Phytoestrogens and breast cancer risk

Transcription

1 Breast Cancer Research and Treatment 77: , Kluwer Academic Publishers. Printed in the Netherlands. Review Phytoestrogens and breast cancer risk Review of the epidemiological evidence P.H.M. Peeters, L. Keinan-Boker, Y.T. van der Schouw, and D.E. Grobbee Julius Center for Health Studies and Primary Care, University Medical Center Utrecht, The Netherlands Key words: breast cancer, dietary intake, enterolactone, isoflavones, lignans, phytoestrogens, review, soy, urinary excretion Summary Phytoestrogens are natural plant substances. The three main classes are isoflavones, coumestans, and lignans. Phytoestrogens have anticarcinogenic potential, but they have also significant estrogenic properties. For an evaluation of the effect of phytoestrogens on breast cancer risk we reviewed the analytical epidemiological data. A total of 18 studies were included [1 18]. Up to now, there are 13 studies that have assessed the direct relation between the individual dietary intake of soy products and the risk of breast cancer [1 13]. Overall, results do not show protective effects, with the exception maybe for women who consume phytoestrogens at adolescence or at very high doses [5, 7, 8]. Only four of these 13 studies are prospective, and none of them found statistically significant breast cancer reductions. Four studies assessed urinary isoflavones excretion in relation to breast cancer [14 17]. Three of these are case control studies [14 16], where excretion was measured after breast cancer occurrence and thus seriously limiting causal interpretation of the results. The only prospective study with urinary measurements before breast cancer occurrence was done in a Dutch postmenopausal population and showed a non-significant breast cancer risk reduction for high excretion [17]. Three studies measured enterolactone (lignan): two case control studies reported a preventive effect on breast cancer risk [14, 18], but the only prospective study did not [17]. In conclusion, few prospective studies (n = 5) were done to assess the effects of phytoestrogens on breast cancer risk. None of them found protective effects. However, these prospective studies did not focus on age at consumption, which seems to be important based on results from dietary case control studies done so far. Introduction Phytoestrogens are natural plant substances [19, 20]. The three main classes are isoflavones, coumestans, and lignans. Isoflavones, the most frequently studied, show structural similarity to mammalian estrogens and are present in large amounts in soybeans, and soy products such as miso and tofu. The most important isoflavones are genistein and daidzein. Biochanin A and formononetin are precursors, present in some legumes, which are metabolized to genistein and daidzein, respectively. Coumestans occur in bean sprouts and fodder crops, and they have received little scientific attention so far. Lignans occur as prelignans and are more widespread in plants where they form the building block in cell walls. Linseed (flaxseed) especially is a rich source, but lignans also occur at relatively high amounts in whole grains and berries [21]. In the 1940s it was recognized that large number of sheep grazing on Australian pastures, containing a particular type of clover, developed infertility. This problem was traced in 1976 to a specific type of phytoestrogen, which converted to daidzein (a major type of isoflavones), with weak non-steroidal estrogenic activity [22]. In the 1980s, soybeans were also responsible for an infertility syndrome in cheetahs in Cincinnati zoo. Removing the soybeans from the feed reversed the syndrome [23]. These examples show that plant compounds have biological capacity; consequently, potentially preventive actions of phytoestrogens on the occurrence of

2 172 PHM Peeters et al. cancer in humans have been suggested and studied in the 1980s and 1990s [24 28]. Isoflavones have received scientific attention largely due to their estrogenic potential: many metabolic conversions occur in the gut, resulting in hormone-like compounds. The postulated anticarcinogenic mechanism involves the weak estrogen-like activity. Isoflavones bind to estrogen receptors, preferentially to the estrogen beta receptor [29]. They initiate only modest response and at the same time they block the binding of more potent estrogens. There is structural similarity to the potent synthetic antiestrogen, tamoxifen, a drug that is effective in breast cancer treatment and prevention [30 32]. There are numerous in vitro studies showing that genistein, and to a lesser extent daidzein, inhibit the growth of a wide range of both hormone-dependent and hormone-independent cancer cells. Genistein has been shown to inhibit growth in many cell lines including those without estrogen receptors. Phytoestrogens may act as antioxidants and/or inhibit angiogenesis [33, 34], the formation of new blood vessels, essential for a tumor to expand. In animal studies, soybean products have generally reduced tumors induced by chemical carcinogens, while diets supplemented with soybean revealed lower number of tumors in rats [35]. Several recent intervention studies in humans (including 8 60 women) assessed effects of soy or flaxseed consumption on hormone levels. Although Martini et al. could not find changes in the urinary excretion of hormones after soy supplementation in 36 premenopausal women [36], five other studies did show effects of soy in premenopausal women. These included a decrease in serum 17beta estradiol [37, 38], a decrease of plasma FSH and LH [39], or an increase in the urinary 2-hydroxyestrone/16alphahydroxyestrone ratio [40, 41]. These data suggest that isoflavones may exert cancer-preventive effects in premenopausal women by altering estrogen metabolism away from genotoxic metabolites toward inactive metabolites. The same effect of soy consumption was recently also reported for postmenopausal women, although less pronounced [42], as well as for flaxseed consumption in pre- and postmenopausal women [43, 44]. Overall the impression is that phytoestrogens may exert anticarcinogenic potential by altering blood levels of relevant hormones and/or exhibit antioxidant or antiangiogenic effects. However, phytoestrogens have also significant estrogenic properties in in vitro and in vivo models [45 48]. A recent study found no differences in later pubertal maturation or growth between infants who had been fed soy-based versus milk-based formula [49]. A closer look at the data however did show small differences in cycle length, theoretically reflecting early life exposure to estrogens [50]. Epidemiological and clinical data support the view that lifetime exposure to estrogens has significant influence on increased breast cancer risk [51 53]. Based on findings of studies addressing the estrogenecity or antiestrogenecity of phytoestrogens, it cannot be concluded whether the net result of their consumption will be an inhibition or a stimulation of carcinogenesis. For an evaluation of their net result we need population data. This review presents the analytical epidemiological studies published so far on phytoestrogens and breast cancer risk. Methods Identification of studies We decided to include analytical epidemiological studies that have data on: (a) dietary soy intake or (b) urinary excretion of isoflavones or lignans or (c) blood measurements of isoflavones or lignans. Soy is used as a marker for isoflavones intake. For lignans there is not a single, clear dietary food group that can be used as a marker. Lignans are more widespread than isoflavones and occur in whole grains, fruit, berries, fiber, and flaxseed [54, 55]. We decided not to review these food groups in relation to breast cancer, because this is beyond the scope of our review. Instead we decided to focus on direct (urinary or blood) measurements of lignans. In order to be included in this review, studies had to have data on individual level and to be published before November We identified studies by computer-aided searches and by consulting review articles. In total, we identified 18 studies that met the above-mentioned criteria [1 18]. Review method The review is divided into three sections: (1) studies that assessed the relation through the dietary intake of soy (as a proxy for the intake of isoflavones) in relation to breast cancer, (2) studies that measured urinary excretion of isoflavones in relation to breast cancer, and finally (3) studies that measured lignans (enterolactone) and breast cancer risk. We present means for

3 Phytoestrogens and breast cancer risk 173 cases and controls, or relative risk (RR) estimates and 95% confidence intervals as they were presented in the original papers. Results are presented according to study design (i.e., case control or prospective) and menopausal status. We decided not to combine the published results into one summary effect estimate, because we believe that the methodological differences between the studies are too large. Studies differed in: (1) study design (case control, prospective); (2) the amounts of isoflavones consumed (Asian, Western populations); (3) the measurement method of isoflavones (questionnaire: soy products, urine or blood: tertiles or quartiles of several subclasses of isoflavones); (4) the types of phytoestrogens included (isoflavones, lignans); (5) the cut-off points used in the analyses of soy consumption (g of soy product, weekly consumption of miso, etc.); and finally (6) menopausal status (pre- or postmenopausal women or mix). All of these issues are relevant for the interpretation of the observed relations, and cause heterogeneity, so that combining studies is not meaningful. Results Dietary measurements of soy intake and breast cancer Studies on geographical differences in breast cancer occurrence suggest preventive effects for soy products [28, 57]. Women in Eastern populations show lower breast cancer rates compared to Western countries and they also consume large amounts of phytoestrogens due to the natural presence of these compounds in their traditional diet. Women from Far-Eastern populations overall also have longer menstrual cycle lengths [58], and have up to roughly 40% lower estrogen levels [59, 60]. Up to now, there are 13 published epidemiological studies that directly assessed relations between dietary intake of soy (as a proxy for isoflavones