Socialstyrelsen 106 30 Stockholm 2006-05-25 Remissvar: nationella riktlinjer för missbruks- och beroendevården. Om myndigheten behöver kontakta mig på grund av detta remissvar: e mail [email protected] Jag har sedan mitten av åttiotalet följt utvecklingen inom svensk beroendevård. Detta har jag gjort genom personalutbildning och handledning av tolvstegskunniga behandlare på Solliden i
- A |
J |K |
U |V |
Dietary phytoestrogen intake and premenopausal breast cancer risk in a german case-control studyInt. J. Cancer: 110, 284 –290 (2004)
Publication of the International Union Against Cancer DIETARY PHYTOESTROGEN INTAKE AND PREMENOPAUSAL BREASTCANCER RISK IN A GERMAN CASE-CONTROL STUDY INSEISEN ,* Regina PILLER , Silke HERMANN and Jenny CHANG-CLAUDE 1Unit of Human Nutrition and Cancer Prevention, Technical University of Munich, Freising-Weihenstephan, Germany2Department of Clinical Epidemiology, Deutsches Krebsforschungszentrum, Heidelberg, Germany A diet high in isoﬂavonoids (soy) is associated with lower
In Western populations with a low intake of isoﬂavonoids, breast cancer risk in Asian populations. Due to the low soy
dietary intake of another class of phytoestrogens, lignans, may be intake, dietary lignans may be the more important phy-
more important. Lignans, mainly secoisolariciresinol and mataires- toestrogen class in Western populations. We used a popula-
inol, are more widespread in plants and plant products (e.g., whole tion-based case-control study of breast cancer by age 50 in
grains, berries) that are frequently included in the usual Western southern Germany to evaluate the association between di-
etary intake of different phytoestrogens and premenopausal
diet.14,15 Analytic data for the lignan content of an acceptable breast cancer risk. Dietary information was collected from
range of foodstuffs have become available only recently, so few 278 premenopausal cases and 666 age-matched controls, us-
studies on the effect of lignan intake on breast cancer risk have ing a validated FFQ. Using multivariate logistic regression,
been performed.16,17 Additionally, lignans undergo bacterial me- the highest vs. lowest intake quartiles of daidzein and
tabolism in the human intestine, and only their metabolites entero- genistein yielded signiﬁcantly reduced ORs (95% CI) for
diol and enterolactone are able to pass the intestinal barrier to exert breast cancer risk of 0.62 (0.40 – 0.95) and 0.47 (0.29 – 0.74),
biologic effects.18,19 Studies using enterolactone concentrations as respectively. The protective effects of daidzein and genistein
biomarkers of lignan intake (and intestinal microbial activity) have were found only for hormone receptor-positive tumors. High
given conﬂicting results: an inverse association with breast cancer intake of other isoﬂavonoids, e.g., formononetin and biocha-
nin A, as well as the sum of isoﬂavonoids were not associated
risk was found in 3 case-control studies,20–22 but no clear answer with a decrease in risk. Breast cancer risk signiﬁcantly de-
creased with a high intake of the plant lignan matairesinol
Therefore, our aim was to assess the association of dietary (OR ؍ 0.58, 95% CI 0.37– 0.94) but not secoisolariciresinol or
intake of different phytoestrogens with premenopausal breast can- the sum of plant lignans. However, both estimated mamma-
cer risk in a German study population with a Western lifestyle, also lian lignans, enterodiol and enterolactone, were inversely
considering the hormone receptor status of breast cancer tissue. In associated with breast cancer risk, with ORs (95% CI) of 0.61
(0.39 – 0.98) and 0.57 (0.35– 0.92), respectively. No effect was
addition, the bioavailable intestinal lignan metabolites enterodiol found for total phytoestrogen intake. Our results suggest an
and enterolactone were calculated and related to breast cancer risk.
important role of dietary intake of daidzein and genistein,
despite low levels, as well as of matairesinol and mammalian
lignans to reduce premenopausal breast cancer risk in this
2004 Wiley-Liss, Inc.
Participants in our population-based case-control study were recruited in the 2 study areas of Freiburg and Rhein-Neckar- Key words: dietary intake; phytoestrogen; isoﬂavonoid; plant lignan;
Odenwald between January 1992 and December 1995. All Ger- mammalian lignan; breast cancer; premenopausal women man-speaking women who were not older than 50 years at the timeof diagnosis of incident in situ or invasive breast cancer wereeligible. Patients were identiﬁed by frequent monitoring of hospital By deﬁnition, phytoestrogens are secondary plant components admissions, surgery schedules and pathology records of 38 hospi- that show structural similarity to mammalian estrogens and, there- tals. Among the 1,020 eligible patients in both study regions, 706 fore, are able to bind to mammalian estrogen receptors.1 Of the 3 (70.2%) completed the study questionnaire. Controls matched by main classes, isoﬂavonoids, lignans and coumestanes, the isoﬂa- exact age and study region were selected from a random list of vonoids, namely daidzein and genistein, have received the most residents derived from the population registries. Of the 2,257 scientiﬁc attention in the last years due to their suggested health eligible controls, 1,381 (61.2%) participated in the study.25,26 The effects.2,3 Daidzein and genistein are found in high concentrationsin soybeans. Consequently, in Asian populations with a fairly highintake of soy and soy products, isoﬂavonoid intake might be high Abbreviations: BMI, body mass index; CI, conﬁdence interval; EPIC, enough to reveal biologic effects in humans.4 Soy consumption has European Prospective Investigation into Cancer and Nutrition; FFQ, food- been suggested to contribute to the lower prevalence of breast frequency questionnaire; FSH, follicle-stimulating hormone; LH, luteiniz- cancer in Asian compared to Western populations.5,6 Since estro- ing hormone; OR, odds ratio; SHBG, sex hormone– binding globulin.
