The Value of a New Cancer Biomarker fHER-2

Proto-oncogene in the Diagnosis of Feline Mammary

Carcinoma

Maria Soares

, Jorge Correia

, José Cabeçadas

, Conceição Peleteiro

and

Fernando Ferreira

CIISA, Faculty of Veterinary Medicine, Technical University of Lisbon

1300-477, Lisbon, Portugal

Anatomical Pathology Service, IPOFG-EPE, 1099-023, Lisbon, Portugal

Abstract. The overexpression of the H

uman Epidermal growth factor Receptor-

2 (HER-2) oncogene in human breast cancer is associated with a poor prognosis

and a specific treatment. Because of its importance and as a first line option for

diagnosis, well established guidelines for its detection are based in

immunohistochemical techniques, Still, in Veterinary Medicine there is little

and inconsistent information about this subject. The aim of our study was to

achieve an optimal immunohistochemical protocol for detection of fHER-2 in

eline Mammary Carcinoma (FMC). Five commercial anti-HER-2 antibodies

were tested using three different protocols. The fHER-2 protein overexpression

was detected in 10 of the 30 FMC cases (33.3%), when the optimized protocol

was performed (associating the A0485 antibody with a longer antigen retrieval

method). These results suggest that fHER-2 may play an important role in

Feline Oncology and that the Cat can be a suitable animal model for human

breast cancer research.

1 Introduction

The HER-2/neu proto-oncogene encodes a 185kD transmembrane glycosylated

protein that belongs to the human epidermal growth factor receptor’s family [10]. In

humans, this gene is located on chromosome 17 and its amplification, identified in 20

to 30% of breast cancers, is an important diagnostic and prognostic marker [9]. In

most of the cases, HER-2 gene amplification leads to an increase in protein

expression levels which results in an increase number of HER-2 receptors in the cell

membrane. Because it is clinical relevance, evaluations of HER-2 status by

immunohistochemical (IHC) and by in situ hybridization assays were recently

validated by the A

merican Society of Clinical Oncology (ASCO). Also in last years,

Gentech/Roche companies engineered a humanized monoclonal antibody that inhibits

the receptor’s dimerization providing a longer survival period in breast cancer patients

[12].

In Feline Oncology, the mammary tumors are very common. Indeed, they are the

third most common tumor in clinical practice and represent 17% of the tumors in

female cats. F

eline Mammary Carcinomas (FMC) have display some particularities

that distinguish them from the dog mammary tumors. They are very aggressive (85%

Soares M., Correia J., Cabeçadas J., Peleteiro C. and Ferreira F..

The Value of a New Cancer Biomarker fHER-2 Proto-oncogene in the Diagnosis of Feline Mammary Carcinoma.

DOI: 10.5220/0003879300090017

In Proceedings of the International Workshop on Veterinary Biosignals and Biodevices (VBB-2012), pages 9-17

ISBN: 978-989-8425-94-2

 2012 SCITEPRESS (Science and Technology Publications, Lda.)

are malignant), showing a poor prognosis and a short survival period [15]; [3]; [4].

Recently, some studies had revealed a wide frequency range of mammary tumors

fHER-2+ (5%-90%) using one or three commercial antibodies [1]; [14]; [10]. Also,

the role of fHER-2 in oncogenic mechanisms remains unknown.



In this study we aim to improve the immunodetection of fHER-2 in order to obtain

a better evaluation of HER-2 status on formalin-fixed, paraffin wax-embedded tissue

sections of FMC. For this we used 5 different commercial antibodies (two of them

never used in feline samples) and three different antigen retrieval (AR) methods. In

the end, we intend to contribute for the characterization of the frequency of FMC-

fHER-2 and compare our results with others authors, for a better understanding

Beyond the potential applications in Veterinary Medicine, the study of the status of

this oncogene could have clinical relevance, while cats can be a suitable natural

model for studying human HER-2 positive breast cancer.

Finally, we also point out the future contributions that Engineering Sciences can

bring to improve the fHER-2 immunodetection and the immunotherapy of FMC’s.

2 Material and Methods

2.1 Sample Collection and Histology

The 30 mammary gland samples used in this study were obtained from the

Anatomical Pathology Diagnostic Service archives, Faculty of Veterinary Medicine,

Lisbon, Portugal and complemented with clinical information provided for each case.

