- Email: [email protected]
Cell turnover in apocrine metaplasia and apocrine adenosis of the breast
Available online at www.sciencedirect.com
Annals of Diagnostic Pathology 14 (2010) 1 – 7
Original Contributions
Cell turnover in apocrine metaplasia and apocrine adenosis of the breast Ghada Elayat, MBChB, MSc, PhDa , Abdel-Ghani A Selim, MBChB, MSc, PhD, FRCPathb,⁎, Clive A Wells, MA MBBChir, FRCPatha a
Department of Histopathology, St Bartholomew's Hospital, West Smithfield, EC1A 7BE London, UK b Department of Histopathology, King's College Hospital, Denmark Hill, SE5 9RS London, UK
Abstract
Keywords:
Apocrine metaplasia (APM) is a common finding in the breast of postmenopausal women and is seen in a broad spectrum of lesions ranging from microscopic cysts to invasive apocrine carcinoma. Apocrine metaplasia within sclerosing adenosis is known as apocrine adenosis (AA) and is considered a benign lesion of the breast. Apocrine metaplasia and AA have been the subject of many studies; however, little is known about the dynamics of cell turnover in these lesions. Recent studies have shown that some forms of AA may show altered degree of proliferation along with altered expression of bcl-2 and bax proteins. In the current study, we investigate further aspects of apoptosis to help understand the mechanisms of cell turnover in AA and APM. To investigate cell turnover in APM and AA, immunohistochemistry was used to study the expression of the apoptotic markers Bak, Mcl-1, Bcl-x, and Bcl-xL in 45 cases of APM (13 cases of nonpapillary APM, 21 cases of simple papillary APM, and 11 cases of complex papillary APM). Also, 34 cases of AA (23 cases of non–atypical AA [NAA] and 11 cases of atypical AA [AAA]) were included in the study. The expression of hTERT and the proliferation marker Ki-67 were also determined. The TdT-mediated dUTP nick-end labeling (TUNEL) technique was used to study the apoptotic status in 28 cases of APM (12 cases nonpapillary APM and 16 cases of papillary APM including simple and complex forms) and 22 cases of AA (15 cases of NAA and 7 cases of AAA). The results showed that all cases studied by immunohistochemistry were positive for the expression of Bak, Mcl-1, Bcl-x, and Bcl-xL showing a pattern of staining similar to that seen in the normal breast epithelium. There was no relation between hTERT positivity and the degree of proliferation in any of the lesions studied. The TUNEL results revealed an apoptotic index (AI) of 0.4% and 0.2% in the papillary and nonpapillary groups of APM, respectively. There was no statistical significance between the AI of these 2 groups and that of the normal breast epithelium (0.3%). The average Ki-67 index in the nonpapillary group was 0.7%, whereas in the papillary group, a value of 4% was recorded. In the cases of AA, an AI of 0.4% and 0.3% in NAA and AAA, respectively, was seen. There was no statistical significance between the AI of these 2 groups and that of the normal breast epithelium (0.3%). The Ki-67 index was 5.2% and 6.6% in the NAA and AAA, respectively. The current results show that apoptosis is not a common event in APM and AA even in the presence of increased proliferation, which may render some of these lesions more susceptible to oncogenic changes. Further studies are needed to study other apoptotic pathways that may be involved in cell turnover in these lesions. © 2010 Elsevier Inc. All rights reserved. TUNEL; Breast; Apocrine; Apoptosis; Proliferation; hTERT
1. Introduction
⁎ Corresponding author. Tel.: +44 0 20 32993605; fax: +44 0 20 32993670. E-mail address: [email protected] (A.-G.A. Selim). 1092-9134/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.anndiagpath.2009.05.001
The Bcl-2 family is a regulator of apoptosis, and both antiapoptotic cell survival proteins and proapoptotic cell death proteins were discovered within the family. Some of the more extensively studied members include the antiapoptotic proteins Bcl-2, Bcl-w, Mcl-1, and Bcl-xlong
2
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7
(Bcl-xL), and the proapoptotic members Bax, Bak, Bad, Bid, and Bcl-xShort (Bcl-xs). The epithelium of the nonlactating human mammary gland undergoes a continual cell turnover, which includes apoptotic cell death and cell proliferation [1,2]. Feuerhake et al [3] studied the expression of Bcl-2 and Bax proteins in the normal non-lactating human mammary gland. Immunoreactivity for Bcl-2 and Bax was present throughout all epithelia, suggesting a balance between pro- and anti apoptotic effects. Krajewski et al [4] studied the expression of Mcl-1 in variable human tissues. The results showed that Mcl-1 was expressed in the mammary epithelium and the intensity of staining ranged from mild to moderate. In one study by Feuerhake et al [5], the authors reported that the frequency of apoptotic cells as well as the low rate of proliferative activity in APM resembled those of normal mammary epithelium. Other studies investigating the role of the Bcl-2 family proteins in APM have shown that apocrine cells are characteristically Bcl-2 negative [6,7]. On the other hand, Bax protein immunostaining was reported in APM [5]. Selim et al [8] studied the expression of Bcl-2 and Bax in lesions of APM and apocrine adenosis and reported that all cases of apocrine adenosis (AA) and APM were negative for the antiapoptotic protein Bcl-2, but all cases of APM and 33.3% of AA cases showed cytoplasmic positivity for Bax. Human telomerase synthesizes new (TTAGGG) repeats onto chromosome ends. As there is no template DNA available on the complementary strand to encode the telomeric sequence, telomerase circumvents this problem by its association with an RNA molecule that acts as a template. Telomerase activity is considered among the factors controlling normal cell turnover. Low to moderate telomerase activity has been reported in normal breast epithelium in a few studies [9,10]. Normal and neoplastic breast tissues were among the first tissues studied using in situ hybirdisation (ISH) for hTERT [9], which was present in some normal breast lobules (irrespective of menopausal status) and showed increased expression in in situ and invasive carcinomas. Yashima et al [10] reported the presence of increased hTR expression in some foci of APM and atypical hyperplasia.
Fig. 1. Mcl-1 cytoplasmic positivity in APM of the breast (immunoperoxidase).