gen levels affect breast cancer,7 the early postulated anticarcino-genic mechanism is based on the weak estrogen-like activity Grant sponsor: Kurt-Eberhard-Bode-Stiftung; Grant sponsor: Deutsche (compared to endogenous estrogen) of genistein and, to a lesser extent, daidzein, as found in in vitro studies.1,8,9 These substanceshave since been shown to have hormonal receptor-independentanticarcinogenic effects in cancer cell lines and animal studies, *Correspondence to: Unit of Human Nutrition and Cancer Prevention, such as inhibition of angiogenesis, effects on endogenous hormone Technical University of Munich, Alte Akademie 16, 85350 Freising- concentrations and metabolism as well as antioxidant activity.10–12 Weihenstephan, Germany. Fax: ϩ49-8161-713931.
Stimulatory effects on carcinogenic processes as observed in invitro studies were not found in vivo.10 According to previous reviews,10,13 analytic epidemiologic stud- Received 4 August 2003; Revised 30 October 2003; Accepted 10 ies in humans support the view of a protective effect of soy consumption or related phytoestrogen intake on breast cancer riskwhen consumed at an early age (adolescence) or at very high doses. So far, only few investigations have been conducted in Published online 26 February 2004 in Wiley InterScience (www.
study was approved by the local ethical committee, and all partic- and total physical activity did not affect the estimates signiﬁcantly and, thus, were not used for adjustment. For trend estimation, Both cases and controls ﬁlled in a self-administered risk factor categorized variables were entered into the regression models as questionnaire, collecting information on demographic factors, an- thropometric measures, menstrual history, reproductive history,breast-feeding history, family history of cancer, smoking history and alcohol consumption. If not otherwise speciﬁed, all informa-tion was truncated at the reference date, which was the date of A total of 278 cases and 666 controls were classiﬁed as pre- diagnosis for cases and the date of questionnaire completion for menopausal at the time of diagnosis (cases) or recruitment (con- controls. Menopausal status was assigned according to the reported trols) and were, therefore, included in the evaluation. A description state half a year before the reference date. The menopausal status of sociodemographic variables and known risk factors for breast of women with previous hysterectomy not accompanied by bilat- cancer is given in Table I. The mean age for cases and controls was eral oophorectomy was not identiﬁable and classiﬁed as unknown.
42.6 years (SD Ϯ5.48, Ϯ5.77, respectively). Total daily energy Information on hormone receptor status (estrogen, progesterone) intake was signiﬁcantly different between cases and controls (Ta- of breast cancer tissue was obtained from pathology reports.
In addition, participants from Rhein-Neckar-Odenwald were Median dietary intake of phytoestrogens in cases and controls is asked to complete a self-administered FFQ, recording nutritional given in Table II. The higher intake of genistein among controls habits in the year prior to the diagnosis of breast cancer (cases) or compared to cases was statistically signiﬁcant. Additionally, cases questionnaire completion (controls). The FFQ, consisting of 176 and controls differed signiﬁcantly in formononetin intake. For all food items, is comparable to the FFQ used for the German part of other phytoestrogens, no signiﬁcant differences existed. Secoiso- the EPIC study; it was validated for food group, energy and lariciresinol contributed nearly 50% of total phytoestrogen intake nutrient intake.27,28 A detailed description of this FFQ, including of about 1,000 g/day. Relevant food sources for different phy- the treatment of missing values, was published previously.29 The toestrogens are listed in Table III.
FFQ was returned by 1,288 of the 1,451 participants from Rhein- Except for genistein (4th vs. 1st quartile), formononetin and Neckar-Odenwald (88.8%). The median time between diagnosis biochanin A (3rd vs. 1st quartile), none of the crude estimates for and FFQ administration was 209 days for patients. Patients who breast cancer risk was statistically signiﬁcant (Table IV). After ﬁlled in the FFQ more than 2 years after diagnosis were excluded adjustment for several risk factors for breast cancer and possible from analysis. Information about portion size and frequency of confounders, high intake of daidzein and genistein was inversely consumption for single food items was used to calculate mean food associated with premenopausal breast cancer risk (Table IV). For intake per day. Nutrient intake was computed by means of the the sum of daidzein and genistein, the risk estimates for both the German food composition table BLS II.3 (Bundesinstitut fu¨r ge- 3rd and 4th intake quartiles were signiﬁcantly decreased compared su˘ndheitlichen Verbraucherschu¨tz und Veterina¨rmedizin, Berlin, to the lowest intake category, with ORs (95% CI) of 0.63 (0.41– Germany). Food items were classiﬁed into 14 groups.29 After 0.96) and 0.56 (0.36 – 0.87), respectively (p exclusion of the top and bottom 1% of the energy intake range formononetin and biochanin A was associated with increased (extreme over- and underreporting), 355 cases and 838 controls breast cancer risk in the 3rd intake quartile but not the 4th quartile; were ﬁnally included in the analyses.
