Reprint of Breast cancer: hormones and other risk factors

Maturitas 61 (1–2) (2008) 203–213

Reprint of

Breast cancer: hormones and other risk factors Barbara S. Hulka a,*, Patricia G. Moorman b,1 a Department of Epidemiology, School of Public Health and Lineberger Comprehensi6e Cancer Center, Uni6ersity of North Carolina, 2104 McGa6ran-Greenberg Building, CBc 7400, Chapel Hill, NC 27599 -7400, USA b Department of Epidemiology and Public Health, Yale Uni6ersity School of Medicine, New Ha6en, CT 06520, USA

Abstract In North America and Northern Europe, breast cancer incidence rates begin increasing in the early reproductive years and continue climbing into the late seventies, whereas rates plateau after menopause in japan and less developed countries. Female gender, age and country of birth are the strongest determinants of disease risk. Family history and mutations in the BRCA1 and BRCA2 genes are important correlates of lifetime risk. Genetic polymorphisms associated with estrogen synthesis and metabolism are currently under study. Atypical hyperplasia and molecular alterations in benign breast lesions appear to be involved in the pathogenesis of invasive carcinoma. In postmenopausal women, increased breast density on mammograms increases risk. Bone density and breast cancer are associated, presumably through the mechanism of endogenous estrogen levels. Serum estrogen levels are higher in breast cancer cases than controls. Many established risk factors for breast cancer may function through and endocrine mechanism. Current use of oral contraceptives and prolonged, current or recent use of hormone replacement therapy moderately increase risk. Tamoxifen and possibly other selective estrogen receptor modulators reduce breast cancer risk in high risk women. Relationships between various dietary micro and macronutrients and breast cancer have been suggested but require evaluation in clinical trials. Whereas alcohol consumption is associated with increased risk, most environmental factors, including polychlorinated compounds and electromagnetic fields, are not. Conclusion: Breast cancer etiology is becoming clearer through the study of molecular alterations in germline and somatic cell genes, and the interaction of these genes with steroid hormones and relevant growth factors. This knowledge should be useful for breast cancer prevention. © 2001 Elsevier Science Ireland Ltd. All rights reserved. Keywords: Breast neoplasms; Genetics; Hormonal contraceptives; Hormone replacement therapy; Hormones; Risk factors

1. Introduction 

Presented at the International Health Foundation Conference on 3– 4 December 1999 at Hoevelaken, The Netherlands. * Corresponding author. Tel.: + 1-919-9667412; fax: + 1919-9662089. E-mail address: barbara – [email protected] (B.S. Hulka). 1 Present address: Cancer Prevention, Detection, and Control Research Program. Box 2949, Duke University Medical Center, Durham, NC 27710, USA.

Cancer is typically considered a disease of aging, as evidenced by ever-increasing rates throughout the lifetime for most of the common organ sites. Breast cancer, however, shows some distinctive features in terms of age-specific incidence rates. Whereas the incidence rates for most com-

0378-5122/01/$ - see front matter © 2001 Elsevier Science Ireland Ltd. All rights reserved. PII: S 0 3 7 8 - 5 1 2 2 ( 0 0 ) 0 0 1 9 6 - 1

*This article is a reprint of a previously published article, for citation purposes please use the original publication details; Maturitas, 38(1), pp. 103–113. **DOI of original article: doi:10.1016/S0378-5122(00)00196-1

204

B.S. Hulka, P.G. Moorman / Maturitas 61 (1–2) (2008) 203–213

adipose cells are important sites for the conversion of androgenic precursors from the adrenal gland to estrone [2]. Postmenopausal women in the USA are on average much heavier than Japanese women and have higher levels of endogenous estrogens, which could contribute to their higher risk of breast cancer [3]. The average age at natural menopause is also later in the USA than in Japan, allowing for a longer lifetime exposure to endogenous estrogens and progesterone [4]. An alternative interpretation is that environmental factors, such as diet or exogenous hormones, have a more pronounced effect on postmenopausal than premenopausal breast cancer. The incidence and mortality rates from breast cancer vary greatly around the world, exhibiting at least a 10-fold variation (10 –110 new cases per 100 000 women per year) [1,5]. In general, the

mon malignancies start to increase in the late 40s, as illustrated by colon-cancer rates in the upper panel of Fig. 1, breast-cancer rates start to increase in the late 20s (lower panel of Fig. 1) [1]. The early increase in age-specific incidence rates for breast cancer has been attributed to the responsiveness of breast tissues to ovarian hormones, which are active from puberty to menopause. Fig. 1 also shows that age-specific breast cancer rates are similar among most countries until age 35 or 40. Thereafter, the rates diverge. Sweden and the USA show increasing rates until the mid 70s; in Colombia, the age-specific rates increase very little after age 45, and in Japan, low and declining rates are observed after age 45. The different shapes of the age-specific incidence curves among countries are consistent with several interpretations. Differences in levels of endogenous hormones may be more pronounced after menopause due to differences in average body weight. When ovarian production of estrogen ends,

Fig. 1. Age-adjusted incidence rates for colon and breast cancer, 1988– 1992, in four countries: USA (SEER registries), Sweden, Colombia (Cali), and Japan (Miyagi Prefecture). From: Cancer Incidence in Five Continents, Volume VII.