These mammary specimens were fixed in formaldehyde and embedded in paraffin

blocks. Only samples of carcinomas fixed for less than 72 hours were considered for

the study (Table 1). For histologic examination, sections of 4μm thickness were

stained with haematoxylin-eosin (HE) and tumors were classified according to the

World Health Organization (WHO) criteria [7].

Table 1. Histologic classification according to the WHO and grading of the samples submitted

to immunohistochemical evaluation.

Histologic classification

Malignant

Grade

Samples (%) Total

Cribiform carcinoma

II 1/30 (3.3%)

11/30 (36.7%)

III 10/30 (33.4%)

Tubulopapillary

carcinoma

II 4/30 (13.3%)

9/30 (30%)

III 5/30 (16.6%)

Tubular

adenocarcinoma

II 2/30 (6.6%)

6/30 (20%)

III 4/30 (13.3%)

Mucinous carcinoma

III 1/30 (3.33%)

4/30 (13.3%)

Simple Carcinoma

I 1/30 (3.33%)

Solid carcinoma

III 1/30 (3.33%)

Squamous cell

carcinoma

III 1/30 (3.33%)

2.2 Immunohistochemical Study

HER-2/neu Antibodies. Five commercial antibodies were tested for fHER-2/neu

immunostaining on paraffin tissue sections: a rabbit polyclonal anti-human HER-2

(A0485 from DAKO, Glostrup, Denmark), two rabbit monoclonal anti-human HER-2

antibodies (4B5 from Ventana, Tucson, Arizona and SP3 from Zytomed, Berlin,

Germany) and two mouse monoclonal anti-human HER-2 antibodies (CB11 from

Zytomed and TAB250 from Invitrogen, Carlsbad, California). Each one of these

antibodies has literature showing that they recognize the HER-2 protein in human

tissues by binding to extracellular domain of the HER-2 receptor (TAB250 and SP3)

or to recognize the intracellular domain (CB11, 4B5 and A0485). We also note that

SP3 and TAB250 antibodies were used for the first time in order to detect fHER-2 in

this work.

Immunohistochemical Technique. Sections were mounted on Starfrost

microscope

slides and dried at 60ºC for one hour. Each slide were deparaffinized and rehydrated

in distilled water through a series of graded alcohols and then submitted to antigen

retrieval with buffer citrate solution (NaCH

COO, pH=6) in a water bath at 95ºC for

30 minutes or for 60 minutes as resumed in Table 1I. In parallel, to improve antigen

recognition of TAB250 antibody, we performed an enzymatic digestion of tissue

samples with Protease K (Zymed) for 10 minutes following manufacturer’s

recommendations. To exhaust endogenous peroxidase activity, a Peroxide-Block

solution (Zytomed) was applied for 10 minutes and each primary antibody was

incubated during 1h at room temperature. After several PBS washes, primary

antibodies were detected with a secondary antibody for 30 minutes (HER2easy kit

IHC from Zytomed) and 3,3’-diaminobenzidin-tetrahydrochlorid (DAB) was used as

the chromogen prior to counterstain with Mayer’s haematoxylin.

Positive and negative controls were obtained from human breast carcinomas

known to overexpress HER-2 receptor and previously classified as 3+ or, classified as

0 without HER-2 expression (see Table 3).

Table 2. Resume of the immunohistochemical protocols used for fHER-2 detection.

Primary antibody

Antigen retrieval

Clone

Dilution Incubation time

CB11

RTU 60’

Buffer citrate solution 95º C for

30’ and 60’

4B5

RTU 60’

A0485

1:250 60’

SP3

1:100 60’

TAB250

1:50 60’

Buffer citrate solution 95º C for

30’ and 60’

Proteinase K for 10’

RTU = Ready to use

Interpretation Criteria. Overexpression of fHER-2 was defined as a membranous

staining in more than 10% of neoplastic cells and staining was examined over the

maximum area of staining intensity according to the DAKO guidelines (Table 3).

Samples classified as 0 or 1+ were considered negative, whereas scores of 2+ or 3+

were considered positive. Cytoplasmic staining was considered nonspecific staining.

All slides were submitted to blind scoring by two independent DVM pathologists and

one DVM clinician. Any discordant interpretation was debated and settled using a

multiviewer microscope.

Table 3. Interpretation Criteria (HercepTest Interpretation Manual from DAKO).

Grade Interpretation

No staining.

Weak, incomplete membranous staining in any proportion of tumor cells.

Complete membrane staining that is either no uniform or weak in intensity but with

obvious circumferential distribution in at least 10% of cells.

