tool in clinical practice. Chang et al. [42] indicated that the ER or PR
expression in dogs was associated with tumour size, clinical stage, and lymph node metastasis
or distant metastases. Dogs with malignant mammary neoplasia and expression of both ER
and PR had a longer survival rate than dogs with malignant mammary tumour that were ER
positive but PR negative. This latest information on PR suggested that the receptor was a
better outcome predictor than ER status alone and that its positive or negative expression
could serve as a prognostic factor, especially in dogs with malignant neoplasia with ER
expressed [42]. The present results disclosed a high prevalence of hormone receptor
expression in the primary tumour, whose positivity was ensured by reactivity to least one of
the two markers (ER and/or PR) (Table 3). The latest results indicate that there are grounds
for the use of anti-hormone therapy in dogs, administering molecules other than those
hitherto used in veterinary medicine (tamoxifen) as their side-effects are already well-known
[24]. A similar analysis in lymph node showed a net loss of hormone receptor expression,
namely ER. ER loss is a known adverse prognostic factor [42], and therefore its lack of
expression in metastases is indicative. In this study, only five out of the 20 cases showed
positivity to both ER and PR in the primary tumour, with persistent positivity in the lymph
node metastasis in only two cases. The remaining three cases showed loss of one or both
hormone receptor staining in the lymph node metastasis: loss of ER (case n°1), loss of PR
(case n° 14) and concomitant loss of ER and PR (case n° 9). According to the literature, these
cases should have a poor prognosis, justified by our phenotypes, but with maintained
luminality in the first two cases, and a shift to c-erbB-2 overexpression in the third meaning a
even worse evolution by the complete loss of ER and PR.
Interestingly, cases n°3 and n°13 (Table 3) in which in lymph node metastasis occurred, lost
61
all the markers expressed in the primary tumour, likely due to the selection of a significantly
aggressive cell subpopulation. The normal-like or multiple markers negative (MMN) subtype
tumour have been shown to be negative for basal markers, such as CK5/6 and CK14 in our
cases, and negative for other molecular markers. The majority of normal-like subtype tumour
express CK8/18, CK19 (used in this study), with an absence of CK5/6 suggesting that these
cells were most probably derived from a luminal gland cell [43]. The normal-like subtype is
also included in the triple negative breast cancer group (TNBC) characterized by an
aggressive clinical course, poor survival rate and, unlike the overexpressing hormone
receptors or c-erbB-2-overexpressing tumour, is not amenable to hormone therapy or c-erbB2-directed agents [44]. Although no correlation has been found between histological type and
phenotype, the normal-like cases represent an exception.
The present study identified five phenotypes in the lymph node metastases: luminal A (25%),
luminal B (15%), c-erbB-2 overexpressing (20%), basal-like (30%) and normal-like (10%).
The novel aspect of this study is the evaluation of the lymph node metastasis phenotypes and
their correlation with the primary tumour, never hitherto applied to canine species. The
relationship between the primary tumour and metastatic phenotype is defined by a
concordance in 65% of cases and a discordance in the remaining 35%, suggesting the two
main metastatic capability theories coexist. All seven discordant cases showed a progressive
behavior, according to the prognostic value of molecular phenotypes reported by Gama et al.
[8], suggesting phenotypic evolution with a worse prognosis from the primary tumour to
lymph node metastasis.
Conclusions
Molecular phenotype classification is a new model urgently needed in veterinary medicine.
This model will fill current gaps regarding prognosis and a targeted therapeutic approach,
62
since the primary tumour phenotype does not always overlap with that of its metastasis.
According to the present findings, the primary tumour phenotype assumes a predictive
therapeutic role only in concordant cases, meaning that there should be a concomitant
evaluation of both the primary tumour and its lymph node metastasis. Treatment planning
based only on the primary tumour phenotype can lead to therapeutic failures if the lymph
node metastatic phenotype differs from that of the primary tumour.
63
Figures
64
65
References
1. Ferguson HR: Canine mammary gland tumours. Vet Clin North Am Small Anim Pract
1985, 15:501–511.
2. Moulton JE: Tumours of the mammary gland. In Tumours in Domestic Animals. 3rd
edition. Edited by Moulton JE. Berkeley: University of California Press; 1990:518–553.
3. Steinman S, Wang J, Bourne P, Yang Q, Tang P: Expression of cytokeratin markers,
ER-alpha, PR, HER-2/neu, and EGFR in pure ductal carcinoma in situ (DCIS) and
DCIS with co-existing invasive ductal carcinoma (IDC) of the breast. Ann Clin Lab Sci
2007, 37:127–134.
4. Guarneri V, Conte P: Metastatic Breast Cancer: Therapeutic Options According to
Molecular Subtypes and Prior Adjuvant Therapy. Oncologist 2009, 14:645–656.
5. Tavassoli FA, Devilee PL: World Health Organization Classification of Tumours. In
Pathology and Genetics. Tumours of the Breast and Female Genital Organs. Edited by IARC
Press: International Agency for Research on Cancer; 2003.
6. Misdorp W, Else RW, Hellmén E, Lipscomb TP: Histological Classification of Mammary
Tumours of the Dog and Cat. Washington DC: Published by the Armed Forces Institute of
Pathology in cooperation with the American Registry of Pathology and the World Health
Organization Collaborating Centre for Worldwide Aderence on Comparative Oncology;
1999.
7. Goldschmidt M, Peña L, Rasotto R, Zappulli V: Classification and grading of canine
mammary tumours. Vet Pathol 2011, 48:117–131.
8. Gama A, Alves A, Schmitt F: Identification of molecular phenotypes in canine mammary
carcinomas with clinical implications: application of the human classification. Virchows Arch
2008, 453:123–132.
66
9. Sassi F, Benazzi C, Castellani G, Sarli G: Molecular-based tumour subtypes of canine
mammary carcinomas assessed by immunohistochemistry. BMC Vet Res 2010, 6:5.
lesioni proliferative
Sorlie T, Tibshirani R, Parker J, Hastie T, Marron JS, Nobel A, Deng S, Johnsen H, Pesich R,
Geisler S, Demeter J, Perou CM, Lønning PE, Brown PO, Børresen-Dale AL, Botstein D
(2003). Repeated observation of breast tumour subtypes in independent gene expression data
sets. Proceedings of the National Academy of Sciences of the United States of America
100:8418-8423.
15. Wu JM, Fackler MJ, Halushka MK, Molavi DW, Taylor ME, Teo WW, Griffin C, Fetting
J, Davidson NE, DeMarzo AM, Hicks JL, Chitale D, Marc L, Sukumar S, Argani P:
Heterogeneity of Breast Cancer Metastases: Comparison of Therapeutic Target
Expression and Promoter Methylation Between Primary Tumours and Their Multifocal
Metastases. Clin Cancer Res 2008, 10:1938–1946.
16. Tarin D: Comparisons of Metastases in Different Organs: Biological and Clinical
Implications. Clin Cancer Res 2008, 14:1923–1925.
17. Klopfleisch R, Lenze D, Hummel M, Gruber AD: Metastatic canine mammary
carcinomas can be identified by a gene expression profile that partly overlaps with
human breast cancer profiles. BMC Cancer 2010, 10:618.
18. Rivera P, von Euler H: Molecular biological aspects on canine and human mammary
tumours. Vet Pathol 2011, 48:132–146.
19. Matos AJ, Baptista CS, Gärtner MF, Rutteman GR: Prognostic studies of canine and
feline mammary tumours: The need for standardized procedures. Vet J 2012, 193:24–31.
24. Morris JS, Dobson JM, Bostock DE: Use of tamoxifen in the control of canine
mammary neoplasia. Vet Rec 1993, 27:539–542.
25. Gilbertson SR, Kurzman ID, Zachrau RE, Hurvitz AI, Black MM: Canine Mammary
67
Epithelial Neoplasms: Biologic Implications of Mophologic Characteristics Assessed in
232 Dogs. Vet Pathol 1983, 20:127–142.
26. Kim MJ, Ro JY, Ahn SH, Kim HH, Kim SB, Gong G: Clinicopathologic significance of
the basal-like subtype of breast cancer: a comparison with hormone receptor and
Her2/neu-over-expressing phenotypes. Hum Pathol 2006, 37:1217–1226.
27. Millanta F, Calandrella M, Citi S, Della Santa D, Poli A: Overexpression of HER-2 in
Feline Invasive Mammary Carcinomas: Immunohistochemical Survey and Evaluation
of Its Prognostic Potential. Vet Pathol 2005, 42:30–34.
28. Millanta F, Calandrella M, Bari G, Niccolini M, Vannozzi I, Poli A: Comparison of
steroid receptor expression in normal, dysplastic, and neoplastic canine and feline
mammary tissues. Res Vet Sci 2005, 79:225–232.
29. Ramalho LNZ, Ribeiro-Silva A, Cassali GD, Zucoloto S: The Expression of p63 and
Cytokeratin 5 in Mixed Tumours of the Canine Mammary Gland Provides New Insights
into the Histogenesis of These Neoplasms. Vet Pathol 2006, 43:424–429.
30. Lester SC: The Breast. In Robbins and Cotran Pathologic Basis of Disease. 8th edition.
Edited by Kumar V, Abbas A, Fausto N, Aster JC. Philadelphia: Saunders Elsevier;
2010:1084–1085.
31. Lu X, Lu X, Wang ZC, Iglehart JD, Zhang X, Richardson AL: Predicting features of
breast cancer with gene expression patterns. Breast Cancer Res Treat 2008, 108:191–201.
32. Nakshatri H, Srour EF, Badve S: Breast cancer stem cells and intrinsic subtypes:
controversies rage on. Curr Stem Cell Res Ther 2009, 4:50–60.
33. Weigelt B, Glas AM, Wessels LF, Witteveen AT, Peterse JL, van't Veer LJ: Gene
expression profiles of primary breast tumours maintained in distant metastases. Proc
Natl Acad Sci USA 2003, 100:15901–15905.
34. Weigelt B, Hu Z, He X, Livasy C, Carey LA, Ewend MG, Glas AM, Perou CM, Van't
68
Veer LJ: Molecular portraits and 70-gene prognosis signature are preserved throughout
the metastatic process of breast cancer. Cancer Res 2005, 65:9155–9158.
35. Hsu HC, Chang HK, Lin YC, Hseu S, Chen JS, Yang TS, Wang HM, Shen WC: The role
of HER2 in metastatic breast cancer treated with a combination of taxanes and
cisplatin. Chang Gung Med J 2009, 32:33–41.
36. Perou CM, Sorlie T, Eisen MB, van de Rijn M, Jeffrey SS, Rees CA, Pollack JR, Ross
DT, Johnsen H, Akslen LA, Fluge O, Pergamenschikov A, Williams C, Zhu SX, Lonning PE,
Borresen-Dale AL, Brown PO, Botstein D: Molecular portraits of human breast tumours.
Nature 2000, 406:747–752.
37. Demir H, Turna H, Can G, Ilvan S: Clinicopathologic and prognostic evaluation of
invasive breast carcinoma molecular subtypes and GATA3 expression. J Buon 2010,
15:774–782.
38. Brunetti B, Asproni P, Beha G, Muscatello LV, Millanta F, Poli A, Benazzi C, Sarli G:
Molecular Phenotype in Mammary Tumours of Queens: Correlation between Primary
Tumour and Lymph Node Metastasis. J Comp Pathol 2012,
http://dx.doi.org/10.1016/j.jcpa.2012.05.012.
39. Gown AM: Current issue in ER and HER2 testing by IHC in breast cancer. Mod
Pathol 2008, 21:8–15.
40. Bardou VJ, Arpino G, Elledge RM, Osborne CK, Clark GM: Progesterone receptor
status significantly improves outcome prediction over estrogen receptor status alone for
adjuvant endocrine therapy in two large breast cancer databases. J Clin Oncol 2003,
21:1973–1979.
41. Martin De Las Mulas J, Millan Y, Dios R: A prospective analysis of
immunohistochemically determined estrogen receptor α and progesterone expression
and host and tumour factors as predictors of disease-free period in mammary tumours of
69
the dog. Vet Pathol 2005, 42:200–212.
42. Chang CC, Tsai MH, Liao JW, Chan JPW, Wong ML, Chang SC: Evaluation of
hormone receptor expression use in predicting survival of female dogs with malignant
mammary gland tumours. JAVMA 2009, 235:391–396.
43. Munirah MA, Siti-Aishah MA, Reena MZ, Sharifah NA, Rohaizak M, Norlia A, Rafie
MK, Asmiati A, Hisham A, Fuad I, Shahrun NS, Das S: Identification of different subtypes
of breast cancer using tissue microarray. Rom J Morphol Embryol 2011, 52:669–677.
44. Pal SK, Childs BH, Pegram M: Triple negative breast cancer: unmet medical needs.
Breast Cancer Res Treat 2011, 125:627–636.
70
Publications and Proceedings
1. Beha G., Brunetti B., Asproni P, Muscatello L.V., Millanta F., Poli A., Sarli G., Benazzi C.
Molecular portrait-based correlation between primary canine mammary tumour and its lymph
node metastasis: possible prognostic-predictive models and/or stronghold for specific
treatments? BMC Veterinary Research 2012, 12;8:219
2. Beha G., Brunetti B., Asproni P., Muscatello L. V., Millanta F., Poli A., Sarli G., Benazzi
C. Molecular portrait-based correlation between primary canine mammary tumour and its
lymph node metastasis: possible prognostic-predictive models and/or stronghold for specific
treatments? (Oral Presentation). Proceedings of the joint meeting of ESVP-ECVP, Leon, 5-8
September 2012, p. 62
3. Beha G., Brunetti B., Asproni P., Muscatello L.V., Millanta F., Poli A., Sarli, G. Benazzi.
C. Correlation, Based on the Protein Expression Profile, Between Primary Canine Mammary
Tumours and Their Lymph Node Metastases. Journal of Comparative Pathology, 01/2013;
148(1):48.
71
2. MOLECULAR PHENOTYPE IN MAMMARY TUMOURS
OF QUEENS: CORRELATION BETWEEN PRIMARY TUMOUR
AND LYMPH NODE METASTASIS
Introduction
The heterogeneity of breast cancer is proverbial and raises the need for a revision of the WHO
classification, which is purely morphological (Eusebi, 2010). Sorlie et al. (2001) laid the
foundations for a new taxonomy demonstrating that changes in patterns of gene expression,
analyzed by cDNA microarray techniques and hierarchical clustering, allow a "molecular
portrait" to be defined for each tumour. The final goal of this system is to classify the breast
cancers into subtypes, Luminal-A, Luminal-B, c-erbB-2 over-expressing, basal-like and
Normal-like, based on the differences between these patterns (Sorlie et al., 2001). The
existence of four different subtypes of breast cancer was confirmed by protein expression
patterns assessed by IHC on tissue microarray (TMA), an efficient and reliable platform for
sub-classifying breast cancers into relevant subtypes, using a limited number of markers
72
(Matos et al., 2005). So far, this molecular classification has been applied in veterinary
medicine only in canine (mammary).
The usefulness of molecular subtypes is their predictive capability for prognosis and targeted
therapy (Peppercorn et al., 2008). The formulation of a molecular-based taxonomy and its
application to feline clinical practice is necessary for a multimodal therapeutic approach and
to increase the survival rate.
Aim of the present study was: 1) to define the molecular phenotype of feline mammary
carcinomas and their lymph node metastases according to a modified algorithm by Sassi et
al. (2010); 2) to demonstrate the concordance or discordance of the molecular profile between
the primary tumour and lymph node metastasis.
Materials and Methods
Samples
Specimens of mammary carcinomas from 21 female cats were collected from the data base of
the Pathology Service of the Department of Veterinary Medical Science of Bologna
University and from the Department of Animal Pathology of Pisa University. Each case
consisted in primary mammary tumour and its related lymph node metastasis. No cases
displayed systemic metastases.
Samples were available as haematoxylin and eosin stained
sections, obtained from formalin-fixed and paraffin wax-embedded tissue block.
