- Email: [email protected]
C-myc oncoprotein expression and gene amplification in apocrine metaplasia and apocrine change within sclerosing adenosis of the breast
The Breast (2002) 11, 466–472 r 2002 Elsevier Science Ltd. All rights reserved. doi: 10.1054/brst.2002.0474, available online at http://www.idealibrary.com on
ORIGINAL ARTICLE
C-myc oncoprotein expression and gene amplification in apocrine metaplasia and apocrine change within sclerosing adenosis of the breast* Abdel-Ghani A. Selim, Ghada El-Ayat, Mahmoud Naase and Clive A. Wells Department of Histopathology, St. Bartholomew’s Hospital, St. Bartholomew’s and the Royal London School of Medicine and Dentistry, Queen Mary and Westfield College, University of London, West Smithfield, London EC1A 7BE, UK S U M M A R Y . Overexpression and/or amplification of c-myc oncogene are known to occur in human breast carcinomas, particularly those of high grade. Apocrine metaplasia (APM) is a common finding within fibrocystic change, and in some cases appears to be associated with an elevated risk of subsequent breast cancer. It has been suggested that apocrine metaplasia within sclerosing adenosis of the breast, also called apocrine adenosis (AA), has a premalignant potential. Little, however, is known about cellular level genetic alterations in either APM or AA of the breast. Because of this, c-myc expression and amplification in APM and AA were studied. Fluorescence in situ hybridisation (FISH) is a methodological approach to detecting these genetic alterations. In this study, APM and AA were studied immunohistochemically to detect c-myc oncoprotein expression, and FISH was employed using a DNA probe for the c-myc gene in archival tissue sections of cases of APM and AA of the breast. Nuclear immunostaining for c-myc was seen in all APM and AA cases studied, but amplification of the c-myc gene was not seen in any cases with APM or AA. The results of this study indicate that c-myc overexpression appears to occur early in breast oncogenesis. Amplification of the c-myc gene does not occur in APM or AA of the breast, however, suggesting that this particular genetic alteration constitutes a late event in the pathogenesis of breast carcinomas. r 2002 Elsevier Science Ltd. All rights reserved.
Fibrocystic change (FCC) is an extremely common finding and is clinically evident in about 50% of women of reproductive age. The basic morphological changes of FCC are the formation of cysts, apocrine metaplasia, stromal fibrosis and epithelial hyperplasia. Any of these changes could be predominant in the lesion. Apocrine metaplasia (APM) is a very common change which is most often seen in dilated cystic structures, but can also appear in normal-sized tubules.5 Although apocrine metaplastic epithelium may present as a single layer of cells, groups of apocrine cells sometimes form papillary configurations (papillary apocrine metaplasia). The terms papillary apocrine change6 and apocrine hyperplasia7 have been applied to such papillary changes. Haagensen8 showed a relationship between large cysts of the breast, which are often lined by apocrine metaplastic epithelium, and an increased risk of subsequent development of breast cancer. The evidence for this association is conflicting, however. While an unexpectedly high
INTRODUCTION The c-myc oncogene encodes for a nuclear phosphoprotein which acts as a transcriptional regulator, controlling cell proliferation, differentiation and apoptosis.1 Amplification of the c-myc gene has been found in approximately 25% of breast carcinomas.2,3 In addition, overexpression of c-myc oncoprotein has been implicated in the early stages of breast cancer development.4
Address correspondence to: Dr A. A. Selim, Department of Histopathology, St. Bartholomew’s Hospital, West Smithfield, London EC1A 7BE, UK. Tel.: +44 20 7601 8451; Fax: +44 20 7601 8543; E-mail: [email protected] Received: 25 July 2002 Revised: 13 August 2002 Accepted: 21 August 2002 *Support was received from the Joint Research Board, St. Bartholomew’s Hospital, London.
466
C-myc in APM and AA of the breast number of cancers have been reported after a short follow-up of women with cystic disease,9,10 by contrast, Dupont and Page,11 in their long-term follow-up study of women with benign breast disease, found an increased relative risk of only 1.7 for those with cystic disease. However, a slightly increased relative risk of 2.4 for subsequent development of breast carcinoma has recently been noted for those with complex patterns of papillary apocrine change.12 Apocrine change within sclerosing adenosis (AA) is a rare breast lesion, defined as the presence of apocrine cytology in a recognisable lobular unit associated with sclerosing adenosis. Wells et al.13 have suggested AA as a possible precancerous lesion for high-grade breast carcinomas, partly on the basis of the expression of cerbB2 oncoprotein and partly on morphological grounds. Moreover, AA has been misdiagnosed in the past as carcinoma because of the cytological atypia of the apocrine cells seen within areas of sclerosing adenosis and, occasionally, in terminal ducts. Sclerosing adenosis has been implicated in an increased risk of subsequent breast carcinoma, with a relative risk of 1.5– 2, by Dupont and Page.11 Subsequently, Seidman14 reported on the follow-up of 37 patients with atypical AA (atypical apocrine cells within sclerosing adenosis), four of whom developed carcinoma with a relative risk of 5.5. Apart from occasional reports in the literature of aneuploidy in apocrine metaplasia cells15,16 and a recent report of comparative genomic hybridisation in papillary apocrine metaplasia,17 the genetic changes occurring in APM and AA of the breast are relatively poorly known. Recently, the technique of fluorescence in situ hybridisation (FISH) has been employed as a molecular tool by which chromosomes in dividing and nondividing cells (metaphase and interphase cytogenetics) can be identified and characterised. The procedure is based on the hybridisation of a DNA probe to its complementary sequence in the cell nucleus, results being visualised as an area of fluorescence or ‘signal’ at the hybridisation site. This quick method has been used extensively in cytogenetics laboratories to characterise chromosomal rearrangements, determine chromosome or gene copy number, and identify microdeletions. In some studies FISH has been used to analyse these genetic abnormalities in breast cancer, such as detection of amplification of c-erbB2 and c-myc oncogenes.18,20 This study was undertaken to investigate c-myc oncogene expression and amplification in apocrine metaplasia and apocrine change within sclerosing adenosis of the breast using immunohistochemical and interphase cytogenetic FISH methods.
467
MATERIALS AND METHODS Case selection Thirty-three cases of apocrine metaplasia, predominantly with papillary change, and 15 cases of apocrine change within sclerosing adenosis (including three cases of atypical AA according to the criteria of Seidman14) were collected from the files of the histopathology department of St Bartholomew’s Hospital, London. Some of the AA cases had been sent to Dr C. A. Wells as referral cases. The age of the APM cases ranged from 21 to 79 years (mean 46.7), and the age of AA cases ranged from 32 to 62 years (mean 50.4 years).
Immunohistochemistry Tissue: Formalin-fixed, paraffin-embedded blocks of tissue from 33 cases of APM and 15 cases of AA were selected from the files and sectioned at a nominal 4 mm. The standard avidin–biotin–peroxidase complex method21 was used. Heat-mediated antigen retrieval using the pressure cooker method22 was used. Appropriate positive and negative controls omitting the primary antibodies were included with each slide run. Antibodies: The prediluted monoclonal antibody against c-myc oncoprotein, clone 9E10 (Zymed Laboratories, San Francisco, USA) was used. The primary antibody was applied overnight at 41C, and a previously stained positive breast carcinoma was used as a positive control. A polyclonal antibody against Ki-67 (Dako Ltd, London, UK) was applied overnight at a dilution of 1/200, and a tonsillar section was used as a positive control. Assessment: Nuclear staining for c-myc in 420% of the cells was taken as positive and cytoplasmic staining was ignored. Cases showing stronger nuclear staining were usually associated with increased levels of cytoplasmic staining. Ki-67 nuclear staining was taken as positive and assessed quantitatively by counting the number of positive nuclei in 200 cells. This was expressed as a percentage, the proliferative index.
Fluorescence in situ hybridisation Tissue: A further series of 4-mm-thick paraffin sections was available from 15 cases of APM and 12 cases of AA and were mounted on silane-coated slides. The first and last sections of each block were stained with haematoxylin and eosin (H&E) to confirm the presence of the lesion in all sections and were also used as templates.