Uniform intense membrane staining of at least 10% of invasive tumor cells.

2.3 Statistical Study

The association between fHER-2 overexpression and grade of malignancy or

histological classification were assessed by Fisher’s Exact Test. Values of p < 0.05

were considered to reflect statistical significance.

3 Results

The mean age of the 30 queens at the time of mastectomy was 10.4 years. Cribiform

carcinoma (36.7%) was the most common type of malignant mammary tumor,

followed by the tubulopapillary (30%) and tubular carcinomas (20%). The histologic

grading reveals that almost all of these tumors (73.3%) were poorly differentiated

carcinomas, showing a grade III.

The immunodetection of fHER-2 by some commercial antibodies (CB11, 4B5 and

A0485) was revealed by a cellular membrane labeling in several FMC showing a

species cross reactivity. Positive (3+) and negative (0) controls show the label

intensity expected in all protocols (Figure 1).

Fig. 1. Images of positive and negative controls using a CB11 antibody after 60’ of antigen

retrieval. (A) Human positive control scored 3+ showing an intensive and continuous label of

cellular membrane (x400); (B) Human negative control scored 0, with no staining (x400).

Results obtained from the employment of TAB250 and SP3 antibodies showed no

staining in all samples even after use a longer antigen retrieval protocol and in regards

of the others, the best results were achieve with A0485, as its shown in Figure 2.

Fig. 2. Results of IHC protocols using two antigen retrieval durations (30’ or 60’). Note that

regardless the protocol used, there are no staining for SP3 and TAB250. It’s also clear the

improvement in the results with 60’ of antigen retrieval in all the remaining three antibodies.

(A) Scores of IHC with the lower antigen retrieval (30 minutes); (B) Scores of IHC with the

longer antigen retrieval (60 minutes).

Table 4 summarizes our results, in which we observe a fHER-2 overexpression in

6.7% of the samples using CB11 antibody, 16.7% with 4B5 antibody and 33.3% with

polyclonal antibody A0485 from DAKO, when we use a longer antigen retrieval

method (Figure 3). When compared, all samples that demonstrated overexpression

with CB11 or 4B5 had the same or a better score with A0485.

Table 4. IHC results using the CB11, 4B5 and A0485 as primary antibodies.

IHC Classification 0 1+ 2+ 3+ TOTAL

Antibody / AR method

CB11

AR 30’ 24 (80%) 6 (20%) 0 (0%) 0 (0%) 30 (100%)

AR 60’ 18 (60%) 10 (33.3%) 2 (6.7%) 0 (0%) 30 (100%)

4B5

AR 30’ 23 (76.7%) 6 (20%) 1 (3.3%) 0 (0%) 30 (100%)

AR 60’ 15 (50%) 10 (33.3%) 5 (16.7%) 0 (0%) 30 (100%)

A0485

AR 30’ 16 (53.3%) 5 (16.7%) 8 (26.7%) 1 (3.3%) 30 (100%)

AR 60’ 15 (50%) 5 (16.7%) 6 (20%) 4 (13.3%) 30 (100%)

Fig. 3. Expression of fHER-2 in the same sample, classified as Cribiform Carcinoma (A) Score

classification 1+ using CB11 and 60’AR (x400). (B) Score classification 1+ using 4B5 and

60’AR (x400). (C) Score classification 3+ using A0485 and 60’AR (x400). (D) Score

classification 2+ using A0485 and 30’AR (x400).

The 4B5 antibody was the one that showed more cytoplasmic staining, with

samples with diffuse homogeneous staining in the cytoplasm and sometimes with dot

artifacts. Also a weak to moderate non-specific cytoplasmic labeling was seen in the

dermal adnexal structures (Figure 4).

Fig. 4. (A) Cribiform carcinoma scored 1+ with 4B5 and 30’ AR showing homogeneous no

specific cytoplasmic staining (400x); (B) Strong staining of a dermal adnexal structure, a

sebaceous gland (x400).

B A

When analyzed, fHER-2 overexpression in FMC did not evidenced significant

correlation with histological classification (p-value = 0.28) neither with malignance

grade (p-value = 0.47).

4 Discussion

In the past decades we have seen an increased attention and development of

companion animals health services, which create new needs that led to obvious

improvements and findings in Veterinary Medicine. If for one hand this made our pets

more resistant to diseases and with larger/bigger life-expectancy, on the other hand it

urged the rise of pathologies usually confined to the geriatric population, where the

neoplasms fit. Our animals share the same lifestyle as humans and start to be seen as

potential models to research, especially for Cancer Research.

