Doi:10.1016/j.maturitas.2005.02.018

Breast cancer risk in the WHI study: The problem of obesity Department of Gynecology and Obstetrics, J. W. Goethe University of Frankfurt, Theodor-Stern-Kai 7, D-60590 Frankfurt am Main, Germany Abstract
In the climacteric, about 40% of the women have occult breast tumors the growth of which may be stimulated by hormones.
Many genetic, reproductive and lifestyle factors may influence the incidence of breast cancer. Epidemiological data suggestthat the increase in the relative risk (RR) of breast cancer induced by hormone replacement therapy (HRT) is comparable withthat associated with early menarche, late menopause, late first birth, alcohol consumption, etc. One of the most important riskfactors is obesity which exceeds the effect of HRT by far, and in overweight postmenopausal women the elevated risk of breastcancer is not further increased by HRT. As in the WHI study the majority of women was overweight or obese, this trial wasunsuitable for the investigation of breast cancer risk. In the women treated with an estrogen/progestin combination, the RR ofbreast cancer rose only in those women who have been treated with hormones prior to the study, suggesting a selection bias.
In the women not pretreated with hormones, it was not elevated. In the estrogen-only arm of the WHI study, there was noincrease but a steady decrease in the RR of breast cancer during 6.8 years of estrogen therapy. This result was unexpected, asestrogens are known to facilitate the development and growth of breast tumors, and the effect is enhanced by the addition ofprogestins.
Obese women are at high risk to develop a metabolic syndrome including insulin resistance and hyperinsulinemia. In post- menopausal women, elevated insulin levels are not only associated with an increased risk for cardiovascular disease, but alsofor breast cancer. This might explain the effects observed in both arms of the WHI study: HRT with relative low doses ofestrogens may improve insulin resistance and, hence, reduce the elevated breast cancer risk in obese patients, whereas thisbeneficial estrogen effect may be antagonized by progestins. The principal options for the reduction of breast cancer risk inpostmenopausal women are the prevention of overweight and obesity to avoid the development of hyperinsulinemia, the medicaltreatment of insulin resistance, the use of low doses of estrogens and the reduction of exposure to progestins. The latter mightinclude long-cycles with the sequential use of appropriate progestins every 3 months for 14 days. There are large inter-individualvariations in the proliferative response to estrogens of the endometrium. Control by vaginalsonography and progestin challengetests may help to identify those women who may be candidates for low-dose estrogen-only therapy.
2005 Elsevier Ireland Ltd. All rights reserved.
Keywords: Breast cancer risk; Postmenopause; Obesity; Hyperinsulinemia; Hormone replacement therapy ∗ Tel.: +49 69 6301 5692; fax: +49 69 6301 5522.
E-mail address: [email protected].
0378-5122/$ – see front matter 2005 Elsevier Ireland Ltd. All rights reserved.
H. Kuhl / Maturitas 51 (2005) 83–97 1. Introduction
and BRCA2) must be accepted as an unchangeablepredisposition. Many other risk factors, e.g., obesity Breast cancer is the most frequent malignant dis- or hormone replacement therapy (HRT) can, how- ease in Western countries and seems to be dependent ever, be avoided or changed. Early menarche and late on lifestyle and nutrition. The development of breast menopause indicate a prolonged exposition to estro- cancer is usually regarded as a multifactorial process gens and progesterone that increase the risk of breast which means that the etiology is unknown. There are cancer, whereas long-term lactation decreases the risk many theories that are based on experimental investi- gation and relatively inconsistent epidemiological data.
sion of ovulation during breastfeeding. Moreover, the There is, however, no doubt that reproductive factors Nurses’ Health Study showed that irregular cycles in play an important role. Concerning the impact of sex young women are associated with a reduced life-time steroids, the cumulative exposure to endogenous and breast cancer risk As in anovulatory cycles there exogenous estrogens and progestins seems to deter- is generally no estrogen deficiency, it may be assumed mine the life-time risk of breast cancer.
that the protective effect of anovulation is associatedwith the lack of progesterone. Obesity, insulin resis-tance, disorders of lipid metabolism and elevated alco- 2. Risk factors for the development of breast
hol consumption seem to increase breast cancer risk.
The impact of long-term shift work is possibly relatedto the prolonged light exposure at night resulting in Certain risk factors for the development of breast a suppression of melatonin levels, while the cumula- cancer like age and gene mutation (e.g., BRCA1 tive association of use of antibiotics with the risk ofbreast cancer might reflect a weakened immune func-tion ( Table 1Risk factors for the development of breast cancer 3. Breast cancer risk and HRT: epidemiological
Many observational studies on the influence of HRT on breast cancer risk revealed contradictory results, and every new case-control study or cohort study will enlarge this long row of inconsistent outcomes. The collaborative reanalysis from 1997 was an attempt to bring together and re-examine the individual data of all relevant studies published so far. It revealed that each year of delayed menopause increases the risk by 2.8% which was in the range of 2.3% for each year of HRT relative risk of breast cancer increased by 35% in postmenopausal women who had used HRT for 11 years on average. The cumulative excess of breast can- cers diagnosed between the ages of 50 and 70 years per 1000 women who began HRT at age 50 and used it for 5, 10, and 15 years, were estimated to be 2, 6, and 12 cases. Within 5 years after discontinuation of treatment, the elevated risk has returned to baseline H. Kuhl / Maturitas 51 (2005) 83–97 relative risk of breast cancer by 24%, but only in thosewomen who were pretreated with hormones prior to the Randomised placebo-controlled trials are regarded start of the WHI study ver, that arm of the as the non-plus-ultra for the investigation of the im- WHI study which investigated the effect of CEE alone pact of drugs on disease risk. Therefore, the results of in hysterectomized women, revealed a highly surpris- the HER study and both arms of the WHI study were ing result: after 6.8 years of treatment the relative risk of breast cancer was 0.77 ven though the result ever, whether the group of women investigated in these narrowly missed statistical significance, the consistent studies, reflects those women who normally receive time course of the Kaplan–Meier estimates suggests HRT. In the WHI study and the HER study the partic- that the estrogen therapy had a protective effect ipating women were selected as to suffer not from cli-macteric symptoms. Consequently, their mean age was 3.3. Effect of different regimens of hormone very high (about 63 years) and a high proportion of the women was obese (35–45% had a BMI ≥ 30 kg/m2).
As obesity is associated not only with an elevated risk In the collaborative reanalysis the type of HRT was for developing a metabolic syndrome and coronary known for about 40% of the users: 80% of them were heart disease, but also with an increased risk of breast treated with conjugated estrogens and only 12% with cancer, the results may be rather questionable.
estrogen/progestin combinations. The relative risk of The HER study on the secondary prevention of breast cancer was found to be 1.34 in women treated coronary heart disease by continuous treatment with for ≥5 years with estrogens alone and 1.53 in women 0.625 mg conjugated equine estrogens and 2.5 mg treated for ≥5 years with estrogen/progestin combina- medroxyprogesterone acetate (CEE/MPA), observed tions. There was no difference in risk between the dose a non-significant 27% increase in the relative risk of of ≤0.625 mg and ≥1.85 mg conjugated estrogens breast cancer after 6.8 years In the WHI study The Million Women Study (MWS) reported on an in- 5.2 years of treatment with CEE/MPA increased the crease in breast cancer risk by 30% using estrogens Table 2Relative risk (RR) of breast cancer during replacement therapy with estrogens only (ERT) or estrogen/progestin combinations (HRT) a Continuous combined estrogen/progestin therapy.
b Sequential or continuous combined estrogen/progestin therapy.