consumption) and the risk of subsequent breast cancer [1 13] (Table 1, adopted from [13, 56]). Nine of these studies are case control studies [1 9]. Two early case control studies, one in 1991 [2, 61] and one in 1995 [3], showed clear protective effects for frequent soy (product) consumption only in premenopausal women. Later studies do not find differential effects for pre- and postmenopausal women [4, 5, 7 9]; and most of these studies do not find a protective effect for frequent soy (food) consumption [4 6, 8]. The case control study of Wu et al. (1996) is interesting: it showed protective effects, but only significant in non-us born Asian Americans (RR per weekly soy product consumption 0.8; 95% CI 0.7; 0.9) [5]. The authors hypothesized that soy intake is protective only when consumed at higher doses and at younger ages. The mean intake of soy products of US born Asian women is much lower, when compared to non-us born Asian women, who migrated to US. Intake of tofu (in times per year) was more than double among Asian Americans who were migrants (62.0) compared to those who were born in the West (29.5) [5]. Relations with breast cancer may be different for different parts of the dose response relation between soy and breast cancer risk [5]. A recent case control study in Shanghai indeed showed protective effects of soy food intake at adolescence [7], and also at older ages but then only when consumed in high amounts (upper decile of intake) [8]. The inverse association between adolescent soyfood intake and breast cancer risk persisted after control for usual soyfood intake in adults, suggesting that an effect of soyfood intake early in life is independent from adult soyfood intake [7, 8]. All these studies used dietary questionnaires to quantify soy intake [1 8]. It is difficult to accurately measure usual soy intake. And further, all studies were case control studies, which may be subjected to an additional problem, that is, recall bias (cases may remember their diet more or less accurate than healthy controls). Up to now, there are only four prospective studies published [10 13] (Table 2). Three studies were done in Asian populations [10, 11, 13], and for the only one in a non-asian population (Iowa Women s Health Study) only an abstract is available [12]. All four studies show small decreases in breast cancer occurrence for those women who consume (frequently) soy (food), but none of the results were statistically significant. Of concern is that soy intake may be homogeneously high in Asia, making it difficult to identify differences in breast cancer risk between moderate and high consumption. The same problem may occur when studying relations in Western populations, where consumption of soy (products) is homogeneously low. The mean intake of soy is on average times higher in Chinese women who are born and living in Asia (36 45 g soy products/day), than by Asian Americans ( g soy products/ day) [56]. The upper quartile of intake in Asian

4 Table 1. Overview of epidemiological studies on dietary soy and breast cancer risk, case control studies Author (Journal, year) Country Design Dietary questionnaire Results (means; or RR (95% CI)); adjustment (for risk factors soy breast cancer or nutrition) different for studies Hirohata (NCIM, 212 cases, 212 Mean g fat from Combined pre- and 1985) Fukuoka, Japan [1] neighborhood controls soybeans/week postmenopausal Cases 26 g/week Controls 26 g/week 174 PHM Peeters et al. Lee (Lancet, 1991) 200 cases, 420 controls, Soy product Premenopausal Postmenopausal Singapore [2] hospital based case control <20 g/day g/day 0.6 (0.3; 1.2) 0.9 (0.4; 1.9) 55 g/day 0.4 (0.2; 0.9) 1.1 (0.5; 2.3) Hirose (JJCR, 1995) 1186 cases, 21,295 controls, Bean curd Premenopausal Postmenopausal Japan [3] outpatients, case control 3 times/month times/week 0.9 (0.7; 1.2) 0.9 (0.6; 1.2) 3 times/week 0.8 (0.6; 1.0) 1.0 (0.7; 1.3) Miso soup Occasionally Daily 1.2 (1.0; 1.4) 1.0 (0.8; 1.2) Yuan (BJC, 1995) Shanghai 534 cases, 534 controls, Soy protein per 18 g/day Pre- and postmenopausal, 1.0 (0.7; 1.4) Shanghai and Tianjin, China [4] Tianjin 300 cases, 300 controls, population based case control Wu (CEBP, 1996) USA, 597 cases, 966 controls, Soy product 1 time/week US born Asians, pre- and No US born Asians, pre- and Chinese, Japanese and Filippino [5] population based case control postmenopausal combined, postmenopausal combined, 0.9 (0.7; 1.2) 0.8 (0.7; 0.9) Witte (BCRT, 1997) 140 cases, 222 healthy sisters Tofu Premenopausal (bilateral) US and Canada [6] <1 per week per week 0.5 (0.2; 1.1)

5 . Shu (CEBP, 2001) 1459 cases, 1556 controls, Soy food at age years Pre- and postmenopausal Results the same for pre- and Chinese in Shanghai [7] population based case control (median intake 5.4 g/day) postmenopausal. Test for Quintile trend significant Quintile (0.6; 0.9) Quintile (0.6; 0.9) Quintile (0.6; 0.9) Quintile (0.4; 0.7) Dai (BJC, 2001) The same study as [7] Adult intake soy protein Pre- and postmenopausal Significant effects only Chinese in Shanghai [8] (median intake 56 g/week) 1.0 in upper 2 deciles of intake Occasionally <1 g/week 0.8 (0.5; 1.2) (RR = 0.7; 0.5; 1.0). Weekly Effect strongest in high BMI (over 25); and for ER and PR positive cancer Horn-Ross (AJE, 2001) 1326 cases, 1657 controls, Total isoflavones intake computed Pre- and postmenopausal Results the same for Non-Asian multiethnic population based from food frequency questionnaire pre- and postmenopausal US women [9] case control study <1.05 mg/day mg/day 1.1 (0.9; 1.4) mg/day 1.2 (0.9; 1.5) >2.77 mg/day 1.0 (0.8; 1.3) Phytoestrogens and breast cancer risk 175