Increased expression was also observed in all carcinoma in situ (CIS) and invasive lesions. In this study, cases of APM were divided into nonpapillary APM (NAPM) and papillary APM (PAPM) based on the presence or absence of papillae. Papillary metaplasia was further subdivided according to the criteria of Page et al [11] into simple papillary (SPAPM), complex papillary, and highly complex papillary. For statistical analysis, the last 2 groups were merged together (CPAPM). We identified cases of AA based on the definition adapted by Wells et al [12]. The cases were divided into AA without atypia (NAA) and atypical AA (AAA). This classification was based on the criteria suggested by Seidman et al [13] for identifying atypia in AA. 2. Materials and methods 2.1. Cases We collected 45 cases of APM. The cases were divided into 13 cases of NAPM, 21 cases of SPAPM and 11 cases of CPAPM (3 cases showed highly complex papillary features). Also, included were 34 cases of AA (23 cases of NAA and 11 cases of AAA). The cases were selected from the files of the Histopathology Department, St Bartholomew's Hospital,
Table 1 Details of antibodies used for immunohistochemistry Antibody against Mono/ polycolonal
Source
Clone/ code
Dilution Antigen retrieval Incubation time Positive control
Ki-67 Mcl-1 Bak Bcl-xl Bcl-x hTERT
Dako Ltd, Glostrup, Denmark Dako Ltd Dako Ltd Zymed, Inc, San Francisco, CA, USA Dako Ltd Novocastra Ltd, Newcastle, UK
MIB-1 A3534 A3538 2h12 A35-10 44F12
1:50 1:500 1:150 1:50 1:100 1:50
Mono Poly Poly Mono Mono Mono
Pc3 Pc3 Pc3 Pc3 Pc3 Pc3
40 min Overnight 40 min 40 min 40 min 40 min
Tonsil Tonsil Tonsil Hodgkin lymphoma Brain Tonsil
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7
3
Fig. 4. Bcl-X cytoplasmic positivity in AA of the breast (immunoperoxidase). Fig. 2. Bak cytoplasmic positivity in AA of the breast (immunoperoxidase).
London, UK. Some cases identified as suitable for the study were referral cases from other hospitals. 2.2. Immunohistochemistry Formalin-fixed paraffin-embedded blocks of APM and AA were selected from the files and sectioned at a nominal 4 μm. The standard avidin-biotin complex method was used [14]. Heat-mediated antigen retrieval using the pressure cooker was applied [15]. Appropriate positive and negative controls omitting the primary antibodies were included with each slide run. Primary antibodies against Bak, Mcl-1, Bcl-x, Bcl-xL, hTERT, and Ki-67 proteins were used and summarized in Table 1. The staining pattern was assessed as follows. 2.2.1. Mcl-1, Bak, Bcl-xl, and Bcl-x Distinct cytoplasmic staining in more than 25% of cells was taken as positive [8]. Distinct cytoplasmic staining could
Fig. 3. Bcl-XLcytoplasmic positivity in APM of the breast (immunoperoxidase).
be seen in normal breast epithelial cells in more than 25% of cells. Occasional stromal cells and inflammatory cells were seen to express these proteins. 2.2.2. hTERT Nuclear staining was taken as positive. Normal breast epithelium generally showed more than 50% positive cells. The nuclear staining is generally diffuse and nucleoli can be seen stained as well. Cytoplasmic staining was noticed occasionally. Positivity can also be seen in occasional stromal cells, blood vessels, and inflammatory cells [16]. 2.2.3. Ki-67 Nuclear staining was taken as positive and assessed semiquantitatively by counting the number of positive nuclei in 100 to 400 cells according to the size of the studied lesion. This was expressed as a percentage; the proliferative index.
Fig. 5. hTERT punctate nuclear staining of apocrine epithelium (immunoperoxidase).
4
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7
Table 2 Immunohistochemistry results Lesion
NAPM SPAPM CPAPM NAA AAAc a b c
No
13 21 11 23 11
MCL-1
Bcl-xLa
Bak
hTERTb
Bcl-x
+
−
+
−
+
−
+
−
+
−
13 21 11 23 11
0 0 0 0 0
13 21 11 23 9
0 0 0 0 0
13 21 11 15 11
0 0 0 0 0
13 21 11 23 11
0 0 0 0 0
13 21 11 23 9
0 0 0 0 0
Fifteen NAA cases only studied for BCL-xL. Puncatate staining pattern detected. Only 5, 9, and 9 AAA cases were studied for p16, Bak, and BCL-x, respectively.
2.3. TdT-mediated dUTP nick-end labeling Twenty-eight cases of APM were selected. The cases were divided as follows: 12 cases NAPM and 16 cases of papillary APM including simple and complex forms (2 cases with highly complex features), and 22 cases of AA (15 cases of NAA and 7 cases of AAA). Formalin-fixed paraffin-embedded blocks of APM and AA were sectioned at a nominal 4 μm. TdT-mediated dUTP nick-end labeling (TUNEL) was used to look for apoptotic cells in paraffin tissue sections based on the ability of the technique to detect nuclear DNA fragmentation characteristic of apoptotic cells, hence detecting apoptosis in situ at the single-cell level. The procedure was done using the DeadEnd Colorimetric Apoptosis Detection System (cat. no. G7130, Promega, Southampton, UK). Manufacturer's guidelines were followed throughout the procedure. To prepare negative control sections, a control incubation buffer without TdT enzyme was prepared. Human breast carcinoma known to have high level of apoptosis was used as positive control. The percentage of apoptotic cells (apoptotic index [AI]) was determined by counting 200 to 400 cells depending on the amount of metaplastic tissue in every case. Apoptotic cells were counted in normal adjacent breast epithelium, if present, and used for
Fig. 6. Apoptosis detection using TUNEL technique in SPAPM of the breast.
comparison. The Kruskal-Wallis test was done to assess significance of the AI values in the groups studied. 3. Results 3.1. Immunohistochemistry 3.1.1. Mcl-1, Bak, Bcl-xl, and Bcl-x All cases studied showed distinct cytoplasmic staining in more than 25% of the cells and were considered as positive (Figs. 1-4). 3.1.2. hTERT Nuclear staining in more than 50% of the cells was considered positive. A characteristic punctate pattern (Fig. 5) was particularly seen in association with apocrine cells rather than with normal breast epithelium, which had the tendency to show a more diffuse pattern of nuclear staining. The results are summarized in Table 2. 3.2. TdT-mediated dUTP nick-end labeling 3.2.1. Papillary APM A total of 16 cases were analyzed. The AI (Fig. 6) ranged from 0% to 0.75% (average, 0.4%; SD, ±0.3). The Ki-67
Fig. 7. Ki-67 nuclear positivity in complex papillary APM of the breast (immunoperoxidase).