Histological diagnosis and immunohistochemistry
Histological diagnosis was achieved according to the WHO classification system (Misdorp et
al., 1999). Seven consecutive 4 μm thick sections were cut from the paraffin wax blocks
containing representative tumour samples and labelled by immunohistochemistry with the
73
following antibodies: anti-ER, -PR, -c-erbB-2, -CK5/6, -CK14, -CK19 and -p63. Data on the
primary antibodies are summarized in Table 1.
Table 1: Primary antibodies, resources and dilutions used in immunohistochemistry
ANTIBODY
CLONE
MANUFACTURER
DILUITION
(-ANTI)
ER
6F11
Novocastra Lab Ltd., Newcastle upon
Tyne, UK
1: 40
PR
PR88
Biogenex, San Ramon, CA, USA
1: 40
c-erbB-2
Policlonal
Dako, Glostrup, Denmark
1: 250
Cytokeratins 5/6
D5/16B4
Zymed (South San Francisco, Ca)
1: 100
Cytokeratin 14
Ab-1
(LL002)
NeoMarkers (Fremont, Ca)
1: 300
Cytokeratin 19
BA17
Dako (Glostrup, Denmark)
1: 50
p63
4A4
Dako (Glostrup, Denmark)
1: 50
Sections were dewaxed in toluene and rehydrated. Endogenous peroxidase was blocked by
immersion in 0.3% hydrogen peroxide for 20 min. Sections were then rinsed in Tris Buffer
and antigen was retrieved with citrate buffer (2.1 g citric acid monohydrate/litre distilled
water), pH 6.0 (except for CK 5/6 which use EDTA, pH 8.0), and heating for two 5-min
periods in a microwave oven at 750 W, followed by cooling at room temperature for 20 min.
All antibodies were incubated with the tissue sections overnight at 4°C, and their binding was
revealed by a commercial streptavidin-biotin-peroxidase technique (LSAB Kit, Dako,
Amsterdam, The Netherlands). Diaminobenzidine (0.05% for 10 min at room temperature)
was used as chromogen. Slides were counterstained with Papanicolaou's haematoxylin.
As negative control, the primary antibody was replaced with an irrelevant, isotype-matched
antibody to control for non-specific binding of the secondary antibody. As positive controls to
assess the cross-reactivity with feline tissues and the specificity of the immunohistochemical
74
stain, sections of normal feline uterus (for anti-ER and -PR antibodies) and feline skin (for
anti-CK5/6, CK14, -CK19 and -p63 antibodies) were used following the same protocols. A
human poorly differentiated invasive ductal mammary carcinoma (kindly provided by P.
Viacava, Department of Oncology, University of Pisa, Italy) known to react with c-erbB-2
antibody was used as positive control.
The staining result was classified semi-quantitatively with a dichotomous evaluation: positive
or negative. The sample was considered positive when presenting:
•
cytoplasmic stain in more than 1% of the invasive tumour cells for anti-CK-5/6 and
anti-CK14 antibodies (Kim et al., 2006);
•
complete membranous stain in more than 10% of tumour cells for anti-c-erbB-2
antibody (Millanta et al., 2005b);
•
nuclear stain in more than 5% of tumour cells for anti-ER and anti-PR antibodies
(Millanta et al., 2005a);
•
nuclear stain in more than 10% of tumour cells for anti-p63 antibody (Ramalho et al.
2006).
Molecular taxonomy
The application of the panel grouped cases into five molecular subtypes according to an
algorithm modified by Sassi et al. 2010 as follows:
•
Luminal-A: ER+ and/or PR+, c-erbB-2-, regardless of CK5/6, CK14, p63 staining.
•
Luminal-B: ER+ and/or PR+, c-erbB-2+, regardless of CK5/6, CK14, p63 staining.
•
c-erbB-2-overexpressing: ER-, PR-, c-erbB-2+ regardless of CK5/6, CK14, p63
staining.
•
Basal-like: ER-, PR-, c-erbB-2- , CK5/6+ and/or CK14+ and/or p63 staining.
•
Normal-like: negative to all markers.
75
Results
Diagnosis
Six of the 21 primary tumours were classified histologically as solid carcinomas and 15 as
simple tubulo-papillary carcinomas.
Immunohistochemistry
The immunohistochemistry for ER, PR, c-erbB-2, CK5/6, CK14, p63 in the primary tumour
and respective lymph node metastasis is summarized in Table 2. In each case the epithelial
origin of cancer was confirmed by CK19 staining. Hormonal receptor positivity (reactivity to
ER and/or PR) was found in 11 out of 21 cases in the primary tumour, two ER and nine PR
positive respectively (Table 2). The typical nuclear staining of ER and PR decreased
substantially in the lymph node metastasis compared with the primary tumour (reactivity in
only two out of 21 cases). C-erbB-2 over-expression was frequently observed both in the
primary tumour (19 positive cases) and in the lymph node metastases (18 positive cases).
As reported in Table 2, CK5/6 expression was less frequent than CK14 expression both in
primary tumours (11/21 and 19/21, respectively) and in their lymph node metastases (9/21
and 14/21, respectively). P63 was poorly expressed in primary tumours (four out of 21
samples) and especially in metastases (one out of 21).
Molecular phenotypes
Based on the algorithm applied, molecular phenotypes were obtained in the primary
mammary tumour and in their lymph nodes metastases (Table 3). Only three phenotypes
(Luminal B (Fig. 1), c-erbB-2 over-expressing (Fig. 2), Basal (Fig. 3)) were diagnosed both
76
in primary tumours (11 (52.4%), eight (38.1%) and two (9.5%), respectively) and in lymph
node metastases (two (9.5%), 16 (76.2%) and three (14.3%), respectively).
Table 2: Summary of immunohistochemical staining
PRIMARY MAMMARY TUMOUR
SAMPLES
ID
ER
PR
LYMPH NODE METASTASIS
c-erbB-2
CK 14
CK 5/6
p63
ER
PR
c-erbB-2
CK 14
CK 5/6
p63
1
0
1
3
1
1
0
0
1
3
1
1
0
2
0
0
2
1
0
0
0
0
2
1
0
0
3
1
0
2
1
0
1
0
0
2
1
0
0
4
0
1
1
1
0
0
0
0
1
0
0
0
5
0
1
2
1
0
0
0
0
2
1
0
0
6
1
0
3
1
1
0
0
0
2
0
0
0
7
0
0
0
1
0
0
0
0
0
1
0
0
8
0
0
1
1
1
0
0
0
1
0
0
0
9
0
0
2
1
1
0
0
0
2
1
1
0
10
0
0
2
1
1
1
0
0
1
1
0
0
11
0
0
3
1
0
0
0
0
3
1
1
0
12
0
1
1
1
1
0
0
0
0
1
1
0
13
0
0
3
1
1
0
0
0
3
1
1
0
14
0
1
3
1
0
0
0
0
3
0
0
0
15
0
1
3
1
1
1
0
0
3
1
1
0
16
0
0
2
1
0
0
0
0
2
1
0
0
17
0
1
2
0
1
0
0
0
2
0
1
0
18
0
1
2
1
1
1
0
0
2
1
1
1
19
0
0
2
0
0
0
0
0
2
0
0
0
20
0
1
2
1
1
0
0
1
1
1
1
0
21
0
0
0
1
0
0
0
0
0
1
0
0
0= 0%; 1= <25%; 2 <50%; 3 >50%
77
Comparison between histological diagnosis and phenotype in primary mammary
tumour
The tubulopapillary pattern displayed three different phenotypes, i.e. Luminal B (seven
cases), c-erbB-2 over-expressing (six cases) and Basal-like (two cases). The solid pattern was
found in two phenotypes, namely Luminal B (five cases) and c-erbB-2 over-expressing (one
case). No association was found between histological diagnosis and phenotype in primary
tumour (Yates corrected Chi-square, P=0.46).
Relationship between molecular phenotype in the primary mammary tumour and
its related lymph node metastasis
Phenotypic concordance was found in 12 of the 21 cases (57.14%) (eight c-erbB-2 overexpressing, two Basal and two Luminal B). The phenotypic profile of primary tumours
showed discordance from regional lymph node metastases in nine cases (42.86%) (eight
Luminal B became c-erbB-2 over-expressing and one Luminal B became Basal-like in the
lymph node metastasis) (Table 3).
Relationship between histological diagnosis and concordance/discordance
The same molecular phenotype (concordance) was found in the primary tumour and its lymph
node metastasis in ten tubulo-papillary carcinomas, whereas a difference (discordance)
between the two sites was found in the other five. As for the solid pattern concordance was
found in two cases and discordance in four. No difference in the percentage of concordant and
discordant cases emerged comparing the two histological patterns mentioned above (Yates
corrected Chi-square, P=0.36).
78
Discussion
To the best of our knowledge, no literature reports have addressed the immunophenotyping of
feline mammary carcinomas. Only single markers such as c-erbB-2 over-expression and
steroid receptors have been used to date (Millanta et al., 2005b; Rasotto et al., 2011). The
present study analysed markers (ER, PR, c-erbB-2, CK5/6, CK14, p63) both singly and
combined in the molecular phenotypes to detect possible changes from the primary tumour to
its regional lymph node metastasis. Immunophenotyping is essentially based on the
presence/absence of hormonal receptors (Fig. 1) and c-erbB-2 (Fig. 2), and basal markers
(CK5/6, CK14 and p63) are the key for the phenotypic definition of negative hormone
receptors (basal (Fig. 3) and normal-like) and c-erbB-2 tumours. Based on the present study,
only if ER, PR and c-erbB-2 are negative is CK necessary (Fig. 3), otherwise it is useless and
expensive. Gama et al. demonstrated in dogs that basal subtypes rarely express just one basal
marker, but they frequently express several concomitant markers (Gama et al., 2008).
The
present study exhibits some overlap with Gama’s results regarding basal phenotypes, and
suggests a more undifferentiated profile. Lymph node metastases did not show any
immunostaining to the primary tumour markers. No protein absent in the tumour was
expressed in its metastasis.
Given the well-known malignancy of feline mammary carcinomas (Misdorp, 1999) and the
high frequency of c-erbB-2 over-expression, particularly in metastases, in the present study, it
can be assumed that the c-erbB-2 over-expressing phenotype (Fig. 2) also leads to a worse
prognosis in the cat as in human breast cancer (Sorlie et al., 2001). Planning an anti-c-erbB-2
targeted therapy in queens bearing mammary carcinoma would therefore have a specific
therapeutic value to counteract metastases.
79
In human medicine Toft et al. (2011) described the Basal-like phenotype as a distinctive
molecular subtype with a basal epithelial gene signature and an aggressive clinical course
characterized by early relapses and poor survival. They also found a prevalence of Basal-like
breast cancer ranging from 12.3 to 36.7 of breast cancer cases (Toft et al., 2011). This range is
similar to the percentage frequency found in this study on mammary tumours in queens, but
the absence of outcome precludes any comment on the similarity between human and feline
Basal-like phenotypes.
The analysis of hormone receptors status disclosed a low expression of these markers both in
the primary tumours and more markedly in their metastases. Indeed, normal mammary tissue
and benign mammary tumours are mostly positive for ER and PR with respect to carcinomas,
which are more often ER negative or positive in only few cases (Misdorp, 2002). High
negative rates of ER seem to be a characteristic feature of feline mammary carcinomas
compared with human and canine tumours, suggesting a lack of oestrogen dependence
(Millanta et al., 2005a). The high number of ER/PR negative cases in cat mammary
carcinomas reduces the frequency of the tumour classified as Luminal type by the applied
tumour classification system: in this study Luminal A carcinomas were lacking and the
frequency of Luminal B (Fig. 1) lesions was lower than c-erbB-2 over-expressing tumours
(Fig. 2). The results suggest an association between a down-regulation of p63 expression and
progression of the tumour mirroring breast ductal carcinoma (Wang et al., 2002; Lindsay et
al., 2011). C-erbB-2 (Fig. 2) in this study was always over-expressed with a high frequency in
the primary site as well as in the metastases associated with ER/PR expression (Luminal B) or
not (c-erbB-2 over-expressing). Therefore anti-hormonal therapy alone is not indicated in
queens bearing mammary carcinomas, even in cases with ER/PR expression, but should be
associated with a target anti-c-erbB-2 therapy.
80
It has been debated for decades how cancer cells acquire metastatic capability. It is unclear
whether metastases are derived from a distinct tumour cell subpopulation with higher
metastatic potential within the primary site, or whether they originate from a random fraction
of tumour cells (Weigelt et al., 2003). Our comparison between the phenotypes of a primary
tumour and its lymph node metastasis displayed concordance in 57.1% cases and discordance
in 42.9%, with a progressive loss of marker immunoreactivity in the lymph node metastasis
with respect to the primary tumour. Human and canine Luminal A tumours are associated
with higher survival rates and Basal-like and c-erbB-2-overexpressing subtypes are related to
lower survival rates and more aggressive clinical behavior (Matos et al., 2005; Gama et al.,
2008). Hence, the progressive phenotypic discordance found in this study may signify a worse
prognosis with worsening of the phenotype passing from the primary tumour to the lymph
node metastasis. Phenotypic instability could be due to a tumoural cell sub-population that
acquires metastatic ability during carcinogenesis. Conversely, phenotypic concordance of the
primary tumour and metastasis could result from an intrinsic and early metastatic capability of
the dominant phenotype in the primary tumour. Weigelt et al. (2005) reported that the ability
to metastasize to distant sites is an early and inherent genetic property of primary breast
cancer, and that gene expression profiles of primary breast tumours were maintained in their
distant metastases (Weigelt et al. 2005).
An overall concordance has been proven in the immunostaining of the single markers
between primary tumours and metastasis, especially for c-erbB-2 over-expressing lesions,
indicating that the cat may develop the selection of an early clone with worsening prognostic
features rather than a malignant tumour progression. The discordance between the single
markers, also reported in human medicine (Aitken et al., 2010), led to therapeutic failure
when the therapy was only based on the primary tumour phenotype. Histological type is not
81
associated with the molecular phenotype, since several phenotypes are related to each
histological type.
Conclusions
From a molecular standpoint, the taxonomic approach represents a powerful tool for the
biological analysis of mammary tumours. The heterogeneity of the tumour is well-known,
therefore taxonomy based on these molecular differences is the key tool to formulate an
appropriate therapy. As for the difference between the primary tumour and the regional lymph
node metastasis, the primary tumour phenotype was not strictly predictive of the metastic
phenotype in a high percentage of cases (43%) in this study. Both tumour sites (primary
tumours and regional lymph node metastases) should be evaluated for accurate identification
of the tumour profile (or profiles) and appropriate therapy designed to target both primary
tumours and metastases in case of varying phenotypes.
82
Figures
Fig. 1: Case 3. Luminal B phenotype: a ER staining, b c-erbB-2 staining. To be classified as
luminal B phenotype, a neoplasm requires positivity to ER and/or PR and to c-erbB-2, while
the results of the other antibodies (CK5/6 +/- , CK14+/-, p63+/-) are irrelevant. 400x.
Fig. 2: Case 8. C-erbB -2-overexpressing phenotype: a ER staining, b c-erbB-2 staining. For
the diagnosis of the c-erbB-2 phenotype, besides positivity to c-erbB-2 antibody, negativity to
ER and PR is necessary, whereas the outcomes of the other antibodies (CK5/6 +/- , CK14+/-,
p63+/-) are irrelevant. 400x
83
Fig. 3: Case 21. Basal-like phenotypes: a PR staining, b c-erbB-2 staining, c CK14 staining.
Negativity to c-erbB-2, ER and PR and positivity to either CK (CK14 in the photo) or p63 is
indicative of a basal phenotype. 400x
84
References
Aitken SJ, Thomas JS, Langdon SP, Harrison DJ, Faratian D (2010) Quantitative analysis of
changes in ER, PR and c-erbB-2 expression in primary breast cancer and paired nodal
metastases. Annals of Oncology, 21, 1254-61.