6 Table 2. Overview of epidemiological studies on dietary soy and breast cancer risk, prospective studies Author (Journal, year) Country Design Dietary questionnaire Results (means; or RR (95%CI)) adjustment (for risk factors soy breast cancer or nutrition) different for studies 176 PHM Peeters et al. Nomura (AJCN, 1978) 6860 males, with 86 married to a Mean g tofu/week (spouse) Combined pre- and postmenopausal Japanese Hawaii [10] woman with breast cancer Cases 117 (s.e. 24) Controls 151 (s.e. 3) Mean g miso/week (spouse) Cases 171 (s.e. 54) Controls 280 (s.e. 6) Hirayama (Book, 1990) 142,857 women, Miso soup Japan [11] 241deaths <1 per day 1.0 Combined pre- and postmenopausal 1 per day 0.9 (0.7; 1.1) Greenstein (AJE, 1996) Iowa Women s Health Never soya or tofu 1.0 Only postmenopausal US [12] Study, 1018 cases Consumers soya or tofu 0.8 (0.5; 1.2) Key (BJC, 1999) 34,759 women, 427 cases, Tofu Combined pre and post Results the same for Japan [13] prospective population based study 1 per week 1.0 pre- and postmenopausal 2 4 times/week 1.0 (0.8; 1.2) 5 times/week 1.1 (0.8; 1.5) Miso soup 1 per week times/week 1.0 (0.8; 1.3) 5 times/week 0.9 (0.7; 1.1)

7 Phytoestrogens and breast cancer risk 177 Americans is still 2 3 times lower compared to the lowest quartile of women in Asia. In non-asian women, consumption of soy (foods) is even lower. In Iowa Women s Health Study less than 3% reported often consumption of soy (foods) [12]. One case control study and one prospective study in non-asian populations did not find significant protection for soy [6, 12]. However, in Western countries, isoflavones are also present in soybean flour, soymilk and soy sauce, and in much smaller amounts in dried beans. Few analyses of isoflavones in food have been done and food composition tables do usually not include values of isoflavones for different food groups. Some work in this area was done and has created few Food Composition Tables [62, 63]. With the use of this table, for the first time in a non-asian US population phytoestrogen intake was computed by using a food frequency questionnaire in breast cancer cases and controls [9]. No protective effects were observed for isoflavones (or phytoestrogens). Urinary isoflavones and breast cancer When consumed, the plant isoflavones (and lignans) undergo metabolism by bowel microflora, and are absorbed to a variable extent [64, 65]. Intake of foods rich in phytoestrogens is followed by a peak urinary excretion in the subsequent 24 h; excretion turns to its previous rate in h [66, 67]. Urinary excretion is dose dependent at low to moderate levels of soy protein intake [68, 69]. Adlercreutz was the first to detect lower urinary excretion of equol (and also of the lignans enterolactone and enterodiol) in 7 postmenopausal breast cancer cases compared to 10 healthy American women [24]. The first large-scale case control study with urinary excretion measurements was published in 1997 [14] (Table 3). It included a mix of pre- and postmenopausal Australian women and showed protective effects for several types of isoflavones. A statistically significant result was only seen for equol (a metabolite of daidzein) (fourth v.s. first quartile: RR = 0.3; 95% CI 0.1; 0.7). This study was criticized because isoflavones are short-term biomarkers of dietary intake. Since breast cancer cases in Ingram s study were recently confronted with the diagnosis when urine was collected, they may have changed their dietary intake. Moreover, the participation rate of controls was below 50% and healthy subjects may have preferentially been selected [70 75]. A second case control study, published by Zheng et al. [15] again showed protective effects of higher excretion levels of the isoflavones daidzein and genistein, although results did not reach statistical significance [15]. The population comprised both pre- and postmenopausal women living in Shanghai, China. Although participation rates were higher than in Ingram s study [14, 15], again urine was collected after breast cancer diagnosis. This was also the case in a very small Australian study: lower urinary mean levels of daidzein and genistein were observed in 18 postmenopausal breast cancer cases compared to 20 controls [16]. It is difficult to interpret the temporality of the observed associations in these three case control studies. One may question the relevance of phytoestrogen levels in urine samples collected between diagnosis and surgery to induction or progression of the disease many years earlier. The only prospective study that was done included postmenopausal women in The Netherlands with urine collected 1 9 years prior to disease diagnosis [17]. A non-significant breast cancer risk reduction was observed for higher levels of genistein (third tertile v.s. first: RR = 0.8; 95% CI 