sign in sign up
<<
Home , Estrogen receptor alpha, Estrogen receptor beta

ANTICANCER RESEARCH 25: 1679-1686 (2005)

Normal and Malignant Human Express Immunohistochemically Alpha (ER-·), Beta (ER-‚) and Receptor (PR)

I. MYLONAS*, U. JESCHKE*, N. SHABANI, C. KUHN, S. KRIEGEL, M. S. KUPKA and K. FRIESE

First Department of Obstetrics and Gynecology, Ludwig, Maximilians, University Munich, Germany

Abstract. Human endometrium expresses estrogen (ER) and Endometrial cancer is becoming the most common progesterone(PR) receptors , which are related to autocrine and gynecologic malignancy in the western world and occurs in paracrine processes that respond to estrogen and progesterone. reproductive and postmenopausal women (1, 2). It mostly The ER and PR expression and distribution pattern may play appears in women with conditions such as estrogen-only an important role in endometrial function and pathogenesis. hormone replacement therapy (HRT), obesity, polycystic The aim of this study was to evaluate the distribution pattern disease, nulliparity, estrogen-producing tumors and of ER-·, ER-‚ and PR in normal (n=15) and malignant anovulation (1-3). (n=11) human endometrial tissue. Commercially available Estrogen (ER) and progesterone (PR) receptors belong monoclonal antibodies against ER-·, ER-‚ and PR were used. to the nuclear receptor superfamily. They are - The distribution of the steroid receptors was evaluated using the dependent transcriptional factors, which can bind to IRS-score and the Mann-Whitney rank-sum test was used to different DNA sites to initiate the expression of specific compare the means. Correlation was assessed with the . In addition to this direct activation of target genes, Spearman factor and linear regression analysis. ER-·, ER-‚ indirect mechanisms through contacts with DNA-bound and PR declined significantly (p<0.05) in normal glandular transcription factors (such as AP1 or NF-Î B) have been from proliferative to late secretory phase, although reported (3). The effect of the steroid hormones estrogen the staining intensity of ER-‚ was lower than that of ER-·. ER-·, and progesterone are thought to be mediated through these ER-‚ and PR were also expressed in malignant endometrial receptors. ER and PR are closely related to the occurrence tissue. A significant correlation by regression analysis of ER-· of autocrine and paracrine processes that respond to and ER-‚ was demonstrated, showing a dependence in the estrogen and progesterone (4). In several studies, both expression of these steroid receptors. The ER-·/ER-‚ ratio receptors were measured in paraffin-fixed human decreased significantly from normal to malignant endometrial endometrium (4-8) and isolated glandular epithelial cells (9) tissue (p<0.05), while the ER-‚/ER-· ratio showed statistical using immunohistochemical assays. However, the exact differences within normal endometrial tissue. These results mechanism for the proliferative effects of on the showed the presence of steroid receptors in normal and endometrium and their role in neoplasia still remain malignant human endometrium, indicating a significant role in unknown. For several years, it was generally believed that endometrial and malignant transformation. just one single ER receptor existed. However, the discovery of a new with specificity for estrogens has induced new insights in the estrogen signaling system (10). *contributed equally to this work. The novel receptor ER-‚ has a high (approx. 95%) homology in the DNA binding domain, but only 55% homology in the Correspondence to: Dr. med. Ioannis Mylonas, 1st Department of ligand binding domain compared to the classical ER (ER-·). Obstetrics and Gynecology, Ludwig Maximilians University Munich, ER-· can bind with a high affinity and react with a (Director: Prof. Dr. med. K. Friese), Maistrasse 11, 80337 Munich, consensus estrogen response element (ERE), stimulating Germany. Tel: ++49-5160-4266, Fax: ++49-5160 4916, e-mail: of ER target genes (11). [email protected] or [email protected] The ER and PR expression and distribution patterns might Key Words: Endometrium, endometrial cancer, immunohistochemistry, play an important role in normal endometrial function and (ER-·), (ER-‚), pathogenesis and the expression and relationship of the two (PR). distinct ER and PR could be of essential clinical implications.