Eusebi V (2010) Classifications and prognosis of breast cancer: from morphology to
molecular taxonomy. Breast Journal, 16, Suppl 1:S15-6.
Gama A, Alves A, Schmitt F (2008) Identification of molecular phenotypes in canine
mammary carcinomas with clinical implications: application of the human classification.
Virchows Arch 453: 123-132
Kim MJ, Ro JY, Ahn SH, Kim HH, Kim SB et al. (2006) Clinicopathologic significance of
the basal-like subtype of breast cancer: a comparison with hormone receptor and C-erbB2/neu-overexpressing phenotypes. Human Pathology, 37, 1217-26.
Lindsay J, McDade SS, Pickard A, McCloskey KD, McCance DJ (2011) Role of
DeltaNp63gamma in epithelial to mesenchymal transition. The Journal of Biological
Chemistry, 286, 3915-24
Matos I, Dufloth R, Alvarenga M, Zeferino LC, Schmitt F (2005). p63, cytokeratin 5, and Pcadherin: three molecular markers to distinguish basal phenotype in breast carcinomas.
Virchows Archiv, 447, 688-94.
Misdorp W, Esle RW, Hellmén E, Lipscomb TP Histological Classification of Mammary
Tumours of the Dog and Cat. Published by the Armed Forces Institute of Pathology in
cooperation with the American Registry of Pathology and the World Health Organization
Collaborating Centre for Worldwide Aderence on Comparative Oncology. Washington, DC;
1999
Misdorp W (2002) Tumours of the mammary gland. In: Tumours in Domestic Animals, 4th
Edit., JM Donald, Iowa State Press, Ames, pp. 576.
85
Millanta F, Calandrella M, Bari G, Niccolini M, Vannozzi I, Poli A (2005a) Comparison of
steroid receptor expression in normal, dysplastic, and neoplastic canine and feline mammary
tissues. Research in Veterinary Science, 79, 225-32.
Millanta F, Calandrella M, Citi S, Della Santa D, Poli A (2005b) Overexpression of C-ERBB2 in feline invasive mammary carcinomas: an immunohistochemical survey and evaluation of
its prognostic potential. Veterinary Pathology, 42, 30-4.
Peppercorn J, Perou CM, Carey LA (2008) Molecular subtypes in breast cancer evaluation
and management: divide and conquer. Cancer Investigation, 26, 1-10.
Ramalho LN, Ribeiro-Silva A, Cassali GD, Zucoloto S (2006) The expression of p63 and
cytokeratin 5 in mixed tumours of the canine mammary gland provides new insights into the
histogenesis of these neoplasms. Veterinary Pathology, 43, 424-9.
Rasotto R, Caliari D, Castagnaro M, Zanetti R, Zappulli V (2011) An immunohistochemical
study of HER-2 expression in feline mammary tumours. Journal of Comparative Pathology,
144(2-3), 170-9.
Sassi F, Benazzi C, Castellani G, Sarli G (2010) Molecular-based tumour subtypes of canine
mammary carcinomas assessed by immunohistochemistry. BMC Veterinary Research, 6: 5.
Sorlie T, Perou CM, Tibshirani R, Aas T, Geisler S et al. (2001) Gene expression patterns of
breast carcinomas distinguish tumour subclasses with clinical implications. Proceedings of the
National Academy of Sciences of the United States of America, 98, 10869-74.
Toft DJ, Cryns VL (2011) Minireview: Basal-like breast cancer: from molecular profiles to
targeted therapies. Molecular Endocrinology, 25, 199-211.
Wang X, Mori I, Tang W, Nakamura M, Nakamura Y et al. (2002) p63 expression in normal,
hyperplastic and malignant breast tissues. Breast Cancer, 9, 216-9.
86
Weigelt B, Glas AM, Wessels LF, Witteveen AT, Peterse JL et al. (2003) Gene expression
profiles of primary breast tumours maintained in distant metastases. Proceedings of the
National Academy of Sciences of the United States of America, 100, 15901-5.
Weigelt B, Hu Z, He X, Livasy C, Carey LA et al. (2005) Molecular portraits and 70-gene
prognosis signature are preserved throughout the metastatic process of breast cancer. Cancer
Research, 65, 9155-8.
87
Publications and Proceedings
1. Brunetti B., Asproni P., Beha G., Muscatello L.V., Millanta F., Poli A., Benazzi C., Sarli
G. Molecular Phenotype in Mammary Tumours of Queens: Correlation between Primary
Tumour and Lymph Node Metastasis. J. Comp. Path. 2013, 148(2-3):206-13).
2. Brunetti B., Asproni P., Beha G., Muscatello L.V, Millanta F., Poli A, Benazzi C., Sarli G.
Fenotipo molecolare dei tumouri mammari della gatta: correlazione tra tumoure primario e
relativa metastasi linfonodale. Atti AIPVET, Perugia 25-26 May 2012, p. 7-12.
88
3. MOLECULAR PHENOTYPE OF PRIMARY MAMMARY
TUMOURS AND DISTANT METASTASES IN FEMALE
DOGS AND CATS
Introduction
In terms of prognosis and treatment approach, it is already well explained in the general
section that the different phenotypes are associated with different clinical outcomes (Carey,
2011; Sorlie et al., 2003).
In the light of published findings, the metastatic process, i.e. the tumour’s ability to spread
from its primary location to distant parts of the body (Banfalvi, 2012), has become one of the
most urgent, important and difficult issues to approach in human (Tarin, 2008) and animal
cancer medicine. Among the possible sites of metastases, human (Tavassoli and Devilee,
2003), canine (Klopfleisch et al., 2010) and feline (Ginn et al., 2007) mammary tumours are
known to spread to the regional lymph nodes, lungs, pleura, liver, adrenal glands, brain,
kidneys and bone (Fountzilas et al., 2012; Kennecke et al., 2010). In human medicine,
investigations conducted to associate breast tumour molecular subtypes with the site of
89
metastatic disease have reported conflicting results (Arnedos et al., 2012; Fountzilas et al.,
2012; Kennecke et al., 2010). Some studies found a low risk of brain metastases with luminal
A and B disease, but a high risk for pleural relapse (Smid et al., 2008). Basal-like and normallike tumours had a distinctive pattern of relapse, with higher frequencies of lung, brain, liver,
locoregional and distant lymph node metastasis (Arnedos et al., 2012; Fountzilas et al., 2012;
Kennecke et al., 2010) compared to the frequency of c-erbB-2 overexpressing phenotype.
Other studies showed that locoregional relapse (Lester, 2010), and distant metastases in brain
(Blows et al., 2010), lung, liver and soft tissues (Aleskandarany et al.; 2012; Fountzilas et al.,
2012) were more common in c-erbB-2 overexpressing tumours compared to basal and
normal-like phenotypes. Bone metastases also presented a higher frequency with luminal B
disease (Fountzilas et al., 2012; Kennecke et al., 2010).
Few veterinary studies have considered the molecular phenotype of lymph node metastases in
comparison with the primary mammary tumour phenotype (Beha et al., 2012; Brunetti et al.,
2013). The phenotype of feline mammary carcinomas and their lymph node metastases was
recently determined (Brunetti et al., 2013), and the same comparison made for canine
mammary tumours (Beha et al., 2012). No data are yet available on the molecular phenotypes
of distant metastases and their correlation with the primary mammary tumour and its related
lymph node metastasis.
Therefore, the present study aimed to evaluate the molecular trend of cancer from its primary
location to metastatic sites in three cats and two dogs with mammary tumours.
Materials and Methods
Samples
90
Tissue samples of mammary carcinomas from two female dogs and three female cats were
collected from the database of the Pathology Service of the Department of Veterinary Medical
Science, University of Bologna, and from the Department of Veterinary Sciences, University
of Pisa. All the animals underwent autopsy.
Cases were selected based on both the primary mammary neoplasm and histological grade III
(grade III: invasive carcinoma with distant metastases) according to a previous study
(Gilbertson et al., 1983). Samples were available as sections stained with hematoxylin and
eosin and obtained from formalin-fixed and paraffin-embedded tissue blocks.
Histological diagnosis and immunohistochemistry
Histological diagnosis was made according to the WHO classification system (Misdorp et al.,
1999). Six consecutive 4-μm thick sections were cut from the paraffin blocks containing
representative tumour samples and three of them were labeled by immunohistochemistry with
the following antibodies: anti-ER, -PR, -c-erbB-2. According to a previous study (Brunetti et
al., 2013), samples presenting negative staining to the above antibodies were labeled by
immunohistochemistry with anti-CK5/6, -CK14, and -p63 antibodies. Data on the primary
antibodies are summarized in Table 1.
Sections were dewaxed in toluene and rehydrated. Endogenous peroxidase was blocked by
immersion in H2O2 0.3% in methanol for 20 min. Sections were then rinsed in Tris buffer and
antigen was retrieved with citrate buffer (2.1 g citric acid monohydrate/liter distilled water),
pH 6.0 (except for CK5/6 which used EDTA, pH 8.0), and heated for two 5-min periods in a
microwave oven at 750 W, followed by cooling at room temperature for 20 minutes.
All antibodies were incubated with the tissue sections overnight at 4°C, and their binding was
revealed by a commercial streptavidin-biotin-peroxidase technique (LSAB Kit, Dako,
Amsterdam, The Netherlands). Diaminobenzidine (0.05% for 10 minutes at room
91
temperature) was used as chromogen. Slides were counterstained with Papanicolaou's
hematoxylin. As a negative control, the primary antibody was replaced with an irrelevant,
isotype-matched antibody to control for non-specific binding of the secondary antibody. As
positive controls to assess the cross-reactivity with canine and feline tissues and the
specificity of the immunohistochemical stain, sections of normal canine and feline uterus (for
anti-ER and –PR antibodies), canine and feline skin (for anti-CK5/6, -CK14, and -p63
antibodies) were used following the same protocols. A human poorly differentiated invasive
ductal mammary carcinoma (kindly provided by P. Viacava, Department of Oncology,
University of Pisa, Italy) known to react with c-erbB-2 antibody was used as positive control.
Table 1: Immunohistochemical panel of antibodies
ANTIBODY
(-ANTI)
CLONE
MANUFACTURER
DILUTION
ER (dog)
B-10
Abcam, Cambridge, UK
1: 300
6F11
Novocastra Laboratories Ltd., Newcastle
upon Tyne, UK
1:40
PR (dog)
PR4-12
Oncogene TM, Boston, MA, USA
1: 100
PR (cat)
PR88
Biogenex, San Ramon, CA, USA
1:40
c-erbB2
Polyclonal
Dako, Glostrup, Denmark
1: 250
Cytokeratins 5/6
D5/16B4
Zymed (South San Francisco, CA, USA)
1: 100
Cytokeratin 14
Ab-1 (LL002)
NeoMarkers (Fremont, CA, USA)
1: 300
Cytokeratin 19
BA17
Dako (Glostrup, Denmark)
1: 50
p63
4A4
Dako (Glostrup, Denmark)
1: 50
ER (cat)
The staining result was classified semi-quantitatively with a dichotomous evaluation: positive
or negative. The sample was considered positive when presenting:
• cytoplasmic stain in more than 1% of the invasive tumour cells for anti- CK5/6 and antiCK14 antibodies (Kim et al., 2006);
• complete membranous stain in more than 10% of tumour cells for anti- c-erbB-2 antibody
according to the Hercept-test (Millanta et al., 2005b);
92
• nuclear stain in more than 5% of tumour cells for anti-ER and anti-PR antibodies (Millanta
et al., 2005a);
• nuclear stain in more than 10% of tumour cells for anti-p63 antibody (Ramalho et al., 2006).
Molecular taxonomy
The application of the panel allowed cases to be grouped into five molecular subtypes
according to a modified algorithm (Sassi et al., 2010) as follows:
• Luminal-A: ER+ and/or PR+, c-erbB-2–, regardless of CK5/6, CK14, p63 staining.
• Luminal-B: ER+ and/or PR+, c-erbB-2+, regardless of CK5/6, CK14, p63 staining.
• c-erbB-2 overexpressing: ER–, PR–, c-erbB-2+ regardless of CK5/6, CK14, p63 staining.
• Basal-like: ER–, PR–, c-erbB-2–, CK5/6+ and/or CK14+ and/or p63+.
• Normal-like: negative to all markers
Results
Diagnosis
All individual data are listed in Table 2. The first three cases were all from European shorthair
feline samples classified as tubulopapillary carcinomas with different metastatic sites. In the
first case metastases were present in locoregional lymph node and lung (Fig. 1). The second
case developed metastases in locoregional lymph node, lung and muscle. The last feline case
presented metastases in locoregional lymph node, lung and spleen. Canine samples were
collected from a German Shepherd and a Hungarian Hound. The two canine mammary
tumours were histologically diagnosed as anaplastic and tubulopapillary carcinoma
respectively with metastases in locoregional, deep, lombo-aortic lymph node, lung, ovary and
adrenal gland in the first case and in the lung and brain in the second carcinoma (Fig. 2). At
93
the time of the investigation no records were available for the involvement of the locoregional
lymph node of the second carcinoma (Table 2).
Table 2: Individual data and molecular phenotypes
1
2
3
4
5
Feline
Feline
Feline
Canine
Canine
European
European
European
German Shepherd
Hungarian Hound
AGE (years)
NA
NA
19
10
11,8
HISTOTYPE
Tubulopapillary
carcinoma
Tubulopapillary
carcinoma
Tubulopapillary
carcinoma
Anaplastic
carcinoma
Tubulopapillary
carcinoma
MAMMARY TUMOUR
c-erbB-2
overexpressing
c-erbB-2
overexpressing
c-erbB-2
overexpressing
c-erbB-2
overexpressing
c-erbB-2
overexpressing
c-erbB-2
overexpressing
Basal-like
c-erbB-2
overexpressing
c-erbB-2
overexpressing
NA
c-erbB-2
overexpressing
c-erbB-2
overexpressing
NP
c-erbB-2
overexpressing
Basal-like
NP
NP
NP
NP
c-erbB-2
overexpressing
NP
NP
NP
c-erbB-2
overexpressing
NP
NP
NP
NP
c-erbB-2
overexpressing
NP
KIDNEY METASTASIS
NP
NP
NP
NP
NP
SPLEEN METASTASIS
NP
NP
c-erbB-2
overexpressing
NP
NP
MUSCLE METASTASIS
NP
c-erbB-2
overexpressing
NP
NP
NP
OVARY METASTASIS
NP
NP
NP
c-erbB-2
overexpressing
NP
NP
NP
c-erbB-2
overexpressing
NP
N°
SPECIES
BREED
LOCOREGIONAL
LYMPH NODE
METASTASIS
PULMONARY
METASTASIS
BRAIN METASTASIS
LOMBO-AORTIC
LYMPH NODE
METASTASIS
DEEP INGUINAL
LYMPH NODE
METASTASIS
ADRENAL GLAND
NP
METASTASIS
NA= not available; NP= not present
Immunohistochemistry
None of the cases showed positivity to hormone receptors ER and PR. C-erbB-2
overexpression was observed in the primary sites of all five cases, in locoregional lymph node
metastasis (3 out of 4 cases: 2 cats and 1 dog) and in distant metastases (9 locations: 4 in cats
and 5 in dogs). Considering the two cases (locoregional lymph node metastasis of case no.2
and pulmonary metastasis of case no.5) with hormone receptors and c-erbB-2 negative
94
staining, the expression of antibody anti-CK5/6, CK14 and p63 was evaluated. The
pulmonary metastasis of case no.5 stained only for CK14, with the other three antibodies not
being expressed. The locoregional lymph node metastasis of case no. 2 showed positive
staining for CK14 and negative staining for CK5/6 and p63.