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7
5
3.2.6. Summary of the data obtained The TUNEL reaction produced positive staining in single nuclei of apocrine cells, which often showed characteristics of apoptotic cells such as condensed chromatin at the margin and a perinuclear halo. The mean AI was (0.3%) in all cases studied. The apoptotic status was neither related to morphology nor to the degree of proliferation in the different groups studied. Normal breast epithelium also showed a mean AI of 0.3% and a mean proliferation index of 0.9%, which reflects its natural slow cell turnover. There was no statistical significance between the AI of the different groups studied (P = .7) as determined by the Kruskal-Wallis test. The results of TUNEL and Ki67 are summarized in Table 3. Fig. 8. Apoptosis detection using TUNEL technique atypical AA of the breast.
(Fig. 7) index in these cases ranged from 0% to 8% (average, 4%; SD, ±2.2). 3.2.2. Nonpapillary APM A total of 12 cases were analyzed. The AI ranged from 0% to 0.5% (average. 0.2%; SD, ±0.2). The Ki-67 index in these cases ranged from 0% to 1.5% (average, 0.7%; SD, ±0.6). 3.2.3. Non–atypical AA A total of 15 cases were analyzed. The AI ranged from 0% to 0.75% (average, 0.4%; SD, ± 0.2). The Ki-67 index in these cases ranged from 2.5% to 8% (average, 5.2%; SD, ±1.7). 3.2.4. Atypical AA A total of 7 cases were analyzed. The AI (Fig. 8) ranged from 0% to 0.5% (average, 0.3%; SD, ±0.2). The Ki-67 index in these cases ranged from 4.5% to 9% (average, 6.6%; SD, ±1.7). 3.2.5. Normal breast epithelium Normal breast tissues were available in 39 of the studied cases. The AI ranged from 0% to 0.75% (average, 0.3%; SD, ±0.3). The Ki-67 index in these cases ranged from 0% to 4% (average, 0.9%; SD, ±0.99).
4. Discussion Normal cell turnover depends on the balance between apoptosis and the rate of proliferation. A few studies have investigated the status of apoptosis in APM. The results of these studies, collectively, reveal that apocrine epithelium is Bcl-2 negative and Bax protein positive [5-8]. Other studies have investigated the rate of proliferation in APM. In one study, the rate of proliferation in APM was shown to be less than 1% in another study [5]; it was shown to be 1.3%, ranging from 0% to 8.5% [8]. None of these 2 studies, however, define the histologic features of the cases of APM studied. In the current study, we investigate the rate of cell turnover in APM with and without papillary features. The study by Page et al [11], dividing APM according to the degree of papillary complexity, showed that the more complex forms of APM had a tendency to be associated with increased risk of subsequent carcinoma development. Other studies investigating APM have shown different molecular abnormalities in APM including c-myc and c-erbB2 overexpression [8]. LOH and CGH studies have shown that some of the genetic abnormalities inherent to breast carcinoma can be found in some cases of APM [17-19]. These findings emphasize the need for identifying forms of APM that may be associated with increased risk and whether papillary features might be a useful feature for this purpose.
Table 3 Summary of TUNEL and Ki-67 results Statistical analysis
NAPM
Papillary APM
Range (%) AI Average/SD Range (%) Ki-67 index Average/SD Total no. of cases
0-0.5
0-0.8
0.2%, SD ±0.2 0-1.5
0.4%, SD ±0.3 1.5-8.5
0.7%, SD ±0.6 5
4.1%, SD ±2.1 16
NAA 0-0.8
AAA
Normal breast epithelium
0-0.5
0-0.8
0.4%, SD ±0.2 2.5-8
0.3%, SD ±0.2 4.5-9
0.3%, SD ±0.3 0-4
5.2%, SD ±1.7 15
6.6%, SD ±1.7 7
0.9%, SD ±0.99 39
6
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7
The current results show that APM generally expresses the apoptotic markers Mcl-1, Bak, Bcl-xl, and Bcl-x showing a similar pattern of staining to normal breast epithelium. The presence of Mcl-1 expression in APM in absence of Bcl-2 has been reported to occur in some tissues before [4,20] where it can be suggested that most tissues normally lacking Bcl-2 expression would show positivity for Mcl-1. Cytoplasmic staining of these apoptotic proteins is, however, difficult to interpret because it is the balance between proand anti apoptotic forces that counts rather than the mere absence or presence of individual proteins. Hence, the TUNEL technique was used to look for the actual presence of apoptotic cells in APM. The results showed that apoptosis is generally a rare event in APM with an AI similar to that of normal breast epithelium. The presence of apoptotic cells showed no relation to the type of APM studied whether papillary or nonpapillary. These findings are similar to those of Feuerhake et al [5], except that in their study, the histologic features of APM were not known. However, in the current study, a borderline significance of Ki-67 index was seen between the groups studied. We believe that papillary features were associated with increased proliferation, although this did not reach statistical significance (average Ki-67 index 4%). This might be due to the small number of cases included in this study. The current results suggest that apoptosis is generally a rare event in APM; however, it has to be remembered that apoptotic pathways are highly complex and hence more studies need to be done to investigate other pathways that might be involved in apocrine cell control. Although a few studies have addressed the risk of subsequent carcinoma development associated with AA [8,12,13,21-24], only 2 studies have actually investigated the expression of the proliferation marker Ki-67 in these lesions [8,12] and showed that a subset of AA had a significant increase in proliferative activity (mean Ki-67 index of 5.0% and 3.6%, respectively). Other studies have also looked at proliferation in these lesions. Unfortunately, the results were included within groups encompassing variable histologic entities ranging from APM, papillomas, adenomas, to adenomyoepithelioma hindering the identification of true biologic criteria for NAA or AAA [25-27]. A total of 22 cases (15 cases NAA and 7 cases AAA) were studied using the TUNEL technique. The Ki-67 index for the NAA group was 5.2% and that for the AAA group was 6.6%. A low AI of 0.45% was found in the NAA group, whereas a value of 0.3% was recorded in the AAA group. There was no association with proliferation, and apoptosis appeared to be a rare event in these lesions. This finding is contrary to the often expressed view that AA is a degenerative lesion [28]. Collectively, the above results of TUNEL and the apoptotic marker study denote that AA has a slow death rate similar to that of normal breast epithelium. Yet, a subset of AA may show increased proliferative capacity; and in the presence of slow rate of apoptosis, this may render these lesions more susceptible to further oncogenic changes. The
current results support the suggestion of [8,12,21,23] that at a subset of AA may have carcinogenic potential. Finally, hTERT showed a characteristic punctate pattern of staining, which was associated with cells with a higher proliferative capacity in a study by Cressey et al (personal correspondence). There was no relation, however, between this pattern of staining and the degree of proliferation in the current study, although Ikeda et al [29], using an in-house antibody, stated that hTERT expression correlated with increased proliferation as indicated by Ki-67. In the current study, hTERT staining was not of any significant help in identifying cells with higher proliferation because it was expressed in normal epithelium, APM of all forms, and some ductal carcinoma in situ (DCIS) and carcinomas used for comparison. Further studies need to be done to assess the significance of the punctate pattern of staining.