0250-7005/2005 $2.00+.40 1679 ANTICANCER RESEARCH 25: 1679-1686 (2005)

Table I. Antibodies used for immunohistochemical characterization of and late secretory phase (days 23-28, n=4), as previously described endometrial glandular cells. (ER=estrogen receptor, PR=progesterone (7, 9). Additionally, endometrioid adenocarcinomas grade 1 (G1; receptor). n=3) and grade 2 (G2; n=8) were analyzed.

Antibody Clone Isotype Dilution Source Immunohistochemistry. Immunohistochemistry was performed using a combination of microwave-oven heating and the standard ER-· 1D5 mouse IgG1 1:150 Immunotech, streptavidin-biotin-peroxidase complex, using the mouse-IgG- Hamburg, Germany Vectastain Elite ABC kit (Vector Laboratories, Burlingame, CA, USA). Mouse monoclonal antibodies used for the experiments are ER-‚ PPG5/10 mouse IgG2 1:50 Serotec, Oxford, a listed in Table I. Sections of human tissue and United Kingdom normal colon tissue samples were used for positive controls. Briefly, paraffin-fixed tissue sections were dewaxed using xylol PR 10A9 mouse IgG2a 1:50 Immunotech, Hamburg, Germany for 15 min, rehydrated in an alcohol raw, and subjected to antigen retrieval on a high setting for 10 min in a pressure cooker in sodium citrate buffer (pH 6.0), containing citrate acid 0.1M and sodium citrate 0.1M in distillate aqua. After cooling, the slides were washed twice in PBS. Endogenous peroxidase activity was quenched by immersion in 3% hydrogen peroxide (Merck) in The ER status is believed to provide prognostic information methanol for 20 min. Non-specific binding of the primary antibodies was blocked by incubating the sections with "diluted independent of tumor stage and grade in women with normal serum" for 20 min at room temperature. Sections were then endometrial carcinoma (12-14). The ER-‚/ER-· mRNA incubated at room temperature for 60 min with the primary ratio was high in advanced invasive carcinoma. Additionally, antibodies. ER-· and PR were diluted in dilution medium (Dako, Western blot analysis demonstrated that ER-‚ was highly Glostrup, Denmark), while ER-‚ was diluted in PBS. After washing expressed in comparison with ER-· in endometrial cancer with PBS, the slides were incubated in "diluted biotinylated serum" with severe myometrial invasion, suggesting that ER-‚ is for another 30 min at room temperature. After incubation with the important in the progression of myometrial invasion (15). avidin-biotin peroxidase complex (reagent ABC) for a further 30 min and a repeated washing step with PBS, visualization was The intact synchronized expression of ER-‚ interacting with performed with substrate and chromagen 3,3'-diaminobenzidine ER-· might be disrupted in neoplastic endometrium (16), (DAB; Dako) for 8-10 min. The slides were further counterstained playing an important role in endometrial pathogenesis. with Mayer’s acidic hematoxylin and washed in an alcohol raw (50- Available data on the expression of ER-‚ protein in 98%). After xylol treatment, the slides were covered. Negative human endometrial tissue have mainly reported the controls were performed by replacing the primary antibody with expression of the corresponding mRNAs in normal (17-19) normal mouse serum. Positive cells showed a brownish color and and neoplastic human endometrium (15, 16, 20-22). negative controls as well as unstained cells were blue. Immunohistochemical detection has been difficult due to Evaluation and statistical analysis. The intensity and distribution of lack of available monoclonal antibodies against ER-‚. the specific immunohistochemical staining reaction was evaluated Recently, the immunohistochemical expression of ER-‚ in using a semi-quantitative method (IRS-score), as previously normal endometrium was demonstrated using a polyclonal described (25) and used in the evaluation of endometrial steroid antibody (19) and a commercially available monoclonal receptor expression (9), as well as the expression of cathepsin D antibody (8), whereas immunohistochemical analysis of well-differentiated endometrial carcinomas has not been achieved. → Therefore, the aim of this study was the evaluation of the Figures 1-8. Immunohistochemical expression of ER-·, ER-‚ and PR. distribution patterns of ER-·, ER-‚ and PR using recently During the proliferative phase, both glandular epithelial and stromal cells available commercial monoclonal antibodies in normal and demonstrated intense nuclear immunoreactivity for ER-· (Figure 1, x125), malignant endometrial tissue. while it decreased continuously till the late secretory phase. Endometrioid adenocarcinomas also expressed ER-·, but with a lower intensity (Figure Materials and Methods 2, x400). The PR immunohistochemical reaction showed a similar declining pattern between proliferative phase (Figure 3, x400) and late secretory phase, while it was also expressed in stromal cells and Tissue samples. Samples of normal human endometrium were endometrial adenocarcinomas (Figure 4, x400). ER-‚ immunostaining obtained from 15 premenopausal, non-pregnant patients patterns were also expressed in endometrial glandular and stromal cells undergoing gynecological surgery either by D&C or hysterectomy (with a minimal intensity) with a lower immunostaining intensity for benign diseases. All women had had a normal and regular compared to the ER-·. The intensity was higher in the proliferative phase with no hormonal treatment for the 3 months prior (Figure 5, x400), whereas in the late secretory phase, a minimal to no to surgery. Endometrial premenstrual samples were classified staining reaction was observed (Figure 6, x250). G1 carcinomas showed a according to anamnestical and histological dating (23, 24) into low ER-‚ intensity (Figure 7, x400), in contrast to G2 carcinomas that proliferative (days 1-14, n=7), early secretory (days 15-22, n=4) showed minimal or no expression (Figure 8, x250).

