HRT and cancer risks

Maturitas 43 Suppl. 1 (2002) S35– S52 www.elsevier.com/locate/maturitas

HRT and cancer risks Hermann P.G. Schneider * Department of Obstetrics and Gynaecology, Uni6ersity of Muenster, Von-Esmarch-Str. 56, ZMBE, D-48149 Muenster, Germany

Abstract Estrogens are important in the growth and differentiation of hormone-responsive tissue. Estrogens will proliferate surface epithelia of the ductal breast, the vagina and others. Undoubtedly, genetic and environmental factors influence estrogen homeostatasis and tissue specific exposure to estrogen and its metabolites. Accumulative lifetime exposure to estrogen may have a bearing on the metabolism of the hormone-responsive organs. Genetic disposition to polymorphism of key metabolic enzymes with a resultant formation of toxic metabolites may be one of the reasons why in some individuals, estrogen exposure might involve carcinogenesis. In otherwise healthy metabolic conditions, sex steroids are not known to damage DNA. Mortality is reduced in breast cancer women with HRT exposure. The increased risk of endometrial cancer following long-term exposure to estrogen can successfully be counteracted by the appropriate addition of progestogens. The apoptotic potential of some progestins can apply to breast tumors. The low lifetime risk of ovarian cancer and its unfavourable outcome because of as yet later stage diagnosis meets with still inconsistent data on ovarian cancer incidence as related to long-term postmeopausal estrogen use. Cancers of the cervix, vulva and vagina do not seem to be related to HRT use. Available data suggest a reduced risk of colorectal adenoma and colon cancer in HRT current users. There is no contraindication to HRT prescription in survivors of colon cancer, endometrial cancer, epithelial ovarian cancer and of breast cancer. However, the effect of various types of progestogens on the breast awaits further clarification. © 2002 Published by Elsevier Science Ireland Ltd. Keywords: HRT; Cancer; Carcinogen

1. Introduction Experimental experience points to the natural hormone 17b-estradiol as a carcinogen in animals. The administration of estradiol to mice and rats will, among others, increase the incidence of mammary and pituitary tumors [1,2]. Such tumor models have been developed using pharmacological doses of estradiol such as to examine the * Tel.: +49-251-835-710 E-mail address: [email protected] (H.P.G. Schneider).

tumorigenic activity of this hormone in a relatively short period of time. However, no animal models have been developed in which tumors are induced by very low doses of estradiol, presumably because of difficulties in long-term maintenance and dosing in view of the varying endogenous levels in cycling females. Therefore, the predictive value of carcinogenicity testing at high doses has been questioned [3]. Estrogen administration is widely accepted as a risk factor of human endometrial carcinoma [4]. Estrogens, when unopposed by progestins, increase the risk of uterine tumors. This risk in-

0378-5122/02/$ - see front matter © 2002 Published by Elsevier Science Ireland Ltd. PII: S 0 3 7 8 - 5 1 2 2 ( 0 2 ) 0 0 1 4 7 - 0

S36

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

creases with increasing doses of estrogen and with the length of treatment [5]. Recent cohort studies have demonstrated strong relationships between endogenous estrogen levels and the breast cancer risk [6,7]. The human epidemiological data point to estradiol and other estrogens as potential weak carcinogens. Only weak carcinogenic activity, if anything, can be expected because estradiol and other steroidal estrogens are endogenous hormones at low picomolar levels; as strong carcinogens would they have provided evolutionary disadvantages that apparently do not exist.

2. Estradiol as genotoxic carcinogen Genetic lesions known to affect growth-control are part of a general genetic instability resulting in tumor development [8]. In his recent review, Liehr [9] has summarized the investigations of DNA damage by steroidal estrogens via catechol estrogen metabolites. In CD one female mice, 4-hydroxy-estradiol, given 5 days after birth, induced a 9-fold higher incidence of uterine adenocarcinoma than was observed with estradiol, whereas 2-hydroxy-estradiol was approximately as carcinogenic as the parent hormone [10]. On this basis, the formation of catechol estrogens and their metabolic activation to reactive intermediates gained major interest in relation to various types of DNA damage. Two-hydroxylation of steroidal estrogens is the major metabolic oxidation of estrogenic hormones in most mammalian species. In the human, this oxidation is catalyzed by P-450 3A enzymes, whereas cytochrome P-450 1A enzymes are the predominant estrogen-2-hydroxylases in extrahepatic tissues [11]. The estrogen-2-hydroxylases convert estradiol to approximately 80– 85% 2-hydroxy-estradiol and, due to a lack of specificity, in the remaining 15– 20% to 4-hydroxy-estradiol [12]. On the other hand, specific estrogen-4-hydroxylases, which convert estradiol mainly to 4-hydroxy-estradiol, have first been identified in rodents [13]. In humans, the predominant conversion of estradiol to 4-hydroxy-estradiol has been detected in microsomes of uterine myometrium and fibroids, in normal mammary

tissues as well as benign and malignant mammary tumors [9]. This human estrogen-4-hydroxylase activity has been identified as cytochrome P-450 1B1, an extrahepatic isoenzyme located in mammary tissue, ovaries, adrenal glands, and the uterus [14]. When estrogen metabolite concentrations are measured in a human breast cancer extract, the ratio of 4-hydroxy-estradiol to 2-hydroxy-estradiol metabolite concentrations was 4:1 [15]. Further investigation on human organs prone to estrogen-associated cancer should include the metabolic conversion of estrogens to 4-hydroxyestradiol; this is particularly apparent for local tissue catechol estrogen concentrations. A metabolic redox cycling does exist for 4-hydroxy-estradiol. The organic hydroxyperoxide-dependent oxidation of the catechol estrogen (the hydroquinone) to the quinone, and the NADPHdependent cytochrome P-450 reductase-catalyzed reduction of the quinone intermediate back to the hydroquinone is the biochemical process involved [16]. The semiquinone-free radicals are intermediate in each of these metabolic conversions. The estrogen semiquinone is a reactive species and may react with molecular oxygen and form quinone and superoxide radicals. It is of particular interest that tamoxifen stimulates quinone reductase (Fig. 1) [16]. This direct reduction of quinones to hydroquinones bypasses the semiquinone radical intermediates and thus decreases free radical generation. Tamoxifen may thus protect from breast cancer by two different mechanisms, one being the inhibition of hormone receptor-mediated proliferation of breast cancer cells, and the other a decrease in toxicity and potential mutagenicity caused by quinones including estrogen quinone metabolites.

Fig. 1. Metabolic redox cycling of catecholestrogens between hydrochinone (catechol) forms and quinone (adapted from [9]).

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

One other fascinating experimental aspect is the experience in transgenic mice. Mice who are expressing the Wnt-1 gene produce elevated amounts of protein important in cell signaling during embryonal development. These mice develop mammary tumors at a high rate within a few months after birth [17]. These transgenic mice have been crossbred with estrogen receptor-alpha knockout (ERKO) mice to examine the role of estrogen receptors in breast tumor incidence [18]. The incidence of mammary tumors was delayed, but not eliminated, in the crossbred animals. When they were ovariectomized, the resultant reduction in estradiol significantly reduced mammary tumor incidence. There are experimental designs for differentiation of the mammary carcinogenesis as related to genotoxicity of estradiol versus hormone receptor pathways. Our current experimental knowledge points to a hormone dependence of proliferation as well as cellular apoptosis as the means by which hormones can interfere with tumor growth. On the other hand, there are also indications for hormone-induced cellular expression of oncogenes as well as cell-specific expression of variable hormone receptors. This knowledge will also allow for further insight in the way in which a hormonal stimulus will promote or inhibit tumorigenesis long-term and uni-directional. Independent of environmental and dietetic factors, age is a clearly defined risk factor of cancer. Breast cancer manifests predominantly at postmenopausal age. Breast development and differentiation, tumorigenesis as well as growth and progression are influenced by sex hormones. Generally speaking, excessive endocrine stimulation of specific organs will induce incremental cell division and thereby lead to accidental accumulation of genetic defects with accelerated cell cycle turnover. As a result, there will be an accumulation of neoplastic phenotypes [19–22].

3. Reproductive tumors and HRT In view of clinical practice, the fact as to whether HRT will influence hormone-dependent tumors can only be answered by randomized,

S37

controlled long-term epidemiological investigation. Can we introduce steroid hormones for replacement in peri- and postmenopausal women independent of or bound to their family history of cancer? For proper decision-making, we would have to consider HRT with respect to general cancer risks and in addition in such women who survived reproductive cancer and require further medical intervention.