0.5; 1.5). Lignans and breast cancer Precursors of lignans occur amongst others in fruits and vegetables. Recently, a pooled analysis of eight prospective studies estimated a borderline significant reduced breast cancer risk for the highest quartile of intake (RR = 0.93; 95% CI 0.86; 1.00) [76]. Several mechanisms have been suggested, involving a reduction of bioactive estrogen levels, due to the presence of lignans in these foods. However, direct assessments of the dietary consumption of lignans and breast cancer risk are not available. This is due primarily to lack of data on the content of these compounds in a wide variety of foods. Three studies, of which only one is of a prospective design, measured enterolactone in urine or serum in breast cancer cases and controls [14, 17, 18] (Table 4). In Ingram s study, women in the fourth quartile of enterolactone excretion (vs. the first quartile) were at decreased breast cancer risk (RR = 0.4; 95% CI 0.2; 0.9) [14]; and in the second case control study women in the fifth quintile of serum levels of enterolactone showed reduced breast cancer risk (RR = 0.4; 95% CI 0.2; 0.8) [18]. Both studies are case control studies with again measurement of lignans after disease occurrence and so seriously limiting causal interpretation

8 Table 3. Overview of epidemiological studies on urinary isoflavones excretion and breast cancer Author (Journal, year) Country Design Isoflavones: urinary excretion Results, (means; or RR, 95% CI); adjustment (for risk factors breast cancer or nutrition) different for studies Ingram (Lancet, 1997), 144 cases, 144 controls, Equol quartile Mix of pre- and Australia [14] case control study (median: 109 nmol/24 h) postmenopausal women; results the same for (0.2; 1.0) daidzein, non-significant (0.2; 1.2) (0.1; 0.7) 178 PHM Peeters et al. Zheng (CEBP, 1999) 60 cases, 60 controls, Genistein tertile Mix of pre- and postmenopausal Shanghai, China [15] population based case control study (median: 2.2 nmol/mg creatinine) women; results the same for daidzein (0.2; 1.1) (0.3; 1.8) Murkies (Menopause, 2000) 18 cases, 20 controls, Mean genistein nmol/day Australia [16] case control study Cases 25 (5; 132) Only postmenopausal Controls 155 (43; 550) Mean daidzein Cases 31 (4; 234) Controls 427 (96; 1906) Tonkelaar den (CEBP, 2001) 88 cases, 268 controls, Genistein tertile The Netherlands [17] prospective (nested case control) study, (median: 0.8 nmol/mg creatinine) Only postmenopausal urine collection 1 9 years prior to diagnosis (0.5; 1.6) (0.5; 1.5)

9 Table 4. Overview of epidemiological studies on lignans and breast cancer risk Author (Journal, year) Country Design Lignans biomarker Results (RR (95% CI)); adjustment (for risk factors breast cancer or nutrition) different for studies Ingram (Lancet, 1997) 144 cases, 144 controls, Enterolactone quartile (urine) Mix of pre- and postmenopausal women; Australia [14] case control study (median: 3098 nmol/24 h) increased risk for matairesinol, n.s (0.4; 2.0) (0.3; 1.4) (0.2; 0.9) Tonkelaar den (CBEP, 2001) 88 cases, 268 controls, Enterolactone tertile (urine) Only postmenopausal women The Netherlands [17] prospective (nested case control) study, (median: 3290 nmol/24 h) urine collected 1 9 years prior to diagnosis (0.6; 1.9) (0.8; 2.6) Pietinen (CBEP, 2001) 194 cases, 208 controls, Enterolactone quintile (serum) Effect the same in pre- Finland [18] case control study 1 mean: 3 nmol/l 1.0 and postmenopausal women (0.3; 1.2) (0.3; 1.1) (0.3; 1.1) 5 mean: 54 nmol/l 0.4 (0.2; 0.8) Phytoestrogens and breast cancer risk 179

10 180 PHM Peeters et al. of the results. Again, the only prospective study could not disclose any protective effect (third v.s. first tertile of enterolactone excretion: RR = 1.4; 95% CI 0.8; 2.6) [17]. Discussion The anticarcinogenic potential of isoflavones on breast cancer risk was evaluated in 13 studies that measured usual dietary soy (product) (or isoflavones) intake in relation to breast cancer [1 13]. Only four of these studies were prospective and none of these prospective studies found significant protective effects [10 13]. Most of these 13 studies were done in Asian populations where soy consumption is homogeneously high. Relations with breast cancer may be difficult to detect if consumption for the total population is above a threshold value. If there is a protective effect in these populations it seems to be present for women who consume high amounts of soy at young ages [5, 7, 8]. Although the first studies reported protective effects only in premenopausal women [2, 3] later studies did not show differential effects between pre- and postmenopausal breast cancer [4, 5, 7 9, 13]. In Western populations usual consumption of isoflavones is much lower. Assessment through the diet is difficult due to lack of food composition data. One case control study so far was done that assessed dietary intake of isoflavones in a mixed ethnic, non- Asian population in relation to breast cancer risk [9]. They were not able to disclose any protective effect. There were four studies that measured urinary excretion of isoflavones in relation to breast cancer risk [14 17]. Only one study was prospective, and reported a non-significant