Although our understanding of the etiology of breast cancer has improved, many well-known risk factors are not modifiable and present knowledge has proved insufficient to allow the disease to be overcome. Indeed, incidence and mortality among Japanese women have increased over the past three decades. Here, we review epidemiological evidence from our cohort and case–control studies among Japanese women in comparison with other published findings. Our studies confirm the important role of established factors derived primarily from Western populations, such as menstrual and reproductive factors, anthropometric factors, physical activity, and alcohol intake, in the development of breast cancer. In addition, we provide further evidence to better understand the role of traditional Japanese foods in the etiology of breast cancer. Our cohort study found that a higher intake of isoflavone and higher levels of plasma genistein, but not daidzein, were associated with a decreased risk of breast cancer. Our case–control studies reveal a dose–response pattern for these compounds; specifically, decreased risk as women move from “no” to “moderate” intake and leveling off thereafter. In addition, gene–environment interactions have been revealed in the effects of isoflavones. The evidence reviewed suggests that isoflavone has a protective effect against breast cancer in Asian populations. Conversely, our cohort study did not observe an inverse association between breast cancer risk and the intake of green tea and/or the plasma level of tea polyphenols, but we did find an association between increased risk and active and passive smoking. In conclusion, based on current knowledge, primary prevention according to individual lifestyle modification should focus on alcohol intake, weight control, physical activity, and tobacco smoking. (Cancer Sci 2011; 102: 1607–1614)
The incidence and mortality rates of breast cancer vary considerably across countries and regions, with a four to fivefold variation in incidence. Rates are highest in Europe and North America and lowest in Asia.(1) Despite Japan’s status as a low-risk country, the incidence and mortality of breast cancer among Japanese women have increased over the past three decades (Fig. 1),(2–5) with age-standardized incidence rates (per 100 000 population) of 17.0 in 1975 compared with 44.4 in 2005 according to the Monitoring of Cancer Incidence in Japan (MCIJ) project.(6) Breast cancer is the most common cancer diagnosis and the fourth-leading cause of cancer death among Japanese women. For example, in 2005 the MCIJ estimated that more than 47 583 Japanese women were diagnosed with breast cancer(6) and that 10 721 died of it.(7) In contrast, mortality rates in the UK and US have been in decline since the early 1990s, possibly attributable to improvements in screening practices and treatment effectiveness.(3,8) Moreover, incidence rates in the US and several other developed countries have decreased since 2002, due, in part, to the results of the Women’s Health Initiative’s randomized trial in July 2002, which saw a rapid fall in the use of hormone-replacement therapy (HRT).(9)
In addition to differences in the incidence and mortality rates of breast cancer between Asian and Western countries, age-specific incidence curves also differ: in Japan, the incidence of breast cancer increases until 50 years of age and decreases or plateaus thereafter, whereas in Western countries the incidence of breast cancer continues to increase after 50 years of age (Fig. 2).(2) This pattern may be explained by differences in the distribution of risk factors for postmenopausal breast cancer, particularly the low prevalence of obesity and HRT use in Japan.(10,11) Of note, the rapid rise in rate with increasing age slows somewhat around 50 years of age, near the time of menopause, which strongly suggests a role for reproductive hormones in the etiology of this disease.
Geographical distribution and secular trends in cancer incidence and mortality, as well as studies of migrants, highlight the relative importance of environmental and lifestyle influences in cancer etiology. Studies in migrants have shown increases in breast cancer incidence and mortality following migration from a lower- to a higher-risk country.(12–14) For example, Japanese immigrants in Los Angeles County had a clearly higher rate of breast cancer than Japanese in Japan.(12) Furthermore, the incidence of breast cancer in first-generation Japanese immigrants in São Paulo from 1968 to 1978 was higher than that among Japanese living in Japan, whereas mortality increased from 1979 to 2001 to a rate intermediate between that of Japanese living in Japan and Brazilians living in the state of São Paulo.(13,14) These findings strongly suggest that breast cancer risk is influenced by factors associated with the lifestyle or environment of the destination country.