H. Kuhl / Maturitas 51 (2005) 83–97 alone and by 100% using estrogen/progestin combi- gen/progestin combinations increased the frequency nations The results may, however, be impaired of estrogen receptor-positive (ER+) and progesterone by detection bias, as, e.g., within 1 year after the first receptor-positive (PR+) invasive breast cancers 2- to mammographic screening the number of breast cancer 2.5-fold, whereas the effect on receptor-negative car- diagnoses (interval cancers) had increased three-fold in postmenopausal women treated continuously with es- of hormones was associated with a higher incidence of trogen/progestin combinations since several years. The tumors with the low malignancy grade1 anal- MWS did not find any difference in the risk of breast ysis of data from the Nurses’ Health Study revealed that cancer regarding type and dose of estrogens, route of postmenopausal women who used HRT had a higher administration, type of progestins, or sequential or con- probability of developing ER+ and PR+ tumors, and a higher body mass index (BMI) was associated with During the last years evidence has accumulated that the increase in breast cancer risk is relatively lowduring use of unopposed estrogens, and is consider-ably enhanced by the addition of progestins ( 4. The role of sex steroids in the development of
breast cancer
control study which found the highest risk with estro-gen only 4.1. Effect of sex steroids on the proliferation of A qualitative review showed the inconsistency of normal and malignant breast tissue results of observational studies, and most of the cohortstudies were not associated with a significant increase Although estrogens may be involved in the initia- in risk Recent studies confirmed the elevated tion of breast cancer, a carcinogenic/mutagenic role of breast cancer risk using estrogen/progestin combina- sex steroids is rather improbable. The available exper- tions. According to the data of the Nurses’ Health imental, clinical and epidemiological data suggest that Study, the use of unopposed estrogens increases the risk the development of breast cancer is closely related to of breast cancer by 23% and of estrogen/progestin com- an accelerated hormone-induced growth of preexisting binations by 67% The randomised, double-blind, occult tumors. In an autopsy study, small occult breast placebo-controlled Womens’s Health Study revealed cancers were found in 39% of women aged 40–50 no increase in risk using estrogens alone or sequential years Epidemiological studies revealed that the estrogen/progestin combinations, but a significant 82% impact of estrogens on the relative risk of breast cancer increase in women treated with continuous combined is modest, but considerably enhanced by the addition estrogen/progestin preparations Three Scandi- of progestins. This corresponds to the proliferative ef- navian cohort studies revealed a considerably higher fects both on normal mammary epithelium and breast relative risk in women treated with estrogen/progestin cancers of estrogens which is enhanced by the pres- ence of MPA or progesterone. The mitosis rate of both ER+/PR+ and ER−/PR− carcinoma was higher in theluteal phase than in the follicular phase , 3.4. Histological types and receptor status of the mitosis rate of healthy breast epithelium was high- est in the luteal phase, and higher during treatmentof postmenopausal women with CEE/MPA as com- In most studies continuous combined HRT was as- sociated with the highest relative risk of breast can- CEE/MPA were observed in the monkey model cer, particularly of hormone receptor-positive carci- In contrast, neither ethinylestradiol plus norethisterone noma. While the use of estrogens alone was as- nor tibolone had a significant effect on the proliferation sociated with no or a slightly elevated risk, estro- of normal breast epithelium ven though both gen/progestin combinations increased the incidence of tibolone and all types of estrogen/progestin combina- lobular cancers to a much greater extent than that of tions were found to be associated with an increased risk H. Kuhl / Maturitas 51 (2005) 83–97 tionable whether the effects of different HRT prepara- mutations of ER− stem cells may cause the tions on healthy mammary epithelium reflect those on differentiation of a subset of cells into ER+ cells. These tumors contain ER+ and ER− cells and may transito-rily respond to HRT and antiestrogens, but would not 4.2. Regulation of growth in benign and malignant have lasting effects, because proliferation of ER− cells continues. Therefore, HRT should not increase signif-icantly the risk of this subtype of breast cancer There are profound differences between healthy A third subtype may arise through transformation of and malignant breast tissue concerning the hormone- ER+ progenitor cells and consists of more differenti- dependent regulation of mitoses. In the resting normal ated cells. Their growth may be slowed down by treat- mammary tissue ER␣ and PR are expressed in very ment with antiestrogens and accelerated during HRT, few epithelial cells, while ER␤ is present in 70% and in both cases this subtype has the best prognosis of the cells. Those 2% of epithelial cells which are proliferating, do not contain ER The mitosesare probably controlled by paracrine interactionsof adjacent epithelial cells containing ER␣ and PR, 5. Interference of overweight with HRT
while ER␤ was suggested to inhibit ER␣-induced concerning breast cancer risk
effects. The effect of progesterone on proliferationand differentiation of the mammary epithelium is 5.1. Relation between breast cancer risk and body primarily dependent on the PRB, whereas PRA has a negative effect on PRB, and overexpression of PRAmay reflect a more aggressive state Obesity is associated not only with an elevated risk While in healthy tissue ER␣ is expressed only in of developing coronary heart disease, but also with an resting cells, the transition of benign to malignant mam- increase in risk of various cancers Moreover, mary tissue is characterized by a switch from paracrine there is a highly significant association between the risk to autocrine regulation of epithelial cell proliferation of breast cancer and BMI, % body fat and weight gain by sex steroids, i.e., in breast tumors ER␣ and PRs are expressed also in proliferating cells The Epidemiological data suggest that a high BMI may development of breast cancer is closely related to the attenuate the effect of estrogens on breast cancer risk.