overall, no signiﬁcant effect was found for the sum of isoﬂa- To estimate dietary phytoestrogen intake, a database of phy- vonoids (Table IV). Coumestrol intake was also not associated toestrogen-containing foods was established using the available analytic data from the literature, focusing on foodstuffs usually Of the lignans, no effect was found for secoisolariciresinol.
consumed in Europe.2,14,30–40 The food content of the isoﬂa- However, premenopausal breast cancer risk decreased with in- vonoids (daidzein, genistein, formononetin, biochanin A), coumes- creasing intake of matairesinol (OR ϭ 0.58, 95% CI 0.37– 0.94 in tanes (coumestrol) and lignans (secoisolariciresinol, matairesinol) 4th vs. 1st intake quartile, p was included as aglycons in micrograms per 100 g edible food derived phytoestrogens did not affect breast cancer risk. A high portion. If more than one value for the same foodstuff was avail- estimated amount of mammalian lignans, enterodiol and enterolac- able, mean values were calculated. Additionally, in vitro data on tone, was inversely associated with breast cancer risk. ORs (95% the intestinal production of enterodiol and enterolactone after CIs) for the highest vs. lowest quartile were 0.61 (0.39 – 0.98, incubation with human feces were used to estimate the bioavail- ϭ 0.034) for enterodiol and 0.57 (0.35–0.92, p able intestinal lignan metabolites per 100 g of food ingested.41 When a particular phytoestrogen was found “in trace amounts” in If considering only patients who completed the FFQ within 1 a certain food, it was assigned to 0.01 g/100 g food. After year after diagnosis (n ϭ 213), adjusted relative risks were only calculation of the phytoestrogen content of each FFQ item, mean slightly modiﬁed with few exceptions: relative risk estimates for daily intake of the different phytoestrogens was computed for each the 3rd quartiles of daidzein intake and total isoﬂavonoid intake became signiﬁcant, while for matairesinol, enterodiol and en- The association between phytoestrogen intake and risk of pre- terolactone the point estimates for the 4th intake quartiles slightly menopausal breast cancer was estimated by means of conditional increased, with ORs of 0.64, 0.66 and 0.63, but were no longer logistic regression modeling. ORs and their corresponding 95% CIs were calculated using the PHREG procedure of the SAS We also examined the combined effect of daidzein and genistein statistical software package, version 8.2 (SAS Institute, Cary, NC) intake and intestinal lignan metabolites. Women were grouped by with stratiﬁcation for exact age. Higher quartiles of phytoestrogen low (1st and 2nd quartiles) and high (3rd and 4th quartiles) consumption were compared to the lowest quartile of consump- consumption of daidzein and genistein and of enterodiol and tion, whereby the distribution of the phytoestrogen intake in con- enterolactone. High consumption of both phytoestrogen classes trols was used for quartile deﬁnition. Results were adjusted for showed the strongest risk reduction, with an adjusted OR (95% CI) ﬁrst-degree family history of breast cancer (no, yes), number of of 0.57 (0.36 – 0.83) compared to low consumption of both groups births (3 categories of 0, 1 or 2, Ն3), duration of breast-feeding (4 categories of 0, 1– 6, 7–12, Ͼ12 months), energy intake (deciles), Further, we examined whether the phytoestrogen effect differed BMI [classiﬁed as underweight (Ͻ18.5 kg/m2), normal weight by hormone receptor status. Differentiated by estrogen and pro- (18.5–30 kg/m2) and obese (Ն30 kg/m2)], education level (low, gesterone receptor status of breast tissue, a signiﬁcant inverse middle, high) and alcohol consumption (3 categories of 0, 1–18 association between daidzein and genistein intake and breast can- and Ͼ18 g ethanol/day). Age at menarche, smoking, sports activity cer risk was found only for breast tumors positive for estrogen or TABLE I – DISTRIBUTION OF SOCIODEMOGRAPHIC VARIABLES AND RISK
The observed effects of genistein and daidzein were largely FACTORS FOR PREMENOPAUSAL BREAST CANCER IN CASES (N ϭ 278) unexpected because the intake levels for these compounds are very low (150 –160 g/day) compared to Asian populations (10 –30 mg/day).4 Concentrations necessary to reveal hormone receptor– mediated effects as determined in in vitro studies are unlikely to bereached by the estimated dietary intake range in this study popu- lation. This suggests a hormone receptor-independent mechanism of action of these isoﬂavonoids, which, however, appears to be contradicted by the clear results showing that the genistein and daidzein effects are restricted to hormone receptor-positive tumors.