468
The Breast
Probes: A directly, spectrum orange, labelled c-myc locus-specific (8q24–q24.3), DNA probe (Vysis, IL, USA) was used. Hybridisation procedures: Tissue pretreatment and hybridisation procedures were performed according to the manufacturer’s instructions with minor modification. Briefly, slides were pretreated in sodium thiocyanate at 801C for 10 min, followed by pepsin digestion (2 mg/ml in 0.2 MHCl, Sigma) at 371C for 10–14 min to expose target DNA. Denaturation was performed at 701C by incubation in formamide solution at 701C. Hybridisation with the DNA probe was carried out in a humidified chamber at 371C overnight. After washing in formamide, sodium chloride/sodium citrate (SSC) and detergent solutions, sections were counterstained with DAPI and examined directly under a fluorescence microscope equipped with a dual-pass filter for the simultaneous detection of spectrum orange and DAPI. Signal enumeration and standardisation: The area of interest was localised with the help of the H&E template whenever needed. The 100 objective lens with oil immersion was used for signal counting. Closely paired signals, which can reflect splitting of signals in nuclei in S and/or G2 phases of the cell cycle, were counted as single signals. The numbers of cells with 2 signals, 42 signals and o2 signals per nucleus were counted. When the number of cells with negative nuclei (zero signals) exceeded 15%, the procedure was regarded as suboptimal and was repeated; this would mean that the gene estimation might be incorrect. As in similar studies,23,24 the signals present in 100 lesion nuclei were counted; signals from 100 epithelial cells in normal breast tissue on the same slide were also counted. Since whole nuclei are not included in 4-mm-thick sections, the signal number is lower in paraffin sections than in complete nuclei (e.g., in cytology imprint preparations). To control for this under-representation, the FISH values for normal nuclei in the same section were used. In some cases where too few normal epithelial cells were present, lymphocyte and/or stromal cell nuclei were used as normal controls. Interphase analysis values in morphologically normal breast epithelium with the c-myc locus-specific probe used in this study were taken as normal values. In all cases some normal cells showed more than two signals (most three and occasionally four) in up to 5% of the nuclei. In addition, fewer than two signals (mainly one and occasionally no signal) were seen in normal cells in up to 44% of the nuclei. Some artefactual loss of signals is to be expected, because thin sections do not encompass the entire nucleus. Moreover, the extent to which signals are lost will vary with the type of tissue in relation to the average nuclear diameter. It was also
anticipated that the spatial location of the particular gene studied in each cell might be variable. Lastly, owing to variability in tissue-DNA preparation the possibility of hybridisation inefficiency had to be borne in mind. Definitions of true gain or loss were formulated on the basis of these analyses. Standards were set so that cells with more than two signals in more than 10% of the nuclei or with fewer than two signals in greater than 50% of nuclei were considered to indicate true gain or loss, respectively. These cut-off values are in keeping with those that have been used in other studies.24–26 We believe that these are conservative thresholds, which are likely to underestimate the true incidence of gene aberrations. Borderline or questionable loss or gain was defined as from 45% to 50% or 6% to 10%, respectively. Statistical analysis To evaluate statistical significance the Mann–Whitney test was applied. A P-value of o0.05 was considered significant.
RESULTS Immunohistochemistry Normal breast epithelium showed nuclear positivity in up to 16% of the cells with or without cytoplasmic staining. Hence, cases of APM or AA (Fig. 1) showing nuclear positivity in more than 20% of the cells were counted as positive whether associated with cytoplasmic staining or not. All the cases of APM (33 cases) and AA (15 cases) were positive for c-myc protein, but the mean percentage of nuclear positivity was higher in AA than
Fig. 1 C-myc nuclear positivity of apocrine adenosis of the breast (immunoperoxidase).
C-myc in APM and AA of the breast in APM cases: 50% (range 28–74%) in AA and 37% (range 21–68%) in APM cases. The mean percentage positivity of Ki-67 in AA cases was 3.6% (range 0– 18.5%). The three atypical cases showed a range of 1.5–4% positivity. In APM cases, a mean Ki-67 positivity of 1.3% (range 0–8.5%) was detected. Normal epithelium was available for assessment in 22 cases and showed a mean percentage positivity of 0.9% (range 0– 7%). There was a highly significant association between the percentage of c-myc and Ki-67 nuclear positivity in both APM and AA (P o 0.0001).
Table 2
469
Analysis of c-myc gene in apocrine adenosis cases
Apocrine adenosis (AA) 12 cases
% of cells with o2 signals/nucleus
% of cells with 2 signals/nucleus
% of cells with 42 signals/nucleus
Normal area(s) Range Mean S.D.
33–41 38.2 72.8
56–66 60.3 72.7
0–3 1.5 70.9
AA area(s) Range Mean S.D.
33–41 37.3 72.3
56–63 59.8 72.0
1–6 2.9 71.4
S.D.= standard deviation.
Fluorescence in situ hybridisation The results of 15 cases of APM (Fig. 2) and 12 cases of AA together with the normal epithelium are summarised in Tables 1 and 2. FISH analysis of the c-myc gene in APM The mean percentages of cells with 2 signals/nucleus were 59.8 (72.7) and 60.1 (72.5) with ranges of 56–66
Fig. 2 C-myc gene copy (red signals) in apocrine metaplasia of the breast. Nuclei (blue) show mainly 2 signals/nucleus. Also cells with 1 – 4 signals/nucleus are seen (FISH).
Table 1 Analysis of c-myc gene in apocrine metaplasia cases Apocrine metaplasia (APM) 15 cases
% of cells with o2 signals/nucleus
% of cells with 2 signals/nucleus
% of cells with 42 signals/nucleus
Normal area(s) Range Mean S.D.
33–42 38.8 72.8
56–66 59.8 72.7
0–3 1.4 71.0
APM area(s) Range Mean S.D.
34–42 37.9 72.6
56–64 60.1 72.5
1–4 2.0 71.1
S.D.=standard deviation.
and 56–64 in normal breast epithelium and APM, respectively. The mean percentages of cells with 42 signals/nucleus were 1.4 (71.0) and 2.0 (71.1) with ranges of 0–3 and 1–4 in normal breast epithelium and APM, respectively. The mean percentages of cells with o2 signals/nucleus were 38.8 (72.8) and 37.9 (72.6) with ranges of 32–42 and 34–42 in normal epithelium and APM, respectively. FISH analysis of the c-myc gene in AA The mean percentages of cells with 2 signals/nucleus were 60.3 (72.7) and 59.8 (72.0) with ranges of 56–66 and 56–63 in normal breast epithelium and AA, respectively. The mean percentages of cells with 42 signals/nucleus were 1.5 (70.9) and 2.9 (71.4), with ranges of 0–3 and 1–6 in normal breast epithelium and AA, respectively. In no case of AA 44 signals/nucleus were found, but one case showed borderline gain (amplification) with 42 signals/nucleus in 6% of the cells. The mean percentages of cells with o2 signals/ nucleus were 38.2 (72.8) and 37.3 (72.3) with ranges of 33–41 and 33–41 in normal breast epithelium and AA, respectively.