Molecular taxonomy
Based on the modified algorithm (Sassi et al., 2010), only two out of the five different
molecular phenotypes were obtained in the primary mammary tumours and their locoregional
lymph node and distant metastases in the dogs and cats examined. In the feline cases, all the
primary tumours, lymph node metastasis and distant metastases were diagnosed as c-erbB-2
overexpressing phenotype (Fig. 1) with the exception of the locoregional lymph node
metastasis in case no. 2 that was diagnosed as basal-like. The two canine cases were classified
as c-erbB-2 overexpressing phenotype in the primary site, whereas the lymph node metastasis
and distant metastases except the pulmonary metastasis of case no. 5 were diagnosed as basallike (Table 2) (Fig. 2).
Relationship between molecular phenotype in the primary mammary tumour, its
related lymph node and distant metastases
Concordance (the same phenotypes in the primary mammary site and its related lymph node
and distant metastases), was found in three out of the five cases (all c-erbB-2 overexpressing
phenotypes) (Fig. 1). Two cases showed discordance with the locoregional lymph node (case
no. 2, cat, Table 2) and pulmonary distant metastases (case no. 5, dog, Table 2) showing a
different phenotypic profile from the primary tumour (c-erbB-2 overexpressing vs basal-like)
(Fig. 2).
95
Discussion
In human medicine, molecular phenotypes are determined in both the primary tumour site and
in metastases to minimize any therapeutic margin of error if the primary phenotype differs
from that of the metastatic lymph node and/or systemic metastases. In breast cancer (Sorlie et
al., 2003) and lately canine (Beha et al., 2012; Gama et al., 2008; Sassi et al., 2010) and
feline (Brunetti et al., 2013) mammary neoplasms, the molecular portrait-based classification
system has been adopted as a valid tool to evaluate predictive-prognostic models and/or target
therapies (Peppercorn et al., 2008). Many investigations have demonstrated that different
molecular phenotypes resulted in different prognoses and target treatments. Overall- and
relapse-free survival times were longer in luminal tumours compared to c-erbB2overexpressing, basal and normal-like phenotypes (Gama et al., 2008; Sorlie et al., 2003).
Metastasis to the regional lymph node is considered one of the most important prognostic
factors, representing an early step in metastatic spread (Lester, 2010) and the development of
distant metastases can be fatal in canine and feline mammary carcinomas (Klopfleisch et al.,
2010; Morris et al., 2008).
The molecular phenotypes of 21 primary feline mammary tumours were determined and
compared to their related lymph node metastases obtaining three out of the five known
phenotypes (luminal B, c-erbB-2 overexpressing, basal-like) both in the primary neoplasm
and in the related lymph node metastasis with 57.1% concordant cases and 42.9% discordant
cases (Brunetti et al., 2013). The same comparison was studied in 20 bitches and the results
disclosed four molecular phenotypes (luminal A, luminal B, c-erbB-2 overexpressing, basallike) in the primary tumour and all five phenotypes in the lymph node metastasis (luminal A,
luminal B, c-erbB-2 overexpressing, basal-like and normal-like) with concordance in 65% of
cases and discordance in the remaining 35% (Beha et al., 2012).
96
The results of the present study in feline and canine carcinomas differed from the above
findings since all five primary mammary tumours displayed the same c-erbB-2
overexpressing phenotype. The phenotype was maintained in all the lymph nodes and distant
metastases except for one lymph node and one lung metastasis that were discordant with
respect to the primary tumour. A possible explanation for the prevalence of the c-erbB-2
overexpressing phenotype in primary tumours is likely to be the small number of cases
reported in the present study, but it could also result from the higher malignancy of the cerbB-2 overexpressing phenotype than the luminal phenotypes, as already demonstrated in
human medicine (Perou et al., 2000; Sorlie et al., 2003). In fact, a previous study
demonstrated that ER and/or PR could be lost when the neoplasm metastasizes (Wu et al.,
2008). The present findings are in accordance with the results reported for cats (Brunetti et
al., 2013; Millanta et al., 2005b) that had a high frequency of c-erbB-2 overexpression
particularly in the lymph node metastases, and the well-known malignancy (Misdorp et al.,
1999) of feline mammary carcinomas. Therefore, it can be assumed that the c-erbB-2
overexpressing phenotype is associated with an increased risk of metastases as in human
breast cancer (Aleskandarany et al., 2012; Arnedos et al., 2012; Fountzilas et al., 2012;
Steinman et al., 2007). Different studies on dog tumour phenotypes (Gama et al., 2008;
Ressel et al., 2013) showed a higher survival rate and lower aggressiveness associated with cerbB-2 overexpressing phenotypes, which differed substantially from the results obtained in
this investigation so that more cases are required before making any assumptions or
comparisons. An overall concordance for c-erbB-2 overexpressing phenotypes between
primary tumours and metastases was found in female dogs and cats. This result is in
agreement with the cited study on feline tumours (Brunetti et al., 2013), where a phenotypic
concordance of c-erbB-2 overexpressing between primary feline mammary tumour and its
related lymph node metastasis was found in 38% of cases with 16 out of 21 primary
97
mammary tumours presenting the c-erbB-2 overexpressing phenotype in their related lymph
node metastases.
Phenotypic concordance of the primary tumour and its related lymph node and distant
metastases could be explained by an intrinsic and early metastatic capability of the dominant
phenotype in the primary tumours that is maintained in their distant metastases (Weigelt et al.,
2005).
Planning an anti-c-erbB-2-targeted therapy in veterinary medicine, particularly in female cats
bearing mammary carcinomas could therefore have a specific therapeutic value to counteract
metastases. Given the affinity of phenotypes to metastasize in different sites, the present
results are in agreement with in literature reports on breast cancer showing that the c-erbB-2
overexpressing phenotype has an increased tendency to metastasize to distant (Aleskandarany
et al., 2012; Fountzilas et al., 2012) and locoregional organs (Blows et al., 2010).
Two cases in our study presented discordance when the primary site phenotype differed from
that or those in the lymph node and/or distant metastases. Case no.2, a cat, presented c-erbB-2
overexpression in both the primary mammary phenotype and distant pulmonary and muscle
metastases, but this changed to the basal-like phenotype in the locoregional lymph node
metastasis. These results coincide with those obtained in a human breast cancer study where
locoregional relapse was associated with the triple negative phenotype (Fountzilas et al.,
2012). The second discordant case, no.5, was a dog presenting the c-erbB-2 overexpressing
phenotype in the primary mammary tumour and the same phenotype in the brain metastasis
that became basal-like in the lung metastasis. In human medicine, brain metastasis is more
common in triple negative phenotypes (basal and/or normal-like phenotype) (Fountzilas et al.,
2012; Kennecke et al., 2010), and in the c-erbB-2 phenotype (Lester, 2010), in accordance
with our case no.5. The basal-like pulmonary metastasis showed overlaps with some reports
of breast cancer (Arnedos et al., 2012; Kennecke et al., 2010; Peng, 2012). However, these
98
similarities do not identify a standard behavior of the different phenotypes in various
metastatic sites since human literature reported indications of cases but not a known behavior.
The two discordant cases may be explained by the loss of one or more receptors or by the
selection of a subpopulation with an intrinsic program of transition to a different phenotype
enhancing their ability for heterotypic interaction and survival proliferation in distant organs
(Aleskandarany et al., 2012) as Darwinian (Nakshatri et al., 2009) evolution.
In canine and feline literature (Brunetti et al., 2013; Beha et al., 2012) a greater rate of
discordance regarding the relationship between the phenotype of the primary tumour and
lymph node metastasis is reported. Otherwise, in this study, among the primary tumour and
distant metastases a prevalence of concordant cases is observed.
Conclusions
In conclusion, this study, although limited to five cases, confirmed the existence of both
biological phenomena of concordance and discordance in metastatic sites. The prevalence of
concordance between primary and metastatic sites supports the predictive therapeutic value of
the primary tumour phenotype, minimizing any margin of error which can occur in rare
discordant cases.
99
Figures
Fig. 1: Cat: Concordance between primary mammary tumour and locoregional and distant
metastases. C-erbB-2 overexpressing concordant case with PR– and c-erbB-2+ [A, D (400x.
Bar, 38μm)] in the primary neoplasia, PR– and c-erbB-2+ [B, E(400x. Bar, 38μm)] in the
lymph node metastasis and PR– and c-erbB-2+ [C, F (400x. Bar, 38μm)] in the pulmonary
metastasis. 200x. Bar, 76μm
100
Fig. 2: Dog: Discordance between primary mammary tumour and distant metastases.
Discordant case showing c-erbB-2 overexpressing phenotype ER–, c-erbB-2+, CK14– (A, D,
G) in the primary tumour becoming basal-like ER–, c-erbB-2–, CK14+ [B, E, H(400x. Bar,
38μm)] in the pulmonary metastasis and c-erbB-2 overexpressing phenotype ER–, c-erbB-2+,
CK14– [C, F (400x. Bar, 38μm), I] in the brain metastasis. 200x. Bar, 76μm
101
References
Aleskandarany MA, Green AR, Benhasouna AA, Barros FF, Neal K et al. (2012) Prognostic
value of proliferation assay in the luminal, HER2-positive, and triple-negative biologic
classes of breast cancer. Breast Cancer Research, 14, R3.
Arnedos M, Bihan C, Delaloge S, Andre F (2012) Triple-negative breast cancer: are we
making headway at least?. Therapeutic Advances in Medical Oncology, 4,195–210.
Banfalvi G (2012) Metastatic view of breast cancer. Cancer Metastasis Reviews, 31, 815–822.
Beha G, Brunetti B, Asproni P, Muscatello LV, Millanta F et al. (2012) Molecular portraitbased correlation between primary canine mammary tumour and its lymph node metastasis:
possible prognostic-predictive models and/or stronghold for specific treatments?. BMC
Veterinary Research, 8, 219.
Blows FM, Driver KE, Schmidt MK, Broeks A, van Leeuwen FE et al. (2010) Subtyping of
breast cancer by immunohistochemistry to investigate a relationship between subtype and
short and long term survival: a collaborative analysis of data for 10,159 cases from 12 studies.
PLoS Medicine, 7, 1–11.
Brunetti B, Asproni P, Beha G, Muscatello LV, Millanta F et al. (2013), Molecular Phenotype
in Mammary Tumours of Queens: Correlation between Primary Tumour and Lymph Node
Metastasis. Journal of Comparative Pathology, 148, 206-13
Carey LA (2011) Directed therapy of subtypes of triple-negative breast cancer. Oncologist,
15, 49–56.
Gama A, Alves A, Schmitt F (2008) Identification of molecular phenotypes in canine
mammary carcinomas with clinical implications: application of the human classification.
Virchows Archiv, 453,123–132.
102
Ginn PE, Mansell JEKL, Rakich PM (2007) Skin and appendages. In: Pathology of Domestic
Animals, 5th Edit., MG Maxie Ed., Saunders Elsevier Press, Philadelphia, pp. 779
Gilbertson SR, Kurzman ID, Zachrau RE, Hurvitz AI, Black MM et al. (1983) Canine
Mammary Epithelial Neoplasms: Biologic Implications of Morphologic Characteristics
Assessed in 232 Dogs. Veterinary Pathology, 20, 127–142.
Kennecke H, Yerushalmi R, Woods R, Cheang MCU, Voduc D et al. (2010) Metastatic
behavior of breast cancer subtypes. Journal Of Clinical Oncology, 28, 3271-3277.
Kim MJ, Ro JY, Ahn SH, Kim HH, Kim SB et al. (2006) Clinicopathologic significance of
the basal-like subtype of breast cancer: a comparison with hormone receptor and Her2/neuoverexpressing phenotypes. Human Pathology, 37, 1217–1226.
Klopfleisch R, Lenze D, Hummel M, Gruber AD (2010) Metastatic canine mammary
carcinomas can be identified by a gene expression profile that partly overlaps with human
breast cancer profiles. BMC Cancer, 10, 618.
Lester SC (2010) The Breast. In: Robbins and Cotran Pathologic Basis of Disease, 8th Edit.,
V Kumar, A Abbas, N Fausto, JC Aster Eds., Saunders Elsevier, Philadelphia, pp. 1084–
1085.
Matos AJ, Baptista CS, Gärtner MF, Rutteman GR (2012) Prognostic studies of canine and
feline mammary tumours: The need for standardized procedures. Veterinary Journal, 193,
24–31.
Millanta F, Calandrella M, Bari G, Niccolini M, Vannozzi I et al. (2005a) Comparison of
steroid receptor expression in normal, dysplastic, and neoplastic canine and feline mammary
tissues. Research in Veterinary Science, 79, 225–232.
Millanta F, Calandrella M, Citi S, Della Santa D, Poli A. (2005b) Overexpression of HER-2
in Feline Invasive Mammary Carcinomas: Immunohistochemical Survey and Evaluation of Its
Prognostic Potential. Veterinary Pathology, 42, 30–34.
103
Misdorp W, Else RW, Hellmén E, Lipscomb TP (1999) Histological Classification of
Mammary Tumours of the Dog and Cat. Washington DC: Published by the Armed Forces
Institute of Pathology in cooperation with the American Registry of Pathology and the World
Health Organization Collaborating Centre for Worldwide Adherence on Comparative
Oncology.
Morris JS, Nixon C, Bruck A, Nasir L, Morgan IM et al. (2008) Immunohistochemical
expression of TopBP1 in feline mammary neoplasia in relation to histological grade, Ki67,
ER alpha and p53. Veterinary Journal, 175, 218-226.
Nakshatri H, Srour EF, Badve S (2009) Breast cancer stem cells and intrinsic subtypes:
controversies rage on. Current Stem Cell Research & Therapy, 4, 50–60.
Peng Y (2012) Potential prognostic tumour biomarkers in triple-negative breast cancer.
Beijing Da Xue Xue Bao, 44, 666-72.
Peppercorn J, Perou CM, Carey LA (2008) Molecular subtypes in breast cancer evaluation
and management: divide and conquer. Cancer Investigation, 26, 1–10.
Perou CM, Sorlie T, Eisen MB, van de Rijn M, Jeffrey SS et al. (2000) Molecular portraits of
human breast tumours. Nature, 406, 747–752.
Ramalho LNZ, Ribeiro-Silva A, Cassali GD, Zucoloto S. et al. (2006) The Expression of p63
and Cytokeratin 5 in Mixed Tumours of the Canine Mammary Gland Provides New Insights
into the Histogenesis of These Neoplasms. Veterinary Pathology, 43, 424–429.
Ressel L, Puleio R, Loria G R, Vannozzi I, Millanta F et al. (2013) HER-2 expression in
canine morphologically normal, hyperplastic and neoplastic mammary tissues and its
correlation with the clinical outcome. Research in Veterinary Science, 94, 299-305.
Sassi F, Benazzi C, Castellani G, Sarli G (2010) Molecular-based tumour subtypes of canine
mammary carcinomas assessed by immunohistochemistry. BMC Veterinary Research, 6, 5.
104
Sleeckx N, de Rooster H, Veldhuis Kroeze EJ, Van Ginneken C, Van Brantegem L (2011)
Canine mammary tumours, an overview. Reproduction in Domestic Animals, 46, 1112-1131.
Smid M, Wang Y, Zhang Y, Sieuwerts AM, Yu J et al. (2008) Subtypes of breast cancer
show preferential site of relapse. Cancer Research, 68, 3108-3114.
Sorlie T, Tibshirani R, Parker J, Hastie T, Marron JS et al. (2003) Repeated observation of
breast tumour subtypes in independent gene expression data sets. Proceedings of the National
Academy of Sciences USA, 100, 8418–8423.
Steinman S, Wang J, Bourne P (2007) Expression of cytokeratin markers, ER-alpha, PR,
HER-2/neu, and EGFR in pure ductal carcinoma in situ (DCIS) and DCIS with co-existing
invasive ductal carcinoma (IDC) of the breast. Annals of Clinical & Laboratory Science, 37,
127–134.
Tarin D (2008) Comparisons of Metastases in Different Organs: Biological and Clinical
Implications. Clinical Cancer Research, 14, 1923–1925.
Tavassoli FA, Devilee PL (2003) World Health Organization Classification of Tumours. In:
Pathology and Genetics. Tumours of the Breast and Female Genital Organs, IARC Press Ed.,
International Agency for Research on Cancer, Lyon, France.