function of the normal, slow dividing, long living, The collaborative reanalysis from 1997 found an asso- undifferentiated stem cells which have both a highly ciation between body mass index (BMI) and the rela- proliferative potential and the ability to differentiate tive risk of breast cancer, increasing by 3.1% per kg/m2 Long-term exposure to genotoxic agents may Moreover, the relative breast cancer risk associ- cause mutations resulting in the formation of breast ated with HRT decreased progressively with increasing cancer stem cells/progenitor cells. They can either lose weight or BMI. It was 1.73 in postmenopausal women their steroid receptors and become rapidly proliferat- with a BMI below 22.5 kg/m2 and 1.02 for BMI of ing ER− cells or they become ER+ progenitor cells which proliferate and in addition stimulate growth of to increase the risk of breast cancer only in women ER− cells by producing paracrine factors The with a BMI of less than 24.5 kg/m2 In the Nurses better prognosis of ER+ breast tumors and the more Health Study, the risk of breast cancer correlated with aggressive behaviour of ER− tumors as well as the the BMI in postmenopausal women without HRT, but effect of HRT on these subtypes are associated with their origin. ER␣ and ER␤ are expressed in 60–75% It is well known that an increase in caloric uptake and energy expenditure leads to a stimulation of ER− tumors which arise from the most primitive adrenal androgen secretion, a decrease in SHBG ER− stem/early progenitor cells, are poorly differenti- and an elevated aromatisation of androgens in the ated, more aggressive, and have a poor prognosis. Their excessive fat tissue. A significant correlation between growth is neither influenced by HRT nor by SERMs the serum levels of total and free estradiol and the H. Kuhl / Maturitas 51 (2005) 83–97 characterized by hypertension, coronary heart disease, Association between risk of breast cancer and obesity-related factors dyslipidemia, insulin resistance and hyperinsulinemia Recent investigations suggest that it is in all Breast cancer risk is elevated in obese postmenopausal probability the elevated insulin level in obese post- menopausal women which is responsible for the in- At very low serum concentrations breast cancer risk corre- creased risk of breast cancer. A specific protein secreted HRT does not increase risk of breast cancer in obese post- by adipocytes, adiponectin, correlates with insulin sen- sitivity. Low levels of adiponectin which precede a de- Breast cancer risk correlates with body mass index crease in insulin sensitivity, are closely and inversely Breast cancer risk correlates with % body fat associated with insulin resistance and hyperinsuline- Breast cancer risk correlates with weight gainPrevalence of metabolic syndrome is elevated in obese mia In postmenopausal women a significant in- verse relation between serum adiponectin and breast Insulin resistance and hyperinsulinemia increase breast can- cancer risk was observed, whereas in premenopausal Serum level of C-peptide correlates with risk of mammary has been suggested to be associated with breast cancer Serum level of adiponectin correlates with insulin sensitivity risk in premenopausal women, but in postmenopausal Serum level of adiponectin correlates negatively with body women no relation between breast cancer risk and the levels of IGF-1 was found significant cor- Serum level of adiponectin correlates negatively with insulin relation of the levels of C-peptide with the occurrence of epithelial hyperplasia of the breast or breast can- Serum level of adiponectin correlates negatively with breast cer suggests a key role of elevated insulin levels in Estrogen replacement therapy reduces fasting insulin and in- the growth of breast cancer in postmenopausal women Postmenopausal patients, but not premenopausal Estrogen/progestin reduces incidence of diabetes mellitus in women with type 2 diabetes had a 16% higher breast cancer risk than women without diabetes ( Estrogen replacement therapy reduces risk of breast cancer The lack of an association between breast cancer risk and hyperinsulinemia in premenopausal womensuggests a modulatory role of sex steroids.
risk of breast cancer in postmenopausal women has Low doses of estrogens have been demonstrated to been reported. At estradiol levels of above 8 pg/ml improve insulin sensitivity in postmenopausal women the risk was three times higher than that at levels and to reduce elevated fasting insulin levels, while below 5 pg/ml, and at levels of above 11 pg/ml it was higher estrogen levels or the use of more potent es- five times higher than that at levels below 8 pg/ml trogens may decrease insulin sensitivity. The addition Correlations do not imply causality, and it of progestins may decrease insulin sensitivity, possibly seems rather improbable that such large differences in by reducing insulin binding to the insulin receptor and breast cancer risk are due to such small differences in the estradiol levels. There must be an additional risk Treatment of non-obese postmenopausal women factor associated with obesity, e.g., insulin resistance with 0.625 mg CEE improved insulin sensitivity by and hyperinsulinemia, which is influenced by sex 25%, whereas 1.25 mg CEE caused a decrease by 25%.
steroids and might be involved in the development of The sequential addition of 10 mg MPA antagonized the beneficial effect of 0.625 mg CEE and caused an18% decrease in insulin sensitivity Treatment of 5.2. Hyperinsulinemia and breast cancer risk postmenopausal women with estrogen-only reducedfasting insulin by 35%, while estrogen/progestin The prevalence of insulin resistance and hyperin- combinations were less effective PEPI study sulinemia increases with age, BMI and estrogen de- revealed that treatment of postmenopausal women with ficiency Obese postmenopausal women are at 0.625 mg CEE with or without additional progestins led a high risk to develop a metabolic syndrome that is to a reduction in fasting insulin and glucose levels H. Kuhl / Maturitas 51 (2005) 83–97 In postmenopausal women with impaired glucose tol- erance continuous combined treatment with 0.625 mg Baseline characteristics of the volunteers participating in the WHIstudy CEE and 2.5 mg MPA reduced insulin resistance andfasting glucose levels, while in women with normal glu- cose tolerance the levels of fasting insulin and glucose were decreased increase in the postchallenge glucose concentrations during OGTT that was ob- served in postmenopausal women under HRT, might be caused by a delayed insulin response to glucose and an increased insulin clearance in the liver It was observed in the WHI study that treatment of postmenopausal women with CEE/MPA for 5.6 years on average caused a significant decrease in the inci- dence of diabetes mellitus by 21%. This was probably mediated by a decrease in insulin resistance, as already after 1 year of treatment fasting glucose and insulin had 6. The WHI study – unsuitable for the
investigation of breast cancer risk
6.1. Characteristics of the women participating in In the estrogen-only study the women were randomly assigned to be Concerning the assessment of breast cancer risk, the treated with either placebo or 0.625 mg conjugated equine estrogens high age of the women enrolled in the WHI study could (CEE) in the estrogen/progestin study either with placebo be regarded as an advantage, because the incidence of or 0.625 mg conjugated equine estrogens plus 2.5 mg medroxypro- invasive breast cancer rises with increasing age. The annual number of breast cancer diagnoses increasesfrom 18/1000 women at age 50 years up to 45/1000 cancer risk. As it was shown that HRT does not influ- women at age 63 years and to 63/1000 women at age ence breast cancer risk in postmenopausal women with 70 n both arms of the WHI study the mean age was about 63 years on average and two third of the women suitable for the investigation of the influence of HRT on breast cancer risk. This was even confirmed by the On the other hand, the extremely high mean body WHI Observational Study with about 86,000 women mass index (30.1 and 28.5 kg/m2) and the high per- centage of overweight (34.8 and 35.3%) and adiposewomen (44.6 and 34.1%) in the CEE arm and the 6.2. The WHI study: effect of CEE/MPA or CEE/MPA arm of the WHI study suggests a high preva- lence of the metabolic syndrome (incidence of the metabolic syndrome increases with The randomised placebo-controlled WHI study was menopause, and is associated not only with an ele- planned for an average time of 8.5 years of exposure vated risk of cardiovascular disease, but also with an in- to either CEE/MPA or placebo The premature creased risk of breast cancer owing to insulin resistance discontinuation of treatment with CEE/MPA after 5.2 years was justified with an increased risk of cardiovas- women participating in the WHI study had both an el- cular disease and a pretendedly elevated risk of breast evated risk of coronary heart disease, and a high breast cancer estimated hazard ratio (HR) for inva- H. Kuhl / Maturitas 51 (2005) 83–97 Fig. 1. Number of breast cancer diagnoses per 1000 women per year during the course of treatment with placebo or CEE/MPA in the WHI study(data from of the WHI study published by Chlebowski et al. The left part of the graph refers to women who werenot treated with hormones prior to the WHI study, the right part of the graph refers to women who have received hormone replacement therapyprior to the WHI study (reproduced from Kuhl 2004 sive breast cancer was calculated as 1.26 which was, 1.7 for women with <5 years of prior use and of 2.27 however, not significant. A subsequent updated sub- for ≥5 years of prior use of HRT was due to the ex- analysis based on a mean follow-up of 5.6 years re- tremely low risk in the 2079 women on placebo which vealed a significantly elevated HR of breast cancer of did not show an age-dependent increase in risk. This 1.24 which just surpassed the border of statistical sig- can be interpreted as a hangover effect of pretreatment.