1680 Mylonas et al: Steroid Hormone Receptors in Malignant Human Endometrium

1681 ANTICANCER RESEARCH 25: 1679-1686 (2005)

Figure 9

Figure 10

Figure 11

1682 Mylonas et al: Steroid Hormone Receptors in Malignant Human Endometrium

(26) and CA-125 (27) in human endometrium. The IRS score was calculated as follows: IRS=SI x PP, where SI is the optical stain intensity (graded as 0 = no, 1 = weak, 2 = moderate and 3 = strong staining) and PP the percentage of positive-stained cells. The PP was estimated by counting approximately 200 cells and it was defined as 0 = no staining, 1 = <10%, 2 = 11-50%, 3 = 51- 80% and 4 = >81%. The samples were evaluated by two different observers and the mean of the results were used. The Mann- Whitney rank-sum test was used to compare the means of the different IRS scores (SPSS, Chigaco, IL, USA). Spearman- and regression analysis was used to assess any correlation between the steroid receptors in endometrial cancer. The ratios of ER-·/ER-‚ and ER-‚/ER-· were calculated and the means were compared using the Mann-Whitney rank-sum test. Significance was assumed at p<0.05.

Results Figure 12. Ratio between ER-·/ER-‚ and ER-‚/ER-·. The ER-·/ER-‚ All three steroid receptors were immunohistochemically showed a significant decline from the normal proliferative phase to expressed in normal and malignant endometrial tissue. malignant endometrial G1 (*) and G2 (**) tumors (p<0.05). The ER- However, adenocarcinomas with grade 1 and grade 2 showed ‚/ER-· ratio showed a just significant change from proliferative to early (*) and late secretory (**) phases, while no statistical differences were observed a lower staining reaction than normal endometrium. During in endometrial carcinomas. Mean±SEM. 1=proliferative, 2=early secretory, the proliferative phase, both glandular epithelial and stromal 3=late secretory, 4=adenocarcinoma G1, 5=adenocarcinoma G2 cells demonstrated nuclear immunoreactivity for ER-·, ER-‚ and PR, confirming previous results (4-8, 19). While the staining reaction of ER-· in biopsies in the proliferative phase was intense (Figure 1), it decreased continuously till the late secretory phase. Additionally, the ER-· expression The IRS score for ER-· declined significantly (p<0.05) was low in adenocarcinomas grade 2 (Figure 2). The PR in the glandular epithelium from the proliferative to early immunohistochemical reaction showed in normal secretory phase (Figure 9), reaching the lowest expression endometrium a similar declining pattern between the in the late secretory phase (p<0.05). Additionally, a proliferative phase (Figure 3) and late secretory phase, with significantly lower expression was demonstrated between lower intensity in endometroid adenocarcinomas (Figure 4). proliferative and early secretory phases compared to The ER-‚ immunostaining patterns were also demonstrated adenocarcinomas grade 2 (p<0.05). The IRS score for ER-‚ in endometrial glandular and stromal cells. However, ER-‚ also declined significantly (p<0.05) in the glandular immunostaining intensity was lower compared to the ER-· epithelium from the proliferative to early and late secretory staining reaction. The intensity of positive epithelial cells was phases (Figure 10). Adenocarcinomas grade 2 also showed higher in the proliferative phase (Figure 5) than in the late a significantly lower immunohistochemical reaction secretory phase (Figure 6). Adenocarcinomas grade 1 also compared to the proliferative phase (p<0.05). The PR showed a low intensity (Figure 7), in contrast to grade 2 expression decreased significantly (p<0.05) between the carcinomas that showed minimal or no expression (Figure 8). proliferative and late secretory phase (Figure 11), with no statistical differences between normal and malignant endometrial tissue. A correlation was demonstrated only for ER-· and ER-‚ → in endometrial carcinomas (p<0.05). Regression analysis Figures 9-11. IRS-score for ER-·, ER-‚ and PR. The IRS score for ER-· revealed a positive correlation between ER-· and ER-‚ and ER-‚ declined significantly (p<0.03 and p<0.01, respectively) in the 2 glandular epithelium from proliferative to early secretory phase (*), reaching (r =0.800, p<0.001), while no correlation was observed the lowest significant expression in the late secretory phase (p<0.01, compared between ER’s and PR (p=0.057). The ratio of ER-·/ER-‚ to samples from the proliferative phase, **). ER-· during the proliferative was high during the proliferative phase, decreasing phase and early secretory phase was higher than adenocarcinomas G2 (*** significantly (p<0.05) as the endometrial neoplastic and ****, respectively). ER-‚ also showed a significant decline between the transformation developed. Interestingly, the ratio between proliferative and early and late secretory phases (p<0.05, *), while the staining reaction of G2 carcinomas was also lower than the prolifeative phase ER-‚/ER-· showed significant differences in normal human (p<0.05, **). The PR expression also decreased significantly (p<0.05) endometrial tissue (p<0.05) between the proliferative and between the proliferative and late secretory phases (*). Mean±SEM. early as well as late secretory phases (Figure 12).