DISCUSSION Apocrine epithelium in the breast has been intensely studied owing to its role in gross cystic disease of the breast and its possible relationship to breast carcinoma. Wellings et al.27 considered apocrine cells to be the progenitors of breast cystic disease. They reported that apocrine epithelium associated with the accumulation of secretions might favour the progressive unfolding of lobules, formation of microcysts and, finally, the appearance of macrocysts. The relationship between apocrine cystic changes and breast carcinoma has always been controversial. Clinical follow-up studies of
470
The Breast
women with breast apocrine cystic disease have shown an increased risk of subsequent breast carcinoma.28–30 Wellings and Alpers28 have discovered that cystic apocrine metaplasia is more common in cancer-associated breasts than in the normal breast, and also that foci of apocrine cysts are more numerous in cases of breast carcinoma. Some hypotheses for the relationship between apocrine epithelium and carcinoma of the breast have been proposed by Haagensen.8 Apocrine epithelium may be a precursor of malignant transformation; it may reflect a response to the same stimulus which promotes carcinomas; or, thirdly, apocrine metaplasia could reflect an instability of the breast epithelium, which causes the development of alterations with a higher propensity for cancer. The data relating to the relationship between apocrine epithelium, cysts and cancer of the breast are unclear, and the stimuli which lead to these processes are unknown. Metaplasia throughout the human body is associated with the development of carcinoma in lung, stomach and cervix. Haagensen31 reported a five-fold increase in the incidence of breast carcinoma associated with papillary apocrine metaplasia in fibrocystic change. Also, in a long-term follow-up study of patients diagnosed with atypical AA, there was an increased relative risk of 5.5 for the subsequent development of breast carcinoma.14 Abnormal oncogene expression has been discovered before in apocrine metaplastic epithelium of the breast by Papamichalis32 using an antibody to c-myc oncoprotein, by Agnantis33 using antibodies to ras and c-myc oncoproteins and by McCann34 and Wells13 using antibodies to c-erbB2 oncoprotein. In keeping with these findings, using the pressure cooker, heat-mediated, antigen retrieval method rather than enzymatic pretreatment of tissue sections, we found positivity for c-myc in a higher percentage of cells in all cases of AA and APM tested than normal epithelium. The percentage of nuclear positivity was higher in AA than in APM. Increased c-myc expression may play a part in the early, intermediate or late stages of malignant transformation of normal cells, as demonstrated by in vitro cell transformation studies and in vivo assays of tumorigenicity and metastasis.1,35,36 It is therefore tempting to speculate that its activation in apocrine metaplastic epithelium, particularly in AA, might have a role in their abnormal proliferation and their conversion to a malignant state. In addition, it has been thought that c-myc acts as a bivalent regulator of both cell proliferation and apoptosis, depending on the availability of growth factors. If cells driven by overactivity of c-myc do not have sufficient growth factors in their environment, they undergo apoptosis depending on the avail-
ability of other apoptosis-related factors.37,38 As apocrine metaplastic epithelium usually lines type-1 breast cysts, which contain high amounts of growth factors,29 this suggests a proliferative role for increased c-myc expression in APM. It is of interest that cases with increased mean c-myc expression also show an increased mean proliferative index with the proliferation antigen Ki-67, which was highly significant statistically. This suggests that in these cases, c-myc expression is indeed related to increased cell proliferation. Investigation of the key genetic events involved in neoplastic development and progression is one of the critical areas of research that may lead to significant progress in our understanding of human carcinogenesis and the development of clinical applications. Although precursor lesions can be readily harvested from some tissues, such as a macroscopically visible adenoma adjacent to a colonic adenocarcinoma, noninvasive breast epithelial lesions such as APM and AA are microscopic and are usually only recognised after the tissue has been fixed in formalin and embedded in paraffin. Until recently, investigations have been limited to epidemiological associations with morphology and immunohistochemical identification of altered protein expression. The advent of FISH analysis of formalinfixed paraffin-embedded tissue sections has made the localisation of cytogenetic aberrations to particular cell types possible because the nuclei are visualised in their normal histological context. The nonuniformity of changes within a cell population provides a measure of genetic instability, which is assumed to be a marker of tumour progression.39,40 In addition, the use of sitespecific DNA probes against chromosomal loci and regions has permitted the identification of structural and numerical genetic aberrations. Identification of such aberrations in potentially premalignant lesions would add further weight to the identification of these lesions as biologically premalignant. Using the FISH methodology, c-myc gene copy amplification has been reported in breast cancer.19,20 In the present study we have performed FISH on archival tissue sections to test whether amplification of the c-myc gene is present in a series of APM and AA. FISH on paraffin sections underestimates the true gene copy number owing to sectioning of the nuclei. Although isolated nuclei give a more accurate assessment of gene copy number41 in some situations, this was impossible in the present study owing to the focal and microscopic nature of APM and AA. Preservation of tissue architecture is essential for their identification. We attempted to control for nuclear sectioning artefacts by assessing corresponding normal areas within the same section for each case.
C-myc in APM and AA of the breast None of the APM and AA cases, apart from one case of AA, showed amplification of the c-myc gene. This one case of AA showed borderline gain (amplification) of the c-myc gene, with 6% of cells showing 3–4 signals/ nucleus. As the percentage of cells showing 3–4 signals/ nucleus was fewer than 10%, we are cautious about interpreting this as definite amplification of the gene, but it may represent a low level of amplification in a subset of cells. The absence of c-myc amplification according to the FISH technique is intriguing and does not explain the finding of c-myc overexpression on immunohistochemistry, which may have been a consequence of amplification of the gene. All the APM and AA cases used for cmyc FISH analysis showed overexpression of c-myc protein in more than 20% of the cell population. This may be explained if the gene is being stimulated or switched on by certain stimuli, such as growth factors or hormones without being amplified. Wolman39 reported that increased expression of the protein in the absence of amplification of the coding gene could result from modification in transcription or translation, possibly coded by genes on other chromosomes. FISH will not detect gene activation if it occurs as a result of increased gene transcription with no change in gene copy number. Interestingly, the study of CGH in papillary apocrine metaplasia17 did not show any genetic abnormalities on chromosome 8, including the c-myc locus, in agreement with our results showing absence of amplification. While studies of c-myc in mouse systems have suggested that alterations to the gene may be an important early event in the development of tumours,42 the evidence from human breast carcinomas has been less conclusive. Amplification of the c-myc gene has been found in approximately 16% of breast carcinomas.43,44 The second of these reviews44 found that c-myc amplification in patients with breast cancer correlates with aggressive features and/or poor prognosis, suggesting that amplification of the gene may occur as a late event in breast cancer. Our finding of lack of c-myc amplification in cases of APM and AA that are showing overexpression of the cmyc protein is in accordance with the findings of Escot45 who reported overexpression of c-myc without amplification of the gene in 80% of primary breast carcinomas. Also, when mRNA and protein expression of c-myc have been determined, poor correlation has been demonstrated between these levels and the extent of expression and amplification of the gene.46,47 Pietilainen4 reported that expression of c-myc at the protein level occurs in well-differentiated rather than poorly differentiated breast carcinomas. This led him to suggest that c-myc protein expression, in contrast to gene
471
amplification, is involved in early stages of breast cancer development. In agreement with these findings in breast cancer it seems that there is overexpression of c-myc at the protein level but no amplification of the gene is the case in APM and AA. In previous studies, lack of amplification of other oncogenes has also been reported, e.g., absence of cerbB2 amplification in some breast48 and bladder49 carcinomas that show overexpression of the encoded protein on immunocytochemistry. In conclusion, the findings of abnormal c-myc oncogene expression in apocrine metaplastic epithelium without detectable amplification of the gene indicate that overexpression of c-myc but not its amplification may occur as an early event in breast oncogenesis. Moreover, amplification of the c-myc oncogene appears to be a late event in the pathogenesis of breast cancer.
Acknowledgements We would like to acknowledge support from the Joint Research Board, St. Bartholomew’s Hospital, London. Thanks also go to P. Wencyk and S. Jordan for their technical assistance,to R. Hamoudi for statistical assistance and to Dr. C. Sowter, Dr. P. Hallam and Dr. S. Wilkes for their help with the fluorescence microscopy and preparation of the photography.