In the Nurses’ Health Study, the HRT-induced eleva- The analysis revealed, however, that in those women tion in breast cancer risk decreased within 2 years after who had never used hormones before initiation of the cessation of treatment and remained lowered during WHI study, treatment with CEE/MPA did not increase the first 5 years without hormones The lacking the risk of breast cancer. It was elevated during treat- effect of CEE/MPA on breast cancer risk in the WHI ment with CEE/MPA only in those patients who re- study corresponds to the results of other studies which ported HRT prior to WHI study graph showing showed that HRT does not increase breast cancer risk the annual number of breast cancers per 1000 women in overweight postmenopausal women with a BMI of during the course of the study (was based 25 kg/m2 or more. This has been observed in the col- on the data depicted in the paper of Chlebowski et laborative study in 1997 n cohort studies al. suspicion that the elevated risk calcu- but also in the large WHI Observational Study lated in this group is an artifact due to a pretreatment-associated selection bias In the group of women 6.3. The WHI study: effect of CEE only or placebo without prior HRT, both the 6277 women treated withCEE/MPA and the 6020 women on placebo showed a The findings of a consistent reduction in the HR of similar age-dependent rise in the rate of breast cancer breast cancer during 6.8 years of treatment of hysterec- which corresponded to a hazard ratio (HR) of 1.09.
tomized postmenopausal women with CEE alone was In the group of women with prior HRT before the highly surprising n total, the relative risk was 0.77 WHI study, treatment of 2225 women with CEE/MPA (95% CI 0.59–1.01), narrowly missing statistical sig- also caused an age-dependent increase which – after nificance. According to the available data on the effect smoothing for fluctuations – was similar to that of the 6020 never users. In this group, the calculated HR of been acceptable if the WHI study had revealed no influ- H. Kuhl / Maturitas 51 (2005) 83–97 ence of estrogens on breast cancer risk. The mean BMI associated with fat mass and high caloric nutrition, and of the women participating in the estrogen-only arm is known as a risk factor for breast cancer. Abdominal was even higher than in the CEE/MPA arm, and was obesity in childhood which is related to early menar- in the range of obesity (30.1 kg/m2). Only 21% had a che, tends to continue into adult life and may be asso- BMI < 25 kg/m2, while 34% had a BMI of 25–29 kg/m2 ciated with an earlier onset of insulin resistance Moreover, late pregnancies are associated with the de- The reduction in breast cancer risk is difficult to velopment of insulin resistance which may persist post explain. Although in postmenopausal women increas- partum in overweight women n the other hand, ing BMI correlates with increasing serum levels of obesity in teenage women may lead to anovulatory cy- estradiol and breast cancer risk correlates with cles which are associated with a reduced risk of breast serum estradiol within a very low concentration range cancer n contrast, the manifestation of obesity af- it is not very probable that the rise of estro- ter teenage increases the risk of postmenopausal breast gen levels during use of 0.625 mg CEE directly protects from the development of breast cancer. On the contrary, In a case-control study with Mexican women char- it is generally believed that high estrogen serum con- acterized by a low fat intake, carbohydrate consump- centrations or high local tissue concentrations stimulate tion was associated with increased breast cancer risk growth of breast tumors. So far, the available epidemi- In another study, a direct association with breast ological data did not show any difference between the cancer risk was observed for glycemic index and effect on breast cancer risk of low and high estrogen glycemic load, but more in postmenopausal than in pre- menopausal women High levels of insulin were Another explanation might be derived from the as- also found to be associated with poorer survival for sociation between breast cancer risk, obesity, insulin postmenopausal women, while higher dietary protein resistance and hyperinsulinemia, as outlined above intake was associated with better survival (According to the high proportion of over- The question is, whether or not a change in di- weight and obese women in the WHI study in whom etary habits leading to weight loss and maintenance a high prevalence of the metabolic syndrome and in- of normal body weight, can normalize the elevated sulin resistance can be assumed, long-term treatment breast cancer risk in overweight women. The results with 0.625 mg CEE might have improved insulin re- of various animal experiments suggest that an energy- sistance and reduced the elevated insulin levels. This restricted state induced by reduced caloric intake and/or might have attenuated the stimulatory effect of insulin an increased energy expenditure might be a suitable on tumor growth resulting in a reduction of breast can- measure to prevent breast cancer In contrast to endocrine treatments this strategy would also includereceptor-negative carcinoma. It might reduce both thecarcinogen-induced initiation and the growth of exist- 7. How to reduce breast cancer risk?
ing tumors. Interestingly, caloric restriction was ac-companied by a persistent reduction in insulin levels case-control study revealed that in adult obesewomen weight loss at younger ages may reduce the risk Western lifestyle is associated with overweight, ab- of postmenopausal breast cancer, whereas weight loss dominal obesity, insulin resistance and low physical after age 45 was ineffective. Fluctuating weight, i.e., activity. Higher age, estrogen deficiency and obesity weight loss followed by weight gain did not influence increase the prevalence of insulin resistance, and di- etary habits may play a critical role. Even in non-obese There are various studies on the association between postmenopausal women the prevalence of fasting hy- diet composition and recurrence rate and survival fol- perinsulinemia is high risk of breast cancer in lowing breast cancer diagnosis. Most studies did not ad- Western countries is five-fold that in Japan, but migra- just for energy intake and the results are contradictory, tion of Japanese women to the USA results in adapta- but suggest an increase in mortality with energy intake tion of risk Early menarche is to a certain degree and a protective effect of elevated intake of protein, H. Kuhl / Maturitas 51 (2005) 83–97 beta-carotene, Vitamin C, fruit and vegetables breast cancer in postmenopausal women. The pro- For postmenopausal women, there is no epidemiolog- tection correlated with the duration and intensity ical evidence for a prophylactic effect of intake of soy of physical activity and was most pronounced in or phytoestrogens concerning the risk of breast can- women with a lower BMI (<24.1 kg/m2) In cer. Two large NIH-funded clinical trials are currently another study with postmenopausal women it was investigating the influence of diet composition on re- shown that vigorous exercise was associated with currence and survival in breast cancer patients.
the lowest plasma insulin levels and the highest in-sulin sensitivity, and this effect was enhanced by HRT 7.2. Medical treatment of hyperinsulinemia There are, however, no clear data on the type and in- Even though diet and exercise are recommended as tensity of exercise necessary for a significant beneficial the primary intervention to improve insulin resistance, the use of insulin-sensitizing agents in patients with in-sulin resistance might be an option to reduce the inci- dence of postmenopausal breast cancer. Metformin hasbeen demonstrated to decrease gluconeogenesis and in- The risk of breast cancer is elevated in women with testinal absorption of glucose, to increase peripheral a mother or sister with breast cancer, and increases glucose uptake and utilization, and to improve insulin further if there are more affected relatives, particularly sensitivity and hyperinsulinemia n combination at young age. In women with genetic mutations with diet it has been shown to improve the symptoms associated with a very high risk for breast cancer, of the metabolic syndrome in women with polycystic prophylactic bilateral mastectomy may reduce the risk ovarian syndrome. Metformin has been used in type by 90% Chemoprevention is also an option to 2 diabetes for many years and is recommended par- reduce the probability of developing the disease early ticularly for overweight patients with type 2 diabetes.
in life. Before long-term treatment with tamoxifen Long-term treatment with insulin-sensitizer may be as- or raloxifene, GnRH analogs or aromatase inhibitors sociated with gastrointestinal side-effects and Vitamin will be considered, the risks, side-effects and benefits B12 deficiency t remains, however, to be proven must be carefully evaluated. It is not clarified whether that long-term metformin-induced normalization of in- and to what extent the use of HRT in carriers of sulin levels leads to a reduction in breast cancer risk.