1683 ANTICANCER RESEARCH 25: 1679-1686 (2005)

Discussion substantially lower expression of steroid receptors in endometrial cancer. Therefore, cancer cells might transform The endometrium is one principal target tissue of the to more aggressive cells types during advancement of pituitary-gonadal axis, but has also been recognized as an endometrial cancer and lose their sex steroidal dependency. endocrine organ. Endometrial carcinomas can arise from Interestingly, a high level of ER-· has been demonstrated to hyperplastic precursor lesions, probably due to unopposed be beneficial in the control of female cancers due to the stimulation by estrogen (2). The estrogen-dependent type of observed inhibitory effect on angiogenic pathways (35). It is endometrial carcinomas consist of histopathologically well- not known if this is true for endometrial carcinomas. differentiated endometrioid carcinomas (G1-G2), although Studies in normal and pathological human endometrium some types of endometrial cancer might be estrogen- have reported mostly on the ER-‚ mRNA expression (15- independent (2). There is a clinical need for a simple and 22). Recently, the expression of ER-‚ by efficient marker of the activity of this tissue, especially with immunohistochemical means in normal human endometrium regard to endometrial pathogenesis and carcinogenesis. was demonstrated using a polyclonal (19) and a monoclonal Several studies have measured ER and PR in normal and (8) antibody. Using a commercially available monoclonal malignant human endometrium using immunohistochemical antibody against ER-‚ antigen, we demonstrated the assays. Our results confirm the cyclical variations of ER and immunolocalization of ER-‚ in malignant human PR in normal endometrial tissue, as described by several endometrial tissue. The presence and distribution pattern of authors (4-9). These immunohistochemical results show the ER-‚ in glandular epithelial cells is in agreement with presence of steroid receptors in human epithelium, indicating previous reports (15, 17-22). Our results indicate that that these cells respond to estrogen and progesterone (4). It is glandular ER-‚ expression predominantly occurs during the generally accepted that estradiol up-regulates ER and PR, proliferative phase, declining as the menstrual phase while progesterone down-regulates both receptors (5). continues, as demonstrated for the ER-· expression (8). Recently, a significant decline of ER-· expression was Differences in tissue distribution of ER-· and ER-‚ during demonstrated in glandular epithelial cells after stimulation the menstrual cycle suggest a substantial role in the with (an ), and the modulation and function of estrogen activity in human , whereas PR increased significantly after stimulation endometrial tissue. However, the intensity of the ER-‚ with both substances (28), suggesting a functional association staining reaction was lower compared to ER-· and PR, as between these two steroid receptors. demonstrated through the combination of a semi- The ER status is believed to provide prognostic information quantitative analysis and statistical regression procedure (8). independent of tumor stage and grade in women with Since the ratio of ER-·/ER-‚ is uniform without statistical endometrial carcinoma (12-14). Alteration in estrogen differences in normal endometrial tissue, several regulatory signaling pathways may occur during endometrial functions between both ER’s may be involved. ER-‚ could tumorigenesis, and provide evidence that ER-· expression have a different action and activity compared to ER-·, since may play an important role in the regulation of PR in normal it exerts opposite transcriptional effects after binding to and malignant endometrium (20). PR has also been estrogens and than its ER-· counterpart (36). implicated in the development of endometrial cancer, exerting Differences in the ER-·/ER-‚ ratio could have important its effects by its two receptors PR-A and PR-B, and a decrease functional implications, probably as a result of the different in PR-B has been observed in poorly-differentiated binding characteristics (11). The immunohistochemical ratio endometrial cancer cell lines (29). These observations suggest ER-‚/ER-· demonstrated statistically significantly differences an association between malignancy and disturbances in the during menstrual phases in normal human endometrium, relative abundance of ER and PR subtypes. It is thought that suggesting an important function. Additionally, the PR-A in the endometrium downregulates estrogen action by significant differences of the immunohistochemical ratio preventing ER-· from transactivation. In contrast, the ER-·/ER-‚ between normal and neoplastic endometrial activated PR-B isoform acts as an endometrial estrogen- tissue suggests an important role in malignant transformation. (30). However, PR, in contrast to ER, is suggested to Therefore, a synchronized expression of ER-· and ER-‚ be more predictive of disease-free survival (31, 32). The ER seems to be essential for estrogen-related transcriptional and PR expression was correlated with the nuclear grade, activity in target organs. A disrupted expression of ER-· and being higher in G1 tumors (33). There was a correlation -‚ might play an important role in endometrial pathogenesis between receptor concentration and grade, with a significant and carcinogenesis (16). We demonstrated a significant difference between grades 1 and 2 versus grade 3 (34). We correlation in the expression pattern of ER-· and ER-‚, demonstrated significant differences of ER-· and ER-‚ suggesting an important role in maintaining the functional between normal human endometrial cells in the proliferative status of the ER’s. Interestingly, ER-· showed a significant phase compared to adenocarcinomas grade 2, suggesting a decline with tamoxifen, an antiestrogen and genistein, a