B.S. Hulka, P.G. Moorman / Maturitas 61 (1–2) (2008) 203–213

rates are highest in the developed countries of North America and Northern Europe and lowest in less developed countries of the Far East, Africa and South America. Japan is an exception, being a highly developed country with low breast-cancer rates. However, rates have been rising in Japan over the last few decades, although this is not reflected in older women who have experienced lower breast-cancer rates throughout their lifetimes [6]. The international differences in breast-cancer incidence and mortality rates indicate important differences in the endogenous hormonal milieu, lifestyle and environmental factors, genetic susceptibility, and mammographic screening activities among countries. Different factors may be influential at different points in the pathogenic pathway to malignancy. Susceptibility of populations differ because of the variability in the frequency of major genes such as BRCA1 and 2, and alleles for the metabolizing genes in the CYP P450 group, as well as many predisposing and protective genes that have yet to be identified. The United States experienced a marked increase in breast-cancer incidence during the 1980s with a small decline during the 1990s to about 110 per 100 000 women annually. Since 1989, there has been a decline in mortality rates averaging 1.8% per year [7]. The increased incidence rates of the 1980s are attributed to the introduction and widespread acceptance of screening mammography [8]. The initial effect of a screening program is an increase in the number of cases detected. After the pool of prevalent cases has been detected and treated, the incidence rate should decline, since only newly arising cases are available for mammographic detection. The decline in breast-cancer mortality has been attributed to both the earlier stage at diagnosis associated with screening and the use of adjuvant therapies [7]. For the USA, 175 000 new cases of invasive breast cancer and 43 300 deaths are predicted for 1999 [9].

2. Risk profile There are several well-established risk factors for breast cancer and a variety of others currently

205

under study. Table 1 presents the established factors, organized in categories according to the magnitude of their effects. The ratio of female to male breast cancer approximates 100:1. The importance of age in determining risk was noted previously. Breast cancer is primarily a disease of middle-aged and older women, with over twothirds of cases in the United States diagnosed in women aged 55 or older [10]. Among women aged 75–79, more than one new case occurs among every 300 women annually.

2.1. Family history/genetic factors Family history is one of the most well-established breast-cancer risk factors. A woman with a mother or sister with breast cancer has an approximately two- to threefold excess risk of developing the disease [11]. Having multiple affected relatives, particularly with early onset or bilateral disease, further increases the risk. Recent advances in molecular biologic tecliniques have contributed greatly to our understanding of inherited susceptibility to breast cancer. It has been estimated that approximately 7% of all breast-cancer cases are due to inherited genes. The first breastcancer susceptibility gene, BRCAl, which is linked to both breast and ovarian cancer, was identified in 1990 and cloned in 1994 [12]. This gene has been associated primarily with early-onset disease (before age 50), although carriers of the gene may develop breast cancer at older ages as well. On the basis of studies conducted among high-risk families (i.e. multiple cases spanning several generations), it had been estimated that individuals with a mutated form of BRCAl have an 80–85% chance of developing breast cancer during their lifetime. However, more recent studies have reported BRCA1 variants among women with breast cancer who do not have a pronounced family history and have calculated lower penetrance estimates in the range of 55–70% [13]. A focus of current research is to identify environmental or other genetic factors that trigger disease expression, to determine if breast turnors with BRCAl mutations have distinct clinical and pathologic characteristics, and to evaluate the effectiveness of preventive strategies, such as tamoxifen, among BRCA1 carriers.

206

B.S. Hulka, P.G. Moorman / Maturitas 61 (1–2) (2008) 203–213

Table 1 Established risk factors for breast cancer

Table 1 (Continued) Factor

Factor

Gender Age Country of birth

High-risk group Relative risk \4.0 Female Old North America Northern Europe Yes Yes

BRCA1/2 Two first-degree relatives with breast cancer diagnosed at an early age History of cancer in Yes one breast Mammographic 75% of breast dense density (postmenopausal) Relative risk 2.1–4.0 One first-degree Yes relative with breast cancer Biopsy-confirmed Yes atypical hyperplasia High-dose radiation Yes to chest Oophorectomy No before age 35 Bone density High (postmenopausal) Hormonal Age at first full-term pregnancy Age at menarche Age at menopause Obesity (postmenopausal) Parity (postmenopausal) Lactation (premenopausal) Hormonal contraceptives recent use Hormone replacement therapy recent and long-term use Other Height History of primary cancer in endometrium, ovary, or colon