REFERENCES 1. Evan GI, Wyllie AH, Gilbert CS et al. Induction of apoptosis in fibroblasts by c-myc protein. Cell 1992; 69: 1–20. 2. Meyers SL, O’Brien MT, Smith T et al. Analysis of the int-1, int-2, c-myc, and neu oncogenes in human breast carcinomas. Cancer Res 1990; 50: 5911–5918. 3. Berns EM, Klijn JG, van Putten et al. C-myc amplification is a better prognostic factor than HER2/neu amplification in primary breast cancer. Cancer Res 1992; 1: 1107–1113. 4. Pietilainen T, Lipponen P, Aaltomaa S et al. Expression of c-myc proteins in breast cancer as related to established prognostic factors and survival. Anticancer Res 1995; 15: 959–964. 5. Rosai J. Ackerman’s Surgical Pathology, 8th Edition. New York: Mosby-Year Book, 1996; 1565–1659. 6. Page DL, Vander Zwaag R, Rogers LW et al. Relation between component parts of fibrocystic disease complex and breast cancer. J Natl Cancer Inst 1978, 61: 1055–1063. 7. Tavassoli FA, Norris HJ. Intraductal apocrine carcinoma: a clinicopathological study of 37 cases. Mod Pathol 1994; 7: 813–818. 8. Haagensen Jr DE Is cystic disease related to breast cancer? Am J Surg Pathol 1991; 15: 687–694. 9. Harrington E, Lesnik G. The association between gross cysts of the breast and breast cancer. Breast 1980; 7: 13–17. 10. Roberts MM, Jones V, Elton RA et al. Risk of breast cancer in women with history of benign disease of the breast. Br Med J 1984; 288: 275–278. 11. Dupont WD, Page DL. Risk factors for breast cancer in women with proliferative disease. N Engl J Med 1985; 312:146–151.
472
The Breast
12. Page DL, Dupont WD, Jensen RA. Papillary apocrine change of the breast: association with atypical hyperplasia and risk of breast cancer. Cancer Epidemiol Biomarkers Prev 1996; 5: 29–32. 13. Wells CA, McGregor IL, Makunura CN et al. Apocrine adenosis: a precursor of aggressive breast disease. J Clin Pathol 1995; 48: 737–742. 14. Seidman JD, Ashton M, Lefkowitz M. Atypical apocrine adenosis of the breast: clinicopathological study of 37 patients with 8.7-year follow up. Cancer 1996; 77: 2529–2537. 15. Izuo M, Okagaki T, Richart R et al. DNA content in ‘apocrine metaplasia’ of fibrocystic disease of the breast. Cancer 1971; 27: 643–650. 16. Vasiu R, Olinici CD. DNA nuclear content of apocrine cells in fibrocystic mastopathy and breast cancer. Morphol Embryol 1990; 36: 43–47. 17. 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: 207–214. 18. Pauletti G, Godolphin W, Press MF et al. Detection and quantitation of HER-2/neu gene amplification in human breast cancer in archival material using fluorescence in situ hybridisation. Oncogene 1996; 13: 63–72. 19. Persons DL, Borelli KA, Hsu PH. Quantitation of HER-2/neu and c-myc gene amplification in breast carcinoma using fluorescence in situ hybridisation. Mod Pathol 1997; 10: 720–727. 20. Visscher DW, Wallis T, Awussah S et al. Evaluation of MYC and Chromosome 8 copy number in breast carcinoma by interphase cytogenetics. Genes Chromosomes Cancer 1997; 18: 1–7. 21. Hsu S-M, Raine L, Fanger H. Use of avidin-biotin-peroxidase complex (ABC) in immmunoperoxidase techniques: a comparison between ABC and unlabelled antibody (PAP) procedures. J Histochem Cytochem 1981; 29: 577–580. 22. Norton AJ, 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: 371–379. 23. Dekken HV, Bosman FT, Teijgeman R et al. Identification of numerical chromosome aberration in archival tumours by in situ hybridization in routine paraffin sections: evaluation of 23 phaeochromocytomas. J Pathol 1993; 171: 161–171. 24. Harrison M, Magee HM, O’Loughin J et al. Chromosome 1 aneusomy, identified by interphase cytogenetics, in mammographically detected ductal carcinoma in situ of the breast. J Pathol 1995, 175: 303–309. 25. Micale MA, Visscher W, Gulino SE et al. Chromosomal aneuploidy in proliferative breast disease. Hum Pathol 1994; 25: 29–35. 26. Visscher DW, Wallis TL, Crissman JD. Evaluation of chromosome aneuoploidy in tissue sections of preinvasive breast carcinoma using interphase cytogenetics. Cancer 1996; 77: 315–320. 27. Wellings SR, Jensen HM, Marcum RG. An atlas of subgross pathology of the human breast with special reference to possible precancerous lesions. J Natl Cancer Inst 1975; 55: 231–243. 28. Wellings SR, Alpers CE. Apocrine cystic metaplasia: subgross pathology and prevalence in cancer associated versus random autopsy breasts. Hum Pathol 1987; 18: 381–386. 29. Ness JC, Sedghinasab M, Moe RE et al. Identification of multiple proliferative growth factors in breast cyst fluid. Am J Surg 1993; 166: 237–243.
30. Bruzzi P, Dogliotti L, Naldoni C et al. Cohort study of association of risk of breast cancer with cyst type in women with gross cystic disease of the breast. BMJ 1997; 314: 925–928. 31. Haagensen CD. Apocrine epithelium. In: Diseases of the Breast, 3rd Edition. Philadelphia: W.B. Saunders, 1986; 82–101. 32. Papamichalis G, Francia K, Karachaliou FE et al. Expression of the c-myc oncoprotein in human metaplastic epithelial cells of fibrocystic disease. Anticancer Res 1988; 8: 1217–1222. 33. Agnantis NJ, Mahera H, Maounis N et al. Immunohistochemical study of ras and myc oncoproteins in apocrine breast lesions with and without papillomatosis. Eur J Gynaecol Oncol 1992; 13: 309–315. 34. McCann A, Johnston PA, Dervan PA et al. c-erbB2 oncoprotein expression in malignant and non-malignant breast tissue. Int J Med Sci 1989; 158: 137–140. 35. Spandidos DA, Anderson MLM. A study of the mechanism of carcinogenesis by gene transfer of oncogenes into mammalian cells. Mutat Res 1987; 185: 271–291. 36. Wyllie AH, Rose KA, Morris RG et al. Rodent fibroblast tumours expressing human myc and ras genes: growth, metastasis and endogenous oncogene expression. Br J Cancer 1987; 56: 251–259. 37. Boise LH, Gottschalk AR, Quintans J et al. Bcl-2 and Bcl-2 related proteins in apoptosis regulation. Curr Top Microbiol Immunol 1995, 200: 107–121. 38. Cummings MC, Winterford CM, Walker NI. Apoptosis. Am J Surg Pathol 1997; 21: 88–101. 39. Wolman SR, Pauley RJ, Mohamed AN et al. Genetic markers as prognostic indicators in breast cancer. Cancer 1992; 70: 1765–1774. 40. Nicolson GL. Tumour cell instability, diversification, and progression to the metastatic phenotype: from oncogene to oncofetal expression. Cancer Res 1987; 47: 1473–1487. 41. Emmerich P, Jauch A, Hofmann MC et al. Interphase cytogenetics in paraffin embedded sections from human testicular germ cell tumour xenografts and in corresponding cultured cells. Lab Invest 1989; 61: 235–242. 42. Edwards PAW, Ward JL, Bradbury JM. Alteration of morphogenesis by the v-myc oncogene in transplants of mouse mammary gland. Oncogene 1988; 2: 407–412. 43. Liao DJ, Dickson RB. C-myc in breast cancer. Endocr Relat Cancer 2000 ; 7: 143–164. 44. Deming SL, Nss SJ, Dickson RB, Trock BJ. C-myc amplification in breast cancer: a meta-analysis of its occurrence and prognostic relevance. Br J Cancer 2000; 83: 1688–1695. 45. Escot C, Theillet C, Lidereau R et al. Genetic alteration of the cmyc protooncogene (MYC) in human primary breast carcinomas. Proc Natl Acad Sci USA 1986; 83: 4834–4838. 46. Locker AP, Dowle CS, Ellis IO et al. C-myc oncogene product expression and prognosis in operable breast cancer. Br J Cancer 1989; 60: 669–672. 47. Walker RA, Senior PV, Jones JL et al. An immunohistochemical and in situ hybridisation study of c-myc and c-erbB-2 expression in primary breast tumours. J Pathol 1989; 158: 97–105. 48. Slamon DJ, Godolphin W, Jones LA et al. Studies of the HER2/ neu proto-oncogene in human breast and ovarian cancer. Science 1989; 244: 707–712. 49. Coombs LM, Pigott DA, Sweeney E et al. Amplification and overexpression of c-erbB-2 in transitional cell carcinoma of the urinary bladder. Br J Cancer 1991; 62: 601–608.