1684 Mylonas et al: Steroid Hormone Receptors in Malignant Human Endometrium phytoestrogen, whereas PR expression increased significantly 11 Kuiper GG, Carlsson B, Grandien K, Enmark E, Haggblad J, (28). Whether ER-‚ expression is influenced by PR and/or Nilsson S and Gustafsson JA: Comparison of the ligand binding progesterone remains to be elucidated. specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinology 138: 863-870, 1997. In conclusion, we demonstrated ER-·, ER-‚ and PR 12 Creasman WT: Prognostic significance of hormone receptors in expressions in normal and neoplastic human endometrium endometrial cancer. Cancer 71: 1467-1470, 1993. using immunohistochemical means. A cyclical variation was 13 Chambers JT, MacLusky N, Eisenfield A, Kohorn EI, Lawrence observed during the menstrual cycle, while endometrioid R and Schwartz PE: Estrogen and progestin receptor levels as adenocarcinomas expressed all three steroid receptors but prognosticators for survival in endometrial cancer. Gynecol with lower intensity than proliferative endometrium. A Oncol 31: 65-81, 1988. significant correlation by regression analysis of ER-· and 14 Pertschuk LP, Masood S, Simone J, Feldman JG, Fruchter RG, ER-‚ was also demonstrated in malignant endometrial Axiotis CA and Greene GL: Estrogen receptor immuno- cytochemistry in endometrial carcinoma: a prognostic marker tissue, showing dependence in the expression of all these for survival. Gynecol Oncol 63: 28-33, 1996. steroid receptors. These results show the presence of steroid 15 Takama F, Kanuma T, Wang D, Kagami I , Mizunuma H: receptors in human epithelium, indicating that these cells Oestrogen receptor beta expression and depth of myometrial respond to estrogen and progesterone, playing a significant invasion in human endometrial cancer. Br J Cancer 84: 545- role in endometrial physiology and tumorigenesis. 549, 2001. 16 Fujimoto J, Sakaguchi H, Aoki I, Toyoki H and Tamaya T: References Clinical implications of the expression of estrogen receptor- alpha and -beta in primary and metastatic lesions of uterine endometrial cancers. Oncology 62: 269-277, 2002. 1 Creasman WT: Endometrial cancer: incidence, prognostic factors, 17 Rey JM, Pujol P, Dechaud H, Edouard E, Hedon B and diagnosis, and treatment. Semin Oncol 24: S1-140-S141-150, 1997. Maudelonde T: Expression of oestrogen receptor-alpha splicing 2 Sherman ME: Theories of endometrial carcinogenesis: a variants and oestrogen receptor-beta in endometrium of multidisciplinary approach. Mod Pathol 13: 295-308, 2000. infertile patients. Mol Hum Reprod 4: 641-647, 1998. 3 Oehler MK, Rees MC and Bicknell R: and the 18 Matsuzaki S, Fukaya T, Suzuki T, Murakami T, Sasano H and endometrium. Curr Med Chem 7: 543-560, 2000. Yajima A: Oestrogen receptor alpha and beta mRNA 4 Snijders MP, de Goeij AF, Debets-Te Baerts MJ, Rousch MJ, expression in human endometrium throughout the menstrual Koudstaal J and Bosman FT: Immunocytochemical analysis of cycle. Mol Hum Reprod 5: 559-564, 1999. oestrogen receptors and progesterone receptors in the human 19 Lecce G, Meduri G, Ancelin