3.1. Breast cancer Ever since 1896, when George Beatson, a Scottish surgeon, reported about his experience of a remission of breast cancer following bilateral ovarectomy in premenopausal women [23], the possible relationship of ovarian function in mammary tumorigenesis never escaped our clinical conscience. In that same year, more than 100 years ago, the first successful HRT from ovarian extracts was introduced in parallel to the clinical breast cancer benefit observed in the year of ovarian ablation [24]. With all the favorable risk– benefit equation attributed to HRT during the last century, the possible relation between estrogen and the risk of breast cancer has remained an oncologic enigma.

3.1.1. Epidemiology and biological plausibility In her last publication, Trudy Bush [25] tried to assess whether recent epidemiologic evidence supports an association between use of estrogen replacement therapy or hormone replacement therapy and risk of breast cancer. For that matter, articles published from 1975 to 2000 in Medline and Dialogweb were searched and only included in the study when peer-reviewed and containing original data. A total of 45 publications were identified on ERT and breast cancer risk (Table 1). These data were found to be inconsistent and the risk estimates were distributed in the way as if there were no association of estrogen with breast cancer. Looking for HRT and breast cancer risk, a total of 20 studies, case–control or pro-retrospective cohorts, produced inconsistent results (Table 2). A further check into breast cancer risk estimates by duration of hormone use resulted in what is summed up in Table 3, incon-

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

S38 Table 1 ERT and breast cancer risk [25]

Table 3 Breast cancer risk estimates by duration of hormone use [25]

A total of 45 publications Of these (%) 20 33 47 None Data are inconsistent

Duration ranging from \5 to \20 years Risk elevated significantly in three One ERT, one studies HRT, one any HT Risk elevated non-significantly in five One ERT, one studies HRT, three any HT Risk not increased in two studies Two any HT Inconsistent results for duration even within studies

Case control or pro-retrospective cohort RR RR RR RR Distribution of risk estimates is what could be expected if there were no association

B0.9 \1.1 0.9–1.1 \2.0

For all cohort studies.

sistency even within studies. Finally, breast cancer mortality and survival in ever-versus never-users of HRT was looked for. These data show consistency for both improved mortality and survival (Table 4). In her conclusion, Trudy Bush pointed out that ‘the evidence did not support the hypothesis that estrogen use increases the risk of breast cancer and that combined hormone therapy increases the risk more than estrogen only. Additional observational studies are unlikely to alter this conclusion’. In his critical review, Speroff [26] was uncertain about whether or not there is a slight risk of breast cancer (in lean women) with long exposure to estrogen–progestogen or whether or not this conclusion may be imprecise due to bias and small numbers of investigated women. Criteria to strengthen any conclusion of epidemiological findings would be: (a) The strength of the association: the relative risks of the case– control and cohort studies with postmenopausal estrogen– progestin Table 2 HRT and breast cancer risk [25] A total of 20 studies Four of these observational studies show statistically significant findings Inconsistent results Many studies (not all) found smaller, lower grade, ER positive Breast Tumors

Case control or pro-retrospective cohort two with higher RR two with protective effects

treatment are recognized by epidemiologists as rather weak associations. (b) Consistency, uniformity, and agreement: among many studies are rather scarce, indicating either very small effects or the impact of founding biases. (c) A dose–response relationship: is seen after increasing the dose and time of exposure; this aspect may have the best supporting evidence. (d) Temporal relationship: the outcome data with respect to improved survival rates in hormone users support the contention that hormonal treatment promotes the detection of pre-existing tumors. This epidemiological dilemma enforces questions as to our current insight into tumor biology and breast tissue hormone metabolism. Pathobiology may provide better understanding of the role of hormones in the development and growth of breast cancer. In which way do genetic and environmental factors influence estrogen homeostasis and tissue-specific exposure to estrogen and its metabolites? Ideally, a relation between exposure to estrogen and risk of breast cancer can be identified in specific groups of women and may allow us to predict risks in the individual. Table 4 Breast cancer mortality and survival in ever vs. never users of HRT [25] Risk estimates for death in all five studies (three significant) Risk for survival in all six studies significant Consistency for both improved mortality and survival

B1.0 B1.0

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

S39

Table 5 Reproductive indicators of the risk of breast cancer (additional references see [24]) Indicator

Risk group

Relative risk

References

Low

High

Sex Age (year) Age at menarche (year) Age at birth of first child (year) Breast-feeding (mo) Parity

Male B50 ]14

Female ]50 B12

150.0 6.5 1.2–1.5

Hulka, 1997 Ries et al., 1999 Rockhill et al., 1998; Bruzzi et al., 1985; Gail et al., 1989

B20

]30

1.9–3.5

]16 ]5

0 0

Age at menopause (year) Estrogen therapy Estrogen–progestin therapy Postmenopausal BMI Family history of breast cancer

B45

]55

Hulka, 1997; Leon et al., 1995; Madigan et al., 1995; Ramon et al., 1996; Lambe et al., 1996 Enger et al., 1998 Hulka, 1997; Madigan et al., 1995; Ramon et al., 1996; Lambe et al., 1996 Hulka, 1997

Never Never

Current Current

1.06–1.4 1.4

B22.9

\30.7

1.6

Hulka, 1997

Yes

2.6

Madigan et al., 1995

No

1.37 1.4 2.0

Speroff, 2000; Grodstein et al., 1997 Grodstein et al., 1997

3.1.2. Estrogen and carcinogenesis of the glandular breast The relatively large body of literature on the association between estrogen and breast cancer is inconsistent. The response of an organ to the proliferative effects of a hormone may be a progression from normal growth to hyperplasia to neoplasia. Thus, the risk of breast cancer could be determined by the cumulative exposure of breast tissue to estrogen [27]. Individual reproductive history suggests that early menarche, late first full-term pregnancy, and late menopause are associated with an increased risk of breast cancer in contrast to reduced risk seen with early menopause. The relative risk of these hormonally mediated indicators is listed in Table 5. The predictive value of these factors is increased when they are combined. If we, for example, take individual age and age at first full-term birth, this would not only reflect the total exposure to estrogen, but also the effect of sex steroids on final differentiation of glandular breast induced by pregnancy and lactation as major determinants of susceptibility to cancer [28]. Indeed, the group of Russo and co-workers created a plausible experi-

mental model of breast carcinogenesis: terminal pregnancy is of ultimate biological importance for cellular differentiation of breast tissue, and further to this, the preventive character of lactation is associated to the promotion of type-IV lobules. The further differentiated a glandular breast will be, the less prone it is to experimental cancerization. Other contributing factors to individual variation in exposure to estrogen are obesity in postmenopausal women, differences in exercise, in dietary intake of certain nutrients. Among the latter, studies of intakes of alcohol, fat, anti-oxidant vitamins, and fiber have produced conflicting results. Phytoestrogens with their structural similarity to physiologic estrogens, when ingested, have both estrogen agonist and antagonist effects in humans. Flax seed, a source of mammalian lignanes and alpha-linoleic acid, has been shown to exert anti-estrogenic effects by binding to estrogen receptors and inhibiting the synthesis of estrogen. Indeed, the incidence of breast cancer is lowest in regions where the intake of soy, an abundant source of phytoestrogens, or of flax seed is high; whether or not this inverse relation is

S40

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

direct or only indicative of other influencing factors, is a matter of debate [29].