ORIGINAL ARTICLE
C-myc oncoprotein expression and gene amplification in apocrine metaplasia and apocrine change within sclerosing adenosis of the breast* Abdel-Ghani A. Selim, Ghada El-Ayat, Mahmoud Naase and Clive A. Wells Department of Histopathology, St. Bartholomew’s Hospital, St. Bartholomew’s and the Royal London School of Medicine and Dentistry, Queen Mary and Westfield College, University of London, West Smithfield, London EC1A 7BE, UK S U M M A R Y . Overexpression and/or amplification of c-myc oncogene are known to occur in human breast carcinomas, particularly those of high grade. Apocrine metaplasia (APM) is a common finding within fibrocystic change, and in some cases appears to be associated with an elevated risk of subsequent breast cancer. It has been suggested that apocrine metaplasia within sclerosing adenosis of the breast, also called apocrine adenosis (AA), has a premalignant potential. Little, however, is known about cellular level genetic alterations in either APM or AA of the breast. Because of this, c-myc expression and amplification in APM and AA were studied. Fluorescence in situ hybridisation (FISH) is a methodological approach to detecting these genetic alterations. In this study, APM and AA were studied immunohistochemically to detect c-myc oncoprotein expression, and FISH was employed using a DNA probe for the c-myc gene in archival tissue sections of cases of APM and AA of the breast. Nuclear immunostaining for c-myc was seen in all APM and AA cases studied, but amplification of the c-myc gene was not seen in any cases with APM or AA. The results of this study indicate that c-myc overexpression appears to occur early in breast oncogenesis. Amplification of the c-myc gene does not occur in APM or AA of the breast, however, suggesting that this particular genetic alteration constitutes a late event in the pathogenesis of breast carcinomas. r 2002 Elsevier Science Ltd. All rights reserved.
Fibrocystic change (FCC) is an extremely common finding and is clinically evident in about 50% of women of reproductive age. The basic morphological changes of FCC are the formation of cysts, apocrine metaplasia, stromal fibrosis and epithelial hyperplasia. Any of these changes could be predominant in the lesion. Apocrine metaplasia (APM) is a very common change which is most often seen in dilated cystic structures, but can also appear in normal-sized tubules.5 Although apocrine metaplastic epithelium may present as a single layer of cells, groups of apocrine cells sometimes form papillary configurations (papillary apocrine metaplasia). The terms papillary apocrine change6 and apocrine hyperplasia7 have been applied to such papillary changes. Haagensen8 showed a relationship between large cysts of the breast, which are often lined by apocrine metaplastic epithelium, and an increased risk of subsequent development of breast cancer. The evidence for this association is conflicting, however. While an unexpectedly high
INTRODUCTION The c-myc oncogene encodes for a nuclear phosphoprotein which acts as a transcriptional regulator, controlling cell proliferation, differentiation and apoptosis.1 Amplification of the c-myc gene has been found in approximately 25% of breast carcinomas.2,3 In addition, overexpression of c-myc oncoprotein has been implicated in the early stages of breast cancer development.4
Address correspondence to: Dr A. A. Selim, Department of Histopathology, St. Bartholomew’s Hospital, West Smithfield, London EC1A 7BE, UK. Tel.: +44 20 7601 8451; Fax: +44 20 7601 8543; E-mail: [email protected] Received: 25 July 2002 Revised: 13 August 2002 Accepted: 21 August 2002 *Support was received from the Joint Research Board, St. Bartholomew’s Hospital, London.
466
C-myc in APM and AA of the breast number of cancers have been reported after a short follow-up of women with cystic disease,9,10 by contrast, Dupont and Page,11 in their long-term follow-up study of women with benign breast disease, found an increased relative risk of only 1.7 for those with cystic disease. However, a slightly increased relative risk of 2.4 for subsequent development of breast carcinoma has recently been noted for those with complex patterns of papillary apocrine change.12 Apocrine change within sclerosing adenosis (AA) is a rare breast lesion, defined as the presence of apocrine cytology in a recognisable lobular unit associated with sclerosing adenosis. Wells et al.13 have suggested AA as a possible precancerous lesion for high-grade breast carcinomas, partly on the basis of the expression of cerbB2 oncoprotein and partly on morphological grounds. Moreover, AA has been misdiagnosed in the past as carcinoma because of the cytological atypia of the apocrine cells seen within areas of sclerosing adenosis and, occasionally, in terminal ducts. Sclerosing adenosis has been implicated in an increased risk of subsequent breast carcinoma, with a relative risk of 1.5– 2, by Dupont and Page.11 Subsequently, Seidman14 reported on the follow-up of 37 patients with atypical AA (atypical apocrine cells within sclerosing adenosis), four of whom developed carcinoma with a relative risk of 5.5. Apart from occasional reports in the literature of aneuploidy in apocrine metaplasia cells15,16 and a recent report of comparative genomic hybridisation in papillary apocrine metaplasia,17 the genetic changes occurring in APM and AA of the breast are relatively poorly known. Recently, the technique of fluorescence in situ hybridisation (FISH) has been employed as a molecular tool by which chromosomes in dividing and nondividing cells (metaphase and interphase cytogenetics) can be identified and characterised. The procedure is based on the hybridisation of a DNA probe to its complementary sequence in the cell nucleus, results being visualised as an area of fluorescence or ‘signal’ at the hybridisation site. This quick method has been used extensively in cytogenetics laboratories to characterise chromosomal rearrangements, determine chromosome or gene copy number, and identify microdeletions. In some studies FISH has been used to analyse these genetic abnormalities in breast cancer, such as detection of amplification of c-erbB2 and c-myc oncogenes.18,20 This study was undertaken to investigate c-myc oncogene expression and amplification in apocrine metaplasia and apocrine change within sclerosing adenosis of the breast using immunohistochemical and interphase cytogenetic FISH methods.
467
MATERIALS AND METHODS Case selection Thirty-three cases of apocrine metaplasia, predominantly with papillary change, and 15 cases of apocrine change within sclerosing adenosis (including three cases of atypical AA according to the criteria of Seidman14) were collected from the files of the histopathology department of St Bartholomew’s Hospital, London. Some of the AA cases had been sent to Dr C. A. Wells as referral cases. The age of the APM cases ranged from 21 to 79 years (mean 46.7), and the age of AA cases ranged from 32 to 62 years (mean 50.4 years).
Immunohistochemistry Tissue: Formalin-fixed, paraffin-embedded blocks of tissue from 33 cases of APM and 15 cases of AA were selected from the files and sectioned at a nominal 4 mm. The standard avidin–biotin–peroxidase complex method21 was used. Heat-mediated antigen retrieval using the pressure cooker method22 was used. Appropriate positive and negative controls omitting the primary antibodies were included with each slide run. Antibodies: The prediluted monoclonal antibody against c-myc oncoprotein, clone 9E10 (Zymed Laboratories, San Francisco, USA) was used. The primary antibody was applied overnight at 41C, and a previously stained positive breast carcinoma was used as a positive control. A polyclonal antibody against Ki-67 (Dako Ltd, London, UK) was applied overnight at a dilution of 1/200, and a tonsillar section was used as a positive control. Assessment: Nuclear staining for c-myc in 420% of the cells was taken as positive and cytoplasmic staining was ignored. Cases showing stronger nuclear staining were usually associated with increased levels of cytoplasmic staining. Ki-67 nuclear staining was taken as positive and assessed quantitatively by counting the number of positive nuclei in 200 cells. This was expressed as a percentage, the proliferative index.
Fluorescence in situ hybridisation Tissue: A further series of 4-mm-thick paraffin sections was available from 15 cases of APM and 12 cases of AA and were mounted on silane-coated slides. The first and last sections of each block were stained with haematoxylin and eosin (H&E) to confirm the presence of the lesion in all sections and were also used as templates.