Low-risk group

Male Young Asia, Africa

High-risk group

Alcohol consumption Yes Socio-economic High status Religion Jewish

No No

No B5% dense

No

No No Yes Low

Relative risk 1.1–2.0 \30 years

B20 years

B12 years \55 years Obese

\14 years B45 years Thin

Nulliparous

Multiparous

None

Several years

Yes

No

Yes

No

Tall Yes

Short No

Low-risk group No Low Seventh Day Adventist, Mormon

BRCA1 is the most fully characterized breastcancer susceptibility gene, but a number of other genetic loci have been identified that appear to have a role in the disease, including BRCA2 (which appears to be linked to male breast cancer), ataxia telengectasia heterozygotes, p53 (which is associated with Li –Fraumeni syndrome), and Ha-ras polymorphisms [14]. Genetic polymorphisms are another area of potential interest in breast cancer. Although these do not infer the large relative risk of major genes such as BRCA1, they could still be important from the standpoint of attributable fraction because the alleles conferring susceptibility often have a high frequency in the population. Genes involved in estrogen synthesis, CYP17 and CYPl9, or metabolism, CYPlAl and OST, are examples. Little can be said as yet about the importance of these or other genetic polymorphisms for breast-cancer pathogenesis. In general, the studies that have been reported relied on selected samples and small numbers of subjects, whereas large population-based studies are needed.

2.2. Mammographic density Mamographic density is identified on mammograms as the non-radiolucent portions of the image. These represent the fibrous and glandular tissues in the breast, whereas, the dark radiolucent areas are primarily fat. Several studies have been consistent in showing that postmenopausal women with a high percentage density in their breasts are at a fourfold or greater increased risk of breast cancer as compared to women with very low percentages of density [15]. The mechanism

B.S. Hulka, P.G. Moorman / Maturitas 61 (1–2) (2008) 203–213

for this association is not known, although insulin-like growth factor (IGF-1) has been suggested to play a role. What accounts for the variation among women in the percentage of the breast that is dense is also not known, but it has been suggested that hormones may be influential. For this reason, we have undertaken two studies, both of which are ancillary to large, randomized, double-blind, placebo-controlled clinical trials. The first study relies on the hormone-replacement component of the Women’s Health Initiative. Use of a computerized interactive thresholding technique allows us to outline the breast and the densities within the breast. These measurements are made and recorded on films taken at baseline (prior to initiation of therapy, active or placebo) and at annual visit 1 and 2. Thus, the change in percentage density is available for each subject, and the difference between active treatment and placebo groups can be identified. If our hypothesis is correct, we will find a greater increase in percentage density in the hormone group as compared to the placebo group. The second study relies on mammographic films taken in the context of the Breast Cancer Prevention Trial where women were randomly allocated to either tamoxifen or placebo. Using the same measurement technique, design and analytic approaches as in the Women’s Health Initiative ancillary study, we will identify changes in the percentage breast density from baseline to year 2 and determine if density is reduced in the tamoxifen arm participants relative to those in the placebo group.

2.3. Histology of benign lesions Biopsy confirmed atypical hyperplasia increases breast-cancer risk some three- to fourfold; without atypical features, hyperplasia has a more modest impact on risk [16,17]. Although the relative risk is high, the attributable risk from atypical hyperplasia is small, since in most biopsy series, less than 5% of women exhibit this lesion.

207

2.4. Ionizing radiation High-dose ionizing radiation to the chest, such as that delivered to children to reduce the size of the thymus gland or repeated flouoscopies used during lung-collapse therapy for tuberculosis, has been shown to increase the subsequent risk of breast cancer [18,19]. These indications, which are not currently relevant to medical care, resulted in multi-rad exposures. Radiation exposure from modern mammographic equipment is in the range of 200 –400 mrad, which has been calculated to have minimal impact on breast-cancer risk [20].

2.5. Endogenous estrogens Although there are many hormonally relevant factors, only two appear in the relative risk category 2.1 –4.0. Bilateral oophorectomy has a significant impact on breast-cancer risk. The earlier in life that the ovaries are removed, the greater the risk reduction. This was one of the first observations suggesting a hormonal role in breast cancer etiology [21]. Bone density is the other factor in this risk category. In postmenopausal women, those with a high bone density are also at the highest risk of breast cancer [22]. Since estrogens of either exogenous or endogenous origin are known to help retain bone mass, the inference is made that bone density and breast-cancer risk are correlated as a function of the amount of estrogen available to the target tissues. Other well-established breast cancer risk factors are associated with endogenous female hormones, but the magnitude of the relative risk estimates is in the 1.1–2.0 range [23]. A younger age at menarche and older age at menopause suggest that the risk of breast cancer is directly related to the length of exposure time to cycling ovarian hormones. A younger age at menopause is protective regardless of whether the menopause was natural or surgical. Pregnancy is generally thought to reduce the risk of breast cancer in subsequent years. The younger a woman is at her first full-term pregnancy, the lower her risk of breast cancer, and additional pregnancies further reduce her risk.