3.1.3. Local estrogen biosynthesis in breast cancer Two-thirds of breast tumors are diagnosed after the menopause. Due to the cessation of ovarian estrogen production at menopause, plasma levels of estradiol decrease to less than 10% of levels found before menopause. The majority of breast tumors, however, are estrogen receptor positive and depend on estrogens for growth at the time of diagnosis. This apparent contradiction indicates that the growth of breast tumors depends on local factors rather than on circulating estrogens. Studies on estrogen levels in breast tumors show that intratumoral estrogen levels, despite lower plasma values, are not different between premenopausal and postmenopausal women [30,31]. There are two potential sources from which higher intratumoral levels of estrogens could originate: 1. increased uptake and retention of plasma estrogens by the tumor; 2. estrogen biosynthesis in the tumor or surrounding tissues. Tumor estradiol levels, however, do not correlate with tumor ER status; therefore, retention of estrogen by the receptor does not play a predominant role [32]. Studies on uptake and conversion of radio-labeled androstenedione and estrone by breast tumors in vitro demonstrate that most of the estrone originates from in-situ biosynthesis rather than from estrone uptake [33]. Androgens, however, are taken up by the tumor tissue, and no such gradient exists for androgens as for estradiol plasma/tumor concentrations [33,34]. Thus, the maintenance of the estradiol gradient is not substrate-driven [34], and the estradiol found in tumors is mainly derived from local synthesis. Tumor aromatase has been found not to correlate with age, tumor size or ER/PR status, and thus seems and independent factor in local endocrine environment [35]. Aromatase activity is present in both tumor tissue and adipose tissue; where is it located in the cancerous breast? Aromatase activity is found in adipose tissue, and the epithelial cancer cells are often intermingled with adipocytes and adipose stroma. Different methods of detection (mono-

clonal versus polyclonal antibody techniques) give more or less opposite findings. Thus, we are still awaiting improved technology and more conclusive results. Macrophages and lymphocytes infiltrate breast tumors and can make up as much as 50% of the tumor [32]. Complex paracrine models have been proposed of interaction of IL-6 and IL-6sR with TNFa [32]. There are also variations in tissue-specific promoters of aromatase gene expression that result in variations of estrogen production [36]. Aromatase messenger RNA (mRNA) in normal breast tissue is stimulated by the promoter I.4. However, in breast cancers a change of promoter from PI.4 to PII and PI.3, which are more active, can result in increased synthesis of aromatase mRNA [37]. The mechanism of promoter-switching is unclear, but it may involve transcription factors specific to breast cancer cells. In-situ aromatization in breast tumors results in increased estrogen in breast tissue, which may contribute to the growth of breast tumors in an autocrine or paracrine fashion [38]. It has been argued that the aromatase gene may act as an oncogene that initiates tumor formation in breast tissue [29]. Breast tumors of postmenopausal patients contain all the enzymes necessary for the formation of estradiol from circulating precursors, besides aromatase this is sulfatase and 17b-hydroxysteroid-dehydrogenase [39,40]. While the ‘aromatase pathway’ transforms androgens into estrogens, the ‘sulfatase pathway’ converts estrone sulfate (E1S) into estrone (E1) which is then transformed into E2 by the reductive 17b-HSD activity. Like aromatase activity, sulfatase activity can vary widely among tumors. The activity of sulfatase is much higher than that of aromatase, E1S via sulfatase is 100–500 times higher a precursor for E2 than are androgens via aromatase [41]. Breast tumors also display relatively high levels of sulfotransferase activity. The regulation of both sulfatase and sulfotransferase enzymes in vivo is largely unknown; in cell cultures, however, progestins seem to inhibit sulfatase and increase sulfotransferase activity [42]. Intratumoral 17bHSD activity can potentiate the effect of any E1 produced locally via either the sulfatase or the

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

aromatase route by converting E1 to E2. This conversion may be the source of the high intratumoral E2 levels [43].

3.1.4. Breast tissue sensiti6ity to estrogen Estrogens may diffuse passively through cellular and nuclear membranes. On the other hand, specific cells and tissues express estrogen receptors to which estrogens would bind and form a ligand–receptor complex in order to activate specific sequences in the regulatory region of genes responsive to estrogen, known as estrogen-response elements. These genes in turn regulate cell growth and differentiation. About 50 –80% of breast tumors are estrogen receptor positive when using in vitro ligand binding assays. The ER status has been used to assess overall prognosis and to predict sensitivity of tumors to endocrine therapy. ER levels are low in normal breast tissue; high levels have been directly correlated with an increased risk of breast cancer [44], increase with age in some ethnic groups and apparently are higher in white women as compared with black or Japanese women. This phenomenon may be related with the function of a tumor-suppressor gene, the loss of which may result in failure to down-regulate estrogen receptor with resultant defects of the cell cycle and finally driving breast carcinogenesis [45]. The human estrogen receptor belongs to the nuclear receptor superfamily of ligand-inducible transcription factors. The recent identification of ERb has indicated that the cellular responses to ER ligands are far more complex. ERa and ERb interact with the same DNA response elements and exhibit similar, but not identical ligand-binding characteristics. ERb binds estrogens with a similar affinity to ERa and activates the expression of reporter genes containing estrogen response elements in an estrogen-dependent manner. In vitro, the a- and b-receptors form heterodimeres with each other, and the b-receptor decreases the sensitivity of the a-form to estrogen, thereby acting as a physiologic regulator of the proliferative effects of the a-receptor [46]. In order to evaluate the role of differentiated ERs in breast cancer, the expression of both ER

S41

isoforms in normal and malignant breast tissue has been investigated [47]. In normal breast tissue, expression of ERb predominated, with 22% of samples exclusively expressing ERb; this was not observed in any of the breast tumor samples. Most tumors expressed ERa, either alone or in combination with ERb.

3.1.5. Clinical response of breast cancer tissue to hormone exposure In an ongoing clinical project, we investigated so far over 100 postmenopausal women with breast cancer [48]. During cancer surgery, tissue samples were preserved for laboratory work-up in terms of homogenization by microdysmembranation, suspension with trasylol, extraction with ethanol/acetone, evaporation of liquid phase and separation, defatting, addition of tracers for recovery and extraction for determination of estrone and estradiol by highly specific radio-immunoassays [49]. Local estrone and estradiol concentrations in terms of fmol/g were compared in cancer tissue versus adjacent or distant normal control tissue. Then these tissue estimates were evaluated in never-users versus ever-users of HRT. Other details of this investigation involve the expression of local enzyme activities as well as the production of steroid metabolites and of estrogen receptors. Preliminary experience is consistent with unvaried levels of estrone and estradiol in cancer tissue as compared with neighboring normal breast tissue. It was also not evident that HRT would produce any remarkable difference in local estrogen concentration. The modes of HRT included sequential and combination-type regimens. Given a concentration gradient of plasma versus breast tissue levels of more than an order of magnitude, our observations would suggest that oral hormone replacement does not have any demonstrable impact on local breast tissue estrogen metabolism. Without stressing the importance of such preliminary observations, they would nevertheless add some piece of evidence as to the inconsistency of epidemiological evidence on the interdependence of HRT and breast cancer risk.

S42

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

3.1.6. Effect of HRT on mortality and in breast cancer sur6i6ors If estrogen replacement were of any major harm to women who survived breast cancer and its treatment, one would expect an unfavorable prognosis in women diagnosed breast cancer while on estrogen treatment. On the contrary, these women have better prognoses [50]. Women with a diagnosis of breast cancer within 1 year following discontinuation of estrogen treatment will survive longer than non-hormone users or women who last took their estrogens for longer than 1 year [51]. In this investigation, Gambrell calculated a breast cancer mortality of 22% in women being diagnosed while under ERT as compared with 46% in non-users (P B0.002). In that situation, 57% of hormone users were lymphonode-negative as compared with 42% of non-users; within the lymphonode-negative group, mortality rated 8% for hormone users and 25% for non-users (PB 0.05). Henderson and co-workers [52] confirmed this experience and reported on a 19% reduction of breast cancer mortality among 4988 women using ERT as compared with 3865 non-users who later on developed breast tumors. Relative breast cancer mortality in women while under HRT is documented from nine different studies in Fig. 2.

Fig. 2. Breast cancer mortality in current HRT users: results of mortality from nine studies. The points indicate relative breast cancer mortality when the disease was diagnosed while under HRT. Ninety five percent confidence intervals are given as far as available (additional reference see [24]).

Table 6 Genotype polymorphisms of estrogen-metabolizing genes and risk of breast cancer [53] Risk factor

Odds ratio

95% CI

CYP 17 encoding P-450 17a-hydroxylase CYP 1A1 encoding cytochrome P-450 IAI COMT Two putative high risk genotypes

1.23

0.67–2.28

1.79

0.86–3.78

4.02 3.52

1.12–19.08 1.06–12.4

Association higher with prolonged estrogen exposure years.