BRCA1 or BRCA2 increases the risk of breastcancer.
Moderate alcohol consumption is associated with a slightly elevated risk which increases with the amount Benign breast disease, especially fibrocystic dis- of consumed alcohol Therefore, abstinence ease, epithelial hyperplasia and the presence of atypia or reduction of alcohol consumption may have a enhance the risk of breast cancer two- to four-fold favourable effect. A slight increase in breast cancer It was highest in young women with breast cysts and risk was found in postmenopausal women who started decreased with age In postmenopausal women smoking before 16 years of age. Current smoking has benign breast disease was associated with a relative a favourable rather than an unfavourable effect on the risk of breast cancer of about 1.6 and the use risk of breast cancer, because the proportion of infer- of HRT may increase the occurrence of atypical hy- tile women is higher among smokers and smokers reach perplasia In premenopausal women with benign breast disease long-term treatment with daily 8–10 mgnorethisterone, but not progesterone derivatives, was found to reduce the risk of breast cancer by 50% Whether or not this was associated with a reduction of The WHI study revealed that an increased phys- blood flow in the breast remains an open ques- ical activity is associated with a reduced risk for H. Kuhl / Maturitas 51 (2005) 83–97 might be reconsidered. Moreover, the general rec-ommendations to individualize HRT may include the Women with an elevated mammographic density need of an indication for the use of a progestin.
have a four to six times higher risk of developing breast There is little doubt that treatment with unopposed cancer. Mammographic density in more than 75% of estrogens increases dose-dependently the risk of en- the breast area was found to be associated with a relative dometrial hyperplasia and cancer. It has been suggested risk of about 14 for hyperplasia and of 9 for atypical that the use of low-dose estrogens might be associated hyperplasia and/or carcinoma in situ Histologi- with a lower relative risk of endometrial cancer, but cal investigation of biopsies revealed that the increase the results of clinical trials are inconsistent. Whereas in mammographic density does not reflect changes in no difference in the risk of endometrial cancer was ductal or lobular epithelium, but a significantly higher found between the use of 0.3 and 0.625 mg of unop- expression of proteoglycans in the stroma which is posed CEE, the incidence of endometrial hyperplasia the major breast tissue compartment by volume did not differ between placebo and 0.3 mg unopposed These proteoglycans are a highly abundant component of breast tissue stroma and may be involved in the It is generally accepted that an endometrial development of benign (e.g. fibrocystic changes) and thickness of 5 mm is an appropriate cut-off level in screening for endometrial hyperplasia. Monitoring aggregate to form collagen fibres, but can also form of endometrial growth during estrogen therapy by macromolecules with a high capacity for water stor- means of vaginalsonography has been suggested as age. Therefore, the increase in mammographic density a suitable diagnostic tool to evaluate the need for the observed in postmenopausal women during treatment addition of progestins in patients treated with low-dose with estrogen/progestin preparations may reflect an in- estrogens-only. The choice of patients suitable for creased water storage in breast stroma which may also this therapy may be facilitated by the outcome of a cause breast tenderness. A similar phenomenon can be observed in younger women with premenstrual syn- In postmenopausal women, there are large varia- drome who show an increased capillary permeability tions in the endometrial response to unopposed estro- gen therapy. Treatment with 0.625 mg CEE, 1–2 mg It is unknown whether or not this reversible phe- estradiol or 50 ␮g transdermal estradiol revealed that nomenon induced by HRT is associated with an ele- about 20% of the women were fast growers with an vated risk for breast cancer. As an increased density increase in endometrial thickness by more than 1 mm may impair the sensitivity and accuracy of mammo- in 5 weeks, whereas 50% were slow growers with an graphic screening, transitory discontinuation of HRT increase by 1 mm or less over a period of more than 20 for 3 weeks may reverse mammographic density in- weeks Only a few women developed hyperpla- crease and improve the diagnostic sensitivity - sia within 2 months, and no hyperplasia was observed ing this time, the administration of low-dose estrogens in women with endometrial thickness of 4 mm or less may prevent the recurrence of climacteric symptoms.
In 11% of the patients, there was no or onlya slow proliferation rate and after 2 years of treatment 7.8. Do we need an indication for the use of endometrial thickness was below 8 mm showing nor- mal biopsies n about two third of the patients theadministration of progestin could be postponed until at The only indication for the addition of progestins least to the fourth month without inducing endometrial to estrogen replacement therapy is the endometrial protection. Besides other specific progestin-related Long-cycle HRT using quarterly progestin may, adverse effects, the progestin component increases therefore, be an option for the majority of post- menopausal women, but the most suitable regimens Considering the emotional and clinical impact of remain to be elucidated. There are several clinical tri- breast cancer as compared with that of endometrial als on the risk of endometrial hyperplasia and cancer in cancer, the demand for a regular addition of progestins postmenopausal women during long-cycle HRT which H. Kuhl / Maturitas 51 (2005) 83–97 revealed contradictory results. Whereas some long- estrogen-only arm of the WHI study. The lacking effect cycle regimens taken for 1–5 years did not increase the in the estrogen/progestin combination arm may be due incidence of endometrial hyperplasia as compared with to the impairment by the progestin component of the the use of normal sequential preparations n beneficial effect of estrogens on insulin resistance.
elevated rate of endometrial hyperplasia and cancer was With regard to the breast cancer risk, the develop- observed in two Scandinavian studies n one ment of overweight and obesity should be avoided, of the latter studies, the progestin phase of 10 days and an appropriate diet and lifestyle should be recom- might have been too short and in the other study, many mended early in life. In postmenopausal women with of the women who developed endometrial cancer, have insulin resistance, treatment with insulin-sensitizing been treated with hormones including unopposed es- agents like metformin might be an option, but a favourable effect on breast cancer risk remains to Long-cycle HRT might be applicable to patients be proven. In healthy postmenopausal women with who respond to estrogens with slow endometrial pro- climacteric symptoms, low-dose estrogen therapy is liferation and have weak or no withdrawal bleeding the treatment of choice, and the exposure to pro- during sequential HRT. The most suitable progestins gestins should be kept minimal in non-hysterectomized are compounds with strong endometrial activity, and women. There are large inter-individual variations in the proliferative response to estrogens of the en-dometrium. Vaginalsonographic surveillance and theintensity of withdrawal bleeding may help to identify 8. Conclusion
those women who may profit from long-cycle regimensof HRT or may be candidates for therapy with low-dose HRT may stimulate growth of occult breast tumors in postmenopausal women. This concerns primarilyhormone receptor-positive cancers, and the effect ofestrogens is enhanced by progestins. Observational References
and randomised studies suggest that HRT with estro-gen/progestin combinations increases the relative risk [1] Kvale G. Reproductive factors in breast cancer epidemiology.
of breast cancer in postmenopausal women more than estrogens alone. Besides many other risk factors, over- [2] Collaborative Group on Hormonal Factors in Breast Cancer.
weight and obesity is associated with an elevated risk Breast cancer and breastfeeding: collaborative reanalysis ofindividual data from 47 epidemiological studies in 30 coun- of breast cancer in postmenopausal women, which is tries, including 50302 women with breast cancer and 96973 women without the disease. Lancet 2002;360:187–95.
