SEARCH

SEARCH BY CITATION

Keywords:

  • estrogen;
  • androgen;
  • prolactin;
  • exercise;
  • postmenopausal

Abstract

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Short List of WHI Investigators
  8. Acknowledgments
  9. References

Objective: Levels of estrogen, androgen, and prolactin have been related to risk of postmenopausal breast cancer. However, the determinants of these hormone concentrations are not established. The purpose of this study was to examine correlates of endogenous sex hormones.

Research Methods and Procedures: Associations among adiposity, physical activity, and diet and concentrations of estradiol, free estradiol, estrone, testosterone, free testosterone, sex hormone-binding globulin (SHBG), androstenedione, dehydroepiandrosterone, dehydroepiandrosterone sulfate, and prolactin were evaluated in 267 postmenopausal women randomly selected from the Women's Health Initiative Dietary Modification Trial.

Results: In multiple regression analyses on log-transformed hormones, BMI was positively associated with estrone (β = 0.031, p < 0.001), estradiol (β = 0.048, p < 0.001), free estradiol (β = 0.062, p < 0.001), free testosterone (β = 0.017, p = 0.02), and prolactin (β = 0.012, p = 0.02) and negatively associated with SHBG (β = −0.02, p = 0.001). Total physical activity (metabolic equivalent tasks per week) was negatively associated with concentrations of estrone, estradiol, and androstenedione (β = −0.006, −0.007, and −0.005, respectively, all p ≤ 0.05). Using a composite variable of BMI and physical activity dichotomized by median values, women with high BMI/low physical activity had a mean estrone concentration of 28.8 pg/mL, compared with 24.1, 19.9, and 18.4 pg/mL for women with high BMI/high physical activity, low BMI/low physical activity, and low BMI/high physical activity, respectively (p trend < 0.001). Similar trends were observed for estradiol and free estradiol and, in inverse, for SHBG.

Discussion: These associations may, in part, explain the positive associations between overweight/obesity and a sedentary lifestyle on breast cancer risk.


Introduction

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Short List of WHI Investigators
  8. Acknowledgments
  9. References

Several variables related to energy balance have been related to increased risk for breast, endometrial, and other hormone-related cancers, but the mechanisms linking energy balance to cancer risk are not defined (1). One possible mechanism is through the effect of adiposity, physical activity, and diet on endogenous sex hormones (2). A recent combined analysis of nested case-control data from nine cohort studies, with data from 663 breast cancer cases and 1765 women without breast cancer, reported that postmenopausal women with elevated estrogen or androgen levels have increased risk for developing breast cancer (3). Women in the top quintile for estradiol, free estradiol, testosterone, and androstenedione were approximately twice as likely to develop breast cancer compared with women with serum hormones in the bottom quintile. Some (4, 5) but not other (6) studies have found that prolactin concentrations are also associated with increased breast cancer risk in postmenopausal women. Elevated sex hormone concentrations have also been linked to increased endometrial cancer risk in postmenopausal women (7).

Despite their purported role in breast cancer development, the determinants of elevated circulating estrogen and androgen concentrations in postmenopausal women are not established (8, 9, 10, 11). Although several reports have found a direct association between increasing adiposity and concentrations of estrone, total estradiol, free estradiol, total testosterone, and free testosterone (12, 13, 14, 15, 16, 17), the association between physical activity and endogenous sex hormones independent of a mutual association with adiposity has not been established. Furthermore, the associations among adiposity, physical activity, and prolactin, a hormone potentially linked to breast cancer risk (5), have not been well studied in postmenopausal women.

In a subsample of postmenopausal women randomly selected from the Women's Health Initiative Dietary Modification clinical trial, we assayed several sex hormones including: estradiol, free estradiol, estrone, testosterone, free testosterone, sex hormone-binding globulin (SHBG),1 androstenedione, dehydroepiandrosterone (DHEA), DHEA sulfate (DHEAS), and prolactin. The purpose of this study was to investigate the associations among various adiposity, physical activity, and dietary factors and endogenous sex hormone concentrations in a cross-sectional analysis.

Research Methods and Procedures

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Short List of WHI Investigators
  8. Acknowledgments
  9. References

Study Population

The study population consisted of a subsample of women enrolled in the Women's Health Initiative (WHI) Dietary Modification Trial, a multicenter, multiethnic clinical trial testing the effect of a low-fat dietary pattern on risk of breast cancer and other chronic diseases in postmenopausal women (18). Women were enrolled into the study between October 1993 and December 1998 through 40 clinical centers across the United States. Women were eligible for the WHI Dietary Modification Trial if they were 50 to 79 years old, postmenopausal, planned to live in the clinical center area for at least three years, and were consuming a diet that consisted of at least 32% of calories from fat, as measured by a 120-item food frequency questionnaire (FFQ) (19). Exclusion criteria for the Dietary Modification Trial included eating ≥10 meals/wk outside the home, previous diagnosis of breast or colon cancer, type 1 diabetes, gastrointestinal conditions that prohibited a high-fiber diet, or any serious health conditions that might reduce survival over 3 years (e.g., class IV congestive heart failure). Details of the scientific rationale, design, eligibility requirements, and baseline characteristics of the WHI cohort have been published elsewhere (18). The Institutional Review Boards at all participating institutions including the coordinating center, subcontractors, and clinical centers approved study protocols and procedures. All participants signed informed consent forms.

Sample Selection

A subset of 300 women was randomly selected to assess sex hormones. The simple random sample was chosen from Diet Modification trial participants who were not in the WHI Hormone trials, not part of the subsample earmarked for testing trial effects on blood biomarkers (to conserve sample in that group), and not taking menopausal hormone therapy. The sampling probability was stratified by clinical center and age. Women were excluded from the present analyses if they were using menopausal hormone therapy within 3 months before blood draw, if they were missing data for the specific blood hormone being assessed, or if their hormone assays indicated that they might be premenopausal or using hormone therapy (e.g., estradiol > 30 pg/mL).

Exposure Assessment

All exposure information in this analysis was collected when women entered the study. A standardized written protocol, centralized training of clinic staff, and periodic quality assurance visits by the coordinating center were used to assure uniform administration of data collection instruments.

At a required baseline screening clinic visit, participants completed several self-administered questionnaires including medical history, reproductive and menstrual history, health behavior including physical activity and diet, and family history of select diseases including breast cancer. Reproducibility of WHI questionnaire data has been evaluated in a random sample of 536 women who had health-related information collected a second time approximately 10 weeks after baseline (20). The test-retest reliability (weighted κ) for the variables included in the present analyses ranged from 0.77 to 0.99.

Staff performed anthropometric measures (height, weight, waist circumference) and interviewed participants regarding lifetime use of menopausal hormone therapy. BMI was calculated as weight (kilograms)/height (meters)2. By self-report, women identified their ethnicity/race, selecting from six offered categories: white, African-American, Hispanic, Native American, Asian/Pacific Islander, and unknown.

Participants provided a blood sample after a 12-hour fast. Participants were asked not to take aspirin or non-steroidal anti-inflammatory medications for 48 hours before the blood draw, to refrain from smoking for at least 1 hour before the draw, and not to perform any vigorous physical activity for at least 12 hours before blood draw. Blood was processed within 2 hours, and serum was aliquoted and stored at −70 °C.

Age at menopause was determined as the youngest age at which the participant experienced any of the following: last menstrual bleeding (all participants were 12 or more months post last menstrual period), removal of both ovaries, or beginning of menopausal hormone therapy.

Several dietary variables were developed from the FFQ data and tested for WHI (19), including average total energy intake (kilocalories per day), percentage calories from fat, total carbohydrates (grams per day), fruits and vegetables (servings per day), fiber (grams per day), and alcohol intake (servings per week).

Physical activity was assessed as a variable “hours exercised per week” described in detail elsewhere (21). Briefly, information on current walking frequency, duration, and speed and frequency and duration of strenuous, moderate, and mild recreational physical activity was collected. Minutes of activity were multiplied by frequency separately for strenuous, moderate, and mild recreational physical activity, and three intensities of walking. Metabolic equivalent task (MET) values were assigned for strenuous, moderate-, and low-intensity activities as 7, 4, and 3 METs, respectively. For average (2 to 3 mph), fast (3 to 4 mph), and very fast (>4 mph) walking, MET values of 3, 4, and 4.5, respectively, were assigned. A current total physical activity variable (MET-hours per week) was then calculated by multiplying the MET level for the activity by the hours exercised per week and summing values for all of the types of activities. Variables were also calculated corresponding to minutes per week of strenuous, moderate/strenuous, and total recreational physical activity.

We also created a composite variable consisting of BMI and MET-hours per week of physical activity, both dichotomized above and below their respective median values. Women were then classified into one of four categories: low BMI/high physical activity, low BMI/low physical activity, high BMI/high physical activity, and high BMI/low physical activity.

Hormone Assays

The following assays were used to quantify estrogens and androgens. Assays were performed at Esoterix Laboratory Services Inc. (Calabasas Hills, CA) between January 2001 and January 2003. Placement of samples into one of five batches was randomly ordered. In addition, each batch included split duplicates and pooled quality control samples.

SHBG was measured using an immunoradiometric assay using plastic beads coated with monoclonal antibody against human SHBG and an I125-labeled soluble antibody against human SHBG. The assay sensitivity was 0.1 μg/dL. Intra- and inter-batch percentage coefficients of variation (CV) were 5.7 and 17.7, respectively. Estradiol and estrone were measured by radioimmunoassay (RIA) after organic extraction with hexane:ethyl acetate and LH20 column chromatography. Minimum reportable level was 0.5 ng/dL for both estrogens. Intra- and inter-batch percentage CV were 7.6 and 18.9 for estradiol and 7.3 and 9.9 for estrone, respectively. Testosterone was measured by RIA after extraction with hexane:ethyl acetate and aluminum oxide column chromatography. Sensitivity of the method was 3 ng/dL. Intra- and inter-batch percentage CV were 8.9 and 19.1, respectively. Androstenedione and DHEA were measured by RIA after extraction with hexane:ethyl acetate. Sensitivity of the method was 10 and 20 ng/dL, respectively. Intra- and inter-batch percentage CV were 6.2 and 11.1 for androstenedione and 7.5 and 20.1 for DHEA, respectively. DHEA sulfate was measured as DHEA by RIA after enzymolysis of the DHEA sulfate. The assay sensitivity was 10 μg/dL. Intra- and inter-batch percentage CV were 6.6 and 25.2, respectively. Prolactin was measured by a double-antibody chemiluminescent sandwich method. All samples were analyzed at two doses to avoid false negatives caused by the high dose hook effect. The assay sensitivity was 0.2 ng/mL. Intra- and inter-batch percentage CV were 7.7 and 18.9, respectively.

Free estradiol and free testosterone were calculated using the measured estradiol, testosterone albumin, and SHBG concentrations (22). This method has been found to have high validity compared with direct measurement (23).

Statistical Analyses

The present analyses included baseline data from study entry only (collected before randomization). Analyses were limited to women for whom data for all variables of interest were available (complete-case analysis).

The analysis was performed using SAS version 9.1 (SAS Institute, Cary, NC). Hormone concentrations were log-transformed to reduce the positive skewness of the distributions. Geometric means and 95% confidence intervals (back transformed to the original scale) were calculated. Tests of linear trend across the exposure variable were performed treating the hormone median of each category as a continuous variable in the model. Multiple regression models were used to evaluate the relationship between the hormone assays and the demographic variables including age, ethnicity, education, income, smoking, alcohol use, BMI, total energy expenditure from physical activity (MET-hours per week), percentage energy from fat, percentage energy from carbohydrate, and total fiber (grams) per 1000 kcal. Due to the colinearity among the independent variables (components of physical activity, dietary variables, BMI), not all were included in the multiple regression models. Assay batch was also evaluated as a covariate but did not affect the results and, therefore, was not included in the final model. All explanatory variables were evaluated simultaneously in the multiple regression models. All p values are two-sided. Due to the number of comparisons in these analyses, some results will be significant by chance.

Results

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Short List of WHI Investigators
  8. Acknowledgments
  9. References

Of the original 300 women in this hormone subsample, 274 were not using menopausal hormone therapy and had all serum hormone assay data available. Of these, seven women had estradiol concentrations >30 pg/mL and were determined likely to be premenopausal or using menopausal hormone therapy; their data were excluded. Therefore, data for 267 women were included in the analyses.

Baseline characteristics for the study group are presented in Table 1. Participants, on average, were 65 years old, obese (mean BMI of 30.1), and highly educated (40% had a college degree or higher). Most were parous, and 56% had had three or more full-term pregnancies. Almost 20% had experienced menopause at age 44 or younger. Twenty-three percent reported a history of hysterectomy, and 8.8% reported a history of bilateral oophorectomy. Alcohol intake was low (mean 2 alcohol servings/wk, with one serving defined as a 5-ounce glass of wine, a 12-ounce beer, or a 1-ounce shot of hard alcohol), and there were few current smokers. Their mean reported daily energy intake was 1800 kcal (low in comparison with their levels of adiposity), with an average percentage calories from fat of 38.8% and a mean number of just under 4 fruit and vegetable servings/d. The participants reported an average 89 minutes/wk of moderate/strenuous physical activity. These values were comparable with those of the overall WHI Dietary Modification participants (24).

Table 1.  Demographic and health characteristics of the study population; subsample of WHI Dietary Modification Trial (N = 267)*
 Total eligible = 267
 N% (Mean ± SD)
  • WHI, Women's Health Initiative; SD, standard deviation; MET, metabolic equivalent test.

  • *

    Excluded participants who were current HRT users at baseline or used menopausal hormone therapy within 3 months of screening or estradiol ≥ 30 pg/mL.

  • Percentages may not sum to 100 due to missing values and/or round-off error.

Age at screening (years)26764.8 ± 6.9
Ethnicity  
 White21078.7
 African-American4215.7
 Hispanic41.5
 Native American31.1
 Asian/Pacific Islander51.9
 Unknown31.1
Education  
 0 to 8 years31.1
 Some high school145.2
 High school diploma/general equivalency degree4516.9
 School after high school9836.7
 College degree or higher10740.1
Height (cm)266162.0 ± 6.3
Weight (kg)26778.9 ± 17.0
BMI (kg/m2)26630.1 ± 6.4
Waist circumference (cm)26590.8 ± 12.9
Hip circumference (cm)266110.0 ± 12.4
Alcohol servings/wk2672.0 ± 3.9
Daily intake (kcal)2671796.6 ± 646.03
Dietary total fat intake (g)26777.9 ± 31.0
Percentage calories from fat26738.85 ± 4.81
Fruits/vegetables (medium servings/d)2673.7 ± 1.7
Dietary fiber (g)/1000 kcal2679.1 ± 2.5
Physical activity (min/wk)251157.2 ± 166.5
Physical activity (moderate/strenuous min/wk)25189.5 ± 136.3
Physical activity (strenuous min/wk)25123.6 ± 64.3
Physical activity (MET-h/wk)25110.7 ± 12.8
Smoking  
 Missing41.5
 Never smoked14152.8
 Past smoker11442.7
 Current smoker83.0
Age at first period  
 ≤1210740.1
 13+15859.2
Number of term pregnancies  
 Never had term pregnancy51.9
 1 to 28632.2
 3 to 410539.3
 5+4516.9
Age last had any menstrual bleeding, years  
 44 or younger5219.5
 45 to 5415256.9
 55+3412.7
History of hysterectomy6523.7
History of bilateral oophorectomy248.8

Geometric mean concentrations of hormones are presented according to anthropometric characteristics in Table 2. Concentrations of estradiol, estrone, and free estradiol increased significantly with increasing quartile of BMI, waist circumference, and hip circumference. Concentrations of free estradiol, for example, were more than 3 times higher among women in the highest categories of BMI, waist circumference, and hip circumference compared with women in the lowest categories of these variables (p trend < 0.0001). Free testosterone also increased with increasing categories of these adiposity measures (p for BMI and hip circumference, p < 0.001), although the degree of increase was less compared with the estrogens.

Table 2.  Hormone concentrations according to anthropometric, diet, and physical activity characteristics
  Estrone (pg/mL)Estradiol (pg/mL)
Risk factorsNGeometric mean95% CIGeometric mean95% CI
  • CI, confidence interval; MET, metabolic equivalent; SHBG, sex hormone-binding globulin; DHEA, dehydroepiandrosterone; DHEAS, dehydroepiandrosterone sulfate.

  • *

    p < 0.01 for the test of linear trend.

  • p < 0.0001 for the test of linear trend.

All eligible26722.821.3 to 24.36.56.1 to 7.0
BMI (kg/m2) quartiles     
 19.45 to 25.396617.5*15.4 to 19.74.5*3.9 to 5.2
 25.40 to 28.936720.618.4 to 23.25.64.9 to 6.3
 28.96 to 33.276722.720.0 to 25.76.86.0 to 7.8
 33.37 to 64.886632.528.6 to 37.110.89.5 to 12.3
Waist circumference (cm) quartiles     
 61.5 to 80.56316.8*14.8 to 19.14.4*3.8 to 5.2
 81.0 to 90.07121.919.9 to 24.25.75.0 to 6.5
 90.1 to 100.06623.320.1 to 27.17.46.5 to 8.5
 100.2 to 124.06530.026.4 to 34.29.78.5 to 11.1
Hip circumference (cm) quartiles     
 84.5 to 100.56317.0*14.7 to 19.54.2*3.6 to 5.0
 101.0 to 107.56621.819.8 to 24.05.54.8 to 6.2
 108.0 to 117.27122.820.0 to 26.07.36.4 to 8.4
 118.0 to 146.06630.426.7 to 34.610.49.3 to 11.7
Waist-to-hip ratio quartiles     
 0.514 to 0.7856620.818.4 to 23.45.85.1 to 6.5
 0.785 to 0.8216623.620.8 to 26.77.16.1 to 8.3
 0.822 to 0.8686723.320.1 to 27.17.16.0 to 8.3
 0.868 to 1.0376623.520.5 to 26.86.45.5 to 7.4
Alcohol use (servings/wk)     
 None10926.3*23.6 to 29.47.5*6.6 to 8.5
 0.212 to 1.3657723.721.3 to 26.37.06.2 to 8.0
 1.423 to 30.0966718.216.3 to 20.45.14.5 to 5.7
Kilocalorie intake/d quartiles     
 613.5 to 1400.26623.420.5 to 26.76.85.9 to 7.9
 1402.1 to 1714.76723.220.5 to 26.46.55.6 to 7.5
 1717.0 to 2145.76721.919.2 to 25.06.25.3 to 7.2
 2147.3 to 4790.76722.619.6 to 26.16.65.6 to 7.7
Physical activity (MET-h/wk) quartiles     
 0.0 to 1.36227.8*24.1 to 32.08.6*7.4 to 9.9
 1.5 to 6.56322.920.0 to 26.27.06.0 to 8.2
 6.8 to 16.36322.219.7 to 25.15.54.7 to 6.3
 16.5 to 64.06318.616.3 to 21.35.74.9 to 6.6
BMI and physical activity (METs)     
 Low BMI (<29.0) and high physical activity (>6.5)7918.4*16.5 to 20.45.0*4.4 to 5.6
 Low BMI (<29.0) and low physical activity (≤6.5)5519.917.3 to 22.85.04.3 to 5.9
 High BMI (≥29.0) and high physical activity (>6.5)4724.120.5 to 28.26.85.6 to 8.2
 High BMI (≥29.0) and low physical activity (≤6.5)8628.825.6 to 32.39.78.8 to 10.8
  Free estradiol (pg/mL)Testosterone (pg/mL)
Risk factorsNGeometric mean95% CIGeometric mean95% CI
All eligible2670.130.12 to 0.15195.4178.8 to 197.0
BMI (kg/m2) quartiles     
 19.45 to 25.39660.08*0.07 to 0.09189.7162.0 to 199.2
 25.40 to 28.93670.110.10 to 0.13176.3177.2 to 213.4
 28.96 to 33.27670.150.13 to 0.17180.7164.2 to 194.7
 33.37 to 64.88660.250.22 to 0.28199.8181.1 to 224.6
Waist circumference (cm) quartiles     
 61.5 to 80.5630.08*0.07 to 0.09188.8163.8 to 203.3
 81.0 to 90.0710.120.10 to 0.13182.1174.7 to 206.5
 90.1 to 100.0660.160.13 to 0.18193.7162.9 to 199.9
 100.2 to 124.0650.230.20 to 0.26181.2174.8 to 214.4
Hip circumference (cm) quartiles     
 84.5 to 100.5630.08*0.07 to 0.09164.2148.5 to 181.5
 101.0 to 107.5660.110.09 to 0.12188.8170.4 to 209.3
 108.0 to 117.2710.150.13 to 0.18194.6179.4 to 211.0
 118.0 to 146.0660.240.21 to 0.27199.9180.0 to 222.0
Waist-to-hip ratio quartiles     
 0.514 to 0.785660.110.09 to 0.12193.3176.7 to 211.4
 0.785 to 0.821660.140.12 to 0.17190.2174.3 to 207.5
 0.822 to 0.868670.150.12 to 0.18189.4172.1 to 208.5
 0.868 to 1.037660.140.12 to 0.17175.4156.1 to 197.1
Alcohol use (servings/wk)     
 None1090.15*0.13 to 0.18201.4181.1 to 210.9
 0.212 to 1.365770.140.12 to 0.17198.0174.0 to 206.8
 1.423 to 30.096670.100.09 to 0.12179.2160.6 to 193.5
Kilocalorie intake/d quartiles     
 613.5 to 1400.2660.140.12 to 0.17179.3164.9 to 198.0
 1402.1 to 1714.7670.140.12 to 0.16193.8180.0 to 221.8
 1717.0 to 2145.7670.130.11 to 0.15199.7172.1 to 207.1
 2147.3 to 4790.7670.130.11 to 0.15210.3164.0 to 202.1
Physical activity (MET-h/wk) quartiles     
 0.0 to 1.3620.17*0.14 to 0.20210.31192.5 to 229.7
 1.5 to 6.5630.150.12 to 0.18184.58167.1 to 203.9
 6.8 to 16.3630.110.10 to 0.13170.53155.7 to 186.8
 16.5 to 64.0630.110.10 to 0.14185.35166.2 to 206.7
BMI and physical activity (METs)     
 Low BMI (<29.0) and high physical activity (>6.5)790.10*0.08 to 0.11182.0166.6 to 198.8
 Low BMI (<29.0) and low physical activity (≤6.5)550.090.08 to 0.11190.4169.7 to 213.5
 High BMI (≥29.0) and high physical activity (>6.5)470.160.13 to 0.19170.8151.7 to 192.4
 High BMI (≥29.0) and low physical activity (≤6.5)860.210.19 to 0.24200.7184.8 to 218.1
  Free testosterone (pg/mL)SHBG (nM)
Risk factorsNGeometric mean95% CIGeometric mean95% CI
All eligible2672.92.7 to 3.161.358.4 to 64.4
BMI (kg/m2) quartiles     
 19.45 to 25.39662.5*2.2 to 2.876.0*68.7 to 84.0
 25.40 to 28.93673.02.6 to 3.466.059.9 to 72.7
 28.96 to 33.27672.72.4 to 3.255.550.8 to 60.7
 33.37 to 64.88663.53.1 to 3.952.247.5 to 57.3
Waist circumference (cm) quartiles     
 61.5 to 80.5632.62.2 to 3.077.8*70.5 to 85.8
 81.0 to 90.0712.82.6 to 3.267.762.0 to 73.9
 90.1 to 100.0662.82.3 to 3.357.953.1 to 63.2
 100.2 to 124.0653.43.1 to 3.847.843.4 to 52.7
Hip circumference (cm) quartiles     
 84.5 to 100.5632.49*2.16 to 2.8771.0*64.1 to 78.7
 101.0 to 107.5662.642.24 to 3.1164.758.0 to 72.2
 108.0 to 117.2713.052.79 to 3.3461.656.8 to 66.8
 118.0 to 146.0663.483.10 to 3.9150.946.3 to 55.9
Waist-to-hip ratio quartiles     
 0.514 to 0.785662.682.39 to 3.0174.6*68.5 to 81.3
 0.785 to 0.821662.932.63 to 3.2866.360.7 to 72.4
 0.822 to 0.868673.092.74 to 3.4958.253.0 to 64.0
 0.868 to 1.037662.872.40 to 3.4350.044.8 to 55.7
Alcohol use (servings/wk)     
 None1093.3*3.0 to 3.759.555.2 to 64.1
 0.212 to 1.365772.72.3 to 3.158.452.9 to 64.6
 1.423 to 30.096672.62.3 to 2.966.561.1 to 72.5
Kilocalorie intake/d quartiles     
 613.5 to 1400.2662.82.5 to 3.260.955.8 to 66.5
 1402.1 to 1714.7673.12.6 to 3.761.455.2 to 68.2
 1717.0 to 2145.7672.82.5 to 3.161.355.7 to 67.6
 2147.3 to 4790.7672.82.5 to 3.261.755.4 to 68.6
Physical activity (MET-h/wk) quartiles     
 0.0 to 1.3623.32.9 to 3.662.757.3 to 68.7
 1.5 to 6.5632.82.5 to 3.258.453.4 to 64.0
 6.8 to 16.3632.62.2 to 3.060.454.0 to 67.7
 16.5 to 64.0633.02.6 to 3.463.056.4 to 70.5
BMI and physical activity (METs)     
 Low BMI (<29.0) and high physical activity (>6.5)792.782.47 to 3.1371.1*64.8 to 78.0
 Low BMI (<29.0) and low physical activity (≤6.5)552.622.29 to 2.9969.962.9 to 77.7
 High BMI (≥29.0) and high physical activity (>6.5)472.732.20 to 3.3849.343.5 to 56.0
 High BMI (≥29.0) and low physical activity (≤6.5)863.333.02 to 3.6656.552.5 to 60.7
  Androstenedione (pg/mL)DHEA (pg/mL)
Risk factorsNGeometric mean95% CIGeometric mean95% CI
All Eligible267603.7578.6 to 630.01548.91458.5 to 1644.9
BMI (kg/m2) quartiles     
 19.45 to 25.3966587.2538.8 to 640.01618.51429.8 to 1832.0
 25.40 to 28.9367632.8581.9 to 688.11632.41453.2 to 1833.7
 28.96 to 33.2767575.6530.6 to 624.41484.51319.1 to 1670.5
 33.37 to 64.8866629.5573.6 to 691.01510.61337.2 to 1706.5
Waist circumference (cm) quartiles     
 61.5 to 80.563585.3536.4 to 638.71559.91363.7 to 1784.4
 81.0 to 90.071598.8551.0 to 650.71584.51402.4 to 1790.3
 90.1 to 100.066618.8567.4 to 674.91639.61466.8 to 1832.8
 100.2 to 124.065612.8560.4 to 670.21409.21253.3 to 1584.5
Hip circumference (cm) quartiles     
 84.5 to 100.563562.6510.5 to 620.01414.01232.5 to 1622.2
 101.0 to 107.566637.0590.0 to 687.81699.21501.3 to 1923.2
 108.0 to 117.271597.2545.9 to 653.31649.41476.7 to 1842.3
 118.0 to 146.066625.1577.5 to 676.71454.61297.0 to 1631.4
Waist-to-hip ratio quartiles     
 0.514 to 0.78566593.2546.0 to 644.41719.61530.0 to 1932.6
 0.785 to 0.82166624.0575.7 to 676.31628.61444.1 to 1836.6
 0.822 to 0.86867575.7530.0 to 625.31376.41206.5 to 1570.2
 0.868 to 1.03766623.9565.3 to 688.51484.71324.3 to 1664.4
Alcohol use (servings/wk)     
 None109608.4567.9 to 651.71489.31351.3 to 1641.5
 0.212 to 1.36577612.3563.2 to 665.71577.01402.4 to 1773.4
 1.423 to 30.09667589.2548.4 to 633.11604.51448.5 to 1777.4
Kilocalorie intake/d quartiles     
 613.5 to 1400.266611.3561.6 to 665.41530.31327.9 to 1763.7
 1402.1 to 1714.767584.0539.9 to 631.81482.91334.9 to 1647.2
 1717.0 to 2145.767568.9514.6 to 628.91489.81310.4 to 1693.8
 2147.3 to 4790.767653.2604.0 to 706.41699.11526.5 to 1891.3
Physical activity (MET-h/wk) quartiles     
 0.0 to 1.362637.7590.6 to 688.51692.41520.0 to 1884.4
 1.5 to 6.563671.7619.2 to 728.81579.31396.5 to 1786.0
 6.8 to 16.363562.4515.0 to 614.31471.51281.8 to 1689.2
 16.5 to 64.063568.0512.2 to 629.81545.41358.2 to 1758.4
BMI and physical activity (METs)     
 Low BMI (<29.0) and high physical activity (>6.5)79568.5524.5 to 616.21560.61382.3 to 1761.9
 Low BMI (<29.0) and low physical activity (≤6.5)55663.0607.3 to 723.81664.61472.6 to 1881.6
 High BMI (≥29.0) and high physical activity (>6.5)47559.6495.8 to 631.61421.21223.5 to 1651.0
 High BMI (≥29.0) and low physical activity (≤6.5)86625.3583.7 to 669.81540.61390.9 to 1706.5
  DHEAS (μg/dL)Prolactin (pg/mL)
Risk factorsNGeometric mean95% CIGeometric mean95% CI
All eligible26742.840.0 to 45.97.67.2 to 7.9
BMI (kg/m2) quartiles     
 19.45 to 25.396642.536.8 to 49.27.26.6 to 7.9
 25.40 to 28.936747.441.1 to 54.88.27.5 to 8.9
 28.96 to 33.276741.937.1 to 47.27.06.5 to 7.6
 33.37 to 64.886639.934.2 to 46.58.07.3 to 8.8
Waist circumference (cm) quartiles     
 61.5 to 80.56340.834.8 to 47.97.16.5 to 7.8
 81.0 to 90.07146.240.4 to 52.97.87.2 to 8.5
 90.1 to 100.06639.435.0 to 44.37.16.5 to 7.8
 100.2 to 124.06544.538.2 to 51.88.07.3 to 8.8
Hip circumference (cm) quartiles     
 84.5 to 100.56337.932.8 to 43.77.06.4 to 7.7
 101.0 to 107.56650.843.9 to 58.78.07.3 to 8.7
 108.0 to 117.27142.437.6 to 47.97.26.7 to 7.9
 118.0 to 146.06640.935.1 to 47.68.17.3 to 8.8
Waist-to-hip ratio quartiles     
 0.514 to 0.7856645.739.8 to 52.47.97.2 to 8.6
 0.785 to 0.8216638.733.3 to 45.06.96.3 to 7.5
 0.822 to 0.8686742.437.1 to 48.47.56.8 to 8.2
 0.868 to 1.0376644.238.4 to 50.98.07.3 to 8.7
Alcohol use (servings/wk)     
 None10939.335.0 to 44.27.77.1 to 8.4
 0.212 to 1.3657743.938.6 to 49.97.87.2 to 8.3
 1.423 to 30.0966746.941.7 to 52.87.26.7 to 7.8
Kilocalorie intake/d quartiles     
 613.5 to 1400.26643.337.4 to 50.07.46.7 to 8.1
 1402.1 to 1714.76738.533.7 to 44.07.77.0 to 8.5
 1717.0 to 2145.76742.536.5 to 49.67.67.0 to 8.3
 2147.3 to 4790.76748.042.0 to 54.77.67.0 to 8.2
Physical activity (MET-h/wk) quartiles     
 0.0 to 1.36247.141.1 to 54.07.77.0 to 8.5
 1.5 to 6.56344.738.5 to 51.97.67.0 to 8.2
 6.8 to 16.36342.336.3 to 49.27.46.8 to 8.1
 16.5 to 64.06340.235.0 to 46.17.97.1 to 8.8
BMI and physical activity (METs)     
 Low BMI (<29.0) and high physical activity (>6.5)7943.838.2 to 50.17.87.2 to 8.6
 Low BMI (<29.0) and low physical activity (≤6.5)5546.639.9 to 54.57.36.8 to 8.0
 High BMI (≥29.0) and high physical activity (>6.5)4737.332.1 to 43.47.26.5 to 8.0
 High BMI (≥29.0) and low physical activity (≤6.5)8643.037.9 to 48.77.67.1 to 8.3

Concentrations of androstenedione, DHEA, and DHEAS were not associated with any of the adiposity variables in univariate analysis, although the level of androstenedione was significantly negatively associated with physical activity level in multivariate analysis (p = 0.03, Table 3). Prolactin concentration was also inconsistently related to BMI and circumference measures, although when tested in the multiple regression model, it was significantly linearly related to BMI (p = 0.02, Table 3). SHBG decreased significantly with increasing quartiles of BMI, waist and hip circumferences, and waist-to-hip ratio.

Table 3.  Multiple regression coefficients* and p values of the associations among adiposity, physical activity, dietary variables, and hormone variables (log-transformed); multiple regression coefficient (bolded values indicate significance at 5% level) p value for testing H0 (in parentheses), β = 0
 EstroneEstradiolFree estradiolTestosteroneFree testosteroneSHBGAndrostenedioneDHEADHEASProlactin
  • SHBG, sex hormone-binding globulin; DHEA, dehydroepiandrosterone; DHEAS, dehydroepiandrosterone sulfate; MET, metabolic equivalent.

  • *

    All independent variables were evaluated simultaneously in the multiple regression models. Also included in the models were age and race/ethnicity (white and African-American participants only). Other dietary, reproductive, and demographic variables and assay batch were not significantly associated with the hormones and did not affect the model results and, hence, are not presented.

BMI (kg/m2)0.031 (<0.001)0.048 (<0.001)0.062 (<0.001)0.006 (0.30)0.017 (0.02)−0.02 (0.001)0.002 (0.69)−0.00 (0.39)0.002 (0.75)0.012 (0.02)
Alcohol servings/wk−0.047 (0.01)−0.003 (0.89)−0.005 (0.82)−0.014 (0.34)−0.024 (0.23)0.015 (0.31)0.013 (0.323)0.029 (0.11)0.023 (0.26)−0.009 (0.48)
Caloric intake (kcal/d)−0.059 (0.45)−0.010 (0.91)−0.105 (0.26)−0.059 (0.34)−0.078 (0.37)0.033 (0.61)−0.012 (0.82)−0.020 (0.80)−0.079 (0.37)0.017 (0.76)
Physical activity (MET-h/wk)−0.006 (0.05)−0.007 (0.05)−0.004 (0.27)−0.001 (0.57)0.002 (0.67)−0.003 (0.22)−0.005 (0.03)−0.004 (0.23)−0.005 (0.21)0.004 (0.13)

Using a composite variable of BMI and physical activity dichotomized by median values, women with high BMI/low physical activity had a mean estrone concentration of 28.8 pg/mL, compared with 24.1, 19.9, and 18.4 pg/mL for women with high BMI/high physical activity, low BMI/low physical activity, and low BMI/high physical activity, respectively (p trend < 0.001). Women classified as high BMI/low physical activity had a mean estradiol level of 9.7 mg/dL vs. a mean of 5.0 mg/dL for women who were classified as low BMI/high physical activity (p trend < 0.001). Similar associations were observed for free estradiol (p trend < 0.001). An inverse association between this composite BMI/physical activity variable with SHBG was observed. Mean SHBG levels were 71.1, 69.9, 49.3, and 56.5, respectively, for women with low BMI/high physical activity, low BMI/low physical activity, high BMI/high physical activity, and high BMI/low physical activity (p trend < 0.01).

In univariate analyses, alcohol intake was inversely associated with increased concentrations of estrone, estradiol, free estradiol, and free testosterone (Table 2). However, in multiple regression analysis, alcohol intake was statistically significantly associated only with estrone concentration (p = 0.01, Table 3).

Macronutrient intakes (total and specific fats, carbohydrates, protein) were inconsistently associated with hormone concentrations (Tables 2 and 3). Concentrations of androstenedione, DHEA, DHEAS, estradiol, estrone, free estradiol, testosterone, and free testosterone decreased with increasing quartile of energy expenditure but were only statistically significant for androstenedione, estradiol, and estrone after multivariate adjustment.

In multiple regression analysis, associations between several adiposity variables and estrogens were high and statistically significant. In multiple regression analyses on log-transformed hormones, BMI was positively and statistically significantly associated with concentrations of estrone (β = 0.031, p < 0.001), estradiol (β = 0.048, p < 0.001), free estradiol (β = 0.062, p < 0.001), free testosterone level (β = 0.017, p = 0.02), and prolactin (β = 0.012, p = 0.02) and negatively associated with SHBG (β = −0.02, p = 0.001) (Table 3). Similar associations were observed with waist and hip circumferences (data not shown).

In multiple regression analysis, alcohol intake was statistically significantly associated only with estrone concentration (p = 0.01) (Table 3). In multiple regression analysis, total physical activity (MET per week) was negatively and significantly associated with concentrations of estrone, estradiol, and androstenedione (β = −0.006, −0.007, and −0.005 respectively, all p ≤ 0.05). Other dietary variables were inconsistently and not statistically significantly related to any hormone concentrations.

Discussion

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Short List of WHI Investigators
  8. Acknowledgments
  9. References

In this cross-sectional analysis, measures of obesity were significantly related to concentrations of several sex hormones. In particular, increased BMI, waist circumference, and hip circumference were associated with increased levels of estrone, estradiol, and free estradiol and negatively associated with SHBG. The association between free testosterone and adiposity was likely due to the association between adiposity and SHBG, however, because total testosterone was not associated with adiposity, and because SHBG was used to calculate free testosterone. The lack of significant association between waist-to-hip ratio and estrogen levels is not surprising, given that the up-regulation of aromatase, which converts testosterone to estradiol, may be higher in peripheral vs. visceral fat (25).

Increased physical activity was negatively associated with concentrations of estrone, estradiol, and free estradiol and after adjusting for adiposity, with concentrations of estrone, estradiol, and androstenedione. A composite variable combining dichotomized BMI and physical activity indicated that there was a synergy between adiposity and physical activity in their association with several hormone levels. Women who were the heaviest and most sedentary had the highest levels of estrone, estradiol, and free estradiol and the lowest levels of SHGB. Women who had the lowest BMI and highest physical activity had the lowest levels of estrone, estradiol, and free estradiol and the highest levels of SHGB.

Several prior reports have evaluated associations among a more limited array of sex hormones in relation to adiposity (12, 13, 15, 16, 17), but such associations with a comprehensive array of reproductive hormones in a population with minority representation have not, to our knowledge, been reported previously. These previous studies have observed increasing concentrations of estrogens and androgens and decreasing SHBG concentrations, with increasing adiposity (13, 14, 15, 17).

One cross-sectional study found a negative association of exercise with estrogen concentrations in postmenopausal women (14), whereas another found that increasing exercise was associated with increased postmenopausal estrogen concentrations (17). A randomized clinical trial of postmenopausal women observed decreased serum estrogens and androgens with a 1-year moderate-intensity exercise program, particularly in women who lost body fat (26, 41).

Several studies have assessed the relationship between adiposity and prolactin levels, with varying results (15, 27, 28). Animal models suggest a modest role of prolactin in adipocyte growth and development; there are prolactin receptors on adipocytes (29), and more recent work suggests that prolactin may inhibit lipoprotein lipase (30). Some (31, 32, 33), but not all (34), studies in young women suggest an acute rise in prolactin levels after a bout of vigorous exercise. One study found a reduced basal level of prolactin after young women increased weekly running by 50 miles but a blunted prolactin response to thyrotropin-releasing hormone (32). It is not clear whether the observed associations among adiposity, physical activity, and prolactin in premenopausal women are relevant to postmenopausal women, however.

Study strengths include the sampling from a large multi-ethnic/racial cohort of postmenopausal women from a large cross-section of U.S. geographic areas and measurement of an array of multiple sex hormones. Height, weight, hip, and waist measures were obtained by trained staff at a WHI clinic and are, therefore, not subject to bias of subject reporting or recall.

Study weaknesses include a limited sample size that precluded ability to separately assess associations with hormones in all racial/ethnic groups and a cross-sectional design that precludes determination of cause and effect. In addition, several comparisons were made that could have resulted in chance statistical significance.

To be eligible for the Dietary Modification trial, women had to have diets consisting of ≥32% of calories from fat, resulting in a relatively homogeneous population of women with high-fat dietary patterns. Dietary variables including total energy and fat intake were inconsistently and not statistically related to hormone concentrations. However, the distribution of dietary variables may not have been large enough to observe associations with hormones (35). Emerging evidence suggests that women with genetic polymorphisms associated with high concentrations of estradiol have higher BMIs and are less likely to lose weight with exercise compared with wild-type individuals (36). As a result, the distribution of dietary variables may not have been large enough to observe associations with hormones.

The self-reported nature of the dietary and physical activity data limits firm conclusions regarding these variables because there is considerable under-reporting of dietary intake and over-reporting of physical activity level, particularly among obese individuals (37, 38). Additionally, the FFQ does not perform well with respect to minorities, who are oversampled in this subsample, and in those with greater weight and less education. Specifically, under-reporting of energy intake among African Americans and Hispanics was ∼646 and 463 kcal/d, respectively, when estimating energy expenditure by the Harris-Benedict equation (39). These inaccuracies of dietary reporting through the FFQ may have obscured any real associations between dietary variables and hormone concentrations.

Only one blood specimen was assayed per woman. However, a single measure of many of these estrogens and androgens has predicted risk of breast cancer (3) and has been shown to reliably reflect usual hormone levels for several estrogens (40).

In summary, the consistent and highly significant associations noted between increased adiposity and elevated concentrations of estrogens and androgens and the associations between increased physical activity and decreased concentrations of several of these hormones suggest that these hormones can be modified by lifestyle changes. One randomized clinical trial in 173 postmenopausal women found that a 1-year moderate-intensity exercise program significantly decreased concentrations of estrone, estradiol, free estradiol, testosterone, and free testosterone, but the effect was limited to women who lost body fat (26, 41). More randomized clinical trial data are needed to define the independent and joint roles of weight loss and physical activity on these sex hormones.

Short List of WHI Investigators

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Short List of WHI Investigators
  8. Acknowledgments
  9. References

Program Office

National Heart, Lung, and Blood Institute (Bethesda, MD): Barbara Alving, Jacques Rossouw, Linda Pottern.

Clinical Coordinating Center

Fred Hutchinson Cancer Research Center (Seattle, WA): Ross Prentice, Garnet Anderson, Andrea LaCroix, Charles L. Kooperberg, Ruth E. Patterson, Anne McTiernan; Wake Forest University School of Medicine (Winston-Salem, NC): Sally Shumaker; Medical Research Laboratories (Highland Heights, KY): Evan Stein; University of California at San Francisco (San Francisco, CA): Steven Cummings.

Clinical Centers

Albert Einstein College of Medicine (Bronx, NY): Sylvia Wassertheil-Smoller; Baylor College of Medicine (Houston, TX): Jennifer Hays; Brigham and Women's Hospital, Harvard Medical School (Boston, MA): JoAnn Manson; Brown University (Providence, RI): Annlouise R. Assaf; Emory University (Atlanta, GA): Lawrence Phillips; Fred Hutchinson Cancer Research Center (Seattle, WA): Shirley Beresford; George Washington University Medical Center (Washington, DC): Judith Hsia; Harbor-University of California Los Angeles Research and Education Institute (Torrance, CA): Rowan Chlebowski; Kaiser Permanente Center for Health Research (Portland, OR): Evelyn Whitlock; Kaiser Permanente Division of Research (Oakland, CA): Bette Caan; Medical College of Wisconsin (Milwaukee, WI): Jane Morley Kotchen; MedStar Research Institute/Howard University (Washington, DC): Barbara V. Howard; Northwestern University (Chicago/Evanston, IL): Linda Van Horn; Rush-Presbyterian St. Luke's Medical Center (Chicago, IL): Henry Black; Stanford Prevention Research Center (Stanford, CA): Marcia L. Stefanick; State University of New York at Stony Brook (Stony Brook, NY): Dorothy Lane; The Ohio State University (Columbus, OH): Rebecca Jackson; University of Alabama at Birmingham (Birmingham, AL): Cora E. Lewis; University of Arizona (Tucson/Phoenix, AZ): Tamsen Bassford; University at Buffalo (Buffalo, NY): Jean Wactawski-Wende; University of California at Davis (Sacramento, CA): John Robbins; University of California at Irvine (Orange, CA): Allan Hubbell; University of California at Los Angeles (Los Angeles, CA): Howard Judd; University of California at San Diego (LaJolla/Chula Vista, CA): Robert D. Langer; University of Cincinnati (Cincinnati, OH): Margery Gass; University of Florida (Gainesville/Jacksonville, FL): Marian Limacher; University of Hawaii (Honolulu, HI): David Curb; University of Iowa (Davenport, IA): Robert Wallace; University of Massachusetts/Fallon Clinic (Worcester, MA): Judith Ockene; University of Medicine and Dentistry of New Jersey (Newark, NJ): Norman Lasser; University of Miami (Miami, FL): Mary Jo O'sullivan; University of Minnesota (Minneapolis, MN): Karen Margolis; University of Nevada (Reno, NV): Robert Brunner; University of North Carolina (Chapel Hill, NC): Gerardo Heiss; University of Pittsburgh (Pittsburgh, PA): Lewis Kuller; University of Tennessee (Memphis, TN): Karen C. Johnson; University of Texas Health Science Center (San Antonio, TX): Robert Brzyski; University of Wisconsin (Madison, WI): Gloria E. Sarto; Wake Forest University School of Medicine (Winston-Salem, NC): Denise Bonds; Wayne State University School of Medicine/Hutzel Hospital (Detroit, MI): Susan Hendrix.

Acknowledgments

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Short List of WHI Investigators
  8. Acknowledgments
  9. References

The WHI program is funded by the National Heart, Lung, and Blood Institute, U.S. Department of Health and Human Services.

Footnotes
  • 1

    Nonstandard abbreviations: SHBG, sex hormone-binding globulin; DHEA, dehydroepiandrosterone; DHEAS, DHEA sulfate; WHI, Women's Health Initiative; FFQ, food frequency questionnaire; MET, metabolic equivalent task; CV, coefficient(s) of variation; RIA, radioimmunoassay.

  • The costs of publication of this article were defrayed, in part, by the payment of page charges. This article must, therefore, be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

References

  1. Top of page
  2. Abstract
  3. Introduction
  4. Research Methods and Procedures
  5. Results
  6. Discussion
  7. Short List of WHI Investigators
  8. Acknowledgments
  9. References
  • 1
    International Agency for Research on Cancer Working Group on the Evaluation of Cancer-Preventive Agents (2002) Weight Control and Physical Activity (IARC Handbooks of Cancer Prevention, Vol. 6). IARC: Lyon, France.
  • 2
    McTiernan, A., Ulrich, C., Slate, S., Potter, J. (1998) Physical activity and cancer etiology: associations and mechanisms. Cancer Causes Control. 9: 487509.
  • 3
    Endogenous Hormones and Breast Cancer Collaborative Group (2002) Endogenous sex hormones and breast cancer in postmenopausal women: reanalysis of nine prospective studies. J Natl Cancer Inst. 94: 606616.
  • 4
    Hankinson, S., Willett, W., Michaud, D., et al (1999) Plasma prolactin levels and subsequent risk of breast cancer in postmenopausal women. J Natl Cancer Inst. 91: 629634.
  • 5
    Tworoger, S., Eliassen, A., Rosner, B., Sluss, P., Hankinson, S. (2004) Plasma prolactin concentrations and risk of postmenopausal breast cancer. Cancer Res. 64: 68146819.
  • 6
    Manjer, J., Johansson, R., Berglund, G., et al (2003) Postmenopausal breast cancer risk in relation to sex steroid hormones, prolactin and SHBG (Sweden). Cancer Causes Control. 14: 599607.
  • 7
    Lukanova, A., Lundin, E., Micheli, A., et al (2004) Circulating levels of sex steroid hormones and risk of endometrial cancer in postmenopausal women. Int J Cancer. 108: 425432.
  • 8
    Judd, HL, Shamonki, IM, Frumar, AM, Lagasse, LD (1982) Origin of serum estradiol in postmenopausal women. Obstet Gynecol. 59: 680686.
  • 9
    Meldrum, D., Davidson, D., Tateryn, I., Judd, H. (1981) Changes in circulating steroids with aging in postmenopausal women. Obstet Gynecol. 57: 624628.
  • 10
    Siiteri, PK (1987) Adipose tissue as a source of hormones. Am J Clin Nutr. 45 (Suppl 1): 277282.
  • 11
    Siiteri, PK, Murai, JT, Hammond, GL, Nisker, JA, Raymoure, WJ, Kuhn, RW (1982) The serum transport of steroid hormones. Recent Prog Horm Res. 38: 457510.
  • 12
    Bezemer, I., Rinaldi, S., Dossus, L., et al (2005) C-peptide, IGF-I, sex-steroid hormones and adiposity: a cross-sectional study in healthy women within the European Prospective Investigation into Cancer and Nutrition (EPIC). Cancer Causes Control. 16: 561572.
  • 13
    Boyapati, S., Shu, X., Gao, Y., et al (2004) Correlation of blood sex steroid hormones with body size, body fat distribution, and other known risk factors for breast cancer in post-menopausal Chinese women. Cancer Causes Control. 15: 305311.
  • 14
    Cauley, JA, Gutai, JP, Kuller, LH, LeDonne, D., Powell, JG (1989) The epidemiology of serum sex hormones in postmenopausal women. Am J Epidemiol. 129: 11201131.
  • 15
    Hankinson, SE, Willett, WC, Manson, JE, et al (1995) Alcohol, height, and adiposity in relation to estrogen and prolactin levels in postmenopausal women. J Natl Cancer Inst. 87: 12971302.
  • 16
    McTiernan, A., Rajan, KB, Tworoger, SS, et al (2003) Adiposity and sex hormones in postmenopausal breast cancer survivors. J Clin Oncol. 21: 19611966.
  • 17
    Verkasalo, PK, Thomas, HV, Appleby, PN, Davey, GK, Key, TJ (2001) Circulating levels of sex hormones and their relation to risk factors for breast cancer: a cross-sectional study in 1092 pre- and postmenopausal women (United Kingdom). Cancer Causes Control. 