Current knowledge of preventive or risk factors
Accumulating evidence obtained mainly from Western countries has established a relatively large number of preventative or risk factors for breast cancer (Table 1).(15–17) Many established risk factors are linked to ovarian hormones, and estrogens in particular, and prospective studies in postmenopausal women have shown a direct association between higher levels of estrogens and their androgen precursors and an increased risk of breast cancer.(18) One possible biological mechanism of the effect of ovarian hormones on risk is that both endogenous and exogenous hormones increase cellular proliferation in the breast, thereby increasing the likelihood of random genetic errors during cell division.(19)
Table 1. Established risk factors for breast cancer and corresponding results from the Japan Public Health Center-based Prospective (JPHC) study
Results from the JPHC study
HR (95% CI)
†All women (both premenopausal and postmenopausal women). ‡Postmenopausal women. §Participation in sports and physical activity in leisure time. Data are from Iwasaki et al.(23,26) and Suzuki et al.(34,37) BMI, body mass index; CI, confidence interval; HR, hazard ratio; NA, not available.
Endogenous and exogenous hormones
Endogenous estrogen levels
Oral contraceptive use
Hormone replacement therapy
Menstrual and reproductive factors
Age at menarche
≥16 years vs <14 years†
Age at menopause
≥54 years vs <48 years‡
Nulliparous vs parous†
Age at first birth
≥30 years vs <22 years†
History of breast feeding
Have history vs no history†
≥160 cm vs 148 cm‡
Body fatness (postmenopausal)
BMI ≥30 kg/m2vs BMI <19 kg/m2‡
Body fatness (premenopausal)
Diet and physical activity
Regular drinkers (>150 g ethanol/week) vs never drinkers†
≥3 days/week vs <3 days/month†§
History of benign breast disease
Mammographically dense breasts
Family history in first-degree relatives
Although our understanding of the etiology of breast cancer has improved, many well-known risk factors, such as menstrual and reproductive factors, are not modifiable for the purpose of reducing risk. In addition, only a few dietary factors have been causally related to the etiology of breast cancer, even thought diet is an environmental factor that may contribute to the population distribution of breast cancer risk (Table 1). Not surprisingly, present knowledge has proved insufficient to allow the disease to be overcome and the identification of other important etiological factors is thus required.
Rational for epidemiological studies among Japanese
Given that the population distribution of breast cancer risk is determined by variations in exposure, the substantial difference in lifestyle and environment between Japan and Western countries leads to the following general hypothesis: if a factor is characterized by high exposure in Japan (a low-risk country), but low exposure in those Western countries that are considered high-risk countries, it may be associated with a decreased risk of breast cancer. Good examples are traditional foods in Japan, such as soy foods and green tea. Similarly, a factor with low exposure in Japan but high exposure in Western countries may be associated with increased risk. We have used this hypothesis to conduct population-based cohort and hospital-based case–control studies among Japanese women with the goal of identifying risk factors and to further our understanding of the etiology of breast cancer, as detailed below.(20,21)
Briefly, the Japan Public Health Center-based Prospective (JPHC) study, which began in 1990 for Cohort I and in 1993 for Cohort II, enrolled 140 420 subjects (68 722 men and 71 698 women) living in municipalities supervised by 11 public health centers.(20) The study population consisted of registered Japanese inhabitants aged 40–59 years in Cohort I and 40–69 years in Cohort II. Approximately 55 000 women returned a self-administered questionnaire (response rate ∼83%) and approximately 25 000 women provided a blood sample (response rate ∼45%) in the baseline survey from 1990 to 1995. We conducted 5- and 10-year follow-up surveys to collect information regarding dietary habits, changes in lifestyle, and disease occurrence, as well as information regarding residential status, mortality, and incidence of cancer and cardiovascular diseases.
Regarding the multicenter, hospital-based case–control studies, these were conducted from 2001 to 2005 at four hospitals in Nagano Prefecture, Japan, and from 2001 to 2006 at eight hospitals in São Paulo, Brazil.(21) Cases were recruited from a consecutive series of female patients aged 20–74 years who were newly diagnosed with histologically confirmed invasive breast cancer. In the Nagano study, healthy controls were selected from medical checkup examinees who were confirmed to be cancer free, with one control matched for each case according to age and residential area. In the São Paulo study, controls were preferentially selected from cancer-free patients who visited the same hospital as the index cases with one control matched for age and ethnicity. Eventually, a total of 877 matched pairs participated (405 Japanese in Nagano, along with 83 Japanese Brazilians and 389 non-Japanese Brazilians in São Paulo).
Here, we review our findings in the JPHC study and case–control studies in Nagano and São Paulo in comparison with those from other Japanese and Western studies.