Weigelt B, Hu Z, He X, Livasy C, Carey LA et al. (2005) Molecular portraits and 70-gene
prognosis signature are preserved throughout the metastatic process of breast cancer. Cancer
Research, 65, 9155–9158.
Wu JM, Fackler MJ, Halushka MK, Molavi DW, Taylor ME et al. (2008) Heterogeneity of
Breast Cancer Metastases: Comparison of Therapeutic Target Expression and Promoter
Methylation Between Primary Tumours and Their Multifocal Metastases. Clinical Cancer
Research, 10, 1938–1946.
105
Publications and Proceedings
1. Beha G, Muscatello LV, Brunetti B, Asproni P, Millanta F, Poli A, Benazzi C, Sarli G.
Molecular Phenotype of Primary Mammary Tumours and Distant Metastases in Female Dogs
and
Cats.
J
Comp
Pathol.
2013
Sep
20.
doi:pii:
S0021-9975(13)00132-1.
10.1016/j.jcpa.2013.07.011. [Epub ahead of print].
2. Muscatello L. V., Beha G., Brunetti B., Asproni P., Millanta F., Poli A., Benazzi C., Sarli
G. Distant metastases have the same molecular subtype as the related primary mammary
tumour and sentinel node in female dogs and cats? Atti AIPVET 2013. Giulianova Lido (TE)
29-31 May.2013
106
FURTHER RESEARCH ON CANINE MAMMARY TUMOURS
107
4. MORPHOLOGY OF THE MYOEPITHELIAL CELL:
IMMUNOHISTOCHEMICAL CHARACTERIZATION FROM RESTING
TO MOTILE PHASE
Introduction
Mammary gland tumours of dogs are formed by both epithelial (epithelium and
myoepithelium) and mesenchymal components. The origin of the mesenchymal cells is still
debated. The elevated frequency of tumours showing myoepithelial or basal cell proliferation
is a unique feature of canine mammary tumours [1]. In the normal mammary gland, the
lumina are delimitated by an inner layer of polarized epithelial cells resting on two outer or
basal layers of epithelial and myoepithelial cells [2]. Both basal and myoepithelial cells
synthesize the basement membrane of ducts and alveoli and form a structural barrier between
the luminal epithelial cells and the surrounding stroma [3]. In ducts, myoepithelial cells form
a nearly continuous layer of cells oriented parallel to the long axis of the ducts. This layer
surrounds the luminal epithelial cells and separates them from the basement membrane and
the stroma. In alveoli, the myoepithelial cells are discontinuous, forming a basket-like
108
network around the alveoli, allowing some luminal epithelial cells to contact the basement
membrane directly [3–5]. Therefore, the myoepithelium is not only located in an ideal
position to communicate between these two compartments, but it is also positioned to provide
important regulatory signals for the maintenance of normal cell structure [5]. Based on
immunohistochemistry, the three layers of cells of the normal mammary gland display
different markers: the luminal epithelium is labeled by CK19, and the basal cells and
myoepithelial cells are stained by CK5/6 [6] and CK14 [2] and p63, Alpha-SMA, and VIM
[2]. Myoepithelial cells are contractile elements exhibiting a combined epithelial and
smoothmuscle immunoprofile. Themarkers mentioned above are expressed in the cytoplasm,
except for p63 which is a nuclear marker [1]. The myoepithelial cell layer is the sole source of
tumour suppressor p63, which is significantly inhibited on proliferation and invasion of
associated tumour cells [7]. In addition, basal myoepithelial cells in the normal mammary
gland are occasionally labeled by ER antibody [8], which is used for the molecular-based
classification of canine mammary tumours [9, 10]. Distinct myoepithelial cell morphologies
can be recognized in canine complex and mixed tumours: resting and proliferative suprabasal
myoepithelial cells and spindle and stellate motile interstitial myoepithelial cells. Suprabasal
cells are located between the basement membrane and the luminal epithelium and exhibit
flattened spindle (resting cells) or polygonal morphologies (proliferative cells). Interstitial
cells are frequently arranged in solid nests apposed to epithelial elements or isolated in the
interstitium [1, 11]. Spindle and stellate myoepithelial cells differentiate toward a more
general contractile phenotype [12]. Interstitial myoepithelial cells may eventually become
fibroblast-like cells, showing only VIM immunoreactivity [11]. The myoepithelial
differentiation may culminate in the formation of various mesenchymal tissues, including
cartilage and bone in canine mammary mixed tumour.
109
The acquisition of typical features of mesenchymal cells is likely to originate through
epithelial-mesenchymal transition (EMT). EMT is a biological phenomenon that allows a
polarized epithelial cell, which normally interacts with the basement membrane via its basal
surface, to undergo multiple biochemical changes enabling it to assume the traits and
functions of mesenchymal cells [13].
Aim
This paper will focus on various aspects of myoepithelial cells and mammary tumours in
dogs, specifically (1) characterization of the four different myoepithelial cell morphological
types in the normal and neoplasticmammary gland using a panel of antibodies and (2) the
immunohistochemical changes in myoepithelial cells from an epithelial to a mesenchymal
phenotype.
Materials and Methods
Samples
Mammary gland specimens of 29 female dogs were retrieved fromthe database of the
Anatomopathological Service of the Faculty of Veterinary Medicine of Bologna. The subjects
belonged to different breeds: mongrel (n = 13), German shepherd (n = 3), Poodle (n = 3),
Yorkshire Terrier (n = 3), Dachshund (n = 2), Setter (n = 1), Pointer (n = 1), Cocker spaniel (n
= 1), Schnauzer (n =1), and Siberian Husky (n = 1); they were all females, with an average
age of 9.20 ± 2.28 years (mean ± SD). The tumours consisted of: 3 benign myoepithelial
tumours, 3 malignant myoepithelial tumours, 7 carcinomas in benign mixed tumours, and 16
complex carcinomas (the last two groups were differentiated by the presence of cartilage
and/or bone in the mixed tumours). In addition, 29 specimens from normal mammary glands
110
of the same tumour line and 3 mammary samples from 3 healthy nonmammary tumour
bearing female dogs were evaluated. Tumours were classified according to Misdorp et al. [14]
and Goldschmidt et al. [15] into benign myoepithelial tumours: a rare neoplasm composed of
myoepithelial cells arranged in short bundles admixed with an extracellular fibrillar
basophilic material; malignant myoepithelial tumours: different from the benign variant with
more polymorphic myoepithelial cells; complex carcinoma: a carcinoma composed of both
luminal epithelial and myoepithelial components; carcinoma in benign tumour: a tumour with
foci of malignant-appearing epithelial cells or distinct nodules of such cells occurring together
with mesenchymal cells that have produced cartilage and/or bone possibly in combination
with fibrous tissue.
Immunohistochemistry
Four μm thick sections were cutfrom formalin-fixed paraffin-embedded blocks containing
representative tumour samples. Immunohistochemistry for the following markers was done on
these tissues: CK19, ER, CK5/6, CK14, VIM, Alpha-SMA, p63. Sections were dewaxed in
toluene and rehydrated. Endogenous peroxidase was blocked by immersion in 0.3% hydrogen
peroxide for 20min. Sections were then rinsed in Tris buffer. Antigen retrieval was performed
with citrate buffer (2.1 g citric acid monohydrate/liter distilled water), pH 6.0 (except for
CK5/6 and ER, which used EDTA, pH 8.0), and heating for two 5 min periods in a
microwave oven at 750 W, followed by cooling at room temperature for 20 min. The primary
antibodies are summarized in Table 1. All primary antibodies were incubated overnight at
4◦C, followed by a commercial streptavidin-biotin-peroxidase technique (LSAB Kit, Dako,
Amsterdam, The Netherlands). Diaminobenzidine (0.05% for 10min at room temperature)
was used as chromogen. Slides were counterstained with Papanicolaou’s hematoxylin.
111
As a negative control, the primary antibody was replaced with an irrelevant, isotype-matched
antibody to control for nonspecific binding of the secondary antibody. Positive tissue
controls using the same IHC protocols included canine normal mammary gland (anti-CK19, ER, -CK14, -VIM, Alpha-SMA, -p63 antibodies) and canine skin (anti-CK5/6). The number
of positive cells by each marker was calculated semiquantitatively: − = no stained cells, ± =
less than 5% positive cells, + = 5–50% positive cells, ++ = more than 50% positive cells.
Cases were considered positive for ER when nuclear staining was observed in at least 5%
tumour cells [16].
Table 1: Primary antibodies, resources and dilutions used in immunohistochemistry
Antibody (anti-)
Clone
Manufacturer
Dilution
P63
4A4
Dako (Glostrup, Denmark)
1:50
Αlpha-SMA
1A4
Dako (Glostrup, Denmark)
1:100
Cytokeratin 19
BA17
Dako (Glostrup, Denmark)
1:50
Cytokeratin 14
Ab-1 (LL002)
NeoMarkers (Fremont, Ca)
1:300
Cytokeratins 5/6
D5/16B4
Zymed (South San Francisco, Ca)
1:100
VIM
V9
Dako (Glostrup, Denmark)
1:100
ER
1D5
Dako (Glostrup, Denmark)
1:25
Results
Four types of myoepithelial cells were recognized on the basis of their morphology. The
resting subtype exhibitedvthe elongated features of spindle cells in close contact withvluminal
epithelial cells as well as proliferating suprabasal cellsvthat instead showed a polygonal shape
(Figure 1(a)). The interstitialmotile cells were observed both forming nests (the spindle type
lined nests and the stellate cells constituted the nest core) and isolated in the interstitium
(Figure 2(a)).
Normal Mammary Gland
112
In the 3 control cases, all resting and proliferative suprabasal myoepithelial cells were
labeled by p63, CK14, Alpha-SMA, and VIM. Resting and proliferative suprabasal
myoepithelial cells did not express CK19 in any of the cases.
Table 2: Immunohistochemical results for suprabasal and motile myoepithelial cells
Mammary Tissue from the Same Line of the Mammary Tumour.
In the 29 normal tissues in the same line as the tumours, all resting and proliferative
suprabasal myoepithelial cells were labeled by p63, CK14, Alpha-SMA, and VIM. CK5/6
was positive in all but four cases and ER was detected in 12 cases. CK19 expression was only
observed in the luminal epithelium. Myoepithelial motile interstitial cells were not observed.
All the results are summarized in Table 2.
113
Mammary Tumours
Immunohistochemical results for suprabasal (Figure 1) and motile cells (Figure 2) using p63,
CK14, CK5/6, CK19, Alpha-SMA, VIM, and ER are summarized in Table 2. CK5/6 labeled:
suprabasal resting cells in 16 of the 29 cases (5 carcinomas in benign mixed tumours and 11
complex carcinomas). CK 5/6 also labeled proliferativesuprabasal and spindle motile cells in
10 cases (3 carcinomas in benign mixed tumours and 7 complex carcinomas). Stellate motile
cells were present in 25 cases (1 benign myoepithelial tumour, 3 malignant myoepithelial
tumours, 5 carcinomas in benign mixed tumours and 15 complex carcinomas) but were
negative for CK5/6. Cartilage in mammary mixed tumours was always negative. CK14
showed positivity in 23 cases in resting cells (with the exception of 1 complex carcinoma and
benign and malignant myoepithelial tumours), in 22 cases in proliferative cells (6 carcinomas
in benign mixed tumours and 15 complex carcinomas) with a trend to lose the expression
when these cells had acquired the more motile phenotype of spindle cells (positive in 2 benign
and 3 malignant myoepithelial tumours, 2 carcinomas in benign mixed tumours and 2
complex carcinomas). Chondrocytes of mixed tumours were negative. VIM was positive in
suprabasal cells in the 23 cases of carcinoma in benign tumour and complex carcinoma and in
motile myoepithelial cells in all 29 cases. Stromal cells were positive in all cases. Cartilage
was VIM positive in all 7 carcinomas in benign-mixed tumours. Alpha-SMA labeled resting
and proliferative suprabasal myoepithelial cells in 23 carcinomas in benign-mixed tumours
and complex carcinomas. The spindle cells in 10 cases (5 carcinomas in benign mixed tumour
and 5 complex carcinomas) showed positivity for Alpha-SMA. In each case, except for
two complex carcinomas, stellate cells were negative. Stroma showed positivity in only 6
cases (1 carcinoma in benignmixed tumour and 5 complex carcinomas). P63 was detected in
resting and proliferative suprabasal myoepithelial cells of the 23 cases of carcinoma in benign
114
mixed tumour and complex carcinoma. All spindle and stellate motile cells were negative.
Stromal cells and cartilage were negative. ER expression was present in 11
suprabasalmyoepithelial cells (4 carcinomas in benign-mixed tumours and 7 complex
carcinomas); spindle and stellate motile cells were positive in 9 cases (3 carcinomas in
benign-mixed tumours and 6 complex carcinomas). Cartilage of mixed tumours was negative.
Resting and proliferative suprabasal and spindle/stellate motile myoepithelial cells did not
express CK19 in any of the tumours examined. Cartilage of mixed tumours was also negative
for CK19.
Discussion
Based on the findings of Gama et al. [1] and Tateyama et al. [11], four morphological types
of myoepithelial cells are present in the mammary gland: resting and proliferative suprabasal
myoepithelial cells lining alveoli and ducts and spindle and stellate interstitial motile cells,
which lie in the interstitial space where they may be arranged in nests. Myoepithelial markers,
such as p63, CK5/6, CK14, Alpha- SMA, and VIM, proved to be valuable diagnostic adjuncts
to facilitate the evaluation of complex and mixed proliferations. CK19 is considered the gold
standard marker for luminal epithelium and was used to avoid any misdiagnosis with
myoepithelial cells. Because of cross-reactivity patterns and the fact that lesional foci are
typically minute, none of the myoepithelial markers enjoyed 100% sensitivity and specificity
for myoepithelial cells. As such, at least 2 markers should be used to evaluate any given focus
[17]. Based on our results, the best marker for suprabasal cells was p63 especially in
association with CK14, which was limited to mature (basal) myoepithelial cells and, to
a lesser extent followed by CK5/6, Alpha-SMA and VIM (Figure 1). However, CK5/6 also
marked luminal epithelial cellsmaking it difficult to distinguish them fromproliferative
suprabasal myoepithelial cells [2]. Morphologically, both epithelial and myoepithelial cells
115
may have a polygonal shape. A characteristic of both CK14 and CK5/6, but not of p63, alphaSMA, and VIM, was their reduced expression in myoepithelial cells in the suprabasal
proliferative state. CK14, CK5/6, and p63 expression was gradually lost in cells in the spindle
and stellate motile state. Alpha-SMA and VIM were present in spindle motile myoepithelial
cells with different degrees of intensity. Only VIM proved to be a consistent marker for
stellate motile myoepithelial cells. In this study, the stellate motile myoepithelium was
arranged in nests and lined by resting cells presumably of alveolar origin. This feature may
support the idea that the nests of stellate motilemyoepithelial cells, which have lost expression
of the main myoepithelial suprabasal markers, but retained affinity for VIM, are the
precursors of cartilage, indicating that these cells have completed their transformation into
mesenchymal elements. In benign and malignant myoepithelial tumours, VIM labeling in all
cases, loss of all other suprabasal myoepithelial markers, and the scant positivity to CK14 in
spindle cells were indicative of a prevailing expression of the myoepithelium motile state
and a possible passage from simple myoepithelial cells to mesenchymal fibroblasts. In our
study, evidence of the myoepithelial cells shifting to a mesenchymal phenotype, shown by the
loss of CK14, CK5/6, and p63 expression, was reinforced by the discontinuous labeling of
spindle cells for Alpha-SMA, a marker of both myoepithelial cells and myofibroblasts, which
was completely lost in stellate motile cells that have supposedly become fibroblasts. Further
confirmation studies by Tsuda et al. [18] reported the occurrence of myofibroblasts with
remnants of CK14 expression (described as “converted myoepithelial cells”). In the cases
examined in the present study, CK14 progressively faded, therefore indicating a loss of the
(myo-)epithelial phenotype. These results support the EMT hypothesis involving a
myoepithelial-like state [19], which undergoes a myoepithelial mesenchymal transition
(MMT). This hypothesis was confirmed in the dog by Gartner et al. [20] who stated that in
116
mammary tumours one of the steps in the evolution ofmesenchymal cells involves the
expression of typical myoepithelial traits. An interesting result of the present study was the
positivity to ER found in 12/29 suprabasalmyoepithelial cells and 9/29 stellate cells of
carcinoma in benign-mixed tumours and complex carcinomas. Two isoforms of ER receptors
have been described, namely, ER-α and ER-β, the latter being the only form expressed in the
nuclei of isolated basalmyoepithelial cells [8]. The antibody used in the present investigation
was inclusive of both isoforms: both luminal and basal/stellate cells were labeled, presumably
luminal cells by ER- β and basal/stellate cells by ER-α.