3.1.7. Outlook Estrogen is important in the maturation and differentiation of normal breast tissue and is associated with most of the epidemiological risk factors of breast cancer. Estrogens will proliferate normal ductal epithelia in the non-cancerous breast during the menstrual cycle and in pregnancy and will act on these cells via two distinct estrogen receptors. Genetic and environmental factors influence estrogen homeostasis and tissuespecific exposure to estrogen and its metabolites. Whether or not cumulative lifetime exposure to estrogen has any bearing on breast tissue metabolism remains unclear. Most available information supports the hypothesis that estrogen and its metabolites may be related to the promotion of pre-existing breast cancer. Genetic disposition to polymorphisms of key metabolic enzymes may dispose the individual to the formation of estrogen metabolites, which, are toxic to DNA strains. Polymorphisms of both cytochrome CYP17 (encoding P-450 17a-hydroxylase) and CYP-19 (encoding P-450-aromatase) have been identified in the general population [52,53]. Women who are heterozygous or homozygous for a cytochrome CYP-17 polymorphism have been shown to produce high serum estradiol concentrations; however, this polymorphism is not unequivocally associated with increased risk of breast cancer [29]. There are, however, ongoing studies demonstrating a link between polymorphisms of the P-450-aromatase gene with increased risk of breast cancer [53] (Table 6). The estrogen production may also be influenced by variation in tissue-

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

specific promoters of aromatase gene expression [36]. Indeed we might envision reasons for individual genetic predisposition to breast cancer. On the other hand, there is no indication so far of generally practiced oral hormone replacement therapy to specifically alter local breast tissue estrogen metabolism both in normal or cancer tissue.

3.2. Endometrial cancer Not taking breast and colon cancer into account, the adenocarcinoma of the endometrium is the most frequent malignancy in women from western industrialized countries. Rather infrequent before menopause, endometrial cancer will only account for 7.5% of all cancers before the age of 50. Between the ages of 40 and 67, the incidence will increase rapidly, after which it will persist at a constant plateau. Experimental and observational studies associate endogenous and exogenous estrogens with an elevated risk of endometrial hyperplasia and endometrial cancer. Clinical situations in which elevated serum estrogen levels are experienced long-term with a concurrent deficit in cyclical secretion of progesterone as seen in adipose women, the syndrome of polycystic ovaries or various ovarian tumors such as granulosa or theca cell tumors, are associated with an increased risk of endometrial cancer. The risk of endometrial cancer increases doseand time-dependently with estrogen monotherapy (Table 7). Estrogens at higher levels and longterm elevate the risk of endometrial cancer 5-fold and beyond [54]. This elevated cancer risk is restricted to early stages and well-differentiated tumors [55]. Estrogen-induced endometrial cancer is followed by up to 95% complete remission. Less well differentiated endometrial cancers differ, however, with respect to hormone dependence, a phenomenon as yet not well understood.

3.2.1. The low estrogen dose concept There are a lot of low-dose estrogen HRT products on the market, raising the question as to the efficiency and safety of these products targeting the endometrium, breast tissue, vasomotor symptoms, bone metabolism or lipid profiles. En-

S43

dometrial hyperplasia with oral conjugated estrogens can be induced in a dose-dependent manner [56]. After 2 years of treatment with a low-dose preparation (0.3 mg/day of conjugated estrogen), only one case of endometrial hyperplasia (1.7% of the investigated cases) was observed; this is identical with the incidence in the placebo group. In another investigation [57], administering 0.3, 0.625 and 1.25 mg of conjugated estrogens orally per day to postmenopausal women, the low-dose estrogen treatment was associated with a profound reduction in endometrial cancer risk. This has also been confirmed by a 12-week investigation of transdermal application of 25, 50 and 100 mg estradiol per day versus placebo [58]. When vaginal PAP smears of originally atrophic vaginal epithelium were taken as demonstrating trophicity, the 25 mg/day treatment group produced only one case of endometrial hyperplasia in a total of 14 probands as compared with ten cases out of 22 with a 50 mg/day dose and a total of 88 cases in the 100 mg/day application group.

3.2.2. Importance of progestogens The effect of progestogens on the endometrium is also dose- and time-dependent and not so much influenced by the type of progestogen. As a result, those progestogens in clinical use will have the same preventive effect on endometrial cancer [59]. In traditional clinical investigation, progestogens have been applied in a rather sequential or cyclic Table 7 Estrogen monotherapy and risk of endometrial cancer in postmenopausal women (additional references see [24]) Authors

Cases (n)

Relative risk

Quint, 1975 Smith et al., 1975 Ziel and Finkle, 1975 Mack et al., 1976 McDonald et al., 1977 Gray et al., 1977 Horwitz and Feinstein, 1978 Antunes et al., 1979 Jick et al., 1979 Shapiro et al., 1985 Persson et al., 1989

291 317 94 63 145 205 119

1.8 4.5 7.6 8.0 4.9 3.1 1.7

451 67 425 74

6.0 20.0 3.5 1.8

S44

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

Table 8 Effective dose of progestogens for endometrial protection in sequential HRT mg/day Oral Progesterone (micronized) Medroxyprogesterone acetate (MPA) Medrogestone Dydrogesterone Cyproterone acetate (CPA) Norethisterone acetate (NETA) DL-norgestrel (NORG) Levonorgestrel (LNG) Desogestrel Transdermal Norethisterone acetate (NETA)

200–300 5–10 5 10–20 1 1–2.5 0.15 0.075 0.15 0.25

fashion. The continuous-combined mode of estrogen and progestogen therapy has gained preference particularly in later postmenopausal years; at this age, two-thirds of all women prefer amenorrhea, and the risk of dysfunctional bleeding by a residual endogenous ovarian estrogen secretion can be neglected. The inter-individual variation of endometrial reaction to progestogens was investigated [60]. A protection of the endometrium is seen in nearly all women with a minimal dose of 0.7 mg norethisterone, 250 mg levonorgestrel, 200 mg progesterone, 10 mg medroxyprogesterone acetate or 20 mg dydrogesterone (Table 8). Estrogen-dependent endometrial hyperplasia will be prevented by lower than the typical progestogen transformation dose, provided co-medication is offered for a minimum of 12 days [60]. There have also been objections to this concept considering 10 days as sufficient duration of progestogen comedication [61]. Clinically speaking, we do observe individual variation of endometrial transformation all the way from 7 to more than 12 days. Other risk factors of endometrial cancer also need to be considered with respect to dose and duration of progestogen co-medication. Among these are obese women with their enlarged capacity to aromatize androstenedione to estrone, who carry a 3-fold risk [62]. Already in 1977, the second most frequent incidence of endometrial cancer was demonstrated to exist in non-replaced

postmenopausal women who abstained from hormone replacement because they never experienced any menopausal symptoms [63]. In this group, many obese and nulliparous women are found [64]. For that reason, the application of progestogens in non-estrogen substituted women of various risk groups was suggested in order to oppose the endogenous overproduction of estrogens [65]. As long as the gestagen test is positive, a progestogen should be provided for a period of 10 to 14 days each month; a negative gestagen test should be repeated within a year. This risk group should not only consist of postmenopausal obese women, but also of higher-weight young women with polycystic ovaries in which a 5-fold increased endometrial cancer risk was found [66]. Current management of postmenopausal hormone replacement is based on a continuous application of estrogens with an additional cyclical or continuous combination of progestogens. Pharmacokinetic and pharmacodynamic variability both of the estrogen as well as the progestogen component and-more importantly-the inter-individual variation do not favor the concept of one estrogen–progestogen combination for all indications. Varying degrees of blood supply and hormone sensitivity as well as other factors can cause aberrant endometrial development. For this reason, morphological variations of endometrial regeneration is seen from endometrial biopsies that D and C will not always completely varify. Hysteroscopy-guided biopsies as a ‘gold standard’ of endometrial surveillance during HRT as well as endometrial ultrasound are the most reliable ways of morphological control in women at risk. In non-hysterectomized women, hormone replacement should be started with a combined estrogen–progestogen preparation in order to prevent uncontrolled bleeding, unless vaginal ultrasound proves an endometrial thickness of less than 5 mm (double-layer).