reduction in breast cancer risk [17]. Lignans are more widespread than isoflavones and may therefore be more important in Western populations. Again, food composition tables lack this information, explaining the absence of studies that estimated dietary intake of lignans and breast cancer risk. Three studies measured enterolactone (in blood or urine) in relation to breast cancer [14, 17, 18]. The only prospective study was inconclusive [17]. Levels of phytoestrogens in urine or blood do not only reflect dietary intake. Urinary lignan excretion correlates with fruit and vegetable intake, as well as with consumption of fiber and tea [54, 55, 77, 78]. The condition of the intestinal flora is also crucial and therefore use of antibiotics may decrease urinary enterolactone levels for longer than 40 days [64]. Genetic influences may also play a role. From the published data we may conclude that overall information about phytoestrogen consumption and breast cancer risk is still scarce: all prospective studies (only five in total) show non-significant results [10 13, 17]. From the case control studies in Asian populations we get the impression that consumption at young ages (adolescence or earlier) and consumption of high amounts may protect both against pre- and postmenopausal cancer [5, 7, 8, 56]. We may also conclude from this overview that none of the studies so far showed evidence for an increased risk of breast cancer with increased phytoestrogen intake. An absence of risk is important, since the putative physiological effects of phytoestrogens have created a market that has been utilized by the nutritional industry. Soy supplements are on the market now and soy flour is increasingly used in the nutritional industry due to its FDA (Food and Drug Administration) approved claim of reducing lipid levels and preventing cardiovascular diseases (October 1999). Women will increasingly consume phytoestrogens, even without knowing it. One should keep in mind that the studies included overall healthy subjects. It is not clear what the effect of increased consumption will be in women at high risk of breast cancer (i.e., previous cancer, genetic predisposition) [79]. Overall, prospective epidemiological studies on phytoestrogens and breast cancer are scarce, limiting our knowledge about the exposure disease relation. Acknowledgement We thank the World Cancer Research Fund (grant WCRF 2000/30) for the financial support. References 1. Hirohata T, Shigematsu T, Nomura AMY, Nomura Y, Horie A, Hirohata I: Occurrence of breast cancer in relation to diet and reproductive history: a case control study in Fukuoka, Japan. Natl Cancer Institute Monogr 69: , Lee HP, Gourley L, Duffy SW, Esteve J, Lee J, Day NE: Dietary effects on breast cancer risk in Singapore. Lancet 331: , Hirose K, Tajima K, Hamajima N, Inoue M, Takezaki T, Kuroishi T, Yoshida M, Tokudome S: A large-scale, hospitalbased case-control study of risk factors for breast cancer according to menopausal status. Jpn J Cancer Res 86: , 1995

11 Phytoestrogens and breast cancer risk Yuan JM, Wang QS, Ross RK, Henderson BE, Yu MC: Diet and breast cancer in Shanghai and Tianjin, China. Br J Cancer 71: , Wu AH, Ziegler RG, Horn-Ross PL, Nomura AMY, West DW, Kolonel LN, Rosenthal JF, Hoover RN, Pike MC: Tofu and risk of breast cancer in Asian-Americans. Cancer Epidemiol Biomarkers Prev 5: , Witte JS, Ursin G, Siemiatycki J, Thompson WD, Paganini- Hill A, Haile RW: Diet and premenopausal bilateral breast cancer: a case-control study. Breast Cancer Res Treat 42: , Shu XO, Jin F, Dai Q, Wen W, Potter JD, Kushi LH, Ruan Z, Gao YT, Zheng W: Soyfood intake during adolescence and subsequent risk of breast cancer among Chinese women. Cancer Epidemiol Biomarkers Prev 10: , Dai Q, Shu X-O, Jin F, Potter JD, Kushi LH, Teas J, Gao Y-T, Zheng W: Population based case control study of soyfood intake and breast cancer risk in Shanghai. Br J Cancer 85: , Horn-Ross PL, John EM, Lee M, Stewart SL, Koo J, Sakoda LC, Shiau AC, Goldstein J, Davis P, Perez-Stable EJ: Phytoestrogen consumption and breast cancer risk in a multiethnic population: the Bay Area Breast Cancer Study. Am J Epidemiol 154: , Nomura A, Henderson BE, Lee J: Breast cancer and diet among the Japanese in Hawaii. Am J Clin Nutr 31: , Hirayama T: Life-style and mortality: a large scale censusbased cohort study in Japan. In: Wahrendorf J (ed) Contributions to Epidemiology and Biostatistics. Vol 6, Karger, Basel, Greenstein J, Kushi L, Zheng W, Fee R, Campbell D, Sellers T, Folsom A: Risk of breast cancer associated with intake of specific foods and food groups. Am J Epidemiol 145: S36, Key TJ, Sharp GB, Appleby PN, Beral V, Goodman MT, Soda M, Mabuchi K: Soya foods and breast cancer risk: a prospective study in Hiroshima and Nagasaki, Japan. Br J Cancer 81: , Ingram D, Sanders K, Kolybaba M, Lopez D: Case-control study of phyto-oestrogens and breast cancer. Lancet 350: , Zheng W, Dai Q, Custer LJ, Shu X-O, Wen W-Q, Jin F, Franke AA: Urinary excretion of isoflavonoids and the risk of breast cancer. Cancer Epidemiol Biomarkers Prev 8: 35 40, Murkies A, Dalais FS, Briganti EM, Burger HG, Healy DL, Wahlqvist ML, Davis SR: Phytoestrogens and breast cancer in postmenopausal