Epidemiological evidence from Japanese studies: established risk factors
Menstrual and reproductive factors. Menstrual and reproductive factors play an important role in the development of breast cancer. A meta-analysis of eight case–control studies in Japan showed that early age at menarche, nulliparity and low parity, and late age at first birth were associated with increased risk.(22) Similar to previous studies from both Western and Asian countries,(15–17) the JPHC study confirmed that early age at menarche, late age at menopause, nulliparity and low parity, and late age at first birth were associated with an increased risk of breast cancer (Table 1).(23) Although a 2007 report of the World Cancer Research Fund (WCRF) and American Institute for Cancer Research (AICR) concluded that lactation protects against breast cancer,(24) the JPHC study failed to replicate this association.(23) Furthermore, although a recent pooled analysis of 35 568 invasive breast cancer cases showed that nulliparity and late age at first birth were more closely associated with hormone receptor-positive than -negative tumors,(25) the JPHC study observed no significant difference in association by hormone receptor-defined breast cancer.(23)
Anthropometric factors. The 2007 WCRF/AICR report identified adult height as a convincing risk factor for postmenopausal breast cancer and a probable factor for premenopausal breast cancer.(24) The causal factor is unlikely to be tallness itself, but factors that promote linear growth in childhood, including energy intake and exposure levels to growth hormone and insulin-like growth factor.(24) Consistent with the WCRF/AICR report, the JPHC study observed an increased risk associated with greater height, primarily among postmenopausal women (Table 1).(26)
The 2007 WCRF/AICR report documented that the association between body fatness and breast cancer risk depends on menopausal status: although greater body fatness probably protects against premenopausal breast cancer, convincing evidence suggests that it is a cause of postmenopausal breast cancer.(24) In addition, adult weight gain is a probable cause of postmenopausal breast cancer. The mechanism of this association likely relates to levels of circulating estrogen: specifically, a decrease in levels due to an increased frequency of anovulatory cycles in premenopausal women and an increase in levels due to both an increase in estrogen production by aromatase in adipose tissue and a decrease in circulating level of sex hormone-binding globulin (SHBG) in postmenopausal women.(27)
In the JPHC study, we found a positive association between body mass index (BMI) and breast cancer risk, with the association being stronger in post- than premenopausal women (Table 1).(26) We also found an association between an increase in BMI from age 20 years to recent age with increased risk among postmenopausal women.(28) These findings generally agree with those of studies in Japan and other Asian countries.(29,30) A recent meta-analysis of cohort studies showed that risk was increased by 16% and 31% per 5 kg/m2 increment of BMI in pre- and postmenopausal Asian women, respectively, but decreased by 9% in premenopausal and increased by 15% in postmenopausal North American women.(30) The lack of an inverse association among premenopausal women may be due to the lower prevalence of overweight women in Asian countries, with few who are sufficiently overweight to likely develop anovulation. Conversely, risk reduction due to greater body fatness in early adulthood appears to continue into the postmenopausal years, which may explain the stronger association among postmenopausal Asian than North American women. In addition, a recent meta-analysis showed a 10% decrease in risk per 5 kg/m2 increment of BMI among premenopausal women and a 33% increase among postmenopausal women for estrogen and progesterone receptor-positive (ER+PR+) tumors, although no association was seen for estrogen receptor-positive and progesterone receptor-negative (ER+PR−) or estrogen and progesterone receptor-negative (ER–PR−) tumors.(31) In the JPHC study, BMI was more strongly associated with estrogen receptor-positive (ER+) than -negative (ER−) tumors in postmenopausal women. These findings may support the involvement of an ER-mediated estrogen-dependent mechanism.
Physical activity. The 2007 WCRF/AICR report concluded that the evidence that any type of physical activity, including occupational, household, transport, and recreational activity, protects against breast cancer is limited-suggestive for premenopausal and probable for postmenopausal breast cancer.(24) A meta-analysis showed a 6% decrease in risk for each additional hour of physical activity per week.(32) The proposed mechanisms behind this association include the beneficial effect of physical activity on body fatness, effects on endogenous sex hormone levels, and possible improvement of immune function.(33)
In the JPHC study, we observed an inverse association between leisure time physical activity and breast cancer risk (Table 1).(34) Compared with women who participated in sports and physical activity on <3 days/month, adjusted hazard ratio (HR) and 95% confidence intervals (CI) for women who participated in sports on >3 days/week was 0.73 (0.54–1.00; Ptrend = 0.037) for overall breast cancer and 0.43 (0.19–1.00; Ptrend = 0.022) for ER+PR+ tumors. Conversely, we did not observe an inverse association between daily total physical activity and risk of overall breast cancer, but did see an inverse association for ER+PR+ tumors. In addition, we also investigated associations between age- and intensity-specific leisure time physical activity and the risk of hormone receptor-defined breast cancer in the case–control study in Nagano.(35) Strenuous, but not moderate, physical activity at age 12 years was inversely associated with breast cancer risk regardless of menopausal status and hormone receptor-defined breast cancer. Among postmenopausal women, moderate physical activity in the previous 5 years was somewhat more closely associated with ER+PR+ than ER+PR− and ER−PR− tumors. Our findings generally agree with those of the WCRF/AICR report and other Japanese studies.(24,36) Moreover, our findings regarding hormone receptor-defined breast cancer may support the involvement of an ER-mediated estrogen mechanism.