Acknowledgments The authors acknowledge the financial support from the Joint Research Board, St Bartholomew's Hospital, London, UK; L Ghaly for technical support; and J Thomas for statistical assistance.
References [1] Ferguson D, Anderson T. Ultrastructural observations on cell death by apoptosis in the resting human breast. Virchows Arch A Pathol Anat Histpathol 1981;393:193-203. [2] Ferguson D. An ultrastructural study of mitosis and cytokinesis in normal resting human breast. Cell Tissue Res 1988;252:581-7. [3] Feuerhake F, Sigg W, Hofter E, et al. Immunohistochemical analysis of bcl-2 and bax expression in relation to cell turnover and epithelial differentiation markers in the non-lactating human mammary gland epithelium. Cell Tissue Res 2000;299:47-58. [4] Krajewski S, Bodrug S, Krajweska M, et al. Immunohistochemical analysis of Mcl-1 protein in human tissues. Differential regulation of Mcl-1 and bcl-2 protein production suggests a unique role for Mcl-1 in control of programmed cell death in-vivo. Am J Pathol 1995;146: 1309-19. [5] Feuerhake F, Unterberger P, Hofter EA. Cell turnover in apocrine metaplasia of the human mammary gland epithelium: apoptosis, proliferation and immunohistochemical detection of bcl-2, bax, EGFR and c-erbB-2 gene products. Acta Histochem 2001;103:53-65. [6] Tavassoli F, Purcell C, Bratthauer G, et al. Androgen receptor expression along with loss of Bcl-2, ER, and PR expression in benign and malignant apocrine lesions of the breast: Implications of therapy. Breast J 1996;2:261-9. [7] Leal C, Henrique R, Monteiro P, et al. Apocrine ductal carcinoma in situ of the breast: histologic classification and expression of biologic markers. Hum Pathol 2001;32:487-93. [8] Selim A, El-Ayat G, Wells C. Expression of c-erbB-2, p53, Bcl-2, Bax, c-myc and ki-67 in apocrine metaplasia and apocrine change within sclerosing adenosis of the breast. Virchows Arch 2002;441:449-55. [9] Kolquist K, Ellisen L, Counter C, Meyerson MM, et al. Expression of TERT in early premalignant lesions and a subset of cells in normal tissues. Nature genetic 1998;19:182-6. [10] Yashima K, Milchgrub S, Gollahon L, et al. Telomerase enzyme activity and RNA expression during the multistage pathogenesis of breast carcinoma. Clin Cancer Res 1998;4:229-34.
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7 [11] Page D, Dupont W, Jensen R. Papillary apocrine change of the breast: associations with atypical hyperplasia and risk of breast cancer. Cancer Epidemiol, Biomarkers Preven 1996;5:29-32. [12] Wells C, McGregor I, Makunura C, et al. Apocrine adenosis: a precursor of aggressive breast cancer? J Clin Pathol 1995;48:737-42. [13] Seidman J, Ashton M, Lefkowitz M. Atypical apocrine adenosis of the breast. A clinicopathologic study of 37 patients with 8.7 year follow up. Cancer 1996;77:2529-37. [14] Hsu SM, Raine L, Fanger H. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabelled antibody (PAP) procedures. J Histochem Cytochem 1981;29:577-80. [15] Norton A, Jordan S, Yeomans P. Brief, high-temperature heat denaturation (pressure cooking): a simple and effective method of antigen retrieval for routinely processed tissues. J Pathol 1994;173: 371-9. [16] Yan P, Benhattar J, Seelentag W, et al. Immunohistochemical localization of hTERT protein in human tissues. Histochem Cell Biol 2004;121:391-7. [17] Washington C, Dalbegue F, Abreo F, et al. Loss of heterozigosity in fibrocystic change of the breast. Genetic relationship between benign proliferative lesions and associated carcinomas. Am J Pathol 2000;157: 323-9. [18] Jones C, Damiani S, Wells D, et al. Molecular cytogenetic comparison of apocrine hyperplasia and apocrine carcinoma of the breast. Am J Pathol 2001;158:207-14. [19] Selim AA, Ryan A, El-Ayat G, et al. Loss of heterozygosity and allelic imbalance in apocrine metaplasia of the breast: microdissection microsatellite analysis. J Pathol 2002;196:287-91.
7
[20] Shigemasa K, Katoh O, Shiroyama Y, et al. Increased Mcl-1 expression is associated with poor prognosis in ovarian carcinomas. Jpn J cancer Res 2002;93:542-50. [21] Selim A, El-Ayat G, Wells C. C-erbB2 oncoprotein expression, gene amplification and chromosome 17 aneusomy in apocrine adenosis of the breast. J Pathol 2000;191:138-42. [22] Endoh Y, Tamura G, Kato N, et al. Apocrine adenosis of the breast: clonal evidence of neoplasia. Histopathology 2001;38:221-4. [23] Selim A, Ryan A, El-Ayat G, et al. Loss of heterozygosity and allelic imbalance in apocrine adenosis of the breast. Cancer Detect prev 2001; 25:262-7. [24] Schmitt F, Reis-Filho J. Oncogenes, granules and breast cancer: what has c-myc to do with apocrine changes? Breast 2002;11:463-5. [25] Ioffe O, Iskander M, Rucker C, et al. Benign, malignant and pseudomalignant apocrine lesions of the breast: another attempt at differential diagnosis. Lab Invest 1999;79:111 (abstr). [26] Ioffe O, Iskander M, Rucker C, et al. A comparative study of c-erbB-2, p27, bcl-2 and Ki-67 expression in apocrine proliferations and benign breast tissue. Lab Invest 1999;79:112 (abst). [27] Moriya T, Sakamoto K, Sasano H, Kawanaka M, et al. Immunohistochemical analysis of Ki-67, p53, p21 and p27 in benign and malignant apocrine lesions of the breast: its correlation to histologic findings in 43 cases. Mod Pathol 2000;13:13-8. [28] Simpson J, Page D, Dupont W. Apocrine adenosis—a mimic of mammary carcinoma. Am J Surg Pathol 1990;3:289-99. [29] Ikeda S, Shibata T, Eishi Y, et al. Correlation between the expression of telomerase reverse transcriptase and proliferative activity in breast cancer cells using an immunohistochemical restaining method. Pathol Int 2003;53:762-8.