As most participants in the WHI study were over- [3] Garland M, Hunter DJ, Colditz GA, et al. Menstrual cycle weight, it was not suitable for the investigation of breast characteristics and history of ovulatory infertility in relation cancer risk. The increase in risk during treatment with to breast cancer risk in a large cohort of US women. Am JEpidemiol 1998;147:636–43.
estrogen/progestin concerned only those women who [4] Collaborative Group on Hormonal Factors in Breast Cancer.
had been pretreated with hormones prior to the WHI Breast cancer and hormone replacement therapy: collaborative study, suggesting a selection bias. Moreover, treatment reanalysis of data from 51 epidemiological studies with 52705 of postmenopausal women with estrogens alone dur- women with breast cancer and 108411 women without breast ing 6.8 years caused a consistent decrease in the in- [5] Hulka BS, Moorman PG. Breast cancer: hormones and other cidence of breast cancer. Overweight women have a risk factors. Maturitas 2002;42(Suppl 1):95–108.
high risk for the development of insulin resistance, [6] Morimoto LM, White E, Chen Z, et al. Obesity, body size, and and the growth-stimulating effect of elevated insulin risk of postmenopausal breast cancer: the Women’s Health Ini- levels may explain the elevated breast cancer risk in tiative (United States). Cancer Causes Contr 2002;13:741–51.
the postmenopause. As low-dose estrogens may im- [7] Adami HO, Adams G, Boyle P, et al. Breast cancer etiology.
Int J Cancer 1990;(Suppl 5):22–39.
prove insulin resistance and hyperinsulinemia, the el- [8] Ewertz M, Duffy SW, Adami HO, et al. Age at first birth, evated breast cancer risk in obese women may be re- parity and risk of breast cancer: a meta-analysis of 8 studies duced. This could explain the favourable results of the from the Nordic countries. Int J Cancer 1990;46:597–603.
H. Kuhl / Maturitas 51 (2005) 83–97 [9] Colditz GA, Rosner B. Cumulative risk of breast cancer to age [26] Schairer C, Lubin J, Rroisi R, Sturgeon S, Brinton L, Hoover 70 according to risk factor status: data from the nurses’ health R. Menopausal estrogen and estrogen-progestin replacement study. Am J Epidemiol 2000;152:950–64.
therapy and breast cancer risk. JAMA 2000;283:485–91.
[10] Collaborative Group on Hormonal Factors in Breast Cancer.
[27] Kirsh V, Kreiger N. Estrogen and estrogen–progestin replace- Breast cancer and hormonal contraceptives: collaborative re- ment therapy and risk of postmenopausal breast cancer in analysis of individual data on 53297 women with breast cancer Canada. Cancer Causes Contr 2002;13:583–90.
and 100239 women without breast cancer from 54 epidemio- [28] Porch JV, Lee IM, Cook NR. Estrogen–progestin replacement logical studies. Lancet 1996;347:1713–27.
therapy and breast cancer risk: the Women’s Health Study [11] Marchbanks PA, McDonald JA, Wilson HG, et al. Oral con- (United states). Cancer Causes Contr 2002;13:847–54.
traceptives and the risk of breast cancer. N Engl J Med [29] Daling JR, Malone KE, Doody DR. Relation of regimens of combined hormone replacement therapy to lobular, duc- [12] Writing Group for the Women’s Health initiative Investi- tal, and other histologic types of breast carcinoma. Cancer gators. Risks and benefits of estrogen plus progestin in healthy postmenopausal women: principal results from the [30] Weiss LK, Burkman RT, Cushing-Haugen KL. Hormone re- Women’s Health Initiative randomised controlled trial. JAMA placement therapy regimens and breast cancer risk. Obstet [13] Chen WY, Colditz GA, Rosner B, et al. Use of postmenopausal [31] Chen CL, Weiss NS, Newcomb P. Hormone replacement ther- hormones, alcohol, and risk for invasive breast cancer. Ann apy in relation to breast cancer. JAMA 2002;287:734–41.
[32] Jernstr¨om H, Bendahl PO, Lidfeldt J, Nerbrand C, Agardh [14] Kaye JA, Meier CR, Walker AM, Jick H. Statin use, hy- CD, Samsioe G. A prospective study of different types of perlipidaemia, and the risk of breast cancer. Br J Cancer hormone replacement therapy use and the risk of subsequent breast cancer: the women’s health in the Lund area (WHILA) [15] McTiernan, Kooperberg C, White E, et al. Recreational phys- study (Sweden). Cancer Causes Contr 2003;14:673–80.
ical activity and the risk of breast cancer in postmenopausal [33] Li CI, Malone KE, Porter PL. Relationship between long du- women (WHI cohort study). JAMA 2003;290:1331–6.
rations and different regimens of hormone therapy and risk of [16] Schernhammer ES, Laden F, Speizer FE, et al. Rotating night breast cancer. JAMA 2003;289:3254–63.
shifts and risk of breast cancer in women participating in the [34] Bakken K, Asaker E, Eggen AE, Lund E. Hormone replace- nurses’ health study. J Natl Cancer Inst 2001;93:1563–8.
ment therapy and incidence of hormone-dependent cancers [17] Velicer CM, Heckbert SR, Lampe JW, Potter JD, Robertson in the Norwegian Women and Cancer Study. Int J Cancer CA, Taplin SH. Antibiotic use in relation to the risk of breast [35] Stahlberg C, Pedersen AT, Lynge E, et al. Increased risk of [18] Hulley S, Furberg C, Barrett-Connor E, et al. Noncardiovas- breast cancer following different regimens of hormone re- cular disease outcomes during 6.8 years of hormone ther- placement therapy frequently used in Europe. Int J Cancer apy. Heart and estrogen/progestin replacement study follow- up (HERS II). JAMA 2002;288:58–66.
[36] Stahlberg C, Pedersen AT, Andersen ZJ, et al. Breast cancer [19] The Women’s Health Initiative Steering Committee. Effects with different prognostic characteristics developing in Dan- of conjugated equine estrogen in postmenopausal women with ish women using hormone replacement therapy. Br J Cancer hysterectomy. The Women’s Health Initiative Randomized Controlled Trial. JAMA 2004;291:1701–12.