3.2.3. Estrogen in sur6i6ors of endometrial cancer The low estrogen dose not only provides a very low risk of endometrial cancer, but also good bleeding control and thereby excellent patients’ compliance. These advantages raise the question as to estrogen replacement in survivors in en-

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

dometrial cancer. Three retrospective studies have analyzed estrogen replacement following surgery of endometrial carcinoma. In 1986, an investigation of Duke University referred to 215 low-risk women with endometrial cancer stage Ia and Ib, grade 1 and 2, of which 47 (22%) were estrogenreplaced for a mean 26 months. Estrogen replacement was begun between 0 and 81 months following definitive cancer treatment with a mean interval of 15 months. As this was a retrospective analysis, the interval was ranging rather widely. The investigators did not find an increased risk of recurrence when controls were adjusted to tumor size, myometrial invasion, lymphonode spread, peritoneal cytology and age; the estrogen-treated group was compared with 174 women without hormone treatment [67]. Very surprisingly, the risk of recurrence in the non-treated group was higher (15%) as compared with the treatment group (2%). Also mortality risk in the no-treatment group was 26 cases (16 as a result of cancer and ten of intercurrent disease) and was higher as compared with tumor-dependent mortality of estrogen-treated women. Some years later, 44 women with a history of endometrial cancer stage I were subsequently treated with estrogens for a mean of 64 months. The majority of these women started ERT within the first postoperative year. No recurrence of mortality was observed in the treatment group [68]. Later on in 1996, another retrospective analysis on 123 women post endometrial cancer surgery was reported, who did not experience any negative influence on survival when ERT women were compared with non-users [69]. Additional smaller reports took care of all feasible selection criteria and could also confirm the beneficial effect of ERT [70,71]. No observations were ever published on exacerbations of endometrial cancer in women on estrogen replacement. As far as the history of endometrial cancer is concerned, the big majority of women with endometrial cancer stages I and II can be cured with long-term survival rate of more than 80%. Following the addition of a progestogen to ERT in otherwise healthy postmenopausal women will leave a relative risk of endometrial cancer of 1.3 (CI 0.8–2.2). With 5 or more years of estrogen

S45

and progestogen replacement therapy, the risk will be 2.5 (CI 1.1–2.5) [72]. A Swedish cohort analysis found a relative risk of 1.0 (CI 0.7–1.4) in women on combined estrogen–progestogen replacement [73]. These investigations clearly point to a protective effect of an added progestogen. The question whether women following surgical therapy of endometrial cancer should be hormone-replaced cannot be answered without referring to the overall benefit-risk equation of combined progestogen and estrogen treatment. As all such investigations so far are of retrospective nature, this would require additional prospective randomized studies in order to gain therapeutic safety.

3.3. O6arian cancer Epithelial ovarian carcinoma is the leading cause of death from gynecologic cancer. There is an estimate of 14 ovarian cancers in 100 000 women, representing one women in about 70 to develop ovarian cancer in her lifetime, and one woman in 100 to die from this disease. The incidence of ovarian cancer increases with age and peaks in the eighth decade. The median age of diagnosis is 63, and almost half of the patients will be 65 years or older.

3.3.1. Reproducti6e factors Parity is the most important risk factor for ovarian cancer. Having been pregnant would reduce the risk of ovarian cancer by 30–60% [74]. One to two pregnancies infer a relative risk of 0.49–0.97 as compared with 1.0 for nulliparous women. More than three pregnancies further decrease the relative risk to 0.35–0.76. Each month of breast-feeding is associated with an additional risk reduction although months of breast-feeding and decreased cancer risk do not correlate [75]. The risk of developing epithelial ovarian cancer of all histologic subtypes in users of oral contraception is reduced by 40% compared with that of non-users [75,76]. This protective effect increases with duration of use and continues for at least 10–15 years, and the benefit is associated with all monophasic formulations, including the low-dose products [77]. Oral contraception is particularly

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

S46

protective in women at high risk of ovarian cancer (nulliparity, positive family history) [78]. Oral contraception can reduce the risk of epithelial ovarian cancer in women with a positive family history or proven BRCA1 or BRCA2 mutations to a level equal to or less than that experienced by women with a negative family history [78,79]. The protective effect of parity, multiple births, history of breast-feeding and oral contraceptive use supports the incessant ovulation’ hypothesis for the etiology of ovarian cancer [80]. According to this hypothesis, ovarian cancer develops from an aberrant repair process of the surface epithelium, which is ruptured and repaired during each ovulatory cycle. The likelihood of ovarian cancer to develop will, therefore, be a function of the total number of ovulatory cycles, together with a genetic predisposition and other, non well-defined environmental factors.

3.3.2. Impact of hormone replacement No clear difference in the reduction of ovarian cancer risk is seen following high- or low-dose pill use [81]. Sufficient reduction of gonadotropin levels appears to be an important mechanism of protection. As exogenous estrogens reduce the high gonadotropin levels during the menopausal transition, replacement with estrogens and progestogens could conceivably reduce the risk of ovarian cancer. However, from an epidemiological point of view, this does not seem to be the case.

European investigation on HRT use for 5 years or shorter, when reviewed, apparently does not influence the relative risk for ovarian cancer, whilst long-term ERT increases this risk with ratios varying from 0.52 to 1.71 [82]. Literature on the interaction of hormone replacement with ovarian tumorigenesis is relatively scarce. The epidemiological data are inconsistent. A moderate or an often non-significant excess risk of ovarian cancer in HRT users was reported in a multi-centric US case–control study [83]. Other studies conducted in Australia [84] and North America [75] did not show an excess risk; pooled RR of ovarian cancer for ever-HRT users was 0.9 (95% CI 0.7–1.3) in hospital-based studies and 1.1 (95% CI 0.9–1.4) in population-based ones without any duration-risk relationship. In an American collaborative analysis [85], the RR of borderline ovarian tumors based on 327 cases was also 1.1 (95% CI 0.7–1.9). To that point, the available data exclude a strong association between HRT and epithelial ovarian cancer, though a moderate association remains open to debate. To provide further information on the issue, a collaborative re-analysis of four European case–control studies, two conducted in Greece and one each in Italy and the United Kingdom, was performed (Table 9). A total of 1470 ovarian cancer cases and 3271 hospital controls were included [86]. This re-analysis revealed a weak positive association with duration, the RR increasing from 1 (baseline) for those

Table 9 Distribution of ovarian cancer cases and controls according to study center and use of HRT [86] Study Greece 1 Cases HRT use Never Ever OR (95% CI) ever use

Greece 2 Controls

100 175 12 13 1.77 (0.76–4.15)

Cases

Italy Controls

146 125 6 4 1.40 (0.38–5.19)

Cases

UK Controls

915 2421 56 82 1.66 (1.16–2.37)

Cases

Total Controls

200 404 35 47 1.68 (0.99–2.80)

Cases

Controls

1361 3125 109 146 1.71 (1.30–2.25)

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

S47

Fig. 3. Summary of risk ratios and 95% confidence intervals from case – control and cohort studies of estrogen replacement therapy and risk of epithelial ovarian cancer (additional references see [90]).

who had never used HRT to 1.67 (95% CI 1.11– 2.51) for those who had used them for less than 2-years and to 1.79 (95% CI 0.91–3.54) for those who had used them for 2-years or more. There was also some evidence that the excess RR for ovarian cancer declined with time since last use, being 1.96 (95% CI 1.20– 3.21) among recent users ( B 10 years) and 1.45 (95% CI 0.86– 2.52) among those who had stopped using HRT for more than 10 years.

Case– control studies are always limited by their comparitors; controls are selective for certain criteria, which never include the full array of possible confounders. Earlier case–control studies report a decreased risk, no association or increased risk. More recent larger case–control studies have suggested increased risk, particularly with long duration of estrogen use. However, even the largest of these investigations have limited statistical power to assess the risk associated with long duration of estrogen use (Fig. 3).