208

B.S. Hulka, P.G. Moorman / Maturitas 61 (1–2) (2008) 203–213

However, parous women who experience their first pregnancy at an older age (\30) are actually at a higher risk than nulliparous women. Furthermore, the protective effect of pregnancy has been observed primarily for postmenopausal breast cancer. Some studies have shown an increase in risk during and after pregnancy, which has been attributed to the high levels of exposure to estrogen and progesterone [24]. After a postpartum period of 10– 15 years, long-term protection is observed, probably as a result of the differentiation of the breast tissue that occurs with pregnancy. Breast-feeding has long been hypothesized to protect against breast cancer. Lactation delays the return of ovulation after pregnancy and is associated with a different hormonal milieu (increased prolactin), which results in less estrogen exposure to the breast. It has also been suggested that potential carcinogens could be excreted in breast milk. Most epidemiologic evidence suggests that the reduction in risk associated with lactation is more pronounced for premenopausal than postmenopausal breast cancer [25]. Risk reduction has been observed primarily with a longer duration of breastfeeding. Studies within the United States are limited in their ability to evaluate the effect of long-term breastfeeding since only a small proportion of women lactate for more than a year. Obesity in postmenopausal women is also thought to act through a hormonal mechanism. Androgenic precursors from the adrenal gland are metabolized to estrogens, primarily estrone, in adipose cells. As a result, heavier postmenopausal women have been shown to have higher circulating levels of estrone than comparable thin women [26]. Less well understood is the observation that obesity may be protective and leanness a risk factor for premenopausal disease. Several recent studies have measured serum levels of various estrogens in relation to risk of breast cancer. In both case control and cohort studies, higher levels were found in breast-cancer cases than controls [27 – 31].

2.6. Exogenous estrogens The relationship of exogenous hormones, primarily hormonal contraceptives and hormone replacement therapy, to breast cancer has been researched extensively. The lack of total consistency among studies may be attributed in part to the fact that these exposures are not static. Changes in patterns of use, reductions in hormone dose, and temporal considerations all contribute to the difficulty in comparing the many studies. Oral contraceptives used today are considerably different from the formulations used in the 1960s. The average dose of hormones in the most common, currently used oral contraceptives are a fraction of those used in the 1960s. Formulations also have evolved over the years from sequential preparations, to constant dosing of estrogen and progestin for 21 out of 28 days, to more recently introduced bi- or tri-phasic dosing, with different doses of estrogen and progestin at different times in the cycle. In addition, the patterns of use have changed since the pill was first introduced. Early on, oral contraceptives were used primarily by married women to space pregnancies or limit family size. This contrasts with the most common current usage pattern in which younger women (often teenagers) take oral contraceptives to delay a first pregnancy. Because oral contraceptives are used by many women world-wide, possible adverse effects must be evaluated. To the extent that these exist, they can be weighed against the benefits of highly effective contraception. The Collaborative Group on Hormonal Factors in Breast Cancer was set up in 1992 to gather and reanalyze data from the many epidemiologic studies that have addressed this issue in an effort to provide more definitive information on the risk associated with oral contraceptive use [32]. The results of their analyses of data pooled from 54 studies were reassuring, with ever use of oral contraceptives associated with a very small increase in risk (relative risk 1.07). The greatest risk was observed among current and recent users (within 4 years of diagnosis), with the risks declining with increasing time since last use. No increased risk was apparent for women who had discontinued use 10 or more years ago. Dura-

B.S. Hulka, P.G. Moorman / Maturitas 61 (1–2) (2008) 203–213

tion of use, age at first use, or dose of the oral contraceptive had little effect on breast-cancer risk, once recency of use was taken into account. The data also indicated that the tumors diagnosed among oral contraceptive users were more likely to be localized to the breast than among nonusers. As with oral contraceptives, menopausal hormone therapy has not remained the same over the years. Unopposed estrogen therapy was commonly used until the mid- to late 1970s, when its association with endometrial cancer was discovered. Use of menopausal estrogens in the United States declined at that point. Since the 1980s, there has been a resurgence in use due to the recognition of the benefits in preventing osteoporosis and heart disease in addition to the long-recognized benefits in relieving hot flushes and other symptoms of menopause. However, current recommendations are for women with an intact uterus to take a progestin along with estrogen to protect the endometrium. Conjugated equine estrogen (Premarin®) has been the most commonly prescribed estrogen in the United States for decades, but the average estrogen dose has been reduced from 1.25 mg to 0.625 mg for most women. Outside of the United States, other types of estrogens (e.g. estradiol, estriol) have been used more commonly. The evidence suggests that menopausal estrogens are associated with a modest increase in breast-cancer risk. Long-term use (5 years or more) among current or recent users appears to be associated with a 30 – 50% increase in breastcancer risk. Reports from both the Collaborative Group on Hormonal Factors in Breast Cancer and the US Nurses Health Study support these risk estimates [33,34]. Among current users in the Nurses Health Study, older women (aged 60–64) who had used estrogens for at least 5 years had double the risk of women who reported no hormone use. Both study groups found a higher proportion of localized disease among hormone users than among non-users. Other reports have suggested that the risk from exogenous hormones may be somewhat higher in certain subgroups such as women with a family history of breast cancer.