Isoﬂavonoid effects on hormonal status and hormone metabolism may help explain the ﬁndings. Besides in vitro results, human intervention studies in premenopausal women and in men have found a decrease in plasma 17␤-estradiol, progesterone, FSH and LH and an increase in plasma SHBG concentrations after admin- istration of soy (isoﬂavonoids) or ﬂaxseed (lignans).11,19,42–44 In addition, some studies have also shown increased ratios of urinary 2-(OH) to 16␣-(OH) and 2-(OH) to 4-(OH) estrogens associated with soy intake.11 The latter ﬁnding suggests reduced formation of genotoxic and potentially carcinogenic estrogen metabolites. Sev- eral enzymes involved in estrogen metabolism can be affected (mainly inhibited) by phytoestrogens, including aromatase and sulfotransferases.12 The effects were less pronounced in postmeno- pausal women.45 Thus, our present ﬁndings are compatible with the biologic mechanisms showing an important role for isoﬂa- vonoids in protecting against hormone-dependent breast cancer through their effects on the generation, transport and removal of endogenous steroid hormones; furthermore, some of these actions are mediated by hormonal receptors and therefore dependent on receptor status. Surprisingly, intake of formononetin and biochanin A as precursors of daidzein and genistein was unrelated to breast cancer risk or even increased risk (3rd quartile, Table IV). This may indicate very limited conversion of these precursors to their In contrast to daidzein and genistein, the protective effect found for enterodiol and enterolactone was not modiﬁed by hormone receptor status. This implies that there may be differences between the 2 classes of phytoestrogens regarding the biologic mechanisms involved in reducing breast cancer risk. Isoﬂavonoids and mam- malian lignans show similar effects, such as competition with estrogen for binding to estrogen receptors (type II), induction of SHBG, inhibition of some steroid-metabolizing enzymes (aro- matase) and antioxidant activity.1,18 However, depending on the test systems used, differences between phytoestrogenic com- pounds have been reported. The possible underlying mechanism for the observed difference warrants investigation.
Two reviews dealing with the health effects of phytoestrogens concluded that protective effects on breast cancer risk might be found in women who consume soy phytoestrogens at an early age (adolescence) or at very high doses.10,13 Most of the included studies evaluated the relationship between intake of soy/soy prod- ucts and disease risk, predominantly in Asian populations. Studies which assessed isoﬂavonoid or lignan metabolite concentrations in plasma or urine samples in relation to breast cancer yielded incon- MET, metabolic equivalents; data available for 187 cases and 525 sistent results. While 2 prospective studies failed to ﬁnd a signif- icant inverse association between high urinary isoﬂavonoid excre-tion or serum enterolactone concentrations and breast cancer progesterone receptor (Table VI). The effects of enterodiol and risk,23,24 5 case-control studies reported a protective effect on enterolactone were independent of hormone receptor status (data Two studies in Asian populations, living in Japan (prospective study)48 and the United States (case-control study),49 that assessed overall isoﬂavonoid intake reported an inverse association with We found a protective effect of a high intake of genistein and breast cancer risk. The effect of dietary intake of isoﬂavonoids and daidzein (isoﬂavonoids) and matairesinol (lignan) on breast cancer lignans on breast cancer risk has been assessed only recently in 2 risk. The effect of genistein and daidzein was limited to hormone studies in the United States but not yet in Europe.16,17 Both US receptor-positive breast tumors. Additionally, a high amount of studies, one with a prospective design,17 failed to ﬁnd a signiﬁcant estimated enterodiol and enterolactone (mammalian lignans) is association with breast cancer or an effect modiﬁcation by meno- inversely related to premenopausal breast cancer risk.
pausal status or hormone receptor status of the tumor.
TABLE II – DIETARY INTAKE (MEDIAN, 25%–75% PERCENTILE) OF SELECTED PHYTOESTROGENS AND ENERGY IN PREMENOPAUSAL
BREAST CANCER CASES (N ϭ 278) AND POPULATION-BASED CONTROLS (N ϭ 666)1 1Units are g/day except for “Energy”, kcal/day.–2Mann-Whitney U-test.
TABLE III – CONTRIBUTION OF FOOD GROUPS AND SUBGROUPS TO THE DIETARY INTAKE OF SELECTED PHYTOESTROGENS IN PREMENOPAUSAL
BREAST CANCER CASES AND CONTROLS (AVERAGE PERCENTAGE OF TOTAL INTAKE IN CASES/CONTROLS) For the calculation of individual phytoestrogen intake, analytic pumpkin seeds) and coffee. Potatoes, leafy vegetables, cabbages, data from the literature were used, as described by other onion/garlic, nuts/seeds, wine, tea and especially bread contributed groups.15,38,50,51 With the exception of daidzein, the dietary intake to the total matairesinol intake. The proportional contribution of estimates for the different phytoestrogenic compounds in our sub- food groups to the mammalian lignans is spread over a broad jects are compatible with the results from a Dutch study in women variety of foods like different kinds of vegetable, potatoes, fruits, 49 –70 years old participating in the Utrecht EPIC cohort.15 Data nuts/seeds and bread, indicating differences in lignan sources.
on isoﬂavonoid intake in 4 European countries (Ireland, Italy, the There are 2 major concerns over possible biases that may have Netherlands, UK) also estimated an average intake of Ͻ1 mg/day, affected the results. First, a case-control study design of nutritional which was somewhat higher than that found in Germany.52 Studies studies may suffer from recall bias in cases. This cannot be from central Europe reporting lignan intake data are not available.