Annals of Diagnostic Pathology 14 (2010) 1 – 7
Original Contributions
Cell turnover in apocrine metaplasia and apocrine adenosis of the breast Ghada Elayat, MBChB, MSc, PhDa , Abdel-Ghani A Selim, MBChB, MSc, PhD, FRCPathb,⁎, Clive A Wells, MA MBBChir, FRCPatha a
Department of Histopathology, St Bartholomew's Hospital, West Smithfield, EC1A 7BE London, UK b Department of Histopathology, King's College Hospital, Denmark Hill, SE5 9RS London, UK
Abstract
Keywords:
Apocrine metaplasia (APM) is a common finding in the breast of postmenopausal women and is seen in a broad spectrum of lesions ranging from microscopic cysts to invasive apocrine carcinoma. Apocrine metaplasia within sclerosing adenosis is known as apocrine adenosis (AA) and is considered a benign lesion of the breast. Apocrine metaplasia and AA have been the subject of many studies; however, little is known about the dynamics of cell turnover in these lesions. Recent studies have shown that some forms of AA may show altered degree of proliferation along with altered expression of bcl-2 and bax proteins. In the current study, we investigate further aspects of apoptosis to help understand the mechanisms of cell turnover in AA and APM. To investigate cell turnover in APM and AA, immunohistochemistry was used to study the expression of the apoptotic markers Bak, Mcl-1, Bcl-x, and Bcl-xL in 45 cases of APM (13 cases of nonpapillary APM, 21 cases of simple papillary APM, and 11 cases of complex papillary APM). Also, 34 cases of AA (23 cases of non–atypical AA [NAA] and 11 cases of atypical AA [AAA]) were included in the study. The expression of hTERT and the proliferation marker Ki-67 were also determined. The TdT-mediated dUTP nick-end labeling (TUNEL) technique was used to study the apoptotic status in 28 cases of APM (12 cases nonpapillary APM and 16 cases of papillary APM including simple and complex forms) and 22 cases of AA (15 cases of NAA and 7 cases of AAA). The results showed that all cases studied by immunohistochemistry were positive for the expression of Bak, Mcl-1, Bcl-x, and Bcl-xL showing a pattern of staining similar to that seen in the normal breast epithelium. There was no relation between hTERT positivity and the degree of proliferation in any of the lesions studied. The TUNEL results revealed an apoptotic index (AI) of 0.4% and 0.2% in the papillary and nonpapillary groups of APM, respectively. There was no statistical significance between the AI of these 2 groups and that of the normal breast epithelium (0.3%). The average Ki-67 index in the nonpapillary group was 0.7%, whereas in the papillary group, a value of 4% was recorded. In the cases of AA, an AI of 0.4% and 0.3% in NAA and AAA, respectively, was seen. There was no statistical significance between the AI of these 2 groups and that of the normal breast epithelium (0.3%). The Ki-67 index was 5.2% and 6.6% in the NAA and AAA, respectively. The current results show that apoptosis is not a common event in APM and AA even in the presence of increased proliferation, which may render some of these lesions more susceptible to oncogenic changes. Further studies are needed to study other apoptotic pathways that may be involved in cell turnover in these lesions. © 2010 Elsevier Inc. All rights reserved. TUNEL; Breast; Apocrine; Apoptosis; Proliferation; hTERT
1. Introduction
⁎ Corresponding author. Tel.: +44 0 20 32993605; fax: +44 0 20 32993670. E-mail address: [email protected] (A.-G.A. Selim). 1092-9134/$ – see front matter © 2010 Elsevier Inc. All rights reserved. doi:10.1016/j.anndiagpath.2009.05.001
The Bcl-2 family is a regulator of apoptosis, and both antiapoptotic cell survival proteins and proapoptotic cell death proteins were discovered within the family. Some of the more extensively studied members include the antiapoptotic proteins Bcl-2, Bcl-w, Mcl-1, and Bcl-xlong
2
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7
(Bcl-xL), and the proapoptotic members Bax, Bak, Bad, Bid, and Bcl-xShort (Bcl-xs). The epithelium of the nonlactating human mammary gland undergoes a continual cell turnover, which includes apoptotic cell death and cell proliferation [1,2]. Feuerhake et al [3] studied the expression of Bcl-2 and Bax proteins in the normal non-lactating human mammary gland. Immunoreactivity for Bcl-2 and Bax was present throughout all epithelia, suggesting a balance between pro- and anti apoptotic effects. Krajewski et al [4] studied the expression of Mcl-1 in variable human tissues. The results showed that Mcl-1 was expressed in the mammary epithelium and the intensity of staining ranged from mild to moderate. In one study by Feuerhake et al [5], the authors reported that the frequency of apoptotic cells as well as the low rate of proliferative activity in APM resembled those of normal mammary epithelium. Other studies investigating the role of the Bcl-2 family proteins in APM have shown that apocrine cells are characteristically Bcl-2 negative [6,7]. On the other hand, Bax protein immunostaining was reported in APM [5]. Selim et al [8] studied the expression of Bcl-2 and Bax in lesions of APM and apocrine adenosis and reported that all cases of apocrine adenosis (AA) and APM were negative for the antiapoptotic protein Bcl-2, but all cases of APM and 33.3% of AA cases showed cytoplasmic positivity for Bax. Human telomerase synthesizes new (TTAGGG) repeats onto chromosome ends. As there is no template DNA available on the complementary strand to encode the telomeric sequence, telomerase circumvents this problem by its association with an RNA molecule that acts as a template. Telomerase activity is considered among the factors controlling normal cell turnover. Low to moderate telomerase activity has been reported in normal breast epithelium in a few studies [9,10]. Normal and neoplastic breast tissues were among the first tissues studied using in situ hybirdisation (ISH) for hTERT [9], which was present in some normal breast lobules (irrespective of menopausal status) and showed increased expression in in situ and invasive carcinomas. Yashima et al [10] reported the presence of increased hTR expression in some foci of APM and atypical hyperplasia.
Fig. 1. Mcl-1 cytoplasmic positivity in APM of the breast (immunoperoxidase).