women: a case control study. Menopause 7: , Tonkelaar den I, Keinan-Boker L, Veer van t P, Arts CJM, Adlercreutz H, Thijssen JHH, Peeters PHM: Urinary phytooestrogens and postmenopausal breast cancer risk. Cancer Epidemiol Biomarkers Prev 10: , Pietinen P, Stumpf K, Mannisto S, Kataja V, Uusitupa M, Adlercreutz H: Serum enterolactone and the risk of breast cancer. A case control study in Eastern Finland. Cancer Epidemiol Biomarkers Prev 10: , Setchell KDR, Adlercreutz H: Mammelian lignans and phytoestrogens. In: Rowland I (ed) The Role of the Gut Flora in Toxicity and Cancer, Academic Press, London, 1988, pp Setchell KDR, Borriello SP, Hulme P, Axelson M: Nonsteroidal estrogens of dietary origin: possible roles in hormonedependent disease. Am J Clin Nutr 40: , Thompson LU, Robb P, Serraino M, Cheung F: Mammalian lignan production from various foods. Nutr Cancer 16: 43 52, Shutt DA: The effects of plant oestrogens in animal reproduction. Endeavour 35: , Setchell KDR, Gosselin SJ, Welsh MB, Johnston JO, Balistreri WF, Kramer LW, Dressner BL, Tart MJ: Dietary estrogens a probable cause of infertility and liver disease in captive cheetah. Gastroenterology 93: , Adlercreutz H, Fotsis T, Heikkinen R, Dwyer JT, Woods M, Goldin BR, Gorbach SL: Excretion of the lignans enterolactone and enterodiol and of equol in omnivorous and vegetarian postmenopausal women and in women with breast cancer. Lancet II: , Adlercreutz H: Western diet and Western diseases: some hormonal and biochemical mechanisms and associations. Scand J Clin Lab Inv 50 (S201): 3 23, Adlercreutz H, Markkanen H, Watanabe S: Plasma concentrations of phyto-estrogens in Japanese men. Lancet 342: , Messina M, Barnes S: The role of soy products in reducing risk of cancer. JNCI 83: , Messina MJ, Persky V, Setchell KDR, Barnes S: Soy intake and cancer risk: a review of in vitro and in vivo data. Nutr Cancer 21: , Kuiper GGJM, Carlsson B, Grandien K, Enmark E, Haggblad J, Nilsson S, Gustafsson JA: Comparison of the ligand binding specificity and transcript tissue distribution of estrogen receptors alpha and beta. Endocrinol 138: , Fisher B, Costantino JP, Wickerham DL, Redmond CK, Kavanah M, Cronin WM, Vogel V, Robidoux A, Dimitrov N, Atkins J, Daly M, Wieand S, Tan-Chiu E, Ford L, Wolmark N: Tamoxifen for prevention of breast cancer: report of the National Surgical Adjuvant Breast and Bowel Project P-1 Study. JNCI 90: , Early Breast Cancer Trialists Collaborative Group. Tamoxifen for early breast cancer: an overview of randomised trials. Lancet 351: , Powles T, Eeles R, Ashley S, Easton D, Chang J, Dowsett M, Tidy A, Viggers J, Davey J: Interim analysis of the incidence of breast cancer in the Royal Marsden Hospital tamoxifen randomised chemoprevention trial. Lancet 352: , Peterson G: Evaluation of the chemical targets of genistein in tumour cells. J Nutr 125: 784S 789S, Fotsis T, Pepper M, Adlercreutz H, Fleischmann G, Hase T, Montesano R, Schweigerer L: Genistein, a dietary derived inhibitor of in vitro angiogenesis. Proc Nat Acad Sci USA 90: , Barnes S, Grubbs, Setchell KDR, Carlson J: Soybeans inhibit mammary tumors in models of breast cancer. In: Pariza MW, Aeschbacher H-U, Felton JS, Sato S (eds) Mutagens and Carcinogens in the Diet. Wiley Liss, New York, 1990, pp Martini MC, Dancisak BB, Haggans CJ, Thomas W, Slavin JL: Effects of soy intake on sex hormone metabolism in premenopausal women. Nutr Cancer 34: , Nagata C, Takatsuka N, Inaba S, Kawakami N, Shimizu H: Effect of soymilk consumption on serum estrogen concentrations in premenopausal Japanese women. J Natl Cancer Inst 90: , Lu L-JW, Anderson KE, Grady JJ, Kohen F, Nagamani M: Decreased ovarian hormones during a soya diet: implications for breast cancer prevention. Cancer Res 60: , 2000

12 182 PHM Peeters et al. 39. Duncan AM, Merz BE, Xu X, Nagel TC, Phipps WR, Kurzer MS: Soy isoflavones exert modest hormonal effects in premenopausal women. J Clin Endocrinol Metab 84: , Xu X, Duncan AM, Merz BE, Kurzer MS: Effects of soy isoflavones on estrogen and phytoestrogen metabolism in premenopausal women. Cancer Epidemiol Biomarkers Prev 7: , Lu L-JW, Cree M, Josyula S, Nagamani M, Grady JJ, Anderson KE: Increased urinary excretion of 2-hydroxyestrone but not 16 alfa-hydroxyestrone in premenopausal women during soya diet containing isoflavones. Cancer Res 60: , Xu X, Duncan AM, Wangen KE, Kurzer MS: Soy consumption alters endogenous estrogen metabolism in postmenopausal women. Cancer Epidemiol Biomarkers Prev 9: , Haggans CJ, Travelli EJ, Thomas W, Martini MC, Slavin JL: The effect of flaxseed and wheat bran consumption on urinary estrogen metabolites in premenopausal women. Cancer Epidemiol Biomarkers Prev 9: , Haggans CJ, Hutchins AM, Olson BA, Thomas W, Martini MC, Slavin JL: Effect of flaxseed consumption on urinary estrogen metabolites in postmenopausal women. Nutr Cancer 33: , Petrakis NL, Barnes S, King EB, Lowenstein J, Wiencke J, Lee MM, Miike R, Kirk M, Coward L: Stimulatory influence of soy protein isolate on breast secretion in pre- and postmenopausal women. Cancer Epidemiol Biomarkers Prev 5: , Bingham SA, Atkinson C, Liggins J, Bluck L, Coward A: Review article. Phyto-oestrogens: where are we now? Br J Nutr 79: , Messina MJ: Legumes and soybeans: overview of their nutritional profiles and health effects. Am J Clin Nutr 70: 439S 450S, Bouker KB, Hilakivi-Clarke L: Genistein: does it prevent or promote breast cancer. Environ Health Persp 108: , Strom BL, Schinnar R, Ziegler EE, Barnhart KT, Sammel MD, Macones GA, Stallings VA, Drulis JM, Nelson SE, Hanson: Exposure to soy-based formula in infancy and endocrinological and reproductive outcomes in young adulthood. JAMA 286: , Goldman LR, Newbold R, Swan SH: Exposure to soybased formula in infancy. JAMA (letter) 286: , Thomas HV, Reeves GK, Key TJ: Endogenous estrogen and postmenopausal breast cancer: a quantitative review. Cancer Causes Control 8: , Key TJ: Serum oestradiol and breast cancer risk. Endocr Relat Cancer 6: , Clemons M, Goss P: Estrogen and the risk of breast cancer. N Engl J Med 344: , Kilkkinen A, Stumpf K, Pietinen P, Valsta LM, Tapanainen H, Adlercreutz H: Determinants of serum enterolactone concentration. Am J Clin Nutr 73: , Horner NK, Kristal AR, Prunty J, Skor HE, Potter JD, Lampe JW: Dietary determinants of plasma enterolactone. Cancer Epidemiol Biomarkers Prev 11: , Wu AH, Ziegler RG, Nomura AMW, West DW, Kolonel LN, Horn-Ross PL, Hoover RN, Pike MC: Soy intake and risk of breast cancer in Asians and Asian Americans. Am J Clin Nutr 68: 1437S 1443S, Adlercreutz H, Mousavi Y, Hockerstedt K: Diet and breast cancer. Acta Oncol 31: , Treloar AE, Boynton RE, Behn BG, Brown BW: Variation of the human menstrual cycle throughout life. Int J Fertil 12: , Goldin BR, Adlercreutz H, Gorbach SL, Woods MN, Dweyer JT, Conlon T, Bohn E, Gerrshoff SN: The relationship between estrogen levels and diets of Caucasian American and Oriental immigrant women. Am J Clin Nutr 44: , Key TJ, Chen J, Pike MC, Boreham J: Sex hormones in women in rural China and Britain. Br J Cancer 62: , Lee HP, Gourley L, Duffy SW, Esteve J, Lee J, Day NE: Risk factors for breast cancer by age and menopausal status: a case-control study in Singapore. Cancer Causes Control 3: , Pillow PC, Duphorne CM, Chang S, Contois JH, Strom SS, Spitz MR, Hursting SD: Development of a database for assessing dietary phytoestrogen intake. Nutr Cancer 33: 3 19, Horn-Ross PL, Barnes S, Lee M, Coward L, Mandel JE, Koo J, John EM, Smith M: Assessing phytoestrogen exposure in epidemiologic studies: development of a database (United States). Cancer Causes Control 11: , Setchell KD, Lawson AM, Borriello SP, Harkness R, Gordon H, Morgan DM, Kirk DN, Adlercreatz H, Anderson LC, Axelson M: Lignan formation in man microbial involvement and possible roles in relation to cancer. Lancet II: 4 7, Setchell KDM, Borriello SP, Hulme P, Axelson M: Nonsteroidal estrogens of dietary origin: possible roles in hormone-dependent disease. Am J Clin Nutr 40: , Kelly GE, Joannou GE, Reeder AY, Nelson C, Waring MA: The variable metabolic response to dietary isoflavones in humans. Proc Soc Exp Biol Med 208: 40 43, Axelson M, Sjovall J, Gutafsson BE, Setchel KD: Soya: a dietary source of the non-steroidal estrogen equol in man and animals. J Endocrinol 102: 49 56, Karr SC, Lampe AM, Slavin JL: Urinary isoflavonoid excretion in human dose-dependent at low to moderate levels of soy protein consumption. Am J Clin Nutr 66: 46 51, Chen Z, Zheng W, Custer LJ, Dai Q, Shu XO, Jin F, Franke AA: Usual dietary consumption of soy foods and its correlation with the excretion rate of isoflavones in overnight urine samples among Chinese women in Shanghai. Nutr Cancer 33: 82 87, Tesarik J, Mendoza C: Phyto-oestrogens and breast cancer. Lancet 351: 138; discussion , Humfrey C: Phyto-oestrogens and breast cancer. Lancet 351: 138; discussion , de Souza RM: Phyto-oestrogens and breast cancer. Lancet 351: ; discussion , Mangtani P, Silva ID: Phyto-oestrogens and breast cancer. Lancet 351: 137; discussion , Heaton KW, Lewis SJ: Phyto-oestrogens and breast cancer. Lancet 351: 137; discussion , Messina M, Barnes S, Setchell KD: Phyto-oestrogens and breast cancer. Lancet 350: , Smith-Warner SA, Spiegelman D, Yaun SS, Adami HO, Beeson WL, van den Brandt PA, Folsom AR, Fraser GE, Freudenheim JL, Goldbohm RA, Graham S, Miller AB, Potter

13 Phytoestrogens and breast cancer risk 183 JD, Rohan TE, Speizer FE, Toniolo P, Willett WC, Wolk A, Zeleniuch-Jacquotte A, Hunter DJ: Intake of fruit and vegetables and risk of breast cancer: a pooled analysis of cohort studies. JAMA 285: , Mazur WM, Wahala K, Rasku S, Salakka A, Hase T, Adlercreutz H: Lignan and isoflavonoid concentrations in tea and coffee. Br J Nutr 79: 37 45, Lampe JW, Gustafsson DR, Hutchins AM, Martini MC, Li S, Wahala K, Grandits GA, Potter J, Slavin J: Urinary isoflavonoid and lignan excretion on a Western diet: relation to soy, vegetable, and fruit intake. Cancer Epidemiol Biomarkers Prev 8: , This P, de la Rochefordiere A, Clough K, Fourquet A, Magdelenat H: The Breast Cancer Group of the Institut Curie: Phytoestrogens after breast cancer. Endocr-Relat Cancer 8: , 2001 Address for offprints and correspondence: Petra H.M. Peeters, Associate Professor of Epidemiology, Julius Center for Health Sciences and Primary Care, Room D01.335, University Medical Center Utrecht, PO Box 85500, 3508 GA Utrecht, The Netherlands; Tel.: ; Fax: ; P.H.M.Peeters@jc.azu.nl