Conclusions
In conclusion, the suprabasal myoepithelial cells were well characterized by p63 and CK14
and to a lesser extent by the other marker used. The motile myoepithelial cells are instead
characterized by Alpha-SMA and VIM and loss of CK14, CK5/6, and p63 (Figure 2). The
present study also demonstrated ER in both luminal epithelial and suprabasal/stellate
myoepithelial cells (the latter in about half of the cases) and that ER expression is not
influenced by the resting/motile phase. Therefore, in serial or multistained sections,
immunohistochemistry to ER in combination with p63 and CK14 may serve to avoid
erroneous identification of luminal or myoepithelial cells in canine mammary tumours.
The trend of preserved Alpha-SMA and VIM expression in spindle cells, and only VIM
positivity in stellate motile cells as well as the decreased p63 expression in both motile types,
supports the hypothesis of the EMT involving a myoepithelial-like state [19] in MMT. The
spindle motile cell could be considered an earlier transformation than the stellate cell towards
a mesenchymal phenotype.
117
Figures
Fig. 1: Suprabasal myoepithelial cells: resting (thick arrows) and proliferative (thin arrows)
cells. Immunohistochemical expression of a panel of antibodies applied by IHC, 63x
(A) Hematoxylin-eosin; (B) anti-CK19 antibodies labeling the cytoplasm; (C) anti-ER
antibodies labeling the nuclei; (D) anti-CK 5/6 antibodies labeling the cytoplasm; (E) antiCK14 antibodies labeling the cytoplasm; (F) anti-VIM antibodies labeling the cytoplasm; (G)
118
anti-Alpha-SMA antibodies labeling the cytoplasmic membrane; (H) anti-p63 antibodies
labeling the nuclei.
Fig. 2: Motile myoepithelial cells: spindle (asterisks) and stellate (stars) cells.
Immunohistochemical expression of a panel of antibodies applied by IHC. 63x
(A) Hematoxylin-eosin; (B) anti-CK19 antibodies labeling the cytoplasm; (C) anti-ER
antibodies labeling the nuclei; (D) anti-CK 5/6 antibodies labeling the cytoplasm; (E) anti119
CK14 antibodies labeling the cytoplasm; (F) anti-VIM antibodies labeling the cytoplasm; (G)
anti-Alpha-SMA antibodies labeling the cytoplasmic membrane; (H) anti-p63 antibodies
labeling the nuclei.
120
References
[1] A. Gama, A. Alves, F. Gartner, and F. Schmitt, “p63: a novel myoepithelial cell marker in
canine mammary tissues,” Veterinary Pathology, vol. 40, no. 4, pp. 412–420, 2003.
[2] L. N. Z. Ramalho, A. Ribeiro-Silva, G. D. Cassali, and S. Zucoloto, “The expression of
p63 and cytokeratin 5 in mixed tumors of the canine mammary gland provides new insights
into the histogenesis of these neoplasms,” Veterinary Pathology, vol. 43, no. 4, pp. 424–429,
2006.
[3] K. Polyak and M. Hu, “Do myoepithelial cells hold the key for breast tumor progression?”
Journal of Mammary Gland Biology and Neoplasia, vol. 10, no. 3, pp. 231–247, 2005.
[4] K. U. Sorenmo, R. Rasotto, V. Zappulli, and M. H. Goldschmidt, “Development,
anatomy, histology, lymphatic drainage, clinical features, and cell differentiation markers of
canine mammary gland neoplasms,” Veterinary Pathology, vol. 48, no. 1, pp. 85–97, 2011.
[5] T. Gudjonsson, M. C. Adriance, M. D. Sternlicht, O. W. Petersen, and M. J. Bissell,
“Myoepithelial cells: their origin and function in breast morphogenesis and neoplasia,”
Journal of Mammary Gland Biology and Neoplasia, vol. 10, no. 3, pp. 261–272, 2005.
[6] T. O. Nielsen, F. D. Hsu, K. Jensen et al., “Immunohistochemical and clinical
characterization of the basal-like subtype of invasive breast carcinoma,” Clinical Cancer
Research, vol. 10, no. 16, pp. 5367–5374, 2004.
[7] Y. H. Hsiao, H. D. Tsai, M. C. Chou, and Y. G. Man, “The myoepithelial cell layer may
serve as a potential trigger factor for different outcomes of stage-matched invasive lobular and
ductal breast cancers,” International Journal of Biological Sciences, vol. 7, no. 2, pp. 147–
153, 2011.
121
[8] J. Mart´ın De Las Mulas, J. Ord´as, M. Y. Mill´an et al., “Immunohistochemical
expression of estrogen receptor β in normal and tumoral canine mammary glands,” Veterinary
Pathology, vol. 41, no. 3, pp. 269–272, 2004.
[9] A. Gama, A. Alves, and F. Schmitt, “Identification ofmolecular phenotypes in canine
mammary carcinomas with clinical implications: application of the human classification,”
Virchows Archiv, vol. 453, no. 2, pp. 123–132, 2008.
[10] F. Sassi, C. Benazzi, G. Castellani, and G. Sarli, “Molecularbased tumour subtypes of
canine mammary carcinomas assessed by immunohistochemistry,” BMC Veterinary
Research, vol. 6, article 5, 2010.
[11] S. Tateyama, K. Uchida, T. Hidaka, M. Hirao, and R. Yamaguchi, “Expression of bone
morphogenetic protein-6 (BMP-6) in myoepithelial cells in canine mammary gland tumors,”
Veterinary Pathology, vol. 38, no. 6, pp. 703–709, 2001.
[12] L. Rønnov-Jessen and O. W. Petersen, “A function for filamentous α-smooth muscle
actin: retardation of motility in fibroblasts,” Journal of Cell Biology, vol. 134, no. 1, pp. 67–
80, 1996.
[13] R. Kalluri and R. A. Weinberg, “The basics of epithelialmesenchymal transition,”
Journal of Clinical Investigation, vol. 119, no. 6, pp. 1420–1428, 2009.
[14] W. Misdorp, R. W. Esle, E. Hellm´en, and T. P. Lipscomb, Histological Classification of
Mammary Tumors of the Dog and Cat, Armed Forces Institute of Pathology in cooperation
with the American Registry of Pathology and the World Health Organization Collaborating
Centre for Worldwide Aderence on Comparative Oncology,Washington, DC, USA, 1999.
[15] M. H. Goldschmidt, L. Pe˜na, R. Rasotto, and V. Zappulli, “Classification and grading of
canine mammary tumors,” Veterinary Pathology, vol. 48, no. 1, pp. 117–131, 2011.
122
[16] F. Millanta, M. Calandrella, G. Bari, M. Niccolini, I. Vannozzi, and A. Poli,
“Comparison of steroid receptor expression in normal, dysplastic, and neoplastic canine and
feline mammary tissues,” Research in Veterinary Science, vol. 79, no. 3, pp. 225–232, 2005.
[17] R. Dewar, O. Fadare, H. Gilmore, and A. M. Gown, “Best practices in diagnostic
immunohistochemistry: myoepithelial markers in breast pathology,” Archives of Pathology
and Laboratory Medicine, vol. 135, no. 4, pp. 422–429, 2011.
[18] H. Tsuda, T. Takarabe, F. Hasegawa, T. Fukutomi, and S. Hirohashi, “Large, central
acellular zones indicating myoepithelial tumor differentiation in high-grade invasive ductal
carcinomas as markers of predisposition to lung and brain metastases,” American Journal of
Surgical Pathology, vol. 24, no. 2, pp. 197–202, 2000.
[19] O. W. Petersen, H. L. Nielsen, T. Gudjonsson, R. Villadsen, L. Rønnov-Jessen, and M. J.
Bissell, “The plasticity of human breast carcinoma cells is more than epithelial to
mesenchymal conversion,” Breast Cancer Research, vol. 3, no. 4, pp. 213–217, 2001.
123
Publications and Proceedings
1. Beha G., Sarli G., Brunetti B., Sassi F., Ferrara D., Morandi F., Benazzi C. Epithelialmesenchymal transition in canine in canine mammary tumors: the role of myoepithelial cells.
Proceedings of the 29th Annual meeting of the ESVP-ECVP Uppsala, 7-10 September 2011,
p. 145
2. Beha G., Sarli G., Brunetti B., Sassi F., Ferrara D., Benazzi C. Morphology of the
Myoepithelial Cell: Immunohistochemical Characterization from Resting to Motile Phase.
Scientific World Journal. 2012a; Published online 2012 August 5. doi: 10.1100/2012/252034
3. Correlator of Michela Levi Thesis:Marcatori Di Differenziazione Delle Cellule
Mioepiteliali Nei Tumori Mammari Della Cagna
124
5. CD117 EXPRESSION INFLUENCES PROLIFERATION BUT NOT
SURVIVAL IN CANINE MAMMARY TUMOURS
Accepted 10/02/2014 Journal of Comparative Pathology
CD117 expression influences proliferation but not survival in canine mammary
tumours
B. Brunetti*, G. Beha*§, C. Benazzi*, L.J. DeTolla and G. Sarli*
*Department of Veterinary Medical Sciences – University of Bologna, Via Tolara di Sopra,
50 40064 Ozzano Emilia (Bologna), †Program of Comparative Medicine, Department of
Pathology and Greenebaum Cancer Center, University of Maryland, School of Medicine, 10
South Pine St., MSTF, Suite G-100,Baltimore, MD (USA) 21201-1192
Introduction
C-Kit is a gene that encodes a membrane-associated tyrosine kinase growth factor receptor
(CD117) (Yarden et al., 1987), composed of an extracellular ligand-binding region (5
immunoglobulin-like domains), a single transmembrane spanning region (hydrophobic
domain), and a cytoplasmic region which includes both the kinase (ATP-binding and
phosphotransferase) and the juxtamembrane domains (Linnekin, 1999; Boissan et al., 2000).
CD117 is expressed in various cell types during embryonic development and it promotes the
migration, differentiation, proliferation, growth, adhesion, chemotaxis and survival of cells
(Chian et al., 2001; Ronnstrand, 2004). CD117 has been shown to be expressed by neoplastic
cells as well; in the dog most notably, in gastrointestinal stromal tumors (GIST) (Bettini et al.,
2003; Frost et al., 2003; Smith et al, 2009), in mast cell tumours (Gil da Costa et al., 2007;
Thompson et al., 2011), neoplastic testes (Bush et al., 2011; Thorvaldsen et al., 2012), and
125
melanocytic tumours (Murakami et al., 2011; Gomes et al., 2012). Few studies have focused
on the expression of CD117 in canine mammary tissue, and the difference in its expression
between normal tissue and neoplastic benign or malignant change (Kubo et al., 1998; Morini
et al., 2004; Sailasuta et al., 2008). Kubo et al. (1998) investigated the expression of the
canine c-Kit oncogene in mammary tumours with the aid of RT-PCR; the results showed a
significant higher level of transcription in adenocarcinoma than in malignant mixed tumours.
Morini et al. (2004) and Sailasuta et al. (2008) evaluated CD117 expression using
immunohistochemistry. Morini et al. (2004) found a weak to moderate cytoplasmic staining
in ductal and acinar epithelial cells of the normal canine mammary gland, and a weak to
moderate cytoplasmic staining in benign and malignant mammary tumours. Sailasuta et al.
(2008) found either membranous, cytoplasmic staining or both, but they did not see a
statistically significant difference between the CD117 expression and histological type,
histological malignancy, nuclear differentiation or grade of mammary gland tumour.
Since c-Kit is a proto-oncogene that encodes a transmembrane tyrosine kinase growth factor
receptor and stimulates cell proliferation, it plays a crucial role in determining existence of a
correlation between c-Kit expression and Ki67 index. A correlation between aberrant CD117
expression and increased cell proliferation in canine mast cell tumours has been reported (Gil
da Costa et al., 2007; Webster et al., 2007; Thompson et al., 2011). Therefore, the aims of the
present study were (1) to characterize the immunohistochemical staining of CD117 in normal
and neoplastic mammary tissue of the dog, and (2) to correlate CD117 immunohistochemical
results with mammary histotype, histological stage (invasiveness), Ki67 index and patient
survival time.
Materials And Methods
Study material
126
In this study 49 samples of normal and neoplastic canine mammary tissue were examined.
The specimens were retrieved from the archive of the Veterinary Pathology Division of the
Department of Veterinary Medical Sciences of the University of Bologna. The samples
included 8 normal, hyperplastic and dysplastic mammary tissue specimens, 11 cases of benign
mammary tumours and 30 cases of malignant mammary neoplasms. Tissue specimens were
fixed in 10% buffered formalin and embedded in paraffin. Sections of 4 µm were stained with
Haematoxylin–Eosin (H-E) and observed microscopically. The samples were then
histologically diagnosed according to the criteria set by Misdorp et al., (1999). In malignant
tumours, the histologic stage of invasion was determined according to Gilbertson et al. (1983)
as follows: stage 0 = tumours without stromal invasion; stage I = tumours with stromal
invasion, stage II = tumours with neoplastic emboli in vessels or lymph node metastases (or
both). As a follow-up the owners/referring veterinarians of all cases of malignant tumours
were contacted in order to obtain information regarding survival time and the cause of death.
Immunohistochemistry
Two consecutive sections were dewaxed in toluene and rehydrated. Endogenous peroxidase
was blocked by immersion in 3% hydrogen peroxide for 30 minutes. Sections were then
rinsed in Tris buffer, immersed in citrate buffer (2.1g citric acid monohydrate/litre of distilled
water), pH 6.0, incubated for 2 cycles of 5 minutes (anti-CD117 antibody) and for 4 cycles of
5 minutes each (anti-Ki67 antibody) in a microwave oven at 750W, and allowed to cool down
at room temperature for approximately 20 minutes.
The antibodies (CD117, polyclonal, Dako Cytomation, 1:100 dilution with phosphatebuffered saline solution and 1% BSA) and anti-Ki67 antibody (clone MIB-1, Dako
Cytomation, 1:30 dilution with phosphate-buffered solution) were applied, and the sections
were incubated overnight at 4ºC. Sites of primary antibody binding were identified using a
127
commercial Streptavidin-biotin-peroxidase kit (LSAB KIT, Dako) and diaminobenzidine as
the chromogen (0.04% for 10 minutes). The sections were then counterstained with
Papanicolaou Haematoxylin, rinsed in tap water, dehydrated and cover-slipped.
Sections of a well differentiated canine mast cell tumour were used as positive controls for
CD117 and the basal layer of the epidermis served as an internal positive control for Ki67.
Negative controls were prepared by incubating the slides with an isotype-matched nonspecific
antibody.
CD117 evaluation
The slides were blindly evaluated by 2 board-certified anatomical pathologists (C.B. and
B.B.) for CD117 immunoreactivity. The results were considered positive when a brown
labelling of the cells was present, and negative when no immunoreactivity was noted on the
section (CD117 expression).