Increased expression was also observed in all carcinoma in situ (CIS) and invasive lesions. In this study, cases of APM were divided into nonpapillary APM (NAPM) and papillary APM (PAPM) based on the presence or absence of papillae. Papillary metaplasia was further subdivided according to the criteria of Page et al [11] into simple papillary (SPAPM), complex papillary, and highly complex papillary. For statistical analysis, the last 2 groups were merged together (CPAPM). We identified cases of AA based on the definition adapted by Wells et al [12]. The cases were divided into AA without atypia (NAA) and atypical AA (AAA). This classification was based on the criteria suggested by Seidman et al [13] for identifying atypia in AA. 2. Materials and methods 2.1. Cases We collected 45 cases of APM. The cases were divided into 13 cases of NAPM, 21 cases of SPAPM and 11 cases of CPAPM (3 cases showed highly complex papillary features). Also, included were 34 cases of AA (23 cases of NAA and 11 cases of AAA). The cases were selected from the files of the Histopathology Department, St Bartholomew's Hospital,
Table 1 Details of antibodies used for immunohistochemistry Antibody against Mono/ polycolonal
Source
Clone/ code
Dilution Antigen retrieval Incubation time Positive control
Ki-67 Mcl-1 Bak Bcl-xl Bcl-x hTERT
Dako Ltd, Glostrup, Denmark Dako Ltd Dako Ltd Zymed, Inc, San Francisco, CA, USA Dako Ltd Novocastra Ltd, Newcastle, UK
MIB-1 A3534 A3538 2h12 A35-10 44F12
1:50 1:500 1:150 1:50 1:100 1:50
Mono Poly Poly Mono Mono Mono
Pc3 Pc3 Pc3 Pc3 Pc3 Pc3
40 min Overnight 40 min 40 min 40 min 40 min
Tonsil Tonsil Tonsil Hodgkin lymphoma Brain Tonsil
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7
3
Fig. 4. Bcl-X cytoplasmic positivity in AA of the breast (immunoperoxidase). Fig. 2. Bak cytoplasmic positivity in AA of the breast (immunoperoxidase).
London, UK. Some cases identified as suitable for the study were referral cases from other hospitals. 2.2. Immunohistochemistry Formalin-fixed paraffin-embedded blocks of APM and AA were selected from the files and sectioned at a nominal 4 μm. The standard avidin-biotin complex method was used [14]. Heat-mediated antigen retrieval using the pressure cooker was applied [15]. Appropriate positive and negative controls omitting the primary antibodies were included with each slide run. Primary antibodies against Bak, Mcl-1, Bcl-x, Bcl-xL, hTERT, and Ki-67 proteins were used and summarized in Table 1. The staining pattern was assessed as follows. 2.2.1. Mcl-1, Bak, Bcl-xl, and Bcl-x Distinct cytoplasmic staining in more than 25% of cells was taken as positive [8]. Distinct cytoplasmic staining could
Fig. 3. Bcl-XLcytoplasmic positivity in APM of the breast (immunoperoxidase).
be seen in normal breast epithelial cells in more than 25% of cells. Occasional stromal cells and inflammatory cells were seen to express these proteins. 2.2.2. hTERT Nuclear staining was taken as positive. Normal breast epithelium generally showed more than 50% positive cells. The nuclear staining is generally diffuse and nucleoli can be seen stained as well. Cytoplasmic staining was noticed occasionally. Positivity can also be seen in occasional stromal cells, blood vessels, and inflammatory cells [16]. 2.2.3. Ki-67 Nuclear staining was taken as positive and assessed semiquantitatively by counting the number of positive nuclei in 100 to 400 cells according to the size of the studied lesion. This was expressed as a percentage; the proliferative index.
Fig. 5. hTERT punctate nuclear staining of apocrine epithelium (immunoperoxidase).
4
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7
Table 2 Immunohistochemistry results Lesion
NAPM SPAPM CPAPM NAA AAAc a b c
No
13 21 11 23 11
MCL-1
Bcl-xLa
Bak
hTERTb
Bcl-x
+
−
+
−
+
−
+
−
+
−
13 21 11 23 11
0 0 0 0 0
13 21 11 23 9
0 0 0 0 0
13 21 11 15 11
0 0 0 0 0
13 21 11 23 11
0 0 0 0 0
13 21 11 23 9
0 0 0 0 0
Fifteen NAA cases only studied for BCL-xL. Puncatate staining pattern detected. Only 5, 9, and 9 AAA cases were studied for p16, Bak, and BCL-x, respectively.
2.3. TdT-mediated dUTP nick-end labeling Twenty-eight cases of APM were selected. The cases were divided as follows: 12 cases NAPM and 16 cases of papillary APM including simple and complex forms (2 cases with highly complex features), and 22 cases of AA (15 cases of NAA and 7 cases of AAA). Formalin-fixed paraffin-embedded blocks of APM and AA were sectioned at a nominal 4 μm. TdT-mediated dUTP nick-end labeling (TUNEL) was used to look for apoptotic cells in paraffin tissue sections based on the ability of the technique to detect nuclear DNA fragmentation characteristic of apoptotic cells, hence detecting apoptosis in situ at the single-cell level. The procedure was done using the DeadEnd Colorimetric Apoptosis Detection System (cat. no. G7130, Promega, Southampton, UK). Manufacturer's guidelines were followed throughout the procedure. To prepare negative control sections, a control incubation buffer without TdT enzyme was prepared. Human breast carcinoma known to have high level of apoptosis was used as positive control. The percentage of apoptotic cells (apoptotic index [AI]) was determined by counting 200 to 400 cells depending on the amount of metaplastic tissue in every case. Apoptotic cells were counted in normal adjacent breast epithelium, if present, and used for
Fig. 6. Apoptosis detection using TUNEL technique in SPAPM of the breast.
comparison. The Kruskal-Wallis test was done to assess significance of the AI values in the groups studied. 3. Results 3.1. Immunohistochemistry 3.1.1. Mcl-1, Bak, Bcl-xl, and Bcl-x All cases studied showed distinct cytoplasmic staining in more than 25% of the cells and were considered as positive (Figs. 1-4). 3.1.2. hTERT Nuclear staining in more than 50% of the cells was considered positive. A characteristic punctate pattern (Fig. 5) was particularly seen in association with apocrine cells rather than with normal breast epithelium, which had the tendency to show a more diffuse pattern of nuclear staining. The results are summarized in Table 2. 3.2. TdT-mediated dUTP nick-end labeling 3.2.1. Papillary APM A total of 16 cases were analyzed. The AI (Fig. 6) ranged from 0% to 0.75% (average, 0.4%; SD, ±0.3). The Ki-67
Fig. 7. Ki-67 nuclear positivity in complex papillary APM of the breast (immunoperoxidase).
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7
5
3.2.6. Summary of the data obtained The TUNEL reaction produced positive staining in single nuclei of apocrine cells, which often showed characteristics of apoptotic cells such as condensed chromatin at the margin and a perinuclear halo. The mean AI was (0.3%) in all cases studied. The apoptotic status was neither related to morphology nor to the degree of proliferation in the different groups studied. Normal breast epithelium also showed a mean AI of 0.3% and a mean proliferation index of 0.9%, which reflects its natural slow cell turnover. There was no statistical significance between the AI of the different groups studied (P = .7) as determined by the Kruskal-Wallis test. The results of TUNEL and Ki67 are summarized in Table 3. Fig. 8. Apoptosis detection using TUNEL technique atypical AA of the breast.