468
The Breast
Probes: A directly, spectrum orange, labelled c-myc locus-specific (8q24–q24.3), DNA probe (Vysis, IL, USA) was used. Hybridisation procedures: Tissue pretreatment and hybridisation procedures were performed according to the manufacturer’s instructions with minor modification. Briefly, slides were pretreated in sodium thiocyanate at 801C for 10 min, followed by pepsin digestion (2 mg/ml in 0.2 MHCl, Sigma) at 371C for 10–14 min to expose target DNA. Denaturation was performed at 701C by incubation in formamide solution at 701C. Hybridisation with the DNA probe was carried out in a humidified chamber at 371C overnight. After washing in formamide, sodium chloride/sodium citrate (SSC) and detergent solutions, sections were counterstained with DAPI and examined directly under a fluorescence microscope equipped with a dual-pass filter for the simultaneous detection of spectrum orange and DAPI. Signal enumeration and standardisation: The area of interest was localised with the help of the H&E template whenever needed. The 100 objective lens with oil immersion was used for signal counting. Closely paired signals, which can reflect splitting of signals in nuclei in S and/or G2 phases of the cell cycle, were counted as single signals. The numbers of cells with 2 signals, 42 signals and o2 signals per nucleus were counted. When the number of cells with negative nuclei (zero signals) exceeded 15%, the procedure was regarded as suboptimal and was repeated; this would mean that the gene estimation might be incorrect. As in similar studies,23,24 the signals present in 100 lesion nuclei were counted; signals from 100 epithelial cells in normal breast tissue on the same slide were also counted. Since whole nuclei are not included in 4-mm-thick sections, the signal number is lower in paraffin sections than in complete nuclei (e.g., in cytology imprint preparations). To control for this under-representation, the FISH values for normal nuclei in the same section were used. In some cases where too few normal epithelial cells were present, lymphocyte and/or stromal cell nuclei were used as normal controls. Interphase analysis values in morphologically normal breast epithelium with the c-myc locus-specific probe used in this study were taken as normal values. In all cases some normal cells showed more than two signals (most three and occasionally four) in up to 5% of the nuclei. In addition, fewer than two signals (mainly one and occasionally no signal) were seen in normal cells in up to 44% of the nuclei. Some artefactual loss of signals is to be expected, because thin sections do not encompass the entire nucleus. Moreover, the extent to which signals are lost will vary with the type of tissue in relation to the average nuclear diameter. It was also
anticipated that the spatial location of the particular gene studied in each cell might be variable. Lastly, owing to variability in tissue-DNA preparation the possibility of hybridisation inefficiency had to be borne in mind. Definitions of true gain or loss were formulated on the basis of these analyses. Standards were set so that cells with more than two signals in more than 10% of the nuclei or with fewer than two signals in greater than 50% of nuclei were considered to indicate true gain or loss, respectively. These cut-off values are in keeping with those that have been used in other studies.24–26 We believe that these are conservative thresholds, which are likely to underestimate the true incidence of gene aberrations. Borderline or questionable loss or gain was defined as from 45% to 50% or 6% to 10%, respectively. Statistical analysis To evaluate statistical significance the Mann–Whitney test was applied. A P-value of o0.05 was considered significant.
RESULTS Immunohistochemistry Normal breast epithelium showed nuclear positivity in up to 16% of the cells with or without cytoplasmic staining. Hence, cases of APM or AA (Fig. 1) showing nuclear positivity in more than 20% of the cells were counted as positive whether associated with cytoplasmic staining or not. All the cases of APM (33 cases) and AA (15 cases) were positive for c-myc protein, but the mean percentage of nuclear positivity was higher in AA than
Fig. 1 C-myc nuclear positivity of apocrine adenosis of the breast (immunoperoxidase).
C-myc in APM and AA of the breast in APM cases: 50% (range 28–74%) in AA and 37% (range 21–68%) in APM cases. The mean percentage positivity of Ki-67 in AA cases was 3.6% (range 0– 18.5%). The three atypical cases showed a range of 1.5–4% positivity. In APM cases, a mean Ki-67 positivity of 1.3% (range 0–8.5%) was detected. Normal epithelium was available for assessment in 22 cases and showed a mean percentage positivity of 0.9% (range 0– 7%). There was a highly significant association between the percentage of c-myc and Ki-67 nuclear positivity in both APM and AA (P o 0.0001).
Table 2
469
Analysis of c-myc gene in apocrine adenosis cases
Apocrine adenosis (AA) 12 cases
% of cells with o2 signals/nucleus
% of cells with 2 signals/nucleus
% of cells with 42 signals/nucleus
Normal area(s) Range Mean S.D.
33–41 38.2 72.8
56–66 60.3 72.7
0–3 1.5 70.9
AA area(s) Range Mean S.D.
33–41 37.3 72.3
56–63 59.8 72.0
1–6 2.9 71.4
S.D.= standard deviation.
Fluorescence in situ hybridisation The results of 15 cases of APM (Fig. 2) and 12 cases of AA together with the normal epithelium are summarised in Tables 1 and 2. FISH analysis of the c-myc gene in APM The mean percentages of cells with 2 signals/nucleus were 59.8 (72.7) and 60.1 (72.5) with ranges of 56–66
Fig. 2 C-myc gene copy (red signals) in apocrine metaplasia of the breast. Nuclei (blue) show mainly 2 signals/nucleus. Also cells with 1 – 4 signals/nucleus are seen (FISH).
Table 1 Analysis of c-myc gene in apocrine metaplasia cases Apocrine metaplasia (APM) 15 cases
% of cells with o2 signals/nucleus
% of cells with 2 signals/nucleus
% of cells with 42 signals/nucleus
Normal area(s) Range Mean S.D.
33–42 38.8 72.8
56–66 59.8 72.7
0–3 1.4 71.0
APM area(s) Range Mean S.D.
34–42 37.9 72.6
56–64 60.1 72.5
1–4 2.0 71.1
S.D.=standard deviation.
and 56–64 in normal breast epithelium and APM, respectively. The mean percentages of cells with 42 signals/nucleus were 1.4 (71.0) and 2.0 (71.1) with ranges of 0–3 and 1–4 in normal breast epithelium and APM, respectively. The mean percentages of cells with o2 signals/nucleus were 38.8 (72.8) and 37.9 (72.6) with ranges of 32–42 and 34–42 in normal epithelium and APM, respectively. FISH analysis of the c-myc gene in AA The mean percentages of cells with 2 signals/nucleus were 60.3 (72.7) and 59.8 (72.0) with ranges of 56–66 and 56–63 in normal breast epithelium and AA, respectively. The mean percentages of cells with 42 signals/nucleus were 1.5 (70.9) and 2.9 (71.4), with ranges of 0–3 and 1–6 in normal breast epithelium and AA, respectively. In no case of AA 44 signals/nucleus were found, but one case showed borderline gain (amplification) with 42 signals/nucleus in 6% of the cells. The mean percentages of cells with o2 signals/ nucleus were 38.2 (72.8) and 37.3 (72.3) with ranges of 33–41 and 33–41 in normal breast epithelium and AA, respectively.