Alcohol intake. We found a significant positive association between alcohol intake and the risk of breast cancer in the JPHC study (Table 1).(37) An increase in consumption of 10 g ethanol/day (continuous) was associated with a 6% (95% CI 1–13; Ptrend = 0.047) increase in the risk of breast cancer. Our findings generally agree with those from the WCRF/AICR report.(24) A meta-analysis of cohort studies reported a 10% increase in risk per 10 g increment of ethanol/day.(24) However, Nagata et al. concluded that epidemiological evidence from Japanese populations remains insufficient, given that a systematic review revealed that only three of three cohort and eight case–control studies observed a positive association.(38)
Several biological mechanisms for this association have been proposed, including an increase in circulating hormone levels, a direct carcinogenic effect of alcohol metabolites (e.g. acetaldehyde, a known mutagen), and an antagonistic effect on folate absorption and metabolism.(39) In the JPHC study, we found positive associations for both ER+PR+ and ER+PR− tumors, but not for ER−PR− tumors, although the associations failed to reach statistical significance. A recent meta-analysis showed that the relative risk (RR) and 95% CI per 10 g increment of ethanol/day was 1.12 (1.08–1.15) for all ER+ tumors, 1.07 (1.00–1.14) for all ER− tumors, and 1.11 (1.07–1.14) for ER+PR+, 1.15 (1.02–1.30) for ER+PR−, and 1.04 (0.98–1.09) for ER−PR− tumors.(40) These findings suggest that the biological mechanism involves both an ER-mediated estrogen-dependent and hormone-independent mechanism.
Notable epidemiological evidence from Japanese studies
Body weight at age 20 years. A number of epidemiological studies have shown that greater body fatness during childhood and adolescence is associated with a decreased risk of breast cancer.(41–43) The proposed biological mechanism behind this risk reduction is that obese women tend to have an increased frequency of menstrual irregularities and anovulatory cycles, which reduces their lifetime number of ovulations and alters their circulating hormone levels.(27) To date, most studies have been conducted in Western countries, where the prevalence of obesity is high, and little is known about whether greater body fatness during childhood and adolescence is associated with a decreased risk of breast cancer among the lean population.
In the JPHC study, we found a significant inverse association between BMI at age 20 years and the risk of breast cancer. This inverse association was not modified by menopausal status or recent BMI level. Adjusted HR for each 5 unit increment was 0.75 (95% CI 0.61–92).(28) Similarly, the Miyagi Cohort Study also observed a decreased risk associated with higher BMI at age 20 years.(44) These findings from a lean population generally agree with those from Western countries. Interestingly, few women are likely to be sufficiently overweight to cause anovulation in Japan. Moreover, the Nurses’ Health Study II reported that the observed inverse association of BMI in early adulthood with risk was not eliminated after adjustment for ovulatory disorders.(41) Therefore, our findings from Japan imply the presence of other biological mechanisms apart from anovulation.
Soy foods and isoflavone. Soy foods, which are rich in isoflavones, are habitually consumed by Asian populations in large amounts. Isoflavones, of which genistein and daidzein are the major examples, are classified as phytoestrogens, which are plant-derived non-steroidal compounds with estrogen-like biological properties. A high intake of isoflavones has been hypothesized to contribute to the lower incidence of breast cancer in Asian compared with Western countries.(45)
In the JPHC study, we observed an approximate 50% decrease in breast cancer risk associated with higher isoflavone intake, as assessed by a food frequency questionnaire.(46) Moreover, a nested case–control study within the JPHC study revealed a decrease in risk associated with a higher level of plasma genistein, but not plasma daidzein (Fig. 3).(47) Although accumulating evidence suggests that risk is reduced with higher isoflavone intake,(48,49) there is little available evidence for a dose–response relationship. In the case–control studies in Nagano and São Paulo, we evaluated the dose–response relationship using the three populations combined, because the respective amount of and variation in isoflavone intake is high and large for Japanese, intermediate and relatively large for Japanese Brazilians, and low and small for non-Japanese Brazilians.(21) We found that breast cancer risk decreased linearly from “no” to “moderate” isoflavone intake (20–30 mg/day) and thereafter leveled off (Fig. 4), suggesting that isoflavones have a risk-reducing rather than risk-enhancing effect on breast cancer within the range achievable from dietary intake alone.
Several biological mechanisms have been proposed to explain how isoflavones may reduce the risk of breast cancer. Isoflavones and human estrogen share similar chemical structures; given the consequent binding affinity of isoflavones to estrogen receptors, they may act as estrogen agonists and antagonists that compete for estradiol at the receptor complex.(50,51) Isoflavones may also influence risk by altering the biosynthesis, metabolism, and bioavailability of endogenous hormones.(52,53) In this regard, isoflavones have been shown to inhibit aromatase(52) and 17β-hydroxysteroid dehydrogenase Type I (17β-HSD1),(52) as well as to increase the synthesis of SHBG.(53) Considering these mechanisms, we tested the hypothesis that polymorphisms in estrogen receptor genes and genes related to the biosynthesis, metabolism, and bioavailability of endogenous hormones may modify the association between isoflavone intake and breast cancer risk in the case–control studies in Nagano and São Paulo.(54,55) The results showed several suggestive interactions between isoflavone intake and polymorphisms of estrogen receptor beta (ESR2), 17β-HSD1, and SHBG: an inverse association between intake and risk in women with the GG genotype of the rs4986938 polymorphism in ESR2 among postmenopausal Japanese, Japanese Brazilians, and non-Japanese Brazilians (Fig. 5);(54) an inverse association in women with at least one A allele of the rs605059 polymorphism in 17β-HSD1 among the three populations;(55) and an inverse association in women with the GG allele of the rs6259 polymorphism in SHBG among Japanese populations and women with at least one A allele among non-Japanese Brazilians.(55) Our findings support the idea that isoflavones may reduce the risk of breast cancer via mechanisms that involve estrogen receptors or the biosynthesis, metabolism, and bioavailability of endogenous hormones.
A recent meta-analysis observed risk reduction with higher isoflavone intake among Asian, but not Western, populations.(49) Overall, our studies suggest that isoflavone intake has a protective effect against breast cancer. Because we found a decreased risk not only in Japanese, but also Japanese Brazilians and non-Japanese Brazilians, our findings are somewhat inconsistent with those of the meta-analysis. This heterogeneity of findings across populations and studies warrants careful consideration. In this regard, Nagata noted that the association between soy isoflavone intake and the risk of breast cancer may be variously modified by the amount of soy isoflavones consumed, the form and food source of the isoflavones, the timing of isoflavone exposure, the estrogen receptor status of tumors, the equol-producer status, and the hormonal profile of individuals.(56)
Green tea. Although rarely consumed in Europe and North America, where black tea is the common tea beverage, green tea is one of the most popular beverages in Japan and China. Green tea has a higher catechin content than black tea, which may contribute to its protective effects against cancer via the strong antioxidant activity of catechin, its inhibition of cell proliferation and angiogenesis, induction of apoptosis, and antiestrogenic properties.(57,58)
In the JPHC study, we found no significant inverse association between green intake and the risk of breast cancer.(59) Compared with women who drank less than one cup of Sencha or Bancha/Genmaicha per week, the adjusted HR for those who drank 10 or more cups per day was 1.02 (95% CI 0.55–1.89; Ptrend = 0.48) for Sencha and 0.86 (0.34–2.17; Ptrend = 0.66) for Bancha/Genmaicha. One noteworthy strength of this study over previous studies is its remarkably wide variation in green tea intake, from women who drank less than one cup per week to those who drank 10 or more cups per day.
Tea polyphenol content in green tea varies according to preparation, the type and amount of green tea leaves, the frequency of renewing the tea batch in the pot, water temperature, and brewing time, among others. To reduce misclassification due to these factors, we conducted a nested case–control study within the JPHC study and measured plasma levels of (−)-epigallocatechin (EGC), (−)-epicatechin (EC), (−)-epigallocatechin-3-gallate (EGCG), and (−)-epicatechin-3-gallate (ECG).(60) We found no significant association between plasma tea polyphenol levels and breast cancer risk. Adjusted odds ratios (OR) for the highest versus lowest group were 0.90 (95% CI 0.42–1.96; Ptrend = 0.98) for EGC, 0.95 (95% CI 0.43–2.08; Ptrend = 0.86) for EC, 1.21 (95% CI 0.52–2.80; Ptrend = 0.53) for EGCG, and 1.75 (95% CI 0.81–3.78; Ptrend = 0.15) for ECG.