(Fig. 7) index in these cases ranged from 0% to 8% (average, 4%; SD, ±2.2). 3.2.2. Nonpapillary APM A total of 12 cases were analyzed. The AI ranged from 0% to 0.5% (average. 0.2%; SD, ±0.2). The Ki-67 index in these cases ranged from 0% to 1.5% (average, 0.7%; SD, ±0.6). 3.2.3. Non–atypical AA A total of 15 cases were analyzed. The AI ranged from 0% to 0.75% (average, 0.4%; SD, ± 0.2). The Ki-67 index in these cases ranged from 2.5% to 8% (average, 5.2%; SD, ±1.7). 3.2.4. Atypical AA A total of 7 cases were analyzed. The AI (Fig. 8) ranged from 0% to 0.5% (average, 0.3%; SD, ±0.2). The Ki-67 index in these cases ranged from 4.5% to 9% (average, 6.6%; SD, ±1.7). 3.2.5. Normal breast epithelium Normal breast tissues were available in 39 of the studied cases. The AI ranged from 0% to 0.75% (average, 0.3%; SD, ±0.3). The Ki-67 index in these cases ranged from 0% to 4% (average, 0.9%; SD, ±0.99).
4. Discussion Normal cell turnover depends on the balance between apoptosis and the rate of proliferation. A few studies have investigated the status of apoptosis in APM. The results of these studies, collectively, reveal that apocrine epithelium is Bcl-2 negative and Bax protein positive [5-8]. Other studies have investigated the rate of proliferation in APM. In one study, the rate of proliferation in APM was shown to be less than 1% in another study [5]; it was shown to be 1.3%, ranging from 0% to 8.5% [8]. None of these 2 studies, however, define the histologic features of the cases of APM studied. In the current study, we investigate the rate of cell turnover in APM with and without papillary features. The study by Page et al [11], dividing APM according to the degree of papillary complexity, showed that the more complex forms of APM had a tendency to be associated with increased risk of subsequent carcinoma development. Other studies investigating APM have shown different molecular abnormalities in APM including c-myc and c-erbB2 overexpression [8]. LOH and CGH studies have shown that some of the genetic abnormalities inherent to breast carcinoma can be found in some cases of APM [17-19]. These findings emphasize the need for identifying forms of APM that may be associated with increased risk and whether papillary features might be a useful feature for this purpose.
Table 3 Summary of TUNEL and Ki-67 results Statistical analysis
NAPM
Papillary APM
Range (%) AI Average/SD Range (%) Ki-67 index Average/SD Total no. of cases
0-0.5
0-0.8
0.2%, SD ±0.2 0-1.5
0.4%, SD ±0.3 1.5-8.5
0.7%, SD ±0.6 5
4.1%, SD ±2.1 16
NAA 0-0.8
AAA
Normal breast epithelium
0-0.5
0-0.8
0.4%, SD ±0.2 2.5-8
0.3%, SD ±0.2 4.5-9
0.3%, SD ±0.3 0-4
5.2%, SD ±1.7 15
6.6%, SD ±1.7 7
0.9%, SD ±0.99 39
6
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7
The current results show that APM generally expresses the apoptotic markers Mcl-1, Bak, Bcl-xl, and Bcl-x showing a similar pattern of staining to normal breast epithelium. The presence of Mcl-1 expression in APM in absence of Bcl-2 has been reported to occur in some tissues before [4,20] where it can be suggested that most tissues normally lacking Bcl-2 expression would show positivity for Mcl-1. Cytoplasmic staining of these apoptotic proteins is, however, difficult to interpret because it is the balance between proand anti apoptotic forces that counts rather than the mere absence or presence of individual proteins. Hence, the TUNEL technique was used to look for the actual presence of apoptotic cells in APM. The results showed that apoptosis is generally a rare event in APM with an AI similar to that of normal breast epithelium. The presence of apoptotic cells showed no relation to the type of APM studied whether papillary or nonpapillary. These findings are similar to those of Feuerhake et al [5], except that in their study, the histologic features of APM were not known. However, in the current study, a borderline significance of Ki-67 index was seen between the groups studied. We believe that papillary features were associated with increased proliferation, although this did not reach statistical significance (average Ki-67 index 4%). This might be due to the small number of cases included in this study. The current results suggest that apoptosis is generally a rare event in APM; however, it has to be remembered that apoptotic pathways are highly complex and hence more studies need to be done to investigate other pathways that might be involved in apocrine cell control. Although a few studies have addressed the risk of subsequent carcinoma development associated with AA [8,12,13,21-24], only 2 studies have actually investigated the expression of the proliferation marker Ki-67 in these lesions [8,12] and showed that a subset of AA had a significant increase in proliferative activity (mean Ki-67 index of 5.0% and 3.6%, respectively). Other studies have also looked at proliferation in these lesions. Unfortunately, the results were included within groups encompassing variable histologic entities ranging from APM, papillomas, adenomas, to adenomyoepithelioma hindering the identification of true biologic criteria for NAA or AAA [25-27]. A total of 22 cases (15 cases NAA and 7 cases AAA) were studied using the TUNEL technique. The Ki-67 index for the NAA group was 5.2% and that for the AAA group was 6.6%. A low AI of 0.45% was found in the NAA group, whereas a value of 0.3% was recorded in the AAA group. There was no association with proliferation, and apoptosis appeared to be a rare event in these lesions. This finding is contrary to the often expressed view that AA is a degenerative lesion [28]. Collectively, the above results of TUNEL and the apoptotic marker study denote that AA has a slow death rate similar to that of normal breast epithelium. Yet, a subset of AA may show increased proliferative capacity; and in the presence of slow rate of apoptosis, this may render these lesions more susceptible to further oncogenic changes. The
current results support the suggestion of [8,12,21,23] that at a subset of AA may have carcinogenic potential. Finally, hTERT showed a characteristic punctate pattern of staining, which was associated with cells with a higher proliferative capacity in a study by Cressey et al (personal correspondence). There was no relation, however, between this pattern of staining and the degree of proliferation in the current study, although Ikeda et al [29], using an in-house antibody, stated that hTERT expression correlated with increased proliferation as indicated by Ki-67. In the current study, hTERT staining was not of any significant help in identifying cells with higher proliferation because it was expressed in normal epithelium, APM of all forms, and some ductal carcinoma in situ (DCIS) and carcinomas used for comparison. Further studies need to be done to assess the significance of the punctate pattern of staining.