209

Because the use of estrogen along with progestin is a relatively recent development, fewer studies have been able to address the impact of using a progestin along with estrogen. Nearly all studies have found that the risk of breast cancer is at least as high among women who used estrogen and progestin as among those who used estrogen alone. A recent Swedish study reported a relative risk of 1.4 after 1–6 years of intake and 1.7 after more than 6 years. The adverse effect of estrogen plus progestin was primarily confined to recent users [35]. Thus, it appears that the response of breast cells to progestin is different from that of the endometrium. The addition of progestin does not protect the breast from the adverse effects of estrogen. Although a history of breast cancer is generally considered a contraindication for estrogen use, some clinicians have questioned whether this prohibition is always appropriate. Since a large majority of breast cancer patients survive at least 5 years, physicians must give consideration to the long-term well-being of the patient in addition to merely treating the cancer. In particular, younger women treated for breast cancer often experience premature menopause as a result of chemotherapy. Without the option of hormone-replacement therapy, these women may have a diminished quality of life due to menopausal symptoms and may be at higher risk of coronary artery disease and osteoporosis. Currently, there are few data available to evaluate whether the use of hormones puts women at higher risk for breast-cancer recurrence after successful treatment. Several observational studies have reported that women who were using estrogen at the time of breast cancer diagnosis had as good or better survival than non-users [36 –38]. However, potential confounding factors might explain these findings. In selected series of breastcancer patients treated with estrogens for menopausal symptoms, the timing and frequency of recurrences appeared no different to those in untreated patients [39,40]. The apparent absence of a detrimental effect on survival from breast cancer, coupled with the benefits on the heart and bones, has led to the initiation of clinical trials to evaluate estrogen therapy after successful treat-

210

B.S. Hulka, P.G. Moorman / Maturitas 61 (1–2) (2008) 203–213

ment for breast cancer. Issues to be evaluated in such trials include whether there is an increased risk of breast-cancer recurrence, how soon after breast-cancer treatment it would be appropriate to initiate estrogen therapy, and whether hormone therapy should include tamoxifen as well as estrogen.

2.7. Diet Diet has long been hypothesized to be one of the primary reasons for the observed differences in breast-cancer rates between countries, with fat intake the component most often thought responsible for the differences [41]. International comparisons show a strong correlation between a country’s average fat intake and breast-cancer rates. In addition, experimental evidence in animals demonstrates that a high fat intake increases the rate of mammary tumor formation. However, prospective epidemiologic studies generally have not found a relationship between fat intake and breast cancer. It is well known that measurement of dietary intake is inexact and prone to misclassification, which would tend to obscure a true relationship. Recognizing the difficulty in interpreting the observational studies, the Women’s Health Initiative (WHI) has been initiated in the US to test this hypothesis in a randomized clinical trial [42]. A more recent hypothesis suggests that dietary soy protein may protect against breast cancer [43]. As with dietary fat, international comparisons provide some of the rationale for the hypothesis. Breast-cancer rates tend to be lower in those countries (e.g. China and Japan) in which soy products are an important component of the diet. The reduction in breast-cancer risk is hypothesized to be due to the effect of phytoestrogens of the isoflavone class contained in the soy protein. These compounds are thought to act as estrogen agonists/antagonists, analogous to the action of tamoxifen. As yet, epidemiologic evidence is inconclusive. Other dietary components that have been investigated include micronutrients such as vitamins A, C, or E, and selenium [44]. Vitamins A, C, and E and selenium are thought to reduce cancer risk

through their antioxidant properties. Vitamin A has been shown to inhibit tumor proliferation in experimental models. Although some laboratory and epidemiologic evidence is suggestive of a protective effect for these nutrients, it is difficult to pinpoint which are most important since dietary intake of the vitamins tends to be highly correlated. It also has not been established whether it is the vitamins themselves or some other component of the foods in which they are found that confers protection. Vitamin D, a steroid hormone in the same family as the sex-steroid hormones, also has been hypothesized to be associated with breast-cancer risk. Formed vitamin D circulates in the blood, allowing it to be hydroxylated at position 25 in the liver and position 1 in the kidney to form 1,25 (OH)2D, the active metabolite. In in-vitro systems, Vitamin D enhances cell differentiation and can reduce cell proliferation. These observations led us to postulate that lower levels of 1,25 (OH)2D might be associated with an increased risk of breast cancer. We examined this question in a clinic-based, case-control study of 157 cases and 184 controls in which levels of vitamin D metabolites were measured in archived blood samples [45]. Among Caucasian women, mean levels of 1,25 (OH)2D were significantly lower among cases than controls, whereas there were no differences in 25 (OH)D values. As compared to women in the highest quartile of 1,25 (OH)2D values, risk of breast cancer increased for each successively lower quartile. The odds ratio for breast cancer was 5.3 (95% CI 2.1 –13.4) for women in the lowest quartile of 1,25 (OH)2D values. Stronger associations with low 1,25 (OH)2D were observed for tumors that were positive for estrogen and/or progesterone receptors, and among women age 55 and over. These findings are sufficiently provocative that the vitamin D hypothesis deserves further testing in additional cohorts or clinical trials.