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

S48

The American cancer society’s cancer prevention study II, a prospective US cohort study with mortality follow-up ranged from 1982 to 1996. This investigation encounters a total of 211 581 postmenopausal women who completed a baseline questionnaire in 1982 and had no history of cancer, hysterectomy, or ovarian surgery at enrollment [87]. The main outcome measure was ovarian cancer mortality, compared among neverusers, users at baseline, and former users as well as by total years of estrogen replacement therapy. The results are presented in Tables 10 and 11. A total of 944 ovarian cancer deaths were observed in 14 years of follow-up. Duration of use was associated with increased risk in both baseline and former users. With 10 or more years of use, death rates, adjusted for annual age per 100 000 women, were 64.4 for baseline users, 38.3 for former users and 26.4 for never-users. Among former users of more than 10 years of use, risk decreased with time since last use reported at study entry. Following treatment of ovarian cancer, women will usually be considerably distressed. Not only did they have surgery, chemotherapy or radiotherapy, but also will they have to adapt to the rapid onset of hormonal deficiency. Remarkably few of these women are offered HRT to improve their quality of life. An Italian group of investigators looked at conservative treatment of ovarian cancer in premenopausal women (younger than 40

years of age) and found that pregnancy did not affect survival in women who conceived after conservative treatment [88]. There is hardly any information about HRT given to women treated for ovarian cancer. A British study [89] failed to demonstrate any difference in outcome between 78 women who received HRT following treatment of ovarian cancer and 295 women who were not treated. However, there was a tendency for women who had endometrioid or clear-cell tumors to do better on hormonal therapy than women who were not given estrogen and progestogen. The results from this paper were reassuring in that while HRT did not have any adverse effects on outcome, it certainly improved the quality of life for those women who were taking hormones. Lifetime risk of ovarian cancer is low. While protection against ovarian cancer is one of the most important benefits of oral contraception, why should HRT would produce opposite effects? There is contradiction in case–control studies worldwide. Taking large prospective studies of ovarian cancer mortality into particular account, these findings add to the inconsistency of previously published data. The US center for disease control recently reported the results of a metaanalysis of data from 15 case–control studies that provided data on ERT and risk of epithelial ovarian cancer [90]. This meta-analysis did not

Table 10 Ovarian cancer mortality by estrogen use and duration and recency of estrogen use, cancer prevention study II, 1982–1996 [87] Estrogen use

Number of deaths Number of person-years

Rate ratio (95% CI)a

Rate ratio (95% CI)b

Never Ever Recency of use Baseline Former Years of use, baseline users B10 ]10 Years of use, former users B10 ]10

689 255

2185 876 625 984

1.00 (Referent) 1.21 (1.05–1.41)

1.00 (Referent) 1.23 (1.06–1.43)

62 193

151 880 474 103

1.45 (1.11–1.88) 1.15 (0.98–1.36)

1.51 (1.16–1.96) 1.16 (0.99–1.37)

31 31

110 379 41 396

1.07 (0.74–1.54) 2.13 (1.48–3.06)

1.14 (0.79–1.65) 2.20 (1.53–3.17)

158 35

416 823 57 281

1.09 (0.92–1.30) 1.55 (1.10–2.18)

1.10 (0.92–1.31) 1.59 (1.13–2.25)

a

Rate ratio estimates adjusted for age and race. CI indicates confidence interval. Models adjusted for age at baseline, race, duration of oral contraceptive use, number of live births, age at menopause, body mass index, age at menarche, and tubal ligation. b

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

S49

Table 11 Ovarian cancer mortality among former estrogen users, by duration and time since last use, cancer prevention study II, 1982–1996 [87] Years since last estrogen usea

Number of deaths Number of person-years

Rate ratio (95% CI)b

Rate ratio (95% CI)c

Duration of use B10 year Never Former Use within 15 year No use for ]15 year

689 158 45 113

2 185 876 416 823 160 278 256 545

1.00 1.09 1.17 1.06

(Referent) (0.92–1.30) (0.85–1.59) (0.87–1.31)

1.00 1.10 1.17 1.07

(Referent) (0.92–1.31) (0.85–1.60) (0.87–1.32)

Duration of use ]10 year Never Former Use within 15 year No use for ]10 year

689 35 19 16

2 185 876 57 281 30 887 26 394

1.00 1.55 1.98 1.27

(Referent) (1.12–2.18) (1.25–3.15) (0.77–2.10)

1.00 1.59 2.05 1.31

(Referent) (1.13–2.25) (1.29–3.25) (0.79–2.17)

a

Years since last use as reported at study entry. Rate ratio estimates adjusted for age and race. CI indicates confidence interval. c Rate ratio estimates adjusted for age at baseline, race, duration of oral contraceptive use, number of live births, age at menopause, body mass index, age at menarche, and tubal ligation. b

find a significant association of ERT with epithelial ovarian cancer. Furthermore, the CDC evaluation found no clear evidence of an increased risk of ovarian cancer based on variation of estrogen dose. The recent report on the large prospective study of US women [87] associated postmenopausal estrogen use for 10 or more years with increased risk of ovarian cancer mortality that persisted up to 29 years after cessation of use. Such data need to be confirmed. Any increase in risk of ovarian cancer mortality due to long-term estrogen use must be considered in the overall balance of potential risks and benefits.

4. Concluding remarks This review of hormone replacement in malignancies of the human glandular breast, the endometrium and the ovaries should demonstrate that estrogens are important in the growth and differentiation of hormone-responsive tissue. While estrogens will proliferate surface epithelia of the ductal breast, the vagina and others, there is no doubt that genetic and environmental factors influence estrogen homeostasis and tissuespecific exposure to estrogen and its metabolites. Accumulative lifetime exposure to estrogen may have a bearing on the metabolism of the hor-

mone-responsive organs. Genetic disposition to polymorphism of key metabolic enzymes with a resultant formation of toxic metabolites may be one of the reasons why in some individuals, estrogen exposure might involve cancerogenesis. Mortality is reduced in breast cancer women with HRT exposure. Women who develop breast cancer within a year of discontinuing estrogen replacement therapy will survive longer than nonusers. This clinical experience points to major HRT benefits. The increased risk of endometrial cancer following long-term exposure to estrogens can successfully be counteracted by the appropriate addition of progestogens. The apoptotic potential of progestins particularly of the 19-norprogesterone variety can apply to breast tumors. There is a low lifetime risk of ovarian cancer; these tumors are still diagnosed at later stages and, therefore, produce unfavorable outcomes. Literature is still inconsistent with respect to ovarian cancer incidence as related to HRT, current meta-analytic data are reassuring. The recent report of the CDC of an association of long-term postmenopausal estrogen use with increased risk of ovarian cancer mortality needs to be confirmed. There is no space allotted to go into details of other reproductive tumors such as cancers of the

S50

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

cervix, vulva and vagina; however, they do not seem to be related to HRT use. Taken altogether, overall cancer mortality is reduced in current or ever HRT users.

References [1] International Agency for Research on Cancer, In: Monographs on the evaluation of carcinogenic risks to humans, Lyon, France: IARC, Suppl 7, 1987. pp. 280 – 85. [2] Inoh A, Kamiya K, Fujii Y, Yokoro K. Protective effects of progesterone and tamoxifen in estrogen-induced mammary carcinogenesis in ovariectomized W/Fu rats. Jpn J Cancer Res 1985;76:699 –704. [3] Gold LS, Slone TH, Stern BR, Manley NB, Ames BN. Rodent carcinogens: setting priorities. Cience 1982;258:261 – 5. [4] Siiteri PK, Nisker JA, Hammond GL. Hormonal basis of risk factors for breast and endometrial cancer. In: Iacobelli S, King RJB, Lindner HR, Lippman ME, editors. Hormones and cancer. New York: Raven Press, 1980:499 – 505. [5] Key TJA, Pike MC. The dose–effect relationship between ‘unopposed’ estrogens and endometrial mitotic rate: its central role in explaining and predicting endometrial cancer risk. Br J Cancer 1988;57:205 – 12. [6] Toniolo PG, Levitz M, Zeleniuch-Jacquotte A, Banerjee S, Koenig KL, Shore RE, Strax P, Pasternack BS. A prospective study of endogenous estrogens and breast cancer in postmenopausal women, J. Natl. Cancer Inst. 1995;87:1076 – 82. [7] Bernstein L, Ross RK, Pike MC, Brown JB, Henderson BE. Hormone levels in older women: a study of postmenopausal breast cancer patients and healthy population controls. Br J Cancer 1990;61:298 –302. [8] Lengauer C, Kinzler KW, Vogelstein B. Genetic instabilities in human cancers. Nature 1998;396:643 –9. [9] Liehr JC. Is estradiol a genotoxic mutagenic carcinogen. Endocr Rev 2000;21:40 –54. [10] Newbold RR, Liehr JG. Induction of uterine adenocarcinoma in CD-1 mice by catechol estrogens. Cancer Res 2000;60:235 – 7. [11] Kerlan V, Dreano Y, Bercovici JP, Beaune PH, Floch HH, Berthou F. Nature of cytochrome P450 involved in the 2/4-hydroxylation of estradiol in human liver microsomes. Biochem Pharmacol 1992;44:1745 – 56. [12] Weisz J. Metabolism of estrogens by target cells. Diversification and amplification of hormone action and the catecholestrogen hypothesis. In: Hochberg R, Naftolin F, editors. New biology of steroid hormones. New York: Raven Press, 1991:201 –12. [13] Bui QD, Weisz J. Monooxygenase mediating catecholestrogen formation by rat anterior pituitary is an estrogen-4-hydroxylase. Endocrinology 1988;124:1085 –7.