M, Bergeron C and Perrot- throughout the menstrual cycle and after the . Applanat M: Presence of estrogen receptor beta in the human J Reprod Fertil 94: 363-371, 1992. endometrium through the cycle: expression in glandular, 5 Lessey BA, Killam AP, Metzger DA, Haney AF, Greene GL stromal, and vascular cells. J Clin Endocrinol Metab 86: 1379- and McCarty KS Jr: Immunohistochemical analysis of human 1386, 2001. uterine estrogen and progesterone receptors throughout the 20 Saegusa M and Okayasu I: Changes in expression of estrogen menstrual cycle. J Clin Endocrinol Metab 67: 334-340, 1988. receptors alpha and beta in relation to progesterone receptor 6 Garcia E, Bouchard P, De Brux J, Berdah J, Frydman R, and pS2 status in normal and malignant endometrium. Jpn J Schaison G, Milgrom E and Perrot-Applanat M: Use of Cancer Res 91: 510-518, 2000. immunocytochemistry of progesterone and estrogen receptors 21 Utsunomiya H, Suzuki T, Harada N, Ito K, Matsuzaki S, Konno for endometrial dating. J Clin Endocrinol Metab 67: 80-87, 1988. R, Sato S, Yajima A and Sasano H: Analysis of estrogen 7 Jones RK, Bulmer JN and Searle RF: Immunohistochemical receptor alpha and beta in endometrial carcinomas: correlation characterization of proliferation, oestrogen receptor and with ER beta and clinicopathologic findings in 45 cases. Int J progesterone receptor expression in endometriosis: comparison Gynecol Pathol 19: 335-341, 2000. of eutopic and ectopic endometrium with normal cycling 22 Sakaguchi H, Fujimoto J, Aoki I, Toyoki H, Khatun S and endometrium. Hum Reprod 10: 3272-3279, 1995. Tamaya T: Expression of oestrogen receptor alpha and beta in 8 Mylonas I, Jeschke U, Shabani N, Kuhn C, Balle A, Kriegel S, uterine endometrial and ovarian cancers. Eur J Cancer 38: S74- Kupka MS and Friese K: Immunohistochemical analysis of 75, 2002. estrogen receptor alpha (ER-·), estrogen receptor beta (ER-‚) 23 Noyes RW, Hertig AT and Rock J: Dating the endometrial and progesterone receptor (PR) in normal human biopsy. Fertil Steril 1: 3-15, 1950. endometrium. Acta Histochem 106: 245-252, 2004. 24 Dallenbach-Hellweg G and Poulsen H: Atlas der Histopathologie 9 Mylonas I, Speer R, Makovitzky J, Richter DU, Briese V, des Endometriums. Springer, 1985. Jeschke U and Friese K: Immunohistochemical analysis of 25 Remmele W and Stegner HE: Vorschlag zur einheitlichen steroid receptors and glycodelin A (PP14) in isolated glandular Definierung eines immunreaktiven Score (IRS) für den epithelial cells of normal human endometrium. Histochem Cell immunhistochemischen Östrogenrezeptornachweis (ER-ICA) Biol 114: 405-411, 2000. im Mammakarzinomgewebe. Pathologe 8: 138-140, 1987. 10 Kuiper GG, Shughrue PJ, Merchenthaler I and Gustafsson JA: 26 Mylonas I, Makovitzky J, Richter DU, Jeschke U, Briese V and The estrogen receptor beta subtype: a novel mediator of estrogen Friese K: Cathepsin D expression in normal, hyperplastic and action in neuroendocrine systems. Front Neuroendocrinol 19: malignant endometrial tissue: an immunohistochemical analysis. 253-286, 1998. Acta Histochem 105: 245-252, 2003.