excluded for the present study. However, limited modiﬁcation of Finnish data regarding dietary lignan intake vary widely but show the diet over time has been indicated, also after a diagnosis of distinctly lower secoisolariciresinol intake levels than we found cancer. The type of food consumed can be regarded as a personal here.50,53 Reported lignan and coumestrol intake values from 1 of dietary preference, so intake levels of most persons are likely to be 3 studies in the United States are similar to our results,51 while relatively stable over time.54 Zheng et al.47 observed that Ͼ90% of others are not;16,17 however, isoﬂavonoid intake was higher in the cases reported no appreciable dietary change during the time period from initial cancer diagnosis to urine collection. Addition- Differences in phytoestrogen intake between studies can be ally, results from EPIC-Heidelberg, obtained with the same nutri- explained by differing dietary habits but may also be affected by tional assessment tool, were comparable to the present results in differences in the phytoestrogen content of food (e.g., soy protein terms of energy and nutrient intake ranges55 and, therefore, give as an ingredient) or differences in the ability of the FFQ employed some reassurance for the validity of the data. Further, dietary to assess relevant food items for phytoestrogen intake. As shown assessment was not the main topic of the original study project, so in Table III, the important (Ͼ5% contribution) sources of daidzein patients may not have been sensitized to nutrition. The descriptive and genistein in the diet of our study population were soy sprouts, data show few differences between the original study population legumes, soy products, sauces and coffee (daidzein). Secoisolar- and the subgroup with FFQ data; thus, selection bias can largely be iciresinol is predominantly provided by nuts and seeds (ﬂaxseeds, excluded as a source of error.29 Second, phytoestrogen data in- TABLE IV – CRUDE AND ADJUSTED1 ORS AND 95% CIS FOR PREMENOPAUSAL BREAST CANCER RISK BY PHYTOESTROGEN INTAKE
1Adjusted for ﬁrst-degree family history of breast cancer, number of births, duration of breast-feeding, energy intake, BMI, alcohol TABLE V – ADJUSTED1 ORS AND 95% CIS FOR PREMENOPAUSAL BREAST
food additives used in food manufacturing before 1995 were used CANCER RISK BY COMBINATIONS OF LOW AND HIGH2 mainly for technologic reasons and pertain largely to soy protein DIETARY INTAKE OF DAIDZEIN AND GENISTEIN AND OF preparations (with widely varying isoﬂavonoid concentrations).
Regarding the lignans secoisolariciresinol and matairesinol, the most important food sources were considered but there may be other foods not analyzed so far. Several other plant lignans (e.g., pinoresinol, syringaresinol, lariciresinol) have been identiﬁed as precursors of enterodiol and enterolactone,57 but their concentra- tions in different foods are not available. This probably explains why the food sources of secoisolariciresinol and matairesinol are (0.46–1.10) (0.54–1.25) (0.36–0.83) distinctly different from those of mammalian lignans (Table III).
The most critical point, however, may be the production of en- Adjusted for ﬁrst-degree family history of breast cancer, number of terodiol and enterolactone from ingested lignans in the human births, duration of breast-feeding, energy intake, BMI, alcohol con-sumption and education.–2Low levels refer to 1st and 2nd quartiles, intestine. The extent of lignan conversion in the gut depends and high levels refer to 3rd and 4th quartiles.
largely on the quantity and composition of the gut microﬂora.18Here, we used data from Thompson et al.41 based on incubatingdifferent foods with freshly gained human feces. It has been cluded in the database and used for intake calculations may be reported that plasma enterolactone concentrations are decreased incomplete for several reasons. In the case of isoﬂavonoids, the for several months after antibiotic therapy.58 Additionally, gut most important though unsolved question is the use of soy and soy transit time and fat content of the diet appear to modulate the components (e.g., soy protein) as food additives.56 The discussion availability of mammalian lignans.18,59 Therefore, when calculat- of possible health effects of soy was not yet popular during the ing the available amount of mammalian lignans in the present ﬁeld period of the study. Thus, it can be assumed that soy-derived study, the reported high inter- and intraindividual differences in the TABLE VI – ADJUSTED1 ORS AND 95% CIS FOR TOTAL DAIDZEIN AND GENISTEIN INTAKE AND PREMENOPAUSAL
BREAST CANCER RISK BY ESTROGEN AND PROGESTERONE RECEPTOR STATUS OF THE TUMOR 1Adjusted for ﬁrst-degree family history of breast cancer, number of births, duration of breast-feeding, energy intake, BMI, alcohol production of mammalian lignans53,60 could not be taken into monal proﬁle.64,65 This supports the view that a “healthy” diet account. Isoﬂavonoids also undergo intestinal metabolism, and the provides more bioavailable mammalian lignans but does not ex- question of bioavailability of dietary isoﬂavonoids and their me- clude the possibility that mammalian lignans exert their own tabolites may also be important for understanding their health effects.61,62 Equol is the intestinal isoﬂavonoid metabolite with the In conclusion, our results give clear epidemiologic evidence for strongest estrogenic activity, but it is produced in only about a protective role against premenopausal breast cancer of a diet high one-third of Caucasian people, depending largely on the intestinal (albeit low compared to Asian populations) in daidzein, genistein or matairesinol as well as estimated mammalian lignans enterodiol Due to the difﬁculty in accounting for sources, metabolism and and enterolactone. The association between isoﬂavonoid intake bioavailability of dietary lignans, it is not surprising that the and breast cancer risk was found only for hormone receptor- association with breast cancer in our study was not the same for positive tumors, while the effect of mammalian lignans was inde- mammalian and plant lignans. Both approaches suffer from im- pendent of receptor status. Whether phytoestrogens are the under- precision: the lignan intake data do not reﬂect the bioavailable lying causal agents of a favorable diet or a favorable gut microﬂora fraction of lignan metabolites, and the enterodiol and enterolactone or both in relation to reduced breast cancer risk remains to be in vitro data used for calculation may be biased due to differences in the physiologic gut ﬂora between subjects.