[14] Hayes CL, Spink DC, Spink BC, Cao JQ, Walker NJ, Sutter TR. 17ß-Estradiol hydroxylation catalyzed by human cytochrome P450 1B1. Proc Natl Acad Sci 1996;93:9776 – 81. [15] Castagnetta LA, Granata OM, Arcuri FP, Polito LM, Rosati F, Cartoni GP. Gas chromatography/mass spectrometry of catecholestrogens. Steroids 1992;57:437 – 43. [16] Roy D, Liehr JG. Temporary decrease in renal quinone reductase activity induced by chronic administration of estradiol to male Syrian hamsters. J Biol Chem 1988;263:3646 – 51. [17] Tsukamoto AS, Grosschedl R, Guzman RC, Parslow T, Varmus HE. Expression of the int-1 gene in transgenic mice is associated with mammary gland hyperplasia and adenocarcinomas in male and female mice. Cell 1988;55:619 – 25. [18] Bocchinfuso WP, Hively WP, Couse JF, Varmus HE, Korach KS. An MMTV-Wnt-1 transgene induces mammary gland hyperplasia and tumorigenesis in mice lacking estrogen receptor-a. Cancer Res 1999;59:1869 – 76. [19] Gusberg SB, Kaplan AL. Precursors of corpus cancer. Am J Obstet Gynecol 1963;87:662. [20] Fearon ER, Vogelstein B. A genetic model for colorectal tumorigenesis. Cell 1990;61:759 – 67. [21] Kurman RJ, Kaminski PF, Norris HJ. The behavior of endometrial hyperplasia. A long-term study of ‘untreated’ hyperplasia in 170 patients. Cancer 1985;56:403 – 12. [22] Jackisch C, Schneider HPG. Biological effects of estrogens and progestogen on human breast cancer carcinogenesis. Menopause Rev 1997;2:26 – 34. [23] Beatson GT. On the treatment of inoperable cases of carcinoma of the mamma: suggestions for a new method of treatment, with illustrative cases. Lancet 1896;2:104 – 7. [24] Schneider HPG, Jackisch J. Reproductive cancer and hormone replacement. In: Fischl FH, editor. Menopause andropause. Gablitz: Krause and Pachernegg, 2001:173 – 204. [25] Bush TL, Whiteman M, Flaws JA. Hormone replacement therapy and breast cancer: a qualitative review. Obstet Gynecol 2001;98:498 – 508. [26] Speroff L. Postmenopausal estrogen-progestin therapy and breast cancer: a clinical response to epidemiological reports. Climacteric 2000;3:3 – 12. [27] Pike MC, Spicer DV, Dahmoush L, Press MF. Estrogens, progestogens, normal breast cell proliferation, and breast cancer risk. Epidemiol Rev 1993;15:17 – 35. [28] Russo J, Russo ICH. Towards a physiological approach to breast cancer prevention. Can Epid Biomed Prev 1994;3:353 – 64. [29] Clemons M, Gloss P. Estrogen and the risk of breast cancer. New Engl J Med 2001;344:276 – 85. [30] van Landeghem AAJ, Poortman J, Nabuurs M, Thijssen JHH. Endogenous concentration and subcellular distribution of estrogens in normal and malignant human breast tissue. Cancer Res 1985;45:2900 – 6. [31] Miyoshi Y, Anto A, Shiba E, Taguchi T, Tamaki Y, Noguchi S. Involvement of up-regulation of 17beta-hy-

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

[32]

[33]

[34]

[35]

[36]

[37]

[38]

[39]

[40]

[41]

[42]

[43]

[44]

droxysteroid dehydrogenase type 1 in maintenance of intratumoral high estradiol levels in postmenopausal breast cancers. Int J Cancer 2001;94:685 –9. Reed MJ, Purohit A. Breast cancer and the role of cytokines in regulating estrogen synthesis: an emerging hypothesis. Endocr Rev 1997;18:701 – 15. Reed MJ, Owen AM, Lai LC, Coldham NG, Ghilchik MW, Shaikh NA, James VHT. In situ oestrone synthesis in normal breast and breast tumour tissues: effect of treatment with 4-hydroxyandrostenedione. Int J Cancer 1989;44:233 – 7. Blankenstein MA, Maitimu-Smeele I, Donker GH, Daroszewski J, Milewicz A, Thijssen JHH. Tissue androgens and the endocrine autonomy of breast cancer. J Steroid Biochem Mol Biol 1992;43:167 –71. Lipton A, Santner SJ, Santen RJ, Harvey HA, Feil PD, White-Hershey D, Bartholomew MJ, Antle CE. Aromatase activity in primary and metastatic human breast cancer. Cancer 1987;59:779 – 82. Harada N, Utsumi T, Takagi Y. Tissue-specific expression of the human aromatase cytochrome P-450 gene by alternative use of multiple exons 1 and promoters, and switching of tissue-specific exons 1 in carcinogenesis. Proc Natl Acad Sci USA 1993;90:11312 –6. Zhou C, Zhou D, Esteban J, et al. Aromatase gene expression and its exon I usage in human breast tumors: detection of aromatase messenger RNA by reverse transcription-polymerase chain reaction. J Steroid Biochem Mol Biol 1996;59:163 –71. Yue W, Wang JP, Hamilton CJ, Demers LM, Santen RJ. In situ aromatization enhances breast tumor estradiol levels and cellular proliferation. Cancer Res 1998;58:927 – 32. Perel E, Wilkins D, Dillinger DW. The conversion of androstenedione to estrone, estradiol, and testosterone in human breast tissue. J Steroid Biochem 1980;13:89 – 94. Dao TL, Hayes C, Libby PR. Steroid sulfatase activities in human breast tumors. Proc Soc Exp Biol Med 1974;146:381– 4. Pasqualini JR, Chetrite G, Blacker C, Feinstein MC, Delalonde L, Talbi M, et al. Concentrations of estrone, estradiol, and estrone sulfate and evaluation of sulfatase and aromatase activities in pre and postmenopausal breast cancer. J Clin Endocr Metab 1996;81:1460 –4. Chetrite GS, Pasqualini JR. The selective estrogen enzyme modulator (SEEM) in breast cancer. J Steroid Biochem Mol Biol 2001;76:95 –104. Blankenstein MA, van de Ven J, Maitimu-Smeele I, Donker GH, de Jong PC, Daroszewski J, Szymaczak J, Milewicz A, Thijssen JHH. Intratumoral levels of estrogens in breast cancer. J Steroid Biochem Mol Biol 1999;69:293 – 7. Khan SA, Rogers MA, Obando JA, Tamsen A. Estrogen receptor expression of benign breast epithelium and its association with breast cancer. Cancer Res 1994;54:993 – 7.