2.8. Other lifestyle factors Alcohol consumption has been reported quite consistently as a breast-cancer risk factor [46,47]. The risk associated with a given level of alcohol

B.S. Hulka, P.G. Moorman / Maturitas 61 (1–2) (2008) 203–213

intake varies considerably between studies. Some have reported an increased risk with as little as a few drinks per week, while others have observed no increase in risk except at consumption levels over three drinks per day. Taken as a whole, the studies suggest that alcohol consumption at a level of one to two drinks per day modestly increases breast-cancer risk (relative risk in the range of about 1.2 – 1.4). Cigarette smoking appears to have little overall effect on breast-cancer risk [48]. Mutagens from cigarette smoke have been detected in breast fluid, suggesting the possibility of a direct carcinogenic effect. However, women who smoke tend to be leaner, have lower urinary estrogen levels, and experience earlier menopause, effects that would tend to lower the risk of postmenopausal breast cancer. These competing mechanisms might explain why, in general, cigarette smoking has not been found to be associated with breast cancer. It has been postulated that there may be a gene/environment interaction between smoking and the N-acetyl transferase (NAT) genotypes. An early report suggested a strong, dose – response relationship between smoking and breast cancer among women with the NAT2 slow genotype, but subsequent studies have not replicated those findings [49]. Socio-economic status is a factor that emerges consistently as being related to the risk of breast cancer [50]. It emerges at the ecologic level in that developing countries have lower rates, whereas Europe and North America have much higher rates. It is also apparent within countries that have economic diversity in their populations. Socio-economic status can be measured in many ways, but educational level serves reasonably well as a proxy. Why educational level is associated with higher breast-cancer rates is not known, but the association is not attributable to reproductive correlates of education because those factors have been controlled for in the studies. In the United States, breast-cancer incidence rates vary by religious group [50]. Jewish women have higher rates, whereas Seventh Day Adventists and Mormons appear to have lower rates. The higher rates among Jewish women may be due to the high prevalence (about 1%) in the

211

population of distinctive mutations in BRCA1 [51]. Other genetic factors including mutations in BRCA2 may also be operative. Mormons and Seventh Day Adventists may be at lower-than-average risk due to their reproductive patterns, which include early age at first pregnancy and multiparity, or to lifestyle choices, such as the avoidance of alcohol. Physical activity in youth and adulthood, whether on the job or during leisure time, is favored by some as having a beneficial effect on risk of breast cancer [52]. The data are not consistent, possibly due to the problems of measurement and subject memory about prior exercise and physical activity. Furthermore, a clear biologic rationale for the relevant phase of life or ‘‘window’’ during which exercise might be beneficial is not well defined. Clinical trials are needed, even though they present significant logistic and sample-size problems, if breast cancer is the outcome. Mammographic density might be an alternative intermediate outcome. Even clinical trials would be constrained to evaluate the effect of current or recent exercise, and not that which occurred in the distant past (adolescence, for example).

2.9. En6ironmental agents Environmental agents, frequently called ‘‘endocrine disruptors’’, have received a great deal of public and scientific attention in recent years. The compounds that have received most study and for which there are data on breast cancer include the organochlorine, DDT, and its persistent metabolite DDE, and the polychlorinated biphenyls, the PCBs. Several significant studies, using blood or adipose tissue concentrations of the various compounds for exposure measurement, have been reported from the United States, Europe and Mexico; all have shown no association between DDE or PCBs and risk of breast cancer [53]. Electromagnetic fields have been studied as a possible causative factor for breast cancer. Electric blankets might be particularly suspect since they tend to be used regularly, and the fields are in close contact with the body. Overall, the studies do not support an association between electric blanket use and breast cancer [54].

212

B.S. Hulka, P.G. Moorman / Maturitas 61 (1–2) (2008) 203–213

In summary, sex-steroid hormones are undoubtedly involved in the development of neoplasms of the breast. The particular stages of life at which they have the greatest influence and the actual levels and combination of hormones that may have an adverse effect are not fully understood. Advances in molecular biology and genetics have revealed the importance of high-penetrance genes, e.g. BRCAl and BRCA2, particularly among women with early age of disease onset. More recently, interest has focused on genetic polymorphisms involved with hormone production and metabolism and the metabolism of xenobiotic carcinogens. These are genes in which the alleles of interest are common in the population but which have a low penetrance. Their impact is small in terms of relative risk but could be large with respect to attributable risk. The search for physical or chemical factors in the environment affecting breast-cancer risk has not been productive. Diet and other aspects of lifestyle are still controversial. Only alcohol consumption appears to confer a small increased risk, and dietary fat is under study. Dense breasts on mammogram and atypical hyperplasia in biopsy specimens have a significant impact on the subsequent risk of breast cancer.