We previously reported from the same study population a sig- niﬁcantly reduced risk of breast cancer associated with high con-sumption of vegetables.29 Since vegetables are sources of entero- We are grateful to all participants for their help and patience as diol and enterolactone, it cannot be excluded that the mammalian well as to the many gynecologists and oncologists in the 38 clinics lignan effect reﬂects a more healthy diet rich in vegetables. Dietary of the study regions for allowing us to contact their patients. We intervention studies have shown that a diet low in animal products also thank Ms. U. Eilber for data coordination and management and fat and high in complex carbohydrates, fruits and vegetables and Ms. D. Zoller for programming and calculating dietary intake increases enterolactone plasma concentrations and affects the hor- Benassayag C, Perrot-Applanat M, Ferre F. Phytoestrogens as mod- 10. Adlercreutz H. Phytoestrogens and breast cancer. J Steroid Biochem ulators of steroid action in target cells. J Chromatogr B Biomed Sci 11. Kurzer MS. Hormonal effects of soy in premenopausal women and Adlercreutz H, Mazur W. Phyto-oestrogens and Western diseases.
12. Kirk CJ, Harris RM, Wood DM, Waring RH, Huges PJ. Do dietary Bingham SA, Atkinson C, Liggins J, Bluck L, Coward A. Phyto- phytoestrogens inﬂuence susceptibility to hormone-dependent cancer oestrogens: where are we now? Br J Nutr 1998;79:393– 406.
by disrupting the metabolism of endogenous oestrogens? Biochem Messina MJ. Isoﬂavone intakes in Japanese were overestimated. Am J 13. Peeters PHM, Keinen Boker L, van der Schouw YT, Grobbee DE.
Messina MJ, Persky V, Stechell KDR, Barnes S. Soy intake and Phytoestrogens and breast cancer risk. Review of the epidemiologic cancer risk—a review of the in vitro and in vivo data. Nutr Cancer evidence. Breast Cancer Res Treat 2003;77:171– 83.
14. Mazur W. Phytoestrogen content in foods. Baillieres Clin Endocrinol Setchell KD, Borriello SP, Hulme P, Kirk DN, Axelson M. Nonste- roidal estrogens of dietary origin: possible roles in hormone-depen- 15. Keinan Boker L, van der Shouw YT, de Kleijn MJJ, Jacques PF, dent disease. Am J Clin Nutr 1984;40:569 –78.
Grobbee DE, Peeters PHM. Intake of dietary phytoestrogens by Dutch Endogenous Hormones and Breast Cancer Collaborative Group. En- dogenous sex hormones and breast cancer in postmenopausal women: 16. Horn-Ross PL, John EM, Lee M, Steward SL, Koo J, Sakoda LC, reanalysis of nine prospective studies. J Natl Cancer Inst 2002;94: Shiau AC, Goldstein J, Davis P, Perez-Stable EJ. Phytoestrogen consumption and breast cancer risk in a multiethnic population. Am J Adlercreutz H. Western diet and Western diseases: some hormonal and biochemical mechanisms and associations. Scand J Clin Lab 17. Horn-Ross PL, Hoggatt KJ, West DW, Krone MR, Stewart SL, Anton-Culver H, Bernstein L, Deapen D, Peel D, Pinder R, Reynolds Lamartiniere CA, Cotroneo MS, Fritz WA, Wang J, Mentor-Marcel P, Ross RK, et al. Recent diet and breast cancer risk: the California R, Elgavish A. Genistein chemoprevention: timing and mechanisms Teachers Study (USA). Cancer Causes Control 2002;13:407–15.
of action in murine mammary and prostate. J Nutr 2002;132:552S– 18. Wang LQ. Mammalian phytoestrogens: enterodiol and enterolactone.
J Chromatogr B Biomed Sci Appl 2002;777:289 –309.
19. Raffaelli B, Hoikkala A, Lepp¨al¨a E, W¨ah¨al¨a K. Enterolignans. J Chro- 43. Haggans CJ, Hutchins AM, Olson BA, Thomas W, Martini MC, matogr B Biomed Sci Appl 2002;777:29 – 43.
Slavin JL. Effect of ﬂaxseed consumption on urinary estrogen metab- 20. Ingram D, Sanders K, Kolybaba M, Lopez D. Case-control study of olites in postmenopausal women. Nutr Cancer 1999;33:188 –95.
phyto-oestrogens and breast cancer. Lancet 1997;350:990 – 4.