In Cancer Research, the Molecular Biology has a fundamental role, and research

in this area continues to rise, with new information and discoveries published every

day. Unfortunately, in Veterinary Medicine there is little information available on

molecular alterations and the biological behavior of tumors in our companion

animals. In FMC, it is usual to request the histological and grading classification, but

the proteomic status of some receptors isn’t routinely performed.

In the present study we compared five different antibodies anti-HER-2, and we

have found different results, with the most promising one being the A0485, mostly

when combined with a longer antigen retrieval method than the time suggested by the

manufacturer (preferably 60 minutes AR duration), where we obtained an

overexpression of 33.3%, a result similar to those published for Human, with ranges

between 10 to 40% [5], [16].

For human patients, ASCO guidelines are very specific and with well defined

exclusion criteria. Among these recommendations the fixation is imperative and all

the samples fixed in fixatives other than neutral buffered formalin should be excluded.

Beyond this, the samples must be fixated for longer than 6h and less than 48hours

[16]. As our samples did not fulfilled all these requirements we can not exclude that

this may have influenced the results, especially for the SP3 and TAB250 antibodies,

that recognize the extracellular domain, which could be more affected by the

inadequate fixation and as consequence did not show any staining; nor can we

disregard the influence of fixation for the discrepant results obtained among the three

antibodies recognizing the intracellular domain.

The CB11 and the 4B5 antibodies are monoclonal while the A0485, with best

results, is a polyclonal antibody. If we associate the fact that the first two antibodies

that recognize a human epitope of the HER-2 protein, with a aminoacid sequence

homology of 93% (when compared to cat) and, the evidence that the latest antibody

recognizes several human epitopes of the same protein we can suspect that the wide

range of results (6.6% to 33.3%) that were obtained can be due, besides the fixation

problems, to the not total homology between the HER-2 and fHER-2, which makes

A0485 more suitable to recognize the protein.

We can also easily conclude that the antigen retrieval method is critical in the

immunohistochemical assessment of HER-2 in feline tissues and that when this step is

shorter (30’ instead of the 60’), it may significantly lower the threshold of positivity.

These findings are contrary to those of a recent study, where they found a decreased

positivity with most prolonged antigen retrieval [10].

If we compare our findings with the results present by other authors, the incidence

of fHER-2 overexpression is similar to some of the studies [1]; [7]; [10] but markedly

lower than in other reports [6]; [14]. Besides the interpretation criteria in one of these

studies being different to the one we have used [14] the other have respected the

tissue proceeding guidelines [6], which reinforces the importance of this step, so often

neglected. Indeed, in the majority of Diagnostic Services of Anatomical Pathology, it

is usual to receive samples that do not fulfill the requisites for a correct

immunohistochemical analysis. So, it is important to sensitize the clinicians and the

surgeons for this problem moreover since the histological classification and malignant

grade show to be insufficient to classify the tumors and, because none of them

demonstrated predictive value for determination of fHER-2 status in our studies,

which is concordant with others publications [6], [10].

The possibilities to use and introduce engineering sciences to improve the fHER-2

evaluation and anti-fHER-2 clinical treatments were studied. However, two extra

obstacles would have to be passed to achieve the total optimization of fHER-2

immunodetection in feline mammary tumors, whereas engineering can give an

important contribution. One of them is the automatization absence of the technique

which leads to different results between different laboratories and the other is the

interobserver subjectivity in scoring the HER-2 expression in formalin-fixed,

paraffin-embedded breast cancer tissues, due to a very high cellular heterogeneity and

to an extensive calcification/necrotic tumor areas [13], [2].

In human oncology, the American Society of Clinical Oncology (ASCO)

recommends the use of standardized operating procedures, the validation of

laboratories methodologies and also suggests that two or more expert pathologists

should score independently the same patient’s tissue sample to avoid wrong

classifications. To minimize the variability of the results and enhance the

reproducibility, automated systems were developed recently for human samples,

where all the steps of the technique can be regulated (from the deparaffinization of the

tissues till the mounting of slides). Also very recently, a new Automated Cellular

Imaging System (ACIS®, Clarient ChromaVision Medical Systems) was announced

to standardize the detection, the counting and the classification of tumor cells based

on recognition of cellular bodies with a specified shape, size and color.