[37] Tjonneland A, Christensen J, Thomsen BL, et al. Hormone [20] Kuhl H. Effects of estrogen-only treatment in postmenopausal replacement therapy in relation to breast carcinoma incidence rate ratios. Cancer 2004;100:2328–37.
[21] Million Women Study Collaborators. Breast cancer and [38] Colditz GA, Rosner BA, Chen WY, Holmes MD, Hank- hormone-replacement therapy in the Million Women Study.
inson SE. Risk factors for breast cancer according to es- trogen and progesterone receptor status. J Natl Cancer Inst [22] Magnusson C, Baron JA, Correia N. Breast cancer risk following long-term oestrogen- and oestrogen-progestin- [39] Black WC, Welch HG. Advances in diagnostic imaging and replacement therapy. Int J Cancer 1999;81:339–44.
overestimations of disease prevalence and the benefits of ther- [23] Chlebowski RT, Hendrix, Langer RD, et al. Influence of es- apy. N Engl J Med 1993;328:1237–43.
trogen plus progestin on breast cancer and mammography in [40] Menard S, Casalini P, Agresti R, Pilotti S, Balsari A. Prolif- healthy postmenopausal women. JAMA 2003;289:3243–53.
eration of breast carcinoma during menstrual phases. Lancet [24] Bush TL, Whiteman M, Flaws JA. Hormone replacement ther- apy and breast cancer: a qualitative review. Obstet Gynecol [41] Potten CS, Watson RJ, Williams GT, et al. The effect of age and menstrual cycle upon proliferative activity of the normal [25] Ross RK, Paganini-Hill A, Wan PC, Pike MC. Effect of human breast. Br J Cancer 1988;58:163–70.
hormone replacement therapy on breast cancer risk: es- [42] Hofseth LJ, Raafat AM, Osuch JR, Pathak DR, Slom- trogen versus estrogen plus progestin. J Natl Cancer Inst ski CA, Haslam SZ. Hormone replacement therapy with estrogen or estrogen plus medroxyprogesterone acetate is H. Kuhl / Maturitas 51 (2005) 83–97 associated with increased epithelial proliferation in the [60] Lindheim SR, Presser SC, Ditkoff EC, Vijod MA, Stanczyk normal postmenopausal breast. J Clin Endocrinol Metab FZ, Lobo RA. A possible bimodal effect of estrogen on insulin sensitivity in postmenopausal women and the attenuating ef- [43] Cline JM, Soderqvist G, von Schoultz E, Skoog L, von fect of added progestin. Fertil Steril 1993;60:664–7.
Schoultz B. Effects of hormone replacement therapy on [61] Carr MC. The emergence of the metabolic syndrome with mammary gland of surgically postmenopausal cynomolgus menopause. J Clin Endocrinol Metab 2003;88:2404–11.
macaques. Am J Obstet Gynecol 1996;174:93–100.
[62] Stefan N, Vozarova B, Funahashi T, et al. Plasma adiponectin [44] Cline JM, Register TC, Clarkson TB. Effects of tibolone and concentration is associated with skeletal muscle insulin re- hormone replacement therapy on the breast of cynomolgus ceptor tyrosine phosphorylation, and low plasma concentra- tion precedes a decrease in whole-body insulin sensitivity in [45] Suparto ICH, Williams JK, Cline JM, Anthony MS, Fox JL. Contrasting effects of two hormone replacement thera- [63] Weyer C, Funahashi T, Tanaka S, et al. Hypoadiponectine- pies on the cardiovascular and mammary gland outcomes in mia in obesity and type 2 diabetes: close association with surgically postmenopausal monkeys. Am J Obstet Gynecol insulin resistance and hyperinsulinemia. J Clin Endocrinol [46] Clarke RB. Human breast cell proliferation and its relationship [64] Mantzoros C, Petridou E, Dessypris N, et al. Adiponectin to steroid receptor expression. Climacteric 2004;7:129–37.
and breast cancer risk. J Clin Endocrinol Metab 2004;89: [47] Conneely OM, Jericevic BM, Lydon JP. Progesterone recep- tors in mammary gland development and tumorigenesis. J [65] Schairer C, Hill D, Sturgeon SR, et al. Serum concentrations Mammary Gland Biol Neoplasia 2003;8:205–14.
of IGF-1, IGFBP-3 and C-peptide and risk of hyperplasia and [48] Roger P, Sahla ME, M¨akel¨a S, Gustafsson JA, Baldet P, cancer of the breast in postmenopausal women. Int J Cancer Rochefort H. Decreased expression of estrogen receptor ␤ protein in proliferative preinvasive mammary tumors. Cancer [66] Michels KB, Solomon CG, Hu FB, et al. Type 2 diabetes and subsequent incidence of breast cancer in the Nurses’ Health [49] Fuqua SA, Schiff R, Parra I, et al. Estrogen receptor ␤ pro- Study. Diabetes Care 2003;26:1752–8.
tein in human breast cancer: correlation with clinical tumor [67] Brown MD, Korytkowski MT, Zmuda JM, McCole SD, parameters. Cancer Res 2003:2434–9.
Moore GE, Hagberg JM. Insulin sensitivity in postmenopausal women. Diabetes Care 2000;23:1731–6.
stem/progenitor cells and the estrogen receptor. Trends En- [68] Espeland MA, Hogan PE, Fineberg SE, et al. Effect of post- menopausal hormone therapy on glucose and insulin concen- [51] Bray GA. Medical consequences of obesity. J Clin Endocrinol trations. Diabetes Care 1998;21:1589–95.
[69] Sumino H, Ichikawa S, Itoh H, et al. Hormone replacement [52] Pan SY, Johnson KC, Ugnat AM, et al. Association of obesity therapy decreases insulin resistance and lipid metabolism in and cancer risk in Canada. Am J Epidemiol 2004;159:259–68.
Japanese postmenopausal women with impaired and normal [53] Schapira DV, Clark RA, Wolff PA, Jarrett AR, Kumar NB, glucose tolerance. Horm Res 2003;60:134–42.
Aziz NM. Visceral obesity and breast cancer risk. Cancer [70] Gelfand MM, Fugere P, Bissonnette F, et al. Conjugated es- trogens combined with sequential dydrogesterone or medrox- [54] Lahmann PH, Lissner L, Gullberg B, Olsson H, Berglund G.
yprogesterone acetate in postmenopausal women: effects on A prospective study of adiposity and postmenopausal breast lipoproteins, glucose tolerance, endometrial histology, and cancer risk: the Malm¨o diet and cancer study. Int J Cancer [71] Margolis KL, Bonds DE, Rodabough RJ, et al. Effect of oe- [55] Lahmann PH, Hoffmann K, Allen N, et al. Body size and strogen plus progestin on the incidence of diabetes in post- breast cancer risk: findings from the European Prospective menopausal women: results from the Women’s Health Initia- Investigation into Cancer and Nutrition (EPIC). Int J Cancer tive Hormone Trial. Diabetologia 2004;47:1175–87.
[72] Kuhl H. Is the elevated breast cancer risk observed in the [56] Toniolo PG, Levitz M, Zeleniuch-Jacquotte A, et al. A WHI study an artifact? (letter to the editor). Climacteric prospective study of endogenous estrogens and breast cancer in postmenopausal women. J Natl Cancer Inst 1995;87:190–7.