44. Haggans CJ, Travelli EJ, Thomas W, Martini MC, Slavin JL. The 21. Pietinen P, Stumpf K, M¨annist¨o S, Kataja V, Uusitupa M, Adlercreutz effect of ﬂaxseed and wheat bran consumption on urinary estrogen H. Serum enterolactone and risk of breast cancer: a case-control study metabolites in premenopausal women. Cancer Epidemiol Biomarkers in eastern Finland. Cancer Epidemiol Biomarkers Prev 2001;10:339 – 45. Xu X, Duncan AM, Wangen KE, Kurzer MS. Soy consumption alters 22. Dai Q, Franke AA, Jin F, Shu XO, Hebert JR, Custer LJ, Cheng J, Gao endogenous estrogen metabolism in postmenopausal women. Cancer YT, Zheng W. Urinary excretion of phytoestrogens and risk of breast Epidemiol Biomarkers Prev 2000;9:781– 6.
cancer among Chinese women in Shanghai. Cancer Epidemiol Bi- 46. Murkies A, Dalais FS, Briganti EM, Burger HG, Healy DL, Wahlqvist ML, Davis SR. Phytoestrogens and breast cancer in postmenopausal 23. Den Tonkelaar I, Keinan-Boker L, Veer PV, Arts CJM, Adlercreutz women: a case control study. Menopause 2000;7:289 –96.
H, Thijssen JH, Peeters PH. Urinary phytoestrogens and postmeno- 47. Zheng W, Dai Q, Custer LJ, Shu XO, Wen WQ, Jin F, Franke AA.
pausal breast cancer risk. Cancer Epidemiol Biomarkers Prev 2001; Urinary excretion of isoﬂavonoids and the risk of breast cancer.
Cancer Epidemiol Biomarkers Prev 1999;8:35– 40.
24. Hulten K, Winkvist A, Lenner P, Johansson R, Adlercreutz H, Hall- 48. Yamamoto S, Sobue T, Kobayashi M, Sasaki S, Tsugane S, Japan mans G. An incident case-referent study on plasma enterolactone and Public Health Center–Based Prospective Study on Cancer Cardiovas- breast cancer risk. Eur J Nutr 2002;41:168 –76.
cular Diseases Group. Soy, isoﬂavones, and breast cancer risk in 25. Chang-Claude J, Eby N, Kiechle M, Bastert G, Becher H. Breastfeed- Japan. J Natl Cancer Inst. 2003;95:906 –13.
ing and breast cancer risk by age 50 among women in Germany.
49. Wu AH, Wan P, Hankin J, Tseng CC, Yu MC, Pike MC. Adolescent Cancer Causes Control 2000;11:687–95.
and adult soy intake and risk of breast cancer in Asian-Americans.
26. Kropp S, Becher H, Nieters A, Chang-Claude J. Low to moderate alcohol consumption and breast cancer risk by age 50 years among 50. Valsta LM, Kilkkinen A, Mazur W, Nurmi T, Lampi AM, Ovask ML, women in Germany. Am J Epidemiol 2001;154:624 –34.
Korhonen T, Adlercreutz H, Pietinen P. Phyto-oestrogen database of 27. Bohlscheid-Thomas S, Hoting I, Boeing H, Wahrendorf J. Reproduc- foods and average intake in Finland. Br J Nutr 2003;89:S31– 8.
ibility and relative validity of food group intake in a food frequency 51. De Kleijn MJJ, van der Schouw YT, Wilson PWF, Adlercreutz H, questionnaire developed for the German part of EPIC project. Int J Mazur W, Grobbee DE, Jaques PF. Intake of dietary phytoestrogens is low in postmenopausal women in the United States: the Framing- 28. Bohlscheid-Thomas S, Hoting I, Boeing H, Wahrendorf J. Reproduc- ham Study. J Nutr 2001;131:1826 –32.
ibility and relative validity of energy and macronutrient intake of a 52. Van Erp-Baart MAJ, Brants HAM, Kiely M, Mulligan A, Turrini A, food frequency questionnaire developed for the German part of EPIC Sermoneta C, Kilkkinen A, Valsta LM. Isoﬂavone intake in four project. Int J Epidemiol 1997;26:S71– 81.
different European countries: the VENUS approach. Br J Nutr 2003; 29. Hermann S, Linseisen J, Chang-Claude J. Nutrition and breast cancer risk by age 50: a population-based case-control study in Germany.
53. Kilkkinen A, Valsta LM, Virtamo J, Stumpf K, Adlercreutz H, Pieti- nen P. Intake of lignans is associated with serum enterolactone con- 30. Mazur W, Fotsis T, W¨ah¨al¨a K, Ojala S, Salakka A, Adlercreutz H.
centration in Finnish men and women. J Nutr 2003;133:1830 –3.
Isotope dilution gas chromatographic-mass spectrometric method for 54. Arai Y, Uehara M, Sato Y. Comparison of isoﬂavones among dietary the determination of isoﬂavonoids, coumestrol, and lignans in food intake, plasma concentration and urinary excretion of phytoestrogen samples. Anal Biochem 1996;233:169 – 80.