In the near future, we think that Engineering Sciences can bring a substantial

contribution by developing/adapting the automatized devices similar to the ones used

in the human tissues processing. Additionally, we see as a very promising tool the

improvement of the adjunctive computer-assisted methodology to feline mammary

carcinomas samples, providing reproducibility in the acquisition and scoring of

immunohistochemical images evaluated by a qualified pathologist, after the

development of new image processing algorithms.

Acknowledgements

The authors thank Dr. Rita Ribeiro for the statistical analysis, Dr. João Matos and Dr.

Sandra Carvalho from the Anatomical Pathology Service. This work was supported

by CIISA and FCT (SFRH/BD/70720/2010).

References

1. De Maria R., Olivero M., Iussich S., Nakaichi M., Murata T. Bartolomeo B. and Di Renzo

M.F.. (2005). Spontaneous feline mammary carcinoma is a model of HER2 overexpressing

poor prognosis human breast cancer. Cancer Research, 65, 907-912.

2. Gancberg D., Järvinen T., Leo A., Rouas G., Cardoso F., Paesmans M., Verhest A., Piccart

M., Isola J. and Larsimont D. (2002). Evaluation of HER-2/NEU protein expression in

breast cancer by immunohistochemistry. Breast Cancer Research and Treatment, 74, 113–

120.

3. Martin de las Mulas, J., Van Niel M., Millán, Y., Blankenstein M.A., Van Mil F. and

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

4. Martin de las Mulas, J., Van Niel M., Millán, Y., Ordás J., Blankenstein M.A., Van Mil F.

and Misdorp W. (2002). Progesterone receptors in normal, dysplastic andtumourous feline

mammary glands. Research in Veterinary Science, 72, 153-161.

5. Ménard S., Tagliabue E., Campiglio M. and Pupa S.M. (2000). Role of HER2 Gene

Overexpression in Breast Carcinoma. Journal of Cellular Physiology, 182, 150–162.

6. Millanta F., Calandrella M., Citi S., Della Santa D. and Poli A. (2005). Overexpression of

HER-2 in feline invasive mammary carcinomas. Veterinary Pathology, 42, 30-34.

7. Misdorp W., Else R.W., Hellmen E. and Lipscomb T.P. (1999). Histologic Classification of

Mammary Tumors of the Dog and the Cat. Armed Force Institute of Pathology and World

Health Organization. Washington, DC.

8. Ordás J., Millán Y., Dios R., Reymundo C. and de Las Mulas J.M. (2007). Proto-oncogene

HER-2 in normal, dysplastic and tumorous feline mammary glands. BMC Cancer, 7(179).

9. Press M., Slamon D., Flom K., Park J., Zhou J. and Bernstein L. (2002). Evaluation of

HER-2/neu Gene Amplification and Overexpression. Journal of Clinical Oncology, 20,

3095-3105.

10. Rasotto R., Caliari D., Castagnaro M., Zanetti R. and Zappulli V. (2011). An

Immunohistochemical Study of HER-2 Expression in FMC. Journal Comparative

Pathology, 144, 170-179.

11. Sáez A., Andreu F.J., Seguí M.A., Baré M.L., Fernández S., Dinarés C. and Rey M. (2006).

HER-2 gene amplification by CISH compared FISH. The Breast, 15, 519-527.

12. Stebbing J., Copson E. and O’Reilly S. (2000). Herceptin (trastuzamab) in advanced breast

cancer. Cancer Treatment Reviews, 26, 287-290.

13. Thomson T., Hayes M., Spinelli J., Hilland E., Sawrenko C., Phillips D., Dupuis B. and

Parker R. (2001). HER-2/neu in Breast Cancer. Modern Pathology, 14 (11), 1079-1086.

14. Winston J., Craft D.M., Scase T.J. and Bergman P.J. (2005). Immunohistochemical

detection of HER-2/neu expression in spontaneous feline mammary tumours. Veterinary

and Comparative Oncology, 3, 8-15.

15. Withrow, S. J. and Vail, D. M. (2007). Withrow & MacEwen's Small Animal Clinical

Oncology. (4

Ed.). Saunders, Philadelphia: Elsevier Inc.

16. Wolff A.C., Hammond E.H., Schwartz J.N., Hagerty K.L., Allred D.C., et al. (2007).

ASCO/CAP guideline recommendations for human epidermal growth factor receptor 2

testing in breast cancer. Journal of Clinical Oncology, 25(1), 118-45.