DISCUSSION Apocrine epithelium in the breast has been intensely studied owing to its role in gross cystic disease of the breast and its possible relationship to breast carcinoma. Wellings et al.27 considered apocrine cells to be the progenitors of breast cystic disease. They reported that apocrine epithelium associated with the accumulation of secretions might favour the progressive unfolding of lobules, formation of microcysts and, finally, the appearance of macrocysts. The relationship between apocrine cystic changes and breast carcinoma has always been controversial. Clinical follow-up studies of
470
The Breast
women with breast apocrine cystic disease have shown an increased risk of subsequent breast carcinoma.28–30 Wellings and Alpers28 have discovered that cystic apocrine metaplasia is more common in cancer-associated breasts than in the normal breast, and also that foci of apocrine cysts are more numerous in cases of breast carcinoma. Some hypotheses for the relationship between apocrine epithelium and carcinoma of the breast have been proposed by Haagensen.8 Apocrine epithelium may be a precursor of malignant transformation; it may reflect a response to the same stimulus which promotes carcinomas; or, thirdly, apocrine metaplasia could reflect an instability of the breast epithelium, which causes the development of alterations with a higher propensity for cancer. The data relating to the relationship between apocrine epithelium, cysts and cancer of the breast are unclear, and the stimuli which lead to these processes are unknown. Metaplasia throughout the human body is associated with the development of carcinoma in lung, stomach and cervix. Haagensen31 reported a five-fold increase in the incidence of breast carcinoma associated with papillary apocrine metaplasia in fibrocystic change. Also, in a long-term follow-up study of patients diagnosed with atypical AA, there was an increased relative risk of 5.5 for the subsequent development of breast carcinoma.14 Abnormal oncogene expression has been discovered before in apocrine metaplastic epithelium of the breast by Papamichalis32 using an antibody to c-myc oncoprotein, by Agnantis33 using antibodies to ras and c-myc oncoproteins and by McCann34 and Wells13 using antibodies to c-erbB2 oncoprotein. In keeping with these findings, using the pressure cooker, heat-mediated, antigen retrieval method rather than enzymatic pretreatment of tissue sections, we found positivity for c-myc in a higher percentage of cells in all cases of AA and APM tested than normal epithelium. The percentage of nuclear positivity was higher in AA than in APM. Increased c-myc expression may play a part in the early, intermediate or late stages of malignant transformation of normal cells, as demonstrated by in vitro cell transformation studies and in vivo assays of tumorigenicity and metastasis.1,35,36 It is therefore tempting to speculate that its activation in apocrine metaplastic epithelium, particularly in AA, might have a role in their abnormal proliferation and their conversion to a malignant state. In addition, it has been thought that c-myc acts as a bivalent regulator of both cell proliferation and apoptosis, depending on the availability of growth factors. If cells driven by overactivity of c-myc do not have sufficient growth factors in their environment, they undergo apoptosis depending on the avail-
ability of other apoptosis-related factors.37,38 As apocrine metaplastic epithelium usually lines type-1 breast cysts, which contain high amounts of growth factors,29 this suggests a proliferative role for increased c-myc expression in APM. It is of interest that cases with increased mean c-myc expression also show an increased mean proliferative index with the proliferation antigen Ki-67, which was highly significant statistically. This suggests that in these cases, c-myc expression is indeed related to increased cell proliferation. Investigation of the key genetic events involved in neoplastic development and progression is one of the critical areas of research that may lead to significant progress in our understanding of human carcinogenesis and the development of clinical applications. Although precursor lesions can be readily harvested from some tissues, such as a macroscopically visible adenoma adjacent to a colonic adenocarcinoma, noninvasive breast epithelial lesions such as APM and AA are microscopic and are usually only recognised after the tissue has been fixed in formalin and embedded in paraffin. Until recently, investigations have been limited to epidemiological associations with morphology and immunohistochemical identification of altered protein expression. The advent of FISH analysis of formalinfixed paraffin-embedded tissue sections has made the localisation of cytogenetic aberrations to particular cell types possible because the nuclei are visualised in their normal histological context. The nonuniformity of changes within a cell population provides a measure of genetic instability, which is assumed to be a marker of tumour progression.39,40 In addition, the use of sitespecific DNA probes against chromosomal loci and regions has permitted the identification of structural and numerical genetic aberrations. Identification of such aberrations in potentially premalignant lesions would add further weight to the identification of these lesions as biologically premalignant. Using the FISH methodology, c-myc gene copy amplification has been reported in breast cancer.19,20 In the present study we have performed FISH on archival tissue sections to test whether amplification of the c-myc gene is present in a series of APM and AA. FISH on paraffin sections underestimates the true gene copy number owing to sectioning of the nuclei. Although isolated nuclei give a more accurate assessment of gene copy number41 in some situations, this was impossible in the present study owing to the focal and microscopic nature of APM and AA. Preservation of tissue architecture is essential for their identification. We attempted to control for nuclear sectioning artefacts by assessing corresponding normal areas within the same section for each case.
C-myc in APM and AA of the breast None of the APM and AA cases, apart from one case of AA, showed amplification of the c-myc gene. This one case of AA showed borderline gain (amplification) of the c-myc gene, with 6% of cells showing 3–4 signals/ nucleus. As the percentage of cells showing 3–4 signals/ nucleus was fewer than 10%, we are cautious about interpreting this as definite amplification of the gene, but it may represent a low level of amplification in a subset of cells. The absence of c-myc amplification according to the FISH technique is intriguing and does not explain the finding of c-myc overexpression on immunohistochemistry, which may have been a consequence of amplification of the gene. All the APM and AA cases used for cmyc FISH analysis showed overexpression of c-myc protein in more than 20% of the cell population. This may be explained if the gene is being stimulated or switched on by certain stimuli, such as growth factors or hormones without being amplified. Wolman39 reported that increased expression of the protein in the absence of amplification of the coding gene could result from modification in transcription or translation, possibly coded by genes on other chromosomes. FISH will not detect gene activation if it occurs as a result of increased gene transcription with no change in gene copy number. Interestingly, the study of CGH in papillary apocrine metaplasia17 did not show any genetic abnormalities on chromosome 8, including the c-myc locus, in agreement with our results showing absence of amplification. While studies of c-myc in mouse systems have suggested that alterations to the gene may be an important early event in the development of tumours,42 the evidence from human breast carcinomas has been less conclusive. Amplification of the c-myc gene has been found in approximately 16% of breast carcinomas.43,44 The second of these reviews44 found that c-myc amplification in patients with breast cancer correlates with aggressive features and/or poor prognosis, suggesting that amplification of the gene may occur as a late event in breast cancer. Our finding of lack of c-myc amplification in cases of APM and AA that are showing overexpression of the cmyc protein is in accordance with the findings of Escot45 who reported overexpression of c-myc without amplification of the gene in 80% of primary breast carcinomas. Also, when mRNA and protein expression of c-myc have been determined, poor correlation has been demonstrated between these levels and the extent of expression and amplification of the gene.46,47 Pietilainen4 reported that expression of c-myc at the protein level occurs in well-differentiated rather than poorly differentiated breast carcinomas. This led him to suggest that c-myc protein expression, in contrast to gene
471
amplification, is involved in early stages of breast cancer development. In agreement with these findings in breast cancer it seems that there is overexpression of c-myc at the protein level but no amplification of the gene is the case in APM and AA. In previous studies, lack of amplification of other oncogenes has also been reported, e.g., absence of cerbB2 amplification in some breast48 and bladder49 carcinomas that show overexpression of the encoded protein on immunocytochemistry. In conclusion, the findings of abnormal c-myc oncogene expression in apocrine metaplastic epithelium without detectable amplification of the gene indicate that overexpression of c-myc but not its amplification may occur as an early event in breast oncogenesis. Moreover, amplification of the c-myc oncogene appears to be a late event in the pathogenesis of breast cancer.
Acknowledgements We would like to acknowledge support from the Joint Research Board, St. Bartholomew’s Hospital, London. Thanks also go to P. Wencyk and S. Jordan for their technical assistance,to R. Hamoudi for statistical assistance and to Dr. C. Sowter, Dr. P. Hallam and Dr. S. Wilkes for their help with the fluorescence microscopy and preparation of the photography.
REFERENCES 1. Evan GI, Wyllie AH, Gilbert CS et al. Induction of apoptosis in fibroblasts by c-myc protein. Cell 1992; 69: 1–20. 2. Meyers SL, O’Brien MT, Smith T et al. Analysis of the int-1, int-2, c-myc, and neu oncogenes in human breast carcinomas. Cancer Res 1990; 50: 5911–5918. 3. Berns EM, Klijn JG, van Putten et al. C-myc amplification is a better prognostic factor than HER2/neu amplification in primary breast cancer. Cancer Res 1992; 1: 1107–1113. 4. Pietilainen T, Lipponen P, Aaltomaa S et al. Expression of c-myc proteins in breast cancer as related to established prognostic factors and survival. Anticancer Res 1995; 15: 959–964. 5. Rosai J. Ackerman’s Surgical Pathology, 8th Edition. New York: Mosby-Year Book, 1996; 1565–1659. 6. Page DL, Vander Zwaag R, Rogers LW et al. Relation between component parts of fibrocystic disease complex and breast cancer. J Natl Cancer Inst 1978, 61: 1055–1063. 7. Tavassoli FA, Norris HJ. Intraductal apocrine carcinoma: a clinicopathological study of 37 cases. Mod Pathol 1994; 7: 813–818. 8. Haagensen Jr DE Is cystic disease related to breast cancer? Am J Surg Pathol 1991; 15: 687–694. 9. Harrington E, Lesnik G. The association between gross cysts of the breast and breast cancer. Breast 1980; 7: 13–17. 10. Roberts MM, Jones V, Elton RA et al. Risk of breast cancer in women with history of benign disease of the breast. Br Med J 1984; 288: 275–278. 11. Dupont WD, Page DL. Risk factors for breast cancer in women with proliferative disease. N Engl J Med 1985; 312:146–151.