31. Mazur W, Adlercreutz H. Natural and anthropogenic environmental intake. J Epidemiol 2000;10:127–35.
oestrogens: the scientiﬁc basis for risk assessment. Naturally occur- 55. Brandstetter BR, Korfmann A, Kroke A, Becker N, Schulze MB, ring oestrogens in food. Pure Appl Chem 1998;70:1759 –76.
Boeing H. Dietary habits in the German EPIC cohorts: food group 32. Mazur WM, Duke JA, W¨ah¨al¨a K, Rasku S, Adlercreutz H. Isoﬂa- intake estimated with the food frequency questionnaire. Ann Nutr vonoids and lignans in legumes: nutritional and health aspects in humans. J Nutr Biochem 1998;9:193–200.
56. Fletcher RJ. Food sources of phyto-oestrogens and their precursors in 33. Mazur WM, W¨ah¨al¨a K, Rasku S, Salakka A, Hase T, Adlercreutz H.
Europe. Br J Nutr 2003;89:S39 – 43.
Lignan and isoﬂavonoid concentrations in tea and coffee. Br J Nutr 57. Heinonen S, Nurmi T, Liukkonen K, Poutanen K, W¨ah¨al¨a K, Deyama T, Nishibe S, Adlercreutz H. In vitro metabolism of plant lignans: new 34. Mazur WM, Uehara M, W¨ah¨al¨a K, Adlercreutz H. Phyto-oestrogen precursors of mammalian lignans enterolactone and enterodiol. J content of berries, and plasma concentrations and urinary excretion of Agric Food Chem 2001;49:3178 – 86.
enterolactone after a single-strawberry meal in human subjects. Br J 58. Kilkkinen A, Pietinen P, Klaukka T, Virtamo J, Korhonen P, Adler- creutz H. Use of oral antimicrobials decreases serum enterolactone 35. Liggins J, Bluck LJ, Runswick S, Atkinson C, Coward WA, Bingham concentration. Am J Epidemiol 2002;155:472–7.
SA. Daidzein and genistein contents of vegetables. Br J Nutr 2000; 59. Lampe JW. Isoﬂavonoid and lignan phytoestrogens as dietary biomar- 36. Liggins J, Bluck LJC, Runswick S, Atkinson C, Coward WA, Bing- 60. Kilkkinen A, Stumpf K, Pietinen P, Valsta LM, Tapanainen H, Adler- ham SA. Daidzein and genistein content of fruits and nuts. J Nutr creutz H. Determinants of serum enterolactone concentration. Am J 37. Liggins J, Mulligan A, Runswick S, Bingham SA. Daidzein and 61. Hendrich S. Bioavailability of isoﬂavones. J Chromatogr B Biomed genistein content of cereals. Eur J Clin Nutr 2002;56:961– 6.
38. Horn-Ross PL, Barnes S, Lee M, Coward L, Mandel EJ, Koo J, John 62. Setchell KDR, Brown NM, Desai P, Zimmer-Nechemias L, Wolfe EM, Smith M. Assessing phytoestrogen exposure in epidemiologic BE, Brashear WT, Kirschner AS, Cassidy A, Heubi JE. Bioavailabil- studies: development of a database (United States). Cancer Causes ity of pure isoﬂavones in healthy humans and analysis of commercial soy isoﬂavone supplements. J Nutr 2001;131:1362S–75S.
39. Franke AA, Custer LJ, Cerna CM, Narala K. Rapid HPLC analysis of 63. Setchell KD, Brown NM, Lydeking-Olesen E. The clinical impor- dietary phytoestrogens from legumes and from human urine. Proc Soc tance of the metabolite equol—a clue to the effectiveness of soy and its isoﬂavones. J Nutr 2002;132:3577– 84.
40. Lapcik O, Hill M, Hampl R, W¨ah¨al¨a K, Adlercreutz H. Identiﬁcation 64. Stumpf K, Pietinen P, Puska P, Adlercreutz H. Changes in serum of isoﬂavonoids in beer. Steroids 1998;63:14 –20.
enterolactone, genistein, and daidzein in a dietary intervention study 41. Thompson LU, Robb P, Serraino M, Cheung F. Mammalian lignan in Finland. Cancer Epidemiol Biomarkers Prev 2000;9:1369 –72.
production from various foods. Nutr Cancer 1991;16:43–52.
65. Berrino F, Bellati C, Secreto G, Camerini E, Pala V, Panico S, Allegro 42. Kumar NB, Cantor A, Allen K, Riccardi D, Cox CE. The speciﬁc role G, Kaaks R. Reducing bioavailable sex hormones through a compre- of isoﬂavones on estrogen metabolism in premenopausal women.
hensive changing diet: the Diet and Androgens (DIANA) randomized trial. Cancer Epidemiol Biomarkers Prev 2001;10:25–33.
The politics of coalition formation and democracy in Lesotho Motlamelle Anthony Kapa Department of Political and Administrative Studies National university of Lesotho Abstract That political party coalitions are becoming almost a permanent feature of pluralist democracies the world over is a fact beyond dispute. Although relatively new to the African continent for they began d