Acknowledgments The authors acknowledge the financial support from the Joint Research Board, St Bartholomew's Hospital, London, UK; L Ghaly for technical support; and J Thomas for statistical assistance.
References [1] Ferguson D, Anderson T. Ultrastructural observations on cell death by apoptosis in the resting human breast. Virchows Arch A Pathol Anat Histpathol 1981;393:193-203. [2] Ferguson D. An ultrastructural study of mitosis and cytokinesis in normal resting human breast. Cell Tissue Res 1988;252:581-7. [3] Feuerhake F, Sigg W, Hofter E, et al. Immunohistochemical analysis of bcl-2 and bax expression in relation to cell turnover and epithelial differentiation markers in the non-lactating human mammary gland epithelium. Cell Tissue Res 2000;299:47-58. [4] Krajewski S, Bodrug S, Krajweska M, et al. Immunohistochemical analysis of Mcl-1 protein in human tissues. Differential regulation of Mcl-1 and bcl-2 protein production suggests a unique role for Mcl-1 in control of programmed cell death in-vivo. Am J Pathol 1995;146: 1309-19. [5] Feuerhake F, Unterberger P, Hofter EA. Cell turnover in apocrine metaplasia of the human mammary gland epithelium: apoptosis, proliferation and immunohistochemical detection of bcl-2, bax, EGFR and c-erbB-2 gene products. Acta Histochem 2001;103:53-65. [6] Tavassoli F, Purcell C, Bratthauer G, et al. Androgen receptor expression along with loss of Bcl-2, ER, and PR expression in benign and malignant apocrine lesions of the breast: Implications of therapy. Breast J 1996;2:261-9. [7] Leal C, Henrique R, Monteiro P, et al. Apocrine ductal carcinoma in situ of the breast: histologic classification and expression of biologic markers. Hum Pathol 2001;32:487-93. [8] Selim A, El-Ayat G, Wells C. Expression of c-erbB-2, p53, Bcl-2, Bax, c-myc and ki-67 in apocrine metaplasia and apocrine change within sclerosing adenosis of the breast. Virchows Arch 2002;441:449-55. [9] Kolquist K, Ellisen L, Counter C, Meyerson MM, et al. Expression of TERT in early premalignant lesions and a subset of cells in normal tissues. Nature genetic 1998;19:182-6. [10] Yashima K, Milchgrub S, Gollahon L, et al. Telomerase enzyme activity and RNA expression during the multistage pathogenesis of breast carcinoma. Clin Cancer Res 1998;4:229-34.
G. Elayat et al. / Annals of Diagnostic Pathology 14 (2010) 1–7 [11] Page D, Dupont W, Jensen R. Papillary apocrine change of the breast: associations with atypical hyperplasia and risk of breast cancer. Cancer Epidemiol, Biomarkers Preven 1996;5:29-32. [12] Wells C, McGregor I, Makunura C, et al. Apocrine adenosis: a precursor of aggressive breast cancer? J Clin Pathol 1995;48:737-42. [13] Seidman J, Ashton M, Lefkowitz M. Atypical apocrine adenosis of the breast. A clinicopathologic study of 37 patients with 8.7 year follow up. Cancer 1996;77:2529-37. [14] Hsu SM, Raine L, Fanger H. Use of avidin-biotin-peroxidase complex (ABC) in immunoperoxidase techniques: a comparison between ABC and unlabelled antibody (PAP) procedures. J Histochem Cytochem 1981;29:577-80. [15] Norton A, Jordan S, Yeomans P. Brief, high-temperature heat denaturation (pressure cooking): a simple and effective method of antigen retrieval for routinely processed tissues. J Pathol 1994;173: 371-9. [16] Yan P, Benhattar J, Seelentag W, et al. Immunohistochemical localization of hTERT protein in human tissues. Histochem Cell Biol 2004;121:391-7. [17] Washington C, Dalbegue F, Abreo F, et al. Loss of heterozigosity in fibrocystic change of the breast. Genetic relationship between benign proliferative lesions and associated carcinomas. Am J Pathol 2000;157: 323-9. [18] Jones C, Damiani S, Wells D, et al. Molecular cytogenetic comparison of apocrine hyperplasia and apocrine carcinoma of the breast. Am J Pathol 2001;158:207-14. [19] Selim AA, Ryan A, El-Ayat G, et al. Loss of heterozygosity and allelic imbalance in apocrine metaplasia of the breast: microdissection microsatellite analysis. J Pathol 2002;196:287-91.
7
[20] Shigemasa K, Katoh O, Shiroyama Y, et al. Increased Mcl-1 expression is associated with poor prognosis in ovarian carcinomas. Jpn J cancer Res 2002;93:542-50. [21] Selim A, El-Ayat G, Wells C. C-erbB2 oncoprotein expression, gene amplification and chromosome 17 aneusomy in apocrine adenosis of the breast. J Pathol 2000;191:138-42. [22] Endoh Y, Tamura G, Kato N, et al. Apocrine adenosis of the breast: clonal evidence of neoplasia. Histopathology 2001;38:221-4. [23] Selim A, Ryan A, El-Ayat G, et al. Loss of heterozygosity and allelic imbalance in apocrine adenosis of the breast. Cancer Detect prev 2001; 25:262-7. [24] Schmitt F, Reis-Filho J. Oncogenes, granules and breast cancer: what has c-myc to do with apocrine changes? Breast 2002;11:463-5. [25] Ioffe O, Iskander M, Rucker C, et al. Benign, malignant and pseudomalignant apocrine lesions of the breast: another attempt at differential diagnosis. Lab Invest 1999;79:111 (abstr). [26] Ioffe O, Iskander M, Rucker C, et al. A comparative study of c-erbB-2, p27, bcl-2 and Ki-67 expression in apocrine proliferations and benign breast tissue. Lab Invest 1999;79:112 (abst). [27] Moriya T, Sakamoto K, Sasano H, Kawanaka M, et al. Immunohistochemical analysis of Ki-67, p53, p21 and p27 in benign and malignant apocrine lesions of the breast: its correlation to histologic findings in 43 cases. Mod Pathol 2000;13:13-8. [28] Simpson J, Page D, Dupont W. Apocrine adenosis—a mimic of mammary carcinoma. Am J Surg Pathol 1990;3:289-99. [29] Ikeda S, Shibata T, Eishi Y, et al. Correlation between the expression of telomerase reverse transcriptase and proliferative activity in breast cancer cells using an immunohistochemical restaining method. Pathol Int 2003;53:762-8.