The pattern of positivity of the cells for CD117 was classified as membranous (M) (brown
labelling of the cytoplasmic membrane), cytoplasmic (C) (diffuse brown labelling of the
cytoplasm) or membranous-cytoplasmic (MC) (simultaneous cytoplasmic and membranous
expressions). CD117 extension was the percentage of positively labelled cells (extension)
evaluated as follows: 0= negative (0%), 1= focal (1-19%), 2= intermediate (20-49%), 3=
diffuse (> 50%) as described by Gomes et al. (2012) modified.
Ki67 index
Ki67 immunoistochemical labelling was blindly evaluated by a single pathologist (G.B.)
using a 40x objective to select 5 fields with the highest Ki67 positivity; areas with necrosis or
inflammation were avoided. Ki67 index was calculated as the percentage of labelled nuclei
128
compared with the total nuclear area of the field according to a previously described method
(Sarli et al., 2002).
Statistics
The correlation between CD117 expression, CD117 pattern or CD117 extension vs type of
mammary lesion, histological classification, histological stage and Ki67 index was tested with
the Pearson’s chi-square (χ2) test. Ki67 index was tested with the Shapiro Wilk’s W analysis
for normality and appeared non-normally distributed. Due to this abnormal distribution, Ki67
index in CD117 expression groups (positive vs negative) and CD117 patterns were tested
with the non-parametric Kruskal Wallis ANOVA Median test. Survival analysis was used to
compare Kaplan Meyer estimated curves in 2 groups (CD117 expression in the 3 groups
(CD117 pattern) or in the 4 groups (CD117 exension) selecting only those cases in which
death was due with no doubt to the tumour. In all tests a value of P<0.05 was considered to be
statistically significant.
Results
The 49 cases were as follows: 3 cases of normal mammary tissue, 5 cases of hyperplasia, 11
benign tumours (4 tubulo-papillary adenomas, 1 simple adenoma, 1 complex adenoma, 3
fibroadenomas, and 2 benign mixed tumours); 30 malignant tumours (10 solid carcinomas, 4
carcinoma in benign tumours, 8 simple tubulo-papillary carcinomas, 5 complex carcinomas, 2
anaplastic carcinomas, and 1 squamous carcinoma). Of these 30 malignant tumours, 2 were
found to be histological stage 0, 12 stage I, and 16 stage II. The mast cell tumour used as
positive control showed a strong membranous positivity in all neoplastic cells for CD117
expression. Ki67 expression was evident in the nuclei of the basal layer of epidermis.
129
Information about the patients’ survival time of 22 out of the 30 malignant tumours studied
was retrieved, and in 13 cases death was correlated with the presence of metastasis (Table 1).
Table 1: Individual data
STAGE
CD117
CD117
CD117
EXPRESSION PATTERN EXTENSION
Ki67
index
SURVIVAL
RATE
(MONTHS)
N°
HISTOTYPE
1
Normal
P
M
3
7,9
2
Normal
P
M/C
3
5,78
3
Normal
P
C
1
4,27
4
ADH
P
C
1
3,34
5
ADH
P
C
1
3,56
6
ADH
N
0
2,84
7
Hyperplasia
N
0
6,5
8
Hyperplasia
P
M
3
6,16
9
Tubpap adenoma
P
C
1
4,15
10
Tubpap adenoma
P
C
2
3,07
11
Tubpap adenoma
P
C
3
2,79
12
Tubpap adenoma
N
0
3,43
13
Complex adenoma
N
0
2,7
14
Simple adenoma
N
0
14,34
15
Fibroadenoma
P
C
1
5,36
16
Fibroadenoma
P
M
3
25,07
17
Fibroadenoma
N
0
1,66
18
BMT
N
0
3,07
19
BMT
N
0
8,45
20
Solid carcinoma
2
P
C
1
9,83
36
21
Solid carcinoma
1
P
C
2
3,85
24
22
Solid carcinoma
2
P
C
2
5,77
NA
130
23
Solid carcinoma
0
P
M/C
3
8,53
NA
24
Solid carcinoma
1
P
M/C
3
6,49
NA
25
Solid carcinoma
2
P
M/C
3
23,18
3
26
Solid carcinoma
2
P
M/C
3
8,18
24
27
Solid carcinoma
1
P
M/C
3
9,39
24
28
Solid carcinoma
1
N
0
4,26
NA
29
Solid carcinoma
2
N
0
1,86
1
30
CBT
2
P
M/C
3
8,14
24
31
CBT
2
P
M/C
3
13,85
48
32
CBT
1
P
C
3
5,95
1
33
CBT
1
N
0
2,65
48
34
Tubpap carc
2
P
C
1
8,21
12
35
Tubpap carc
1
P
C
2
2,78
6
36
Tubpap carc
1
P
M/C
2
5,58
NA
37
Tubpap carc
2
P
C
3
6,67
1
38
Tubpap carc
2
P
M/C
3
9,22
NA
39
Tubpap carc
2
N
0
1,96
2
40
Tubpap carc
1
N
0
2,45
12
41
Tubpap carc
1
N
0
6,26
NA
42
Complex carc
2
P
2
3,67
24
43
Complex carc
0
N
0
2,56
5
44
Complex carc
2
N
0
2,95
6
45
Complex carc
1
N
0
2,13
A
46
Complex carc
1
N
0
1,97
N/A
47
Anaplastic carc
2
P
C
3
8,25
2
48
Anaplastic carc
2
P
C
2
5,51
2
49
Squamous carc
2
P
C
3
8,16
A
C
ADH: atypical ductal hyperplasia. Tubpap adenoma: tubulo-papillary adenoma. BMT: benign in mixed tumour. CBT:
carcinoma in benign mixed tumours. Tubpap carc: tubulo-papillary carcinoma. Comlex carc: complex carcinoma. Anaplastic
carc: anaplastic carcinoma. Squamous carc: squamous carcinoma. P: positive. N: negative. M: membranous. C: cytoplasmic.
M/C: membranous and cytoplasmic. NA: available. A: alive
131
Normal mammary tissue and hyperplasia
CD117 expression was observed in the alveolar and ductal mammary epithelium in 6 of the 8
examined normal and hyperplastic mammary tissues. No expression was detected in the
myoepithelial and in the stromal cells in any of the cases. The remaining 2 cases (mammary
hyperplasia) were negative. The CD117 pattern was M in two cases (1 normal mammary
tissue and 1 hyperplasia), C in 3 cases (1 normal mammary tissue and 2 hyperplasias), M/C in
1 normal mammary tissue (Fig. 1a). CD117 extension was focal in 3 cases (1 normal
mammary tissue and 2 hyperplasias) and diffuse in the remaining (1 normal mammary tissue,
2 hyperplasias) (Table 1, Fig. 2a, 2d, 2g).
Benign mammary tumours
In benign mammary tumours CD117 was expressed exclusively in the epithelial cells in 5 out
of 11 cases. Regarding CD117 pattern, 3 of the 4 tubulo-papillary adenomas were positive
with C labelling and 1 was negative. Two positive cases out of 3 of the fibroadenoma type
showed a C and a M pattern respectively (Fig. 1b). The remaining fibroadenoma as well the
simple adenoma, the complex adenoma, and the 2 benign mixed tumours did not express
CD117 . In the positive cases the labelling extension was focal in 2 cases (1 tubulo-papillary
adenoma and 1 fibroadenoma), intermediate in a tubulo-papillary adenoma and diffuse in 2
cases (1 tubulo-papillary adenoma and 1 fibroadenoma). All the data are reported in Table 1
and Fig. 2a, 2d, 2g.
Malignant mammary tumours
CD117 immunoexpression was present in 20 of the 30 malignant tumours, namely 8 solid
carcinomas, 3 carcinoma in benign mixed tumours, 5 tubulo-papillary carcinomas, 1 complex
132
carcinoma, 2 anaplastic carcinomas and 1 squamous cell carcinoma. For the labelling
location, the positive cells were all epithelial cells except for one case that presented both
epithelial and myoepithelial staining (carcinoma in benign tumour). The 8 solid carcinomas
displayed different CD117 pattern: C reactivity in 3 cases and M/C in 5 cases (Fig. 1d, 1e, 1f).
The 3 carcinomas in benign mixed tumours showed C positivity in one case, and M/C in the
other 2 cases (Fig. 1c). The cartilaginous component of the latter 3 cases was negative for
CD117 (Fig. 1c). The 5 tubulo-papillary carcinomas displayed C in 3 cases and M/C pattern
in 2 cases. The 2 anaplastic carcinoma, complex carcinoma and squamous cell carcinoma
showed C expression. The labelling extension was focal in 3 cases (1 solid carcinoma, 1
tubulo-papillary, 1 complex carcinoma), intermediate in 5 cases (2 solid carcinomas, 2 tubulopapillary carcinomas, 1 anaplastic carcinoma) and diffuse in 12 cases (5 solid carcinomas, 3
carcinomas in benign mixed tumours, 2 tubulo-papillary carcinomas, 1 anaplastic carcinoma,
1 squamous cell carcinoma) (Table 1, Fig. 2).
Correlation between the three CD117 variables and type of mammary lesion,
malignant histotypes and histological stage
None of the 3 CD117 variables (CD117 expression, CD117 pattern and CD117 extension)
appeared to be associated with the different mammary lesion types (normal/hyperplasia,
benign tumours, malignant tumours), histotypes and invasiveness. Only in malignant tumours
was an association between CD117 pattern and histological type evident (R=0.37, P=0.043,
Pearson test) (Table 2).
Correlation between CD117 expression, pattern, extension and Ki67
All the 3 CD117 variables were significantly associated with Ki67 index (Table 2) (Fig. 3a,
3c, 3e). Comparing the Ki67 value of the CD117 positive cases (CD117 expression) vs those
133
negative, only in malignant tumours was a significantly higher Ki67 value apparent (P<0.001,
Kruskall-Wallis test) (Fig. 3a). The tumours with both cytoplasmic and membranous staining
patterns showed a significantly higher range and higher median value of the Ki67 index
compared to cases with only cytoplasmic expression (P<0.001, Kruskall-Wallis test), and
were both significantly higher than negative cases (P<0.01, Kruskall-Wallis test) (Fig. 3c).
Ki67 showed values increased progressively from negative cases to those presenting diffuse
extension staining of CD117 (P<0.001, R=0.61, Spearman test) (Fig. 3e) (Table 2).
Correlation between CD117 expression, pattern, extension and patient survival
time
Survival time was available for 22 female dogs bearing malignant tumours. In only 13 cases
death was due to the tumour and only in these cases survival analysis was performed. Survival
analysis did not reveal any differences in the comparison of the 2 groups of CD117
expression (positive vs negative) (P=0.91 Survival analysis) (Fig. 3b) or the 3 groups of
CD117 pattern (membranous, cytoplasmic, cytoplasmic+membranous) (P=0.46 Survival
analysis) (Fig. 3d) and the four groups of CD117 extension (absent, focal, intermediate,
diffuse) (P=0.85 Survival analysis) (Fig. 3f) (Table 2).
Table 2: R and P values of the Pearson chi square test
Nmg/Hyp vs BT
Histotypes of
vs MT
MT
Invasiveness
Ki67 index
R
P
R
P
R
P
R
P
-0.004
0.97
0.21
0.25
-0.30
0.10
-0.35
0.014
CD117 pattern
0.10
0.47
0.37
0.043
-0.16
0.37
-0.55
<0.001
CD117 extension
0.14
0.33
-0.15
0.40
0.25
0.17
0.51
<0.001
CD117 expression
Nmg = normal mammary gland; Hyp = hyperplasia; BT = benign tumour; MT = malignant tumour.
134
Discussion
Normal human breast tissues strongly express the c-Kit proto-oncogene product on the cell
membrane and/or cytoplasm of alveolar and ductal cells (Chui et al., 1996; Tomasino et al.,
2009). Regarding benign and malignant human breast diseases discordant results are reported
in literature; in fact, in several studies the c-Kit proto-oncogene product was detected
heterogeneously with a reduced immunoreactive score in benign lesions and even less
represented in breast cancer (Chui et al., 1996; Tomasino et al., 2009; Kondi-Pafiti et al.,
2010). Other Authors found an increased expression of CD117 in poorly differentiated breast
cancer (Diallo et al., 2006; Kanapathy Pillai et al., 2012).
By now, only few studies have investigated the expression of CD117 in canine mammary
tumours (Kubo et al., 1998; Morini et al., 2004; Sailasuta et al., 2008), but with no clear
results about the importance of this receptor.
The present study evaluated CD117 expression, pattern and extension in canine mammary
tissue (normal and neoplastic), correlating the results not only to different mammary lesions,
histotypes, histological stage and survival, but also with a marker of proliferation Ki67.
Considering CD117 expression, the present results, also in agreement with Sailasuta et al.,
(2008), showed no statistically significant differences between the expression of CD117
(positivity/negativity) and different histotypes. Our results demonstrated a distinctive decrease
in CD117 expression in benign tumours comparing with normal tissue (similarly to what seen
in human studies), but with no such a distinctive decrease in CD117 expression in malignant
tumours as reported in some human breast cancer studies (Chui et al., 1996; Tomasino et al.,
2009; Kondi-Pafiti et al., 2010). This may be linked to canine tumour heterogenicity or
perhaps because malignant mammary tumours are usually very well differentiated and present
a less malignant behaviour than their of the human counterparts.
135
Regarding the CD117 pattern, in normal mammary glands all three patterns were found. This
result may be indicative of functional activity of CD117 irrespective of the type of positivity.
In benign tumours, the present results highlighted the absence of simultaneous M/C
expression, as the cases were completely C or M. With the progression of the lesions
(malignant tumours), C and M/C, but not M only, were present. The decreasing trend of M
only expression to the simultaneous M/C expression or even only C can be interpreted as an
accumulation of receptor proteins that did not undergo final maturation, or an excessive
internalization of activated c-Kit and/or an abnormal synthesis of the receptor, preventing it
from reaching the transmembrane region as suggested by different authors (Torres-Cabala et
al., 2009; Gomes et al., 2012). A significant correlation between CD117 pattern and
histotypes of malignant tumours was found (R=0.37, P=0.043, Pearson test) indicating that
the localization of this type of receptor may be associated with the morphological type of the
malignant mammary carcinoma with an increasing percentage of “only cytoplasmic” pattern
from the well differentiated to the less differentiated histotypes.
Analyzing the CD117 extension (percent of positive cells) it can be observed that all the
extension types (i.e. decrease of positivity) are present in the different type of mammary
lesion except for the “intermediate” in the group of normal mammary tissue and hyperplasia,
which could be due to the low number of cases; in addition, the extension of the positivity
was not related to the type of mammary lesion and malignant histotypes.
None of the three parameters (CD117 expression, pattern and extension) were shown to be
correlated with invasiveness, although, this association has been observed in pancreatic
cancer cells (Yasuda et al., 2006).
Similar negative findings occurred correlating the three parameters (CD117 expression,
pattern and extension) with survival time, i.e. there was no correlation of CD117 and patient
survival.
136
The usefulness of CD117 results changes when these are correlated with the Ki67 index.
Proliferative activity is strictly associated not only to positive/negative CD117 expression, but
also to the type of expression (CD117 pattern) and to the percentage of labelling (CD117
extension), suggesting in canine mammary tumours the existence of a link between the
presence of the receptor and the proliferative activity. This also suggests that the M, M/C as
well as only C expression are able to induce proliferation, with decreasing grades of signal
from M to M/C to C only, demonstrated by our results in mammary malignant neoplasms. Gil
da Costa et al. (2007) as well, in their paper about canine mast cell tumours results have
highlighted a strong correlation between C (altered) CD117 immunoexpression and increased
cell proliferation. These results suggest that even a C expression pattern is able to partially
increase cellular proliferation as assessed by the Ki67 labelling index. As noted above, the
different patterns of expression, which have been identified in the juxtamembrane domain of
c-Kit in certain canine mast cell tumours which are attributed to point mutations, deletions,
and duplications (Preziosi et al., 2004), seem to apply to mammary neoplasms as well. It is
well-known (Regan et al., 2012) that the c-Kit signaling pathway is required for the normal
function of mammary epithelial progenitors.