S51

[45] Shoker BS, Jarvis C, Clarke RB, et al. Estrogen receptorpositive proliferating cells in the normal and precancerous breast. Am J Pathol 1999;155:1811 – 5. [46] Hall JM, McDonnell DP. The estrogen receptor beta-isoform (ERbeta) of the human estrogen receptor modulates ERalpha transcriptional activity and is a key regulator of the cellular response to estrogens and anti-estrogens. Endocrinology 1999;140:5566 – 78. [47] Speirs V, Parkes AT, Kerin MJ, Walton DS, Carleton PJ, Fox JN, et al. Coexpression of estrogen receptor a and ß poor prognostic factors in human breast cancer. Cancer Res 1999;59:525 – 8. [48] Jackisch C, van der Ven J, Thijssen JHH, Schneider HPG. Effects of HRT on estrogen metabolism in normal and cancerous breast tissue, in preparation. [49] van Landeghem AA, Poortman J, Nabuurs M, Thijssen JH. Endogenous concentration and subcellular distribution of androgens in normal and malignant human breast tissue. Cancer Res 1985;45:2907 – 12. [50] Colditz GA, Stampfer MJ, Willett WC, et al. Prospective study of estrogen replacement therapy and risk of breast cancer in postmenopausal women. J Am Med Assoc 1990;264:2648 – 53. [51] Gambrell DR. Proposal to decrease the risk and improve the prognosis in breast cancer. Am Obstet Gynecol 1984;150:119 – 28. [52] Sigelmann-Danieli N, Buetow KH. Constitutional genetic variation at the human aromatase gene (Cyp19) and breast cancer risk. Br J Cancer 1999;79:456 – 63. [53] Huang CS, Chern HD, Chang KJ, Cheng CW, Hsu SM, Shen CY. Breast cancer risk associated with genotype polymorphism of the estrogen-metabolizing genes CYP17, CYP1A1, and COMT: a multigenic study on cancer susceptibility. Cancer Res 1999;59:4870 – 85. [54] Hulka BS. Effect of exogenous estrogen on postmenopausal women: the epidemiologic evidence. Obstet Gynecol Surv 1980;35(6):389 – 99. [55] Samsioe G. The endometrium: Effects of estrogen and estrogen – progestogen replacement therapy. Int J Fertil 1994;39(Suppl. 2):84 – 92. [56] Genant HK, Luca J, Weiss S, et al. Low-dose esterified estrogen therapy. Effects on bone, plasma estradiol concentrations, endometrium, and lipid levels. Arch Int Med 1997;157:2609 – 15. [57] Notelovitz M, Varner RE, Rebar RW, et al. Minimal endometrial proliferation over a 2-year period in postmenopausal women taking 0.3 mg of unopposed esterified estrogens. Menopause 1997;4:80 – 8. [58] The Esclim Study Group, Utian WH, Burry KA, Archer DF, Gallagher JC, Boyett RL, Guy MP, Tachon GJ, Chadha-Boreham HK, Bouvet AA. Efficacy and safety of low, standard, and high dosage of an estradiol transdermal system (Esclim) compared with placebo on vasomotor symptoms in highly symptomatic menopausal patients. Am J Obstet Gynecol 1999;181:71 – 9. [59] Schneider HPG, Birkha¨ user M. Does HRT modify risk of gynecological cancers. Int J Fertil 1995;40(Suppl. 1):40 – 53.

S52

H.P.G. Schneider / Maturitas 43 (2002) S35– S52

[60] Whitehead MI, Hillard TC, Crook D. The role and use of progestogens. Obstet Gynecol 1990;75:59 –76. [61] Sturdee DW, et al. Abstract. Seventh international congress on the menopause, Stockholm, 1993. [62] Gusberg SB. The individual at high-risk for endometrial carcinoma. Obstet Gynecol 1976;126:535 –42. [63] Gambrell RD Jr. Estrogens, progestogens and endometrial cancer. J Reprod Med 1977;18:301 –6. [64] Gambrell RD Jr. Prevention of endometrial cancer with progestogens. Maturitas 1986;8:159 –68. [65] Gambrell RD Jr, Massey FM, Castaneda TA, et al. Use of progestogen challenge test to reduce the risk of endometrial cancer. Obstet Gynecol 1980;55:732 –8. [66] Fletcher R, Kaufman RH. Endometrial adenocarcinoma in Stein– Leventhal syndrome. Cancer 1974;34:444 – 52. [67] Creasman WT, Henderson D, Hinshaw W, Clarke-Pearson DL. Estrogen replacement therapy in the patient treated for endometrial cancer. Obstet Gynecol 1986;67:326 – 30. [68] Lee RB, Burke TW, Park RC. Estrogen replacement therapy following treatment for stage I endometrial carcinoma. Obstet Gynecol 1990;67:326 –30. [69] Chapman JA, Berman ML, Gelotte DL, Roth PD, Osann K, DiSaia PK. Estrogen replacement therapy in surgical stage I and II endometrial cancer survivors. Am J Obstet Gynecol 1996;175:1195 –200. [70] Baker DD. Estrogen replacement therapy in patients with previous endometrial carcinoma. Comp Ther 1990;16:28 – 35. [71] Bryant GW. Administration of estrogen to patients with a previous diagnosis of endometrial adenocarcinoma. South Med J 1990;83:725 –6. [72] Beresford SAA, Weiss NS, Voigt LF, McKnight W. Risk of endometrial cancer in relation to estrogen combined with cyclic progestogen therapy in postmenopausal women. Lancet 1997;349:458 –61. [73] Persson I, Adami H-O, Bergqvist L, et al. Risk of endometrial cancer after treatment with estrogen alone or in conjunction with progestogens. Br Med J 1989;298:147 – 51. [74] Greene MH, Clark JW, Blayney DW. The epidemiology of ovarian cancer. Semin Oncol 1984;11:209. [75] Collaborative Ovarian Cancer Group, Whittemore AS, Harris R, Itnyre J. Characteristics relating to ovarian cancer risk: collaborative analysis of 12 US case – control studies. II: Invasive epithelial ovarian cancers in white women. Am J Epidemiol 1992;136:1184. [76] Vessey MP, Painte R. Endometrial and ovarian cancer and oral contraceptives — findings in a large cohort study. Br J Cancer 1995;71:1340. [77] Rosenberg L, Palmer JR, Zauber AG, Warshauer ME, Lewis JL Jr, Strom BL, Harlap S, Shapiro S. A case –control study of oral contraceptive use and invasive epithelial ovarian cancer. Am J Epidemiol 1994;139:654.

[78] Gross TP, Schlesselman JJ. The estimated effect of oral contraceptive use on the cumulative risk of epithelial ovarian cancer. Obstet Gynecol 1994;83:419. [79] Hereditary Ovarian Cancer Clinical Study Group, Narod SA, Risch H, Moslehi R, Dørum A, Neuhausen S, Olsson H, Provencher D, Radice P, Evans G, Bishop I-S, Brunet JS, Ponder BAJ. Oral contraceptives and the risk of hereditary ovarian cancer. New Engl J Med 1998;339: 424. [80] Fathalla MF. Incessant ovulation — a factor in ovarian neoplasia. Lancet 1971;ii:163. [81] SHARE Study Group, Ness RB, Grisso JA, Klapper J, Schlesselman JJ, Silberzweig S, Vergona R, Morgan M, Wheeler JE. Risk of ovarian cancer in relation to estrogen and progestin dose and use characteristics of oral contraceptives. Am J Epidemiol 2000;152:233. [82] Burger CW, Koomen I, Peters NAJB, van Leeuwen FE, Kenemans P. Postmenopausal hormone replacement therapy and cancer of the female genital tract and the breast. Eur Menopause J 1997;4/1:23. [83] Kaufman DW, Kelly JP, Welch WR, Rosenberg L, Stolley PD, Warshauer ME, Lewis J, Woodruff J, Shapiro S. Noncontraceptive estrogen use and epithelial ovarian cancer. Am J Epidemiol 1989;130:1142. [84] Survey of Women’s Health Study Group, Purdie D, Green A, Bain C, Siskind V, Ward B, Hacker N, Quinn M, Wright G, Russel P, Susil B. Reproductive and other factors and risk of epithelial ovarian cancer: an Australian case – control study. Int J Cancer 1985;62:678. [85] Collaborative Ovarian Cancer Group, Harris R, Whittemore AS, Itnyre J. Characteristics relating to ovarian cancer risk: collaborative analysis of 12 US case – control studies. III. Epithelial tumors of low malignant potential in white women. Am J Epidemiol 1992;136:1204. [86] Negri E, Tzonou A, Beral V, Lagiou P, Trichopoulos D, Parazzini F, Franceschi S, Booth M, La Vecchia C. Hormonal therapy for menopause and ovarian cancer in a collaborative re-analysis of European studies. Int J Cancer 1999;80:848. [87] Rodriguez C, Alpa VP, Calle EE, Jacob EJ, Thun MJ. Estrogen replacement therapy and ovarian cancer mortality in a large prospective study of US women. J Am Med Assoc 2001;285:1460. [88] Colombo N, Chiari S, Maggioni A, Bocciolone L, Torri V, Maggioni C. Controversial issues in the management of early ovarian epithelial cancer: conservative surgery and role of adjuvant therapy. Gynecol Oncol 1994;55:47. [89] Eeles RA, Tan S, Wiltshaw E, et al. Hormone replacement therapy and survival after surgery for ovarian cancer. Br Med J 1991;302:259. [90] Coughlin SS, Giustozzi A, Smith SJ, Lee NC. A metaanalysis of estrogen replacement therapy and risk of epithelial ovarian cancer. J Clin Epidemiol 2000;53:367.