To our knowledge, four cohort and three case–control studies have been published on the association between green tea intake and breast cancer, but findings have been inconsistent.(61–67) Our findings generally agree with those of three of the cohort studies, including two Japanese cohorts, which found no association between green tea intake and risk,(64–66) but contradict those of the three case–control studies, which showed an inverse association between green tea intake and risk.(61–63) Possible explanations for these apparent discrepancies in results include the influence of recall and selection bias stemming from the case–control design; differences in the type of tea and drinking methods; and possible effect modification by dietary and genetic factors.(59,63,66) Moreover, among studies investigating the association between circulating tea polyphenol levels and breast cancer risk using prediagnostic biological specimens, the Shanghai Women’s Health Study found no dose–response relationship between urinary levels of tea polyphenols and their metabolites and the risk of breast cancer,(68) which is similar to the results of our JPHC study.
Smoking and passive smoking. The JPHC study found that both active and passive smoking were associated with an increased risk of breast cancer among premenopausal women.(69) When the reference group was defined as never-active smokers without passive smoking, adjusted HR (95% CI) for ever-smokers were 3.9 (1.5–9.9) and 1.1 (0.5–2.5) in pre- and postmenopausal women, respectively. In never-active smokers, the adjusted HR (95% CI) for passive smoking was 2.6 (1.3–5.2) in premenopausal women and 0.6 (0.4–1.0) in postmenopausal women. Subsequently, Nagata et al.(70) concluded that tobacco smoking possibly increases the risk of breast cancer in the Japanese population, considering that a systematic review of evidence showed a positive association in five of three cohort and eight case–control studies in Japan.
In 2004, the International Agency for Research on Cancer (IARC) endorsed the “lack of carcinogenicity of tobacco smoking in humans for cancers of the female breast”.(71) However, large cohort studies published since 2002 have observed an increased risk associated with a long duration and/or high number of pack-years of smoking.(72) Moreover, a meta-analysis found a significant interaction between smoking, N-acetyltransferase 2 (NAT2) genotype, and risk of breast cancer: higher pack-years were associated with an increased risk among women with the NAT2 slow genotype, but not among rapid acetylators.(73) Recent reappraisals have therefore suggested an increased risk of breast cancer and the IARC concluded that there is limited evidence that tobacco smoking causes breast cancer.(74) With regard to passive smoking, a meta-analysis published in 2007 showed that this was associated with a 60–70% increase in breast cancer risk among younger, primarily premenopausal women who had never smoked.(75) However, a more recent meta-analysis found an increased risk associated with passive smoking based on case–control, but not cohort, studies.(76)
Evidence establishing menstrual and reproductive factors, anthropometric factors, physical activity, and alcohol intake as risk factors for breast cancer was derived primarily from Western countries, but only a few dietary factors have been causally related to this disease.(15–17,24) Our studies among Japanese women have confirmed that these previously established factors play an important role in the development of breast cancer.(23,26,34,37) In addition, we have provided further evidence of the role of traditional Japanese foods in the etiology of breast cancer.(21,46,47,54,55,59,60) In particular, our studies of isoflavones and breast cancer have clarified a dose–response relationship and gene–environment interactions.(21,54,55) Given the evidence reviewed above, we suggest that isoflavones exert a protective effect against breast cancer in Asian populations. Finally, current knowledge of protective and risk factors for breast cancer suggest that primary prevention by lifestyle modification in individuals should focus on alcohol intake, weight control, physical activity, and tobacco smoking.
The authors sincerely thank the members and coworkers of the Japan Public Health Center-based Prospective Study Group, Nagano Breast Cancer Study Group, and São Paulo-Japan Breast Cancer Study Group. The authors’ work reported herein was supported by Management Expenses Grants from the Government to the National Cancer Center, Grant-in-Aid for the Third-Term Comprehensive 10-Year Strategy for Cancer Control from the Ministry of Health, Labor and Welfare of Japan, and Grants-in-Aid for Scientific Research on Innovative Areas (221S0001) and for Young Scientists (B) (22700934) from the Ministry of Education, Culture, Sports, Science, and Technology of Japan, the Japan Society for the Promotion of Science, and the Foundation for Promotion of Cancer Research in Japan.