Conclusions
The M, M/C and C expression, existing in the normal mammary gland as well as in neoplastic
growth, suggest that in transformed mammary cells the c-Kit signalling network may be
activated downstream of the receptor. The demonstration that CD117 expression, pattern and
extension in canine mammary tumours is correlated with proliferative activity may provide
evidence for the utility of tyrosine kinase inhibitors in the therapy of neoplastic mammary
disease.
137
Figures
Fig. 1: Immunohistochemistry (IHC) of CD117 expression, pattern and extension.
A) Dog, normal mammary gland. Normal mammary ducts and lobules show a cytoplasmic
and membranous pattern and diffuse IHC CD117extension. Bar, 50µm. b) Dog, mammary
gland, fibroadenoma. Epithelial cells show a membranous pattern and diffuse extension to
IHC CD117. Bar, 100µm. c) Dog, mammary gland, carcinoma in benign mixed tumour.
Evidence of membranous and cytoplasmic pattern and diffuse extension to IHC CD117 in the
epithelial component of the tumour. The cartilaginous component is negative. Bar, 500 µm. d)
Dog, mammary gland, solid carcinoma. The cells to the left with squamous metaplasia show
cytoplasmic positivity to IHC CD117, while the spindle-shaped epithelial cells on the right
express membranous positivity. Bar, 200 µm. e) Dog, mammary gland, solid carcinoma. The
epithelial cells present a cytoplasmic and membranous pattern and intermediate extension to
IHC CD117. Bar, 100 µm. f) Dog, mammary gland, solid carcinoma. The epithelial cells
138
present a membranous and cytoplasmic pattern and diffuse extension to IHC CD117. Bar, 100
µm.
Fig. 2: CD117 expression, pattern and extension.
CD117 expression (i.e. positive or negative immunohistochemical staining) in the 3
categories of tissues (a) in the histotypes of malignant tumours (b) and in the 3 histological
stages (c). CD117 pattern (i.e. the localization of the immunohistochemical stain as
membranous and/or cytoplasmic) in the 3 categories of tissues (d), in the histotypes of
malignant tumours (e) and in the 3 histological stages (f). CD117 extension (i.e. the amounts
of tissue stained) in the 3 categories of tissues (g), in the histotypes of malignant tumours (h)
and in the 3 histological stages (i).
139
Fig. 3: Correlation between Ki67 index and CD117 expression, pattern, extension and
survival rate.
Ki67 index of the malignant tumours in the two groups (positive; negative) of CD117
expression (a), in the 3 groups (negative; membranous/cytoplasmic; C=cytoplasmic) of
CD117 pattern (b) and in the 4 groups (negative, focal, intermediate, diffuse) of CD117
expression (c). Kaplan-Meyer estimated curves in the two groups (positive; negative) of
CD117 expression (d), in the 3 groups (negative; membranous/cytoplasmic; cytoplasmic) of
CD117 pattern (e) and in the 4 groups (negative, focal, intermediate, diffuse) of CD117
expression (f).
140
References
Bettini G, Morini M, Marcato PS (2003) Gastrointestinal spindle cell tumours of the dog:
histological and immunohistochemical study. Journal of Comparative Pathology, 129, 283293.
Boissan M, Feger F, Guillosson JJ, Arock M (2000) C-Kit and c-Kit mutations in
mastocytosis and other hematological diseases. Journal of Leukocyte Biology, 67, 135-148.
Bush JM, Gardiner DW, Palmer JS, Rajpert-De Meyts E et al. (2011) Testicular germ cell
tumours in dogs are predominantly of spermatocytic seminoma type and are frequently
associated with somatic cell tumours. International Journal of Andrology, 34, 288-295.
Chian R, Young S, Danilkovitch-Miagkova A, Ronnstrand L, Leonard E, et al. (2001)
Phosphatidylinositol 3 kinase contributes to the transformation of hematopoietic cells by the
D816V mutant. Blood, 98, 1365-1373.
Chui X, Egami H, Yamashita J, Kurizaki T, Ohmachi H et al. (1996) Immunohistochemical
expression of the c-kit proto-oncogene product in human malignant and non-malignant breast
tissues. British Journal of Cancer, 73, 1233-1236.
Diallo R., Ting E, Gluz O, Herr A, Schütt G et al. (2006) C-Kit expression in high-risk breast
cancer subgroup treated with high-dose or conventional dose-dense chemotheraphy.
Verhandlungen der Deutschen Gesellschaft für Pathologie, 90, 177-185.
Frost D, Lasota J, Miettinen M (2003) Gastrointestinal stromal tumors and leiomyomas in the
dog: a histopathologic, immunohistochemical, and molecular genetic study of 50 cases.
Veterinary Pathology, 40, 42-54.
Gil da Costa RM, Matos E, Rema A, Lopes C, Pires AM et al. (2007) CD117
immunoexpression in canine mast cell tumours: correlations with pathological variables and
proliferation markers. BMC veterinary research, 3, 19.
141
Gilbertson SR, Kurzman ID, Zachrau RE, Hurvitz AI, Black MM (1983) Canine Mammary
Epithelial Neoplasms: Biologic Implications of Mophologic Characteristics Assessed in 232
Dogs. Veterinary Pathology, 20, 27-42.
Gillespie V, Baer K, Farrelly J, Craft D, Luong R (2011) Canine gastrointestinal stromal
tumors: Immunohistochemical expression of CD34 and examination of prognostic indicators
including proliferation markers KI67 and AgNOR. Veterinary Pathology, 48, 283-291.
Gomes J, Queiroga FL, Prada J, Pires I (2012) Study of c-kit immunoexpression in canine
cutaneous melanocytic tumors. Melanoma Research, 22, 195-201.
Kanapathy Pillai SK, Tay A, Nair S, Leong C (2012) Triple-negative breast cancer is
associated with EGFR, CK5/6 and c-KIT expression in Malaysian women. BMC Clinical
pathology, 12, 18.
Kondi-Pafiti A, Arkadopoulos N, Gennatas C, Michalaki V, Frangou-Plegmenou M et al.
(2010) Expression of c-kit in common benign and malignant breast lesions. Tumori, 96, 978984.
Kubo K, Matsuyama S, Katayama K, Tsutsumi C, Yonezawa K, et al. (1998) Frequent
expression of the c-kit proto-oncogene in canine malignant mammary tumor. The Journal of
Veterinary Medical Science, 60, 1335-1340.
Linnekin D (1999) Early signaling pathways activated by c-kit in haematopoietic cells. The
International Journal of Biochemistry & Cell Biology, 31, 1053-1074.
Misdorp W, Else RW, Hellmén E, Lipscomb TP (1999) Histological Classification of
Mammary Tumors of the Dog and Cat. Published by the Armed Forces Institute of Pathology
in cooperation with the American Registry of Pathology and the World Health Organization
Collaborating Centre for Worldwide Aderence on Comparative Oncology: Washington DC
142
Morini M, Bettini G, Preziosi R, Mandrioli L (2004) C-Kit Gene Product (CD117)
Immunoreactivity in Canine and Feline Paraffin Sections. Journal of Histochemistry &
Cytochemistry, 52, 705-708.
Murakami A, Mori T, Sakai H, Murakami M, Yanai T et al. (2011) Analysis of KIT
expression and KIT exon 11 mutations in canine oral malignant melanomas. Veterinary and
Comparative Oncology, 9, 219-224.
Preziosi R, Morini M, Sarli G (2004) Expression of the KIT protein (CD117) in primary
cutaneous mast cell tumors of the dog. Journal of Veterinary Diagnostic Investigation, 16,
554-561.
Regan JL, Hendrick H, Magnay FA, Vafaizadeh V, Groner B et al. (2012) C-Kit is required
for growth and survival of the cells of origin of Brca-1-mutation-associated breast cancer.
Oncogene, 31, 869-883.
Ronnstrand L (2004) Signal transduction via the stem cell factor/c-Kit. Cellular and
Molecular Life Sciences, 61, 2535-2548.
Sailasuta A, Thomrongsuwannakij T, Romphotiyok K, Theeratammakom T, Wangnaitham S
(2008) Expression of c-Kit Oncogene Product in Mammary gland tumors in dogs. Proc 15th
FAVA congress-OIE Joint Symposium on Emerging Diseases, Bangkok, Thailand, October
27-30, pp 339
Sarli G, Preziosi R, Benazzi C, Castellani G, Marcato PS (2002) Prognostic value of
histologic stage and proliferative activity in canine malignant mammary tumors. Journal of
Veterinary Diagnostic Investigation, 14, 25-34.
Smith AJ, Njaa BL, Lamm CG (2009) Immunohistochemical expression of c-Kit protein in
feline soft tissue fibrosarcomas. Veterinary Pathology, 46, 934-939.
143
Thompson JJ, Yager JA, Best SJ, Pearl DL, Coomber BL et al. (2011) Canine subcutaneous
mast cell tumors: cellular proliferation and KIT expression as prognostic indices. Veterinary
Pathology, 48, 169-181.
Thorvaldsen TE, Nødtvedt AW, Grotmol T, Gunnes G (2012) Morphological and
immunohistochemical characterisation of seminomas in Norwegian dogs. Acta Veterinaria
Scandinavica, 17, 54:52.
Tomasino RM, Morello V, Gullo A, Pompei G, Agnese V et al (2009) Assessment of
“Grading” with Ki-67 and c-kit immunohistochemical expressions may be a helpful tool in
management of patients with flat epithelial atipia (FEA) and columnar cell lesions (CCLs) on
core breast biopsy. Journal of cellular physiology, 221, 343-349.
Torres-Cabala CA, Wang WL, Trent J, Yang D, Chen S et al. (2009) Correlation between
KIT expression and KIT mutation in melanoma: a study of 173 cases with emphasis on the
acral-lentiginous/mucosal type. Modern Pathology, 22, 1446:1456.
Webster JD, Yuzbasiyan-Gurkan V, Miller RA, Kaneene JB, Kiupel M (2007) Cellular
proliferation in canine cutaneous mast cell tumors: associations with c-Kit and its role in
prognostication. Veterinary Pathology, 44, 298-308.
Yarden Y, Kuang W, Yang-Feng T, Coussens L, Munemitsu S et al. (1987) Human protooncogene c-Kit: a new cell surface receptor tyrosine kinase for an unidentified ligand. The
EMBO Journal, 6, 3341-3351.
Yasuda A, Sawai H, Takahashi H, Ochi N, Matsuo Y et al. (2006) The stem cell factor/c-Kit
receptor pathway enhances proliferation and invasion of pancreatic cancer cells. Molecular
Cancer, 5, 46.
144
6. CONCLUSIONS
In Human Medicine, the discovery of molecular subtypes of breast cancer has produced
further proofs of the fact that biological diversity consequently denies a unique therapeutic
approach (Peppercorn et al., 2008). The therapeutic approach to breast cancer varies
according to the different molecular phenotypes and in the recent past it was highlighted the
necessity to identify the molecular phenotypes also for the lymph node metastasis and
eventual systemic metastases (Aitken et al., 2010).
The information acquired from the research conducted in the past three years regarding the
identification of molecular phenotypes in feline and canine mammary tumours revealed the
importance of the phenotyping to fill current gaps regarding prognosis and a targeted
therapeutic approach, since the primary tumour phenotype does not always overlap with that
of its metastasis. The primary tumour phenotype assumes a predictive therapeutic role only in
concordant cases, meaning that there should be a concomitant evaluation of both the primary
tumour and its lymph node metastasis. Treatment planning based only on the primary tumour
phenotype can lead to therapeutic failures if the lymph node metastatic phenotype differs
from that of the primary tumour. To the best of our knowledge, this was the first report that
identified molecular phenotypes in feline mammary tumours.
Regarding the correlation between primary mammary site phenotypes and the phenotypes of
systemic metastases, the existence of both biological phenomena of concordance and
discordance in metastatic sites has been confirmed. The prevalence of concordance between
primary and metastatic sites supports the predictive therapeutic value of the primary tumour
phenotype, minimizing any margin of error which can occur in rare discordant cases.
The research focused on various aspects of myoepithelial cells allowed to better characterized
the four different myoepithelial cell morphological types in the normal and neoplastic
145
mammary gland using a panel of antibodies and confirmed the changes of myoepithelial cells
towards mesenchyme from an myoepithelial to a mesenchymal phenotype.
The investigation that was conducted for the study of CD117 demonstrated that its expression,
pattern and extension in canine mammary tumours is correlated with proliferative activity
may provide evidence for the utility of tyrosine kinase inhibitors in the therapy of neoplastic
mammary disease.
146
References
Aitken SJ, Thomas JS, Langdon SP, Harrison DJ, Faratian D (2010). Quantitative analysis of
changes in ER, PR and HER2 expression in primary breast cancer and paired nodal
metastases. Annals of Oncology 21(6):1254-61.
Peppercorn J, Perou CM, Carey LA (2008). Molecular subtypes in breast cancer evaluation
and management: divide and conquer. Cancer Investigation 26:1-10.
147
7. OTHER PUBLICATIONS AND PROCEEDINGS FROM JANUARY
2011 TO JANUARY 2014
1. Isani G., Sarli G., Andreani G., Morandi F., Brunetti B., Marrocco R., Carpené E., Beha
G., Benazzi C. Effects Of Waterborne Copper On Gills Catalase And Blood Biochemistry In
Gilthead Seabream (Sparus Aurata L.) Journal Of Elementology, 2012, 17(2), 255-267,.
2. Beha G., Pisoni L., Mandrioli L., Bombardi C., Muscedere D., Del Magno S., Cinti F.,
Benazzi C. Analisi Istologica Di Tessuto Discale Erniato Prelevato Chirurgicamente. Studio
Preliminare Su 18 Cani. AIPVET, PERUGIA 25-26 May 2012, pp. 34-39.
3. Muscatello L.V., Dondi F., Beha G., Morbidelli E., Brunetti B.: Intossicazione Cronica Da
Glicole Etilenico In Un Cane: “Risolutivo" L’esame Istologico. SCIVAC, Rimini 8-10 June
2012.
4. Beha G, Brunetti B, Muscatello LV, Dondi F, Sarli G, Benazzi C. A Severe Case Of
Dermotosparaxisis In A Shorthair Cat: AIPVET 2013. Giulianova Lido (TE) 29-31 May
5. Pritchard L., Beha G., Smith K, Brunetti B., Sarli G., Benazzi C., Mcgonnell I., Expression
Of Sox9 And Snail2 In Canine Mammary Tumours. Proceedings Of 31st Meeting Of The
ESVP-ECVP, London (UK), 4th-7th September 2013, pp. 65.
6. Beha G., Pisoni L., Bombardi C., Avallone G., Sarli G., Del Magno S., Cinti F., Mandrioli
L., Gandini G., Benazzi C., Histochemical And Immunohistochemical Analysis Of Herniated
Disc Tissue Surgically Removed From Dogs. Proceedings Of 31st Meeting Of The ESVPECVP, London (UK), 4th-7th September 2013, pp. 97
148
7. Muscatello L, Sarli G, Beha G, Asproni P, Millanta F, Poli A, Benazzi C, Brunetti B.
Validation Of Tissue Microarray For Canine And Feline Mammary TUMORS Molecular
Profiling. LXVII S.I.S.Vet. Italian Society of Veterinary Sciences Brescia 17/09/13 19/09/13.
8. Beha Germana, Muscatello Luisa Vera, Giancarlo Avallone, Benazzi Cinzia, Sarli
Giuseppe, Dondi Francesco, Brunetti Barbara. Astenia cutanea in un gatto di 10 mesi:
descrizione di un caso clinico. Veterinaria SCIVAC. Accepted March 2014
9. Del Magno S., Cinti F., Foglia A., Beha G., Zanoni R., Pisoni L. Necrotizing fasciitis in ten
dogs: retrospective evaluation of surgical treatment and long-term outcome. Accepted as a
Poster to the European College of Veterinary Surgeons (ECVS), Annual Scientific Meeting,
Copenhagen, Denmark, 3-5 July 2014.
149
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement