Hormonal and Body-Size Factors in Relation to Breast Cancer Risk: A Prospective Study of 11,889 Women in a Low-Incidence Area

Transcription

1 Hormonal and Body-Size Factors in Relation to Breast Cancer Risk: A Prospective Study of 11,889 Women in a Low-Incidence Area MEI-HSUAN WU, MPH, YU-CHING CHOU, MPH, JYH-CHERNG YU, MD, CHENG-PUNG YU, PHD, CHO-CHIEH WU, MPH, CHI-MING CHU, PHD, TSAN YANG, MPH, CHING-HUANG LAI, PHD, CHANG-YAO HSIEH, MD, SAN-LIN YOU, PHD, CHIEN-JEN CHEN, SCD, AND CHIEN-AN SUN, SCD PURPOSE: This prospective cohort study of 11,889 women was conducted to determine significant factors associated with the risk of breast cancer among Chinese women in Taiwan, a low-incidence area. METHODS: In-person interviews were completed for subjects to solicit information on hormonal factors. Measurements of height, weight, and waist-and- hip circumferences were performed by well-trained assistants using standardized techniques. Cox proportional hazards models were employed to estimate relative risks (RRs) and 95% confidence intervals (CIs). RESULTS: During an average follow-up time of 10.3 years (range: 1 to 11 years) with an accumulation of 134,063 person-years, 104 incident breast cancer cases were identified through data linkage with national cancer registry profile. There was a significant elevation in breast-cancer risk with increasing duration of the interval between age at menarche and age at first full-term pregnancy (FFTP). Additionally, central adiposity reflected by hip circumference was a significant predictor of breast cancer in this Chinese female population. CONCLUSIONS: The findings of this study indicated common mechanisms responsible for the higher incidence of breast cancer in Western populations may also explain the risk of breast cancer development in Taiwan, a low-incidence area. Ann Epidemiol 2006;16: Ó 2006 Elsevier Inc. All rights reserved. KEY WORDS: Body Size, Breast Cancer, Hormonal Risk Factors, Prospective Study. INTRODUCTION Breast cancer is the most frequent cancer among women worldwide (1). Chinese women historically have a lower risk of breast cancer compared to their counterparts in Western societies. During the past several decades, age-adjusted breast cancer incidence has been remarkably increasing in Taiwan, rising from 5.94 per 100,000 in 1979 to per 100,000 in 2000, a six-fold increase. It is now the second From the School of Public Health, National Defense Medical Center, Taipei 114, Taiwan, Republic of China (M.H.W, C.C.W, C.M.C., T.Y., C.H.L., C.A.S.); the Graduate Institute of Medical Sciences, National Defense Medical Center, Taipei 114, Taiwan, Republic of China (Y.C.C.); the Department of Surgery, Tri-Service General Hospital, Taipei 114, Taiwan, Republic of China (J.C.Y.); The Department of Pathology, Tri-Service General Hospital, Taipei 114, Taiwan, Republic of China (C.P.Y.); the Department of Oncology, National Taiwan University Hospital, Taipei 100, Taiwan, Republic of China (C.Y.H.); and the Graduate Institute of Epidemiology, College of Public Health, National Taiwan University, Taipei 100, Taiwan, Republic of China (S.L.Y., C.J.C.). Address correspondence to: Chien-An Sun, Sc.D., National Defense Medical Center, School of Public Health, No.161, Section 6 Min-Chuan East Road, Taipei 114, Taiwan, R.O.C. Fax: C sunca@mail.ndmctsgh.edu.tw. This work was supported by a grant from the Department of Health, Executive Yuan, Taiwan, Republic of China. Received November 11, 2004; accepted February 7, highest incidence of cancer and the fourth leading cause of cancer death among females in Taiwan (2). Although the incidence rates of breast cancer in Taiwan are much lower than those in Western countries, the age of onset in Taiwan is much younger than is the case with their counterparts in Western societies (3, 4). This pattern of younger age of onset observed in Taiwan may be due to the cohort effect, as observed in Japan (5). While numerous epidemiological follow-up studies have been conducted in Western countries to assess the relationships of hormonal and body-size factors with breast cancer risk, there have been limited prospective follow-up studies of Chinese population. Thus, the question arises as to whether breast cancer in Chinese women is influenced by factors established for high-risk populations. The present long-term follow-up study was undertaken in Taiwan, an area of low incidence but of early-onset of breast cancer, to investigate this subject. METHODS Study Population and Recruitment A community-based cancer-screening program was carried out in Taiwan in July The recruitment and follow up Ó 2006 Elsevier Inc. All rights reserved /06/$ see front matter 360 Park Avenue South, New York, NY doi: /j.annepidem

2 224 Wu et al. AEP Vol. 16, No. 3 HORMONE, BODY SIZE, AND BREAST CANCER March 2006: Selected Abbreviations and Acronyms BMI Z body mass index CI Z confidence interval FFTP Z first full-term pregnancy HRT Z hormonal replacement therapy PAR Z population attributable risk RR Z relative risk SD Z standard deviation WHR Z waist-to-hip ratio of this cancer-screening cohort have been described previously (6, 7). Briefly, the target population included individuals born between 1926 and 1960 and resided in seven townships in Taiwan including Sanchi, Chutung, Potzu, and Kaohsu, located on Taiwan Island; and Makung, Huhsi, and Paihsa, located on Penghu Islets. There were 42,263 eligible females who were invited by letter to participate, and a total of 11,917 female adults were enrolled; approximately 25% agreed to participate. Nonsmokers, the elderly, and those with higher levels of education showed higher rates of response (8). At the time of enrollment, subjects were screened for cancer by interviews and data linkage with the National Cancer Registry System in Taiwan. Individuals were considered to be free of cancer on the basis of either the absence of personal history of cancer or the absence of linkage with the National Cancer Registry. In this study, we have defined prevalent breast cancer cases as tumors occurred prior to the date of recruitment or those detected within the first year of enrollment. Accordingly, incident cases of women with invasive breast cancer were defined as women for whom invasive breast cancer was newly diagnosed between one year after entry into the cohort and the end of follow up on December 31, Among enrolled subjects, there were 4 women who died before being contacted, 17 had breast cancer diagnosed before the date of recruitment, 4 had breast cancer diagnosed within 1 year after enrollment, and 3 had missing data on date of recruitment. Consequently, a total of 11,889 females were included in the present cohort study. At baseline recruitment, well-trained research assistants personally interviewed female participants to solicit information on sociodemographic characteristics, history of cigarette smoking and alcohol consumption, personal and family history of cancer, age at menarche, parity, age at first full-term pregnancy (FFTP), menopausal status, and age at menopause. All participants were also measured by welltrained assistants using standardized techniques for current weight, circumferences of the waist and hip, and standing height. Subsequently, body mass index (BMI), weight (in kg)/height (in m 2 ) and waist-to-hip ratio (WHR) were calculated. All subjects gave informed consent, and this community-based cancer-screening program was supported and approved by the Department of Health, Executive Yuan. Follow Up for Vital Status and Breast Cancer Incidence In this cohort study, computer files from the National Cancer Registry System in Taiwan served as a productive and cost-effective means of tracing cohort members in terms of breast cancer incidence (9). In Taiwan, if a patient is diagnosed with or treated for cancer in a hospital with 50 or more beds, that hospital has a legal obligation to report the cases to the National Cancer Registry. The data are evaluated on an annual basis for completeness and accuracy. Case ascertainment by the registry through the hospital system is estimated to be 94% complete (10). With respect to the diagnosis of breast cancer, microscopic confirmation has been accomplished in 98% of reported breast cancer cases (10). In addition, it is mandatory to register any vital event, including migration and death, with the local housing office in Taiwan. All deceased residents in Taiwan are included in the computerized national death certificate data file. Thus, even if subjects emigrated to other areas in Taiwan, their vital status was completely followed up through data linkage with the household registration and death certification systems. Therefore, we linked data with information obtained from the National Death Certification System to trace vital status among subjects. Statistical Analysis The period of observation used in calculating incidence rate began from the date of enrollment to the date of one of the following events, listed in descending order of priority: the date of diagnosis of breast cancer, the date of death, or the date of the end of follow up (December 31, 2002). Subjects who remained unaffected by breast cancer before death during the follow-up period or by the end of follow up were considered censored at the date of death or the last date of follow up. The association between risk determinants, including hormonal and body-size factors, and the subsequent development of breast cancer were estimated using the Cox proportional hazards model to calculate relative risks (RRs) and corresponding 95% confidence intervals (CIs). In the current study, cigarette smoking, alcohol drinking, and the family history of breast cancer were not considered in the analyses of risk factors for breast cancer because the prevalences of cigarette smoking (1.0%), alcohol drinking (0.6%), and family history of breast cancer (0.5%) were remarkably lower among our female subjects. In the present study, an indicator category for the critical estrogen exposure period associated with breast carcinogenesis was defined by the duration between age at menarche and age at FFTP (11). Of note, the analyses of the effects of FFTP and the duration between age at menarche and age at FFTP on breast-cancer risk were restricted to parous women because there were limited numbers of nulliparous women (n Z 144 with a total of 1480 person-years of follow-up) in

3 AEP Vol. 16, No. 3 March 2006: Wu et al. HORMONE, BODY SIZE, AND BREAST CANCER 225 our study population. The exclusion of nulliparous women, however, did not materially change the main findings (data not shown). Given the facts that only a few subjects were nulliparous women in this study and multiparity has little effect on breast cancer risk among parous women (12), the effect of parity on the risk of developing breast cancer was not evaluated in this study. The cumulative incidence of breast cancer by year of follow up was estimated for subjects with various time periods of estrogen exposures by means of the Nelson-Aalen method, a nonparametric method of calculating cumulative hazards (the Kaplan-Keier estimator may be used when the interest is in survival function [13,14]). In this study, menopausal status was defined as last menstruation after one year free of menstrual cycle. Because there was a lower proportion of artificial menopause caused by surgery (2.2%) among studied postmenopausal women, no attempt was made to distinguish between women with artificial and those with natural menopause in this study. To quantify the risk of breast cancer attributable to a specific factor in the study population, population attributable risk (PAR) fraction was calculated following Madigan and colleagues. (15). We performed all analyses using either Statistical Analysis System v. 8.0 (SAS Institute Inc., NC) or STATA (STATA Corporation, College Station, Texas) statistical software. RESULTS The mean age (G standard deviation [SD]) at study entry was 46.6 (G 9.8) years for the study population. The majority of subjects (73%) were less educated (they attended elementary school only). Overall, the mean duration (SD) of follow up was 10.3 (1.2) years (range: 1 11 years). In 134,063 person-years of follow-up, a total of 104 incident breast cancer occurred among subjects, resulting in an incidence rate of 77.6 per 100,000 personyears (95% CI Z per 100,000 person-years). Table 1 presents RRs and 95% CIs for hormonal risk factors in association with breast cancer risk. There was a significantly inverse association between age at menarche and breast cancer risk (p value for trend test Z ). Compared to women with menarchal age equal to or more than 17 years, women with menarchal age equal to or less than 14 years had a 2- fold increased risk of breast cancer (95% CI Z ). With respect to age at FFTP among parous women, older age at FFTP was associated with a significantly elevated risk of breast cancer (p value for trend test Z ). Compared to subjects whose FFTP occurred before or at age 22, those whose FFTP occurred at or after age 25 had a 4-fold increased risk of breast cancer (95% CI Z ). TABLE 1. Incidence rate and relative risk (RR) of breast cancer in relation to hormonal risk factors, Taiwan, No. of No. of Incidence rate Relative risk Variable a,b subjects cases (/100,000/ year) Crude Adjusted e (95% CI d ) Trend test p value Age at menarche (years) R ( ) % ( ) Age at FFTP c,d (years) % ( ) R ( ) Duration between age at menarche and age at FFTP (years) c % ! ( ) R ( ) Menopausal status Premenopausal Postmenopausal ( ) % 50 years ( ) 0.60 O 50 years ( ) Use of oral contraceptives Never Ever ( ) a Variables except for menopausal status shown in the table were categorized based on the tertile distribution among study subjects. b Number of subjects for data analysis of variables studied does not correspond to total number of subjects because of missing value. c Among parous women. d CI Z confidence interval; FFTP Z first full-term pregnancy. e The relative risks are adjusted for age at enrollment.

4 226 Wu et al. AEP Vol. 16, No. 3 HORMONE, BODY SIZE, AND BREAST CANCER March 2006: Accordingly, duration of the interval between age at menarche and age at FFTP was significantly associated with an enhanced risk of breast cancer (p value for trend test Z ). Compared to women with a duration less than 7 years, women with a duration equal to or more than 10 years had a 4-fold increased risk of breast cancer (95% CI Z ). Indeed, our data was in agreement with the argument made by Colditz and Frazier that the time between menarche and FFTP was the time when the breast tissue was most vulnerable to mutagenesis (11). We further analyzed cumulative risk of breast cancer in relation to the time interval between menarche and FFTP, and the results are shown in Figure 1. Women with the time interval equal to or more than 10 years had a much higher cumulative incidence of breast cancer (1.65%) than did those with the time interval between 7 and 9 years (0.77%), and they had an even higher cumulative risk than did those with a time interval of less than 7 years (0.45%) (p Z for both comparisons). Cumulative incidence Years 1: 6 years 2: 7 9 years 3: 10 years FIGURE 1. Cumulative incidence of breast cancer during follow up among 11,889 Chinese women in Taiwan according to the duration between age at menarche and age at first full-term pregnancy. The cumulative incidence was estimated with the use of the Nelson-Aalen method. Menopausal status and use of oral contraceptives, however, were not significantly related to the risk of breast cancer (Table 1) TABLE 2. Incidence rate and relative risk of breast cancer in relation to body-size factors, Taiwan, Relative Risk Body-size factor a No. of subjects No. of cases Incidence rate (/100,000/ year) Crude Adjusted b (95% CI c ) Trend test p value Height (cm) % ( ) ( ) R ( ) Weight (kg) % ( ) ( ) O ( ) Body mass index (kg/m 2 )! ( ) ( ) O ( ) Waist circumference (cm)! ( ) ( ) O ( ) Hip circumference (cm)! ( ) ( ) O ( ) Waist-to-hip ratio! ( ) ( ) O ( ) a Body-size factors were categorized based on the quartile distribution among study subjects. b The relative risks are adjusted for age at enrollment and waist-to-hip ratio for analyses of height, weight, and body mass index, and adjusted for age at enrollment and body mass index for analyses of waist and hip circumferences and waist-to-hip ratio. c CI Z confidence interval.

5 AEP Vol. 16, No. 3 March 2006: Wu et al. HORMONE, BODY SIZE, AND BREAST CANCER 227 Next, we analyzed breast cancer risk in relation to bodysize factors (Table 2). After controlling for WHR, Higher weight and BMI at enrollment were associated with a higher risk of breast cancer (p value for trend test was and , respectively). Compared to women whose weight % 50 kg, the adjusted RR for those whose weight O 63 kg was 2.0 (95% CI Z ). In addition, compared with BMI! 21.6 kg/m 2, the adjusted RR for BMI O 26.2 kg/m 2 was 1.9 (95% CI Z ). Furthermore, the relationships between weight and BMI and the risk of breast cancer were consistent across the strata defined by menopausal status (data not shown). However, there was no significant association of height at enrollment with subsequent development of breast cancer after adjustment for WHR. More interestingly, we observed that hip circumference at enrollment was significantly associated with an increased risk of breast cancer after adjustment for BMI (p value for trend test Z ). Women whose hip circumference more than 100 cm had a 2.9-fold (95% CI Z ) increased breast cancer risk as compared with those whose hip circumference less than 90 cm. However, neither waist circumference nor WHR was significantly related to breast cancer risk when BMI was accounted for. Finally, the results of the Cox proportional hazards regression analysis, which simultaneously assesses multiple risk factors for breast cancer, are shown in Table 3. Age at FFTP played the most important role in the determination of breast cancer risk. The multivariate-adjusted RR was as high as 4.2 (95% CI Z ) for women whose FFTP occurred at or after age 25, and 2.8 (95% CI Z ) for those whose hip circumference was more than 100 cm. The effects of BMI and waist circumference, however, were not found to be significantly associated with breast cancer risk. The PAR percent estimates for those individual risk factors are also presented in Table 3. The reproductive variable age at FFTP accounted for the greatest proportion of breast cancer cases (57.1%), and the anthropometric variable hip circumference contributed the second highest PAR percent (47.0%), while age at menarche contributed 30.2%. DISCUSSION Similar to most epidemiological studies conducted in Western countries (16 21), this large prospective cohort study of 11,889 Chinese women, conducted in a low-risk area, found that early age at menarche and later age at FFTP were related to increased risk of breast cancer. The most salient finding in this study was the consistently elevated risk of breast cancer in relation to the time interval between menarche and FFTP. This time period could be prolonged as a result of earlier menarche and/or delayed FFTP. This observation was in accordance with previous findings that women with an earlier menarche tend to have higher lifetime cumulative estrogen levels than do those with a later menarche (22), and the hormonal milieu of pregnancy at a younger age increases cell proliferation, including the proliferation of the already-initiated cells, yet at the same time inducing terminal differentiation of the mammary gland, rendering it less susceptible to carcinogen stimuli (23). In fact, Colditz and Frazier have argued that the time period between menarche and FFTP had the highest rate of breast tissue aging, and this time interval was the time when the breast tissue was most vulnerable to mutagenesis (11). Many previous studies have demonstrated that body size is an important risk factor for breast cancer (18 21, 24 26). Of particular note, it is generally accepted that the relationship between generalized obesity reflected by BMI and breast cancer is modified by menopausal status, with higher BMI associated with decreased risk for premenopausal women and increased risk for postmenopausal women (27). Although Asian women are generally shorter and lighter than women in Western countries (28), higher BMI was moderately associated with an increased risk of breast cancer in this Chinese female population, and a similar association was observed when strata defined by menopausal status were examined. It is noteworthy that, after adjustment for BMI, hip circumference was significantly associated with elevated breast cancer risk in our study women, which was also observed in Chinese women in Shanghai (29). Indeed, other investigators have reported the importance of central body obesity as a risk determinant of breast cancer in Chinese women (30). Abdominal adiposity might contribute to the risk of breast cancer beyond that attributable to generalized obesity alone. The highest aromatase activity associated with estrogen metabolism was found in the hip (31). In addition, waist and hip circumferences have been found to be associated with higher levels of androgens, insulin, and reduced levels of sex-hormone binding globulin (27, 32), factors that have been previously linked with breast cancer risk (33, 34). Overall, these observations may suggest a different role of body/build measurements in breast cancer etiology between populations, and this may merit future exploration. Our study presents strengths and limitations. The strengths include those inherent to a prospective cohort study, such as the fact that all hormonal factors and anthropometric measures were directly assessed at baseline, avoiding recall bias and inverse causality. On the other hand, the following possible limitations warrant consideration: It cannot be ruled out that some confounding bias is still present, due to the lack of inclusion of other potential risk factors, such as dietary intake. In addition, we did not collect detailed information on anthropometric measures during different periods of life, which limits our interpretation of

6 228 Wu et al. AEP Vol. 16, No. 3 HORMONE, BODY SIZE, AND BREAST CANCER March 2006: TABLE 3. Cox proportional hazards regression analysis of multiple risk factors of breast cancer Variable Multivariateadjusted relative risk (95% CI d ) Population attributable risk percent Age at menarche (years) a R % Age at FFTP (years) a,b % R Body mass index (kg/m 2 ) c! O Waist circumference (cm) c! O Hip circumference (cm) c! O a Variables were categorized based on the tertile distribution among study subjects. b Among parous women. c Variables were categorized based on the quartile distribution among study subjects. d CI Z confidence interval. the role of the critical timing of body-size factors in determining breast cancer risk. Furthermore, data on hormonal replacement therapy (HRT) use did not obtain from study subjects in this study. Use of HRT can obscure the effect of adiposity on breast cancer risk by influencing estrogen levels (35). Accordingly, the effect of HRT use on the associations of body measures and breast cancer risk cannot be assessed in the present study. Overall, it has been estimated that about 41% of breast cancer cases in the United States were attributable to wellestablished risk factors, including later age at FFTP and nulliparity (15). Comparatively, later age at FFTP accounted for the greatest proportion and hip circumference contributed the second highest fraction of breast cancer cases in this Chinese population with multiparity. The significant associations of age at menarche, age at FFTP, and hip circumference with the risk of breast carcinoma revealed in the current study further supports the unifying hypothesis that underlying endocrine mechanisms play a critical role in the genesis of breast cancer. Breast cancer remains highly hormone dependent in low-risk Chinese women, as it also does in high-risk Western populations. The assimilation of westernized lifestyle and dietary patterns in Taiwan, as reflected by increasing prevalence of central obesity (36), later age at FFTP (mean age at FFTP in 1976 and 2002 was 22.6 and 26.7 years, respectively), and the decrease of parity number (average number of parity in 1976 and 2002 was 2.3 and 1.6, respectively) (37), may contribute to the increasing incidence of breast cancer occurring in Chinese women in Taiwan. REFERENCES 1. Parkin DM, Pisani P, Ferlay J. Estimates of the worldwide incidence of 25 major cancers in Int J Cancer. 1999;80: Department of Health. Cancer Registry Annual Report Taipei: Department of Health, Executive Yuan; Cheng SH, Tsou MH, Liu MC, Jian JJ, Cheng JC, Leu SY, et al. Unique features of breast cancer in Taiwan. Breast Cancer Res Treat. 2000;63: Lo YL, Yu JC, Huang CS, Tseng SL, Chang TM, Chang KJ, et al. Allelic loss of the BRCA1 and BRCA2 genes and other regions on 17q and 13q in breast cancer among women from Taiwan (area of low incidence but early onset). Int J Cancer. 1998;79: Minami Y, Tsubono Y, Nishino Y, Ohuchi N, Shibuya D, Hisamichi S. The increase of female breast cancer incidence in Japan: Emergence of birth cohort effect. Int J Cancer. 2004;108: Yang HI, Lu SN, Liaw YF, You SL, Sun CA, Wang LY, et al. Hepatitis B e antigen and the risk of hepatocellular carcinoma. N Engl J Med. 2002;347: Sun CA, Wu DM, Lin CC, Lu SN, You SL, Wang LY, et al. Incidence and cofactors of hepatitis C virus-related hepatocellular carcinoma: a prospective study of 12,008 men in Taiwan. Am J Epidemiol. 2003;157: Chen CJ, You SL, Pwu RF, Wang LY, Lin YP, Hsi GC, et al. Communitybased hepatocellular carcinoma screening in seven townships in Taiwan. J Formos Med Assoc. 1995;94(Suppl 2):S94 S Kelsey JL, Whittemore AS, Evans AS, Thompson WD. Methods in Observational Epidemiology. 2 nd ed. New York: Oxford University Press; Department of Health. Cancer Registry:Annual Report, Department of Health, Executive Yuan, Republic of China, Colditz GA, Frazier AL. Models of breast cancer show that risk is set by events of early life: Prevention efforts must shift focus. Cancer Epidemiol Biomarkers Prev. 1995;4: Pathak DR, Whittemore AS. Combined effects of body size, parity, and menstrual events on breast cancer incidence in seven countries. Am J Epidemiol. 