472
The Breast
12. Page DL, Dupont WD, Jensen RA. Papillary apocrine change of the breast: association with atypical hyperplasia and risk of breast cancer. Cancer Epidemiol Biomarkers Prev 1996; 5: 29–32. 13. Wells CA, McGregor IL, Makunura CN et al. Apocrine adenosis: a precursor of aggressive breast disease. J Clin Pathol 1995; 48: 737–742. 14. Seidman JD, Ashton M, Lefkowitz M. Atypical apocrine adenosis of the breast: clinicopathological study of 37 patients with 8.7-year follow up. Cancer 1996; 77: 2529–2537. 15. Izuo M, Okagaki T, Richart R et al. DNA content in ‘apocrine metaplasia’ of fibrocystic disease of the breast. Cancer 1971; 27: 643–650. 16. Vasiu R, Olinici CD. DNA nuclear content of apocrine cells in fibrocystic mastopathy and breast cancer. Morphol Embryol 1990; 36: 43–47. 17. 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: 207–214. 18. Pauletti G, Godolphin W, Press MF et al. Detection and quantitation of HER-2/neu gene amplification in human breast cancer in archival material using fluorescence in situ hybridisation. Oncogene 1996; 13: 63–72. 19. Persons DL, Borelli KA, Hsu PH. Quantitation of HER-2/neu and c-myc gene amplification in breast carcinoma using fluorescence in situ hybridisation. Mod Pathol 1997; 10: 720–727. 20. Visscher DW, Wallis T, Awussah S et al. Evaluation of MYC and Chromosome 8 copy number in breast carcinoma by interphase cytogenetics. Genes Chromosomes Cancer 1997; 18: 1–7. 21. Hsu S-M, Raine L, Fanger H. Use of avidin-biotin-peroxidase complex (ABC) in immmunoperoxidase techniques: a comparison between ABC and unlabelled antibody (PAP) procedures. J Histochem Cytochem 1981; 29: 577–580. 22. Norton AJ, 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: 371–379. 23. Dekken HV, Bosman FT, Teijgeman R et al. Identification of numerical chromosome aberration in archival tumours by in situ hybridization in routine paraffin sections: evaluation of 23 phaeochromocytomas. J Pathol 1993; 171: 161–171. 24. Harrison M, Magee HM, O’Loughin J et al. Chromosome 1 aneusomy, identified by interphase cytogenetics, in mammographically detected ductal carcinoma in situ of the breast. J Pathol 1995, 175: 303–309. 25. Micale MA, Visscher W, Gulino SE et al. Chromosomal aneuploidy in proliferative breast disease. Hum Pathol 1994; 25: 29–35. 26. Visscher DW, Wallis TL, Crissman JD. Evaluation of chromosome aneuoploidy in tissue sections of preinvasive breast carcinoma using interphase cytogenetics. Cancer 1996; 77: 315–320. 27. Wellings SR, Jensen HM, Marcum RG. An atlas of subgross pathology of the human breast with special reference to possible precancerous lesions. J Natl Cancer Inst 1975; 55: 231–243. 28. Wellings SR, Alpers CE. Apocrine cystic metaplasia: subgross pathology and prevalence in cancer associated versus random autopsy breasts. Hum Pathol 1987; 18: 381–386. 29. Ness JC, Sedghinasab M, Moe RE et al. Identification of multiple proliferative growth factors in breast cyst fluid. Am J Surg 1993; 166: 237–243.
30. Bruzzi P, Dogliotti L, Naldoni C et al. Cohort study of association of risk of breast cancer with cyst type in women with gross cystic disease of the breast. BMJ 1997; 314: 925–928. 31. Haagensen CD. Apocrine epithelium. In: Diseases of the Breast, 3rd Edition. Philadelphia: W.B. Saunders, 1986; 82–101. 32. Papamichalis G, Francia K, Karachaliou FE et al. Expression of the c-myc oncoprotein in human metaplastic epithelial cells of fibrocystic disease. Anticancer Res 1988; 8: 1217–1222. 33. Agnantis NJ, Mahera H, Maounis N et al. Immunohistochemical study of ras and myc oncoproteins in apocrine breast lesions with and without papillomatosis. Eur J Gynaecol Oncol 1992; 13: 309–315. 34. McCann A, Johnston PA, Dervan PA et al. c-erbB2 oncoprotein expression in malignant and non-malignant breast tissue. Int J Med Sci 1989; 158: 137–140. 35. Spandidos DA, Anderson MLM. A study of the mechanism of carcinogenesis by gene transfer of oncogenes into mammalian cells. Mutat Res 1987; 185: 271–291. 36. Wyllie AH, Rose KA, Morris RG et al. Rodent fibroblast tumours expressing human myc and ras genes: growth, metastasis and endogenous oncogene expression. Br J Cancer 1987; 56: 251–259. 37. Boise LH, Gottschalk AR, Quintans J et al. Bcl-2 and Bcl-2 related proteins in apoptosis regulation. Curr Top Microbiol Immunol 1995, 200: 107–121. 38. Cummings MC, Winterford CM, Walker NI. Apoptosis. Am J Surg Pathol 1997; 21: 88–101. 39. Wolman SR, Pauley RJ, Mohamed AN et al. Genetic markers as prognostic indicators in breast cancer. Cancer 1992; 70: 1765–1774. 40. Nicolson GL. Tumour cell instability, diversification, and progression to the metastatic phenotype: from oncogene to oncofetal expression. Cancer Res 1987; 47: 1473–1487. 41. Emmerich P, Jauch A, Hofmann MC et al. Interphase cytogenetics in paraffin embedded sections from human testicular germ cell tumour xenografts and in corresponding cultured cells. Lab Invest 1989; 61: 235–242. 42. Edwards PAW, Ward JL, Bradbury JM. Alteration of morphogenesis by the v-myc oncogene in transplants of mouse mammary gland. Oncogene 1988; 2: 407–412. 43. Liao DJ, Dickson RB. C-myc in breast cancer. Endocr Relat Cancer 2000 ; 7: 143–164. 44. Deming SL, Nss SJ, Dickson RB, Trock BJ. C-myc amplification in breast cancer: a meta-analysis of its occurrence and prognostic relevance. Br J Cancer 2000; 83: 1688–1695. 45. Escot C, Theillet C, Lidereau R et al. Genetic alteration of the cmyc protooncogene (MYC) in human primary breast carcinomas. Proc Natl Acad Sci USA 1986; 83: 4834–4838. 46. Locker AP, Dowle CS, Ellis IO et al. C-myc oncogene product expression and prognosis in operable breast cancer. Br J Cancer 1989; 60: 669–672. 47. Walker RA, Senior PV, Jones JL et al. An immunohistochemical and in situ hybridisation study of c-myc and c-erbB-2 expression in primary breast tumours. J Pathol 1989; 158: 97–105. 48. Slamon DJ, Godolphin W, Jones LA et al. Studies of the HER2/ neu proto-oncogene in human breast and ovarian cancer. Science 1989; 244: 707–712. 49. Coombs LM, Pigott DA, Sweeney E et al. Amplification and overexpression of c-erbB-2 in transitional cell carcinoma of the urinary bladder. Br J Cancer 1991; 62: 601–608.