[73] Colditz GA, Hankinson SE, Hunter DJ, et al. The use of es- [57] Thomas HV, Key TJ, Allen DS, et al. A prospective study of trogens and progestins and the risk of breast cancer in post- endogenous serum hormone concentrations and breast cancer menopausal women. N Engl J Med 1995;332:1589–93.
risk in postmenopausal women on the island of Guernsey. Br [74] Endogenous Hormones and Breast Cancer Collaborative Group. Body mass index, serum sex hormones, and breast [58] Cauley JA, Lucas FL, Kuller LH, et al. Elevated serum estra- cancer risk in postmenopausal women. J Natl Cancer Inst diol and testosterone concentrations are associated with a high risk for breast cancer. Ann Intern Med 1999;130:270–7.
[75] Huang Z, Hankinson SE, Colditz GA, et al. Dual effects [59] Stoll BA. Upper abdominal obesity, insulin resistance and of weight and weight gain on breast cancer risk. JAMA breast cancer risk. Int J Obesity 2002;26:747–53.
H. Kuhl / Maturitas 51 (2005) 83–97 [76] Romieu I, Lazcano-Ponce E, Sanchez-Zamorano LM, Willett vessels in women with premenstrual tension. Am J Obstet W, Hernandez-Avila M. Carbohydrates and the risk of breast cancer among Mexican women. Cancer Epidemiol Biomark [92] Colacurci N, Fornaro F, de Franciscis P, Mele D, Palermo M, del Vecchio W. Effects of a short-term suspension of hormone [77] Augustin LSA, Dal Maso L, La Vecchia C, et al. Dietary replacement therapy on mammographic density. Fertil Steril glycemic index and glycemic load, and breast cancer risk: a case-control study. Ann Oncol 2001;12:1533–8.
[93] Trabal JF, Lenihan JP, Melchione TE, et al. Low-dose [78] Borugian MJ, Sheps SB, Kim-Sing C, et al. Insulin, unopposed estrogens: preliminary findings on the fre- macronutrient intake, and physical activity: are potential quency and duration of vaginal bleeding in postmenopausal indicators of insulin resistance associated with mortality women receiving esterified estrogens over a two-year period.
from breast cancer? Cancer Epidemiol Biomark Prevdkjdot [94] Notelovitz M, Varner RE, Reaber RW, et al. Minimal endome- [79] Thrompson HJ, Zhu Z, Jiang W. Dietary energy restriction in trial proliferation over a two-year period in postmenopausal breast cancer prevention. J Mammary Gland Biol Neoplasia women taking 0.3 mg of unopposed esterified estrogens.
[80] Trentham-Dietz A, Newcomb PA, Egan KM, et al. Weight [95] Cushing KL, Weiss NS, Voigt LF, McKnight B, Beresford change and risk of postmenopausal breast cancer (United SAA. Risk of endometrial cancer in relation to use of low- States). Cancer Causes Contr 2000;11:533–42.
dose, unopposed estrogens. Obstet Gynecol 1998;91:35–9.
[81] Rock CL. Diet and breast cancer: can dietary factors influence [96] Meuwissen JHJM, van Langen H, Moret E, Navarro- survival? J Mammary Gland Biol Neoplasia 2003;8:119–32.
Morquecho I. Monitoring of oestrogen replacement ther- [82] Kirpichnikov D, McFarlane SI, Sowers JR. Metformin: an apy by vaginosonography of the endometrium. Maturitas update. Ann Intern Med 2002;137:25–33.
[83] Biglia N, Defabiani E, Ponzone R, Mariani L, Marenco [97] Meuwissen JHJM, Oddens BJ, Klinkhamer PJJM. Endome- D, Sismondi P. Management of risk of breast carci- trial thickness assessed by transvaginal ultrasound insuffi- noma in postmenopausal women. Endocrine-related Cancer ciently predicts occurrence of hyperplasia during unopposed oestrogen use. Maturitas 1996;24:21–30.
[84] Adami HO, Adams G, Boyle P, et al. Breast-cancer etiology.
[98] Ettinger B, Selby J, Citron JT, Vangessel A, Ettinger Int J Cancer 1990;(Suppl 5):22–39.
VM, Hendrickson MR. Cyclic hormone replacement ther- [85] Dixon JM, McDonald C, Elton RA, Miller WR. Risk of breast apy using quarterly progestin. Obstet Gynecol 1994;83:693– cancer in women with palpable breast cysts: a prospective [99] Hirvonen E, Salmi T, Puolakka J, et al. Can progestin be lim- [86] Gayet A, Esteve J, Seradour B, Piana L, Jacquemier J. Does ited to every third month only in postmenopausal women tak- hormone replacement therapy increase the frequency of breast ing estrogen? Maturitas 1995;21:39–44.
atypical hyperplasia in postmenopausal women? Results from [100] Boerrigter PJ, van de Weijer PHM, Baak JPA, Fox H, Haspels the Bouches du Rhone district screening campaign. Eur J Can- AA, Kenemans P. Endometrial response in estrogen replace- ment therapy quarterly combined with a progestogen. Matu- [87] Plu-Bureau G, Le MG, Sitruk-Ware R, Thalabard JC, Mauvais-Jarvis. Progestogen use and decreased risk of breast [101] Pinto AB, Binder EF, Kohrt WM, Bronder DR, Williams cancer in a cohort study of premenopausal women with benign DB. Effects of trimonthly progestin administration on the en- breast disease. Br J Cancer 1994;70:270–7.
dometrium in elderly postmenopausal women who receive [88] Madjar H, Vetter M, Pr¨ompeler H, Breckwoldt M, Pfleiderer hormone replacement therapy: a pilot study. Am J Obstet Gy- A. Doppler measurement of breast vascularity in women under pharmacologic treatment of benign breast disease. J Reprod [102] Erkkola R, Kumento U, Lehmuskoski S, Mattila L, Mustonen M. No increased risk of endometrial hyperplasia with fixed [89] Boyd NF, Jensen HM, Cooke G, Lee Han H, Lockwood GA.
long-cycle oestrogen–progestogen therapy after five years. J Mammographic densities and the prevalence and incidence of histological types of benign disease. Eur J Cancer Prev [103] Bjarnason K, Cerin A, Lindgren R, Weber T. Adverse endome- trial effects during long cycle hormone replacement therapy.
[90] Alowami S, Troup S, Al-Haddad S, Kirkpatrick I, Watson PH. Mammographic density is related to stroma and stromal [104] Pukkala E, Tulenheimo-Silfvast A, Leminen A. Incidence proteoglycan expression. Breast Cancer Res 2003;5:R129–35.
of cancer among women using long versus monthly cycle [91] Wong WH, Freedman RI, Levan NE, Hyman C, Quilligan hormonal replacement therapy, Finland 1994–1997. Cancer EJ. Changes in the capillary filtration coefficient of cutaneous

Source: http://win.menopausaitaliana.it/maturitas%20maggio%202005%20rischio%20ca%20seno%20nel%20whi%20il%20problema%20della%20obesita.pdf