References [1] Parkin DM, Whelan SL, Ferlay J, Raymond L, Young J, editors. Cancer Incidence in Five Continents. Oxford: International Agency for Research on Cancer, Oxford University Press, 1997. [2] Simpson ER, Mendelson CR. The role of adipose tissue in estrogen biosynthesis in the human. In: Frisch RE, editor. Adipose Tissue and Reproduction. New York: Krager, 1990. [3] Shimizu H, Ross R, Bernstein L, Pike MC, Henderson BE. Serum estrogen levels in postmenopausal women: comparison of US whites and Japanese in Japan. Br J Cancer 1990;62:451–4. [4] Osei-Hyiaman D, Satoshi T, Ueji M, Hideto T, Kano K. Timing of menopause, reproductive years, and bone mineral density: A cross-sectional study of postmenopausal Japanese women. Am J Epidemiol 1998;148:1055– 61. [5] Pisani P, Parkin DM, Ferlay J. Estimates of the worldwide mortality from eighteen major cancers in 1985: Implications for prevention and projections of future burden. Int J Cancer 1993;55:891–903.

[6] Coleman MP, Est’eve J, Damiecki P, Arslan A, Renard H. Trends in Cancer Incidence and Mortality. Lyon, France: IARC Publications, 1993 No. 121. [7] Chu K, Tarone R, Kessler L, et al. Recent trends in U.S. breast cancer incidence, survival, and mortality rates. J Natl Cancer Inst 1996;88:1571– 9. [8] Tarone R, Chu K, Gaudette L. Birth cohort and calendar period trends in breast cancer mortality in the United States and Canada. J Natl Cancer Inst 1997;89:251–6. [9] Landis SH, Murray T, Bolden S, Wingo PA. Cancer statistics. CA Cancer J Clin 1999;49:8– 31. [10] Ries LAG, Kosary CL, Hankey BF, Miller BA, Clegg LX, Edwards BK, editors. SEER Cancer Statistics Review, 1973 – 96. Bethesda, MD: National Cancer Institute, 1999 NIH Pub. No. 99-2789. [11] Thompson WD. Genetic epidemiology of breast cancer. Cancer 1994;74(Suppl. 1):279– 87. [12] Miki Y, Swensen J, Shattuck-Eidens D, et al. A strong candidate for the breast and ovarian cancer susceptibility gene BRCAl. Science 1994;266:66– 71. [13] Blackwood MA, Weber BL. BRCA1 and BRCA2: From molecular genetics to clinical medicine. J Clin Oncol 1998;16:1969–77. [14] Newman B, Millikan RC, King M-C. Genetic epidemiology of breast and ovarian cancers. Epidemiol Rev 1997;19:69– 79. [15] Boyd N, Lockwood G, Byng J, Tritchler D, Yaffe M. Mammographic Densities and Breast Cancer Risk. Cancer Epid Biomark Prevent 1998;7:1133– 44. [16] London S, Connolly J, Schnitt S, Colditz G. A prospective study of benign breast disease and the risk of breast cancer. J Am Med Assoc 1992;267:941–4. [17] Dupont W, Page D. Risk factors for breast cancer in women with proliferative breast disease. New Engl J Med 1985;312:146–51. [18] Hildreth NG, Shore RE, Dvoretsky PM. The risk of breast cancer after irradiation of the thymus in infancy. New Engl J Med 1989;321:1281– 4. [19] Miller AB, Howe GR, Sherman GJ, Lindsay JP, Jaffe MJ, Dinner PJ, et al. Mortality from breast cancer after irradiation during fluoroscopic examinations in patients being treated for tuberculosis. New Engl J Med 1989;321:1285– 9. [20] Bassett LW, Hendrick RE, Bassford T, Butler PF, Carter D, et al. Quality Determinants of Mammography. Rockville, MD: US Department of Health and Human Services, Agency for Health Care Policy and Research, 1994. [21] Trichopoulos D, MacMahon B, Cole P. Menopause and breast cancer risk. J Natl Cancer Inst 1972;48:605– 13. [22] Kuller L, Cauley J, Lucas L, Cummings S, Browner W. Sex steroid hormones, bones mineral density, and risk of breast cancer. Environ Health Perspect 1997;105:1–7. [23] Hulka BS, Stark A. Breast cancer: cause and prevention. Lancet 1995;346:883– 7. [24] Lambe M, Hsieh C, Trichopoulos D, Ekbom A, Pavia M, Adami H. Transient increase in the risk of breast cancer after giving birth. New Engl J Med 1994;331:5– 9.