1992;135: Aalen O. Nonparametric inference for a family of counting processes. Ann Stat. 1978;6: Pena E, Rohatgi VK. Small sample and efficiency results for the Nelson- Aalen Estimator. J Stat Plann Inference. 1993;37: Madigan MP, Ziegler RG, Benichou J, Byrne C, Hoover RN. Proportion of breast cancer cases in the United States explained by well-established risk factors. J Natl Cancer Inst. 1987;87: MacMahon B, Cole P, Lin TM, Lowe CR, Mirra AP, Ravnihar B, et al. Age at first birth and breast cancer risk. Bull World Health Organ. 1970;43: Adami HO, Hansen J, Jung B, Rimsten AJ. Age at first birth, parity and risk of breast cancer in a Swedish population. Br J Cancer. 1980;42: Hsieh CC, Trichopoulos D, Katsouyanni K, Yuasa S. Age at menarche, age at menopause, height and obesity as risk factors for breast cancer:

7 AEP Vol. 16, No. 3 March 2006: Wu et al. HORMONE, BODY SIZE, AND BREAST CANCER 229 Associations and interactions in an international case-control study. Int J Cancer. 1990;46:796N Pathak DR, Whittemore AS. Combined effects of body size, parity, and menstrual events on breast cancer incidence in seven countries. Am J Epidemiol. 1992;135: De Stavola BL, Wang DY, Allen DS, Giaconi J, Fentiman IS, Reed MJ, et al. The association of height, weight, menstrual and reproductive events with breast cancer: results from two prospective studies on the island of Guernsey (United Kingdom). Cancer Causes Control. 1993;4: Colditz GA, Rosner B. Cumulative risk of breast cancer to age 70 years according to risk factor status: data from the Nurses Health Study. Am J Epidemiol. 2000;152: Apter D, Reinila M, Vihko R. Some endocrine characteristics of early menarche, a risk factor for breast cancer, are preserved into adulthood. Int J Cancer. 1989;44: Russo J, Tay LK, Russo IH. Differentiation of the mammary gland and susceptibility to carcinogenesis. Breast Cancer Res Treat. 1982;2: Trentham-Dietz A, Newcomb PA, Storer BE, Longnecker MP, Baron J, Greenberg ER, et al. Body size and risk of breast cancer. Am J Epidemiol. 1997;145: Morimoto LM, White E, Chen Z, Chlebowski RT, Hays J, Kuller L, et al. Obesity, body size, and risk of postmenopausal breast cancer: The Women s Health Initiative (United States). Cancer Causes Control. 2002;13: Weiderpass E, Braaten T, Magnusson C, Kumle M, Vainio H, Lund E, et al. A prospective study of body size in different periods of life and risk of premenopausal breast cancer. Cancer Epidemiol Biomarkers Prev. 2004;13: Hunter DJ, Willett WC. Diet, body size, and breast cancer. Epidemiol Rev. 1993;15: Hirose K, Tajima K, Hamajima N, Takezaki T, Inoue M, Kuroishi T, et al. Effect of body size on breast cancer risk among Japanese women. Int J Cancer. 1999;80: Shu XO, Jin F, Dai Q, Shi JR, Potter JD, Brinton LA, et al. Association of body size and fat distribution with risk of breast cancer among Chinese women. Int J Cancer. 2001;94: Ng EH, Gao F, Ji CY, Ho GH, Soo KC. Risk factors for breast carcinoma in Singaporean Chinese women: The role of central obesity. Cancer. 1997;80: Nelson LR, Bulun SE. Estrogen production and action. J Am Acad Dermatol. 2001;45:S116 S Bruning PF. Endogenous estrogens and breast cancer: A possible relationship between body fat distribution and estrogen availability. J Steroid Biochem. 1987;27: Dorgan JF, Longcope C, Stephenson HE, Falk RT, Miller R, Franz C, et al. Relation of prediagnostic serum estrogen and androgen levels to breast cancer risk. Cancer Epidemiol Biomarkers Prev. 1996;5: Thomas HV, Key TJ, Allen DS, Moore JW, Dowsett M, Fentiman IS, et al. A prospective study of endogenous serum hormone concentrations and breast cancer risk in post-menopausal women on the island of Guernsey. Br J Cancer. 1997;76: Collaborative Group on Hormonal Factors in Breast Cancer. Breast cancer and hormonal 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: Huang KC, Lin WY, Lee LT, Chen CY, Lo H, Hsia HH, et al. Four anthropometric indices and cardiovascular risk factors in Taiwan. Int J Obesity. 2002;26: Ministry of Interior. Demographic Facts. Ministry of Interior, Executive Yuan, Republic of China, 2003.