1685 ANTICANCER RESEARCH 25: 1679-1686 (2005)

27 Mylonas I, Makovitzky J, Richter DU, Jeschke U, Briese V and 33 Gehrig PA, Van Le L, Olatidoye B and Geradts J: Estrogen Friese K: Immunohistochemical expression of the tumour receptor status, determined by immunohistochemistry, as a marker CA-125 in normal, hyperplastic and malignant predictor of the recurrence of stage I endometrial carcinoma. endometrial tissue. Anticancer Res 23: 1075-1080, 2003. Cancer 86: 2083-2089, 1999. 28 Mylonas I, Jeschke U, Makovitzky J, Winkler L, Richter DU, 34 Friberg LG and Noren H: Prognostic value of steroid hormone Friese K and Briese V: Immunohistochemical expression of receptors for 5-year survival in stage II endometrial cancer. steroid receptors and glycodelin A in isolated proliferative Cancer 71: 3570-3574, 1993. human endometrial glandular cells after stimulation with 35 Ali SH, O'Donnell AL, Balu D, Pohl MB, Seyler MJ, Mohamed tamoxifen and (genistein and ). S, Mousa S and Dandona P: Estrogen receptor-alpha in the Anticancer Res 23: 1119-1125, 2003. inhibition of cancer growth and angiogenesis. Cancer Res 60: 29 Kumar NS, Richer J, Owen G, Litman E, Horwitz KB and 7094-7098, 2000. Leslie KK: Selective down-regulation of progesterone receptor 36 Paech K, Webb P, Kuiper GG, Nilsson S, Gustafsson J, isoform B in poorly differentiated human endometrial cancer Kushner PJ and Scanlan TS: Differential ligand activation of cells: implications for unopposed estrogen action. Cancer Res estrogen receptors ERalpha and ERbeta at AP1 sites [see 58: 1860-1865, 1998. comments]. Science 277: 1508-1510, 1997. 30 Oehler MK, Brand A and Wain GV: Molecular genetics and endometrial cancer. J Br Menopause Soc 9: 27-31, 2003. 31 Fukuda K, Mori M, Uchiyama M, Iwai K, Iwasaka T and Sugimori H: Prognostic significance of progesterone receptor immunohistochemistry in endometrial carcinoma. Gynecol Oncol 69: 220-225, 1998. 32 Kleine W, Maier T, Geyer H and Pfleiderer A: Estrogen and progesterone receptors in endometrial cancer and their Received August 2, 2004 prognostic relevance. Gynecol Oncol 38: 59-65, 1990. Accepted February 4, 2005

1686