B.S. Hulka, P.G. Moorman / Maturitas 61 (1–2) (2008) 203–213

[25] Newcomb P, Storeer B, Longnecker M, et al. Lactation and a reduced risk of premenopausal breast cancer. New Engl J Med 1994;330:81– 7. [26] Cauley U, Gutai JP, Kuller LH, LeDonne D, Powell JG. The epidemiology of serum sex hormones in postmenopausal women. Am J Epidemiol 1989;129:1120– 31. [27] Thomas H, Reeves G, Key T. Endogenous estrogen and postmenopausal breast cancer: a quantitative review. Cancer Causes Control 1997;8:922–8. [28] Dorgan J, Longcope C, Stephenson H, et al. Relation of prediagnostic serum estrogen and androgen levels to breast cancer risk, cancer epidemiology. Biomark Prevent 1996;5:533–9. [29] Toniolo P, Levitz M, Zeleniuch-Jacquotte A. A prospective study of endogenous estrogens and breast cancer in postmenopausal women. J Natl Cancer Inst 1995;87:190– 7. [30] Hankinson S, Willett W, Manson J, et al. Plasma sex steroid hormone levels and risk of breast cancer in postmenopausal women. J Natl Cancer Inst 1998;90:1292– 9. [31] Bernstein L, Yuan J, Ross R, et al. Serum hormone levels in pre-menopausal Chinese women in Shanghai and white women in Los Angeles: results from two breast cancer case-control studies. Cancer Causes Control 1990;1:51– 8. [32] Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal contraceptives: collaborative reanalysis of individual data on 53 297 women with breast cancer and 100 239 women without breast cancer from 54 epidemiological studies. Lancet 1996;347:1713–27. [33] Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormone replacement therapy: collaborative reanalysis of data from 51 epidemiological studies of 52 705 women with breast cancer and 108 411 women without breast cancer. Lancet 1997;350:1047– 59. [34] Colditz G, Hankinson S, Hunter D, et al. The use of estrogens and progestins and the risk of breast cancer in postmenopausal women. New Engl J Med 1995;332:1589–93. [35] Persson I, Weiderpass E, Bergkvist L, Bergstrom R, Schairer C. Risks of breast and endometrial cancer after estrogen and estrogen–progestin replacement. Cancer Causes Control 1999;10:253– 60. [36] Hunt K, Vessey M, McPherson K, Coleman M. Longterm surveillance of mortality and cancer incidence in women receiving hormone replacement therapy. Br J Obstet Gynecol 1987;94:620– 35. [37] Henderson BE, Paganini-Hill A, Ross RK. Decreased mortality in users of estrogen replacement therapy. Arch Intern Med 1994;151:75–8. [38] Bergkvist L, Adami HO, Persson I, Bergstrotn R, Krusemo UB. Prognosis after breast cancer diagnosis in women exposed to estrogen and estrogen– progestogen replacement therapy. Am J Epidemiol 1992;130:221– 8. [39] DiSaia P, Odicino F, Grosen E, Cowan B, Pecorelli S, Wile A. Hormone replacement therapy in breast cancer. Lancet 1993;342:1232.

213

[40] Cobleigh MA, Berris RE, Bush T, et al. Estrogen replacement therapy in breast cancer survivors. J Am Med Assoc 1994;272:540– 6. [41] Hunter DJ, Willett WC. Nutrition and breast cancer. Cancer Causes Control 1996;7:56– 8. [42] The Women’s Health Initiative Study Group. Design of the Women’s Health Initiative clinical trial and observational study. Control Clin Trials 1998;19:61– 109. [43] Wu AH, Ziegler RG, Nomura AMY, et al. Soy intake and risk of breast cancer in Asians and Asian Americans. Am J Clin Nutr 1998;68(Suppl.):1437S– 43S. [44] Garland M, Willett WC, Manson JE, Hunter DJ. Antioxidant micronutrients and breast cancer. J Am Coll Nutr 1993;12:400– 11. [45] Janowsky E, Lester G, Weinberg C, et al. The association between low levels of 1,25-dihydroxyvitamin D and breast cancer risk. Publ Health Nutr 1999;2:283– 91. [46] Schatzkin A, Longnecker MP. Alcohol and breast cancer. Where are we now and where do we go from here? Cancer 1994;74:1101– 10. [47] Smith-Warner SA, Spiegelman D, Yaun SS, et al. Alcohol and breast cancer in women: a pooled analysis of cohort studies. J Am Med Assoc 1998;279:535– 40. [48] Palmer J, Rosenberg L, Clarke E, et al. Breast cancer and cigarette smoking: a hypothesis. Am J Epidemiol 1991;134:1– 13. [49] Dunning AM, Healey CS, Pharoah PDP, Teare MD, Ponder BAJ, Easton DF. A systematic review of genetic polymorphisms and breast cancer risk. Cancer Epidemiol Biomark Prev 1999;8:843– 54. [50] Kelsey JL, Horn-Ross PL. Breast cancer: magnitude of the problem and descriptive epidemiology. Epidemiol Rev 1993;15:7– 16. [51] Egan KM, Newcomb PA, Longnecker MP, et al. Jewish religion and risk of breast cancer. Lancet 1996;347:1645– 6. [52] Gammon MD, John EM, Britton JA. Recreational and occupational physical activities and risk of breast cancer. J Natl Cancer Inst 1998;90:100– 17. [53] Laden F, Hunter DJ. Environmental risk factors and female breast cancer. Annu Rev Publ Health 1998;19:101– 23. [54] Gammon MD, Schoenberg JB, Britton JA, et al. Electric blanket use and breast cancer risk among younger women. Am J Epidemiol 1998;148:556– 63.

Discussion Den Tonkelaar: We have just finalized a cohort nested case-control prospective study on phyto-oestrogens and breast cancer; we measured genisteine and enterolactone in urine. In 100 cases and 300 controls we did not see any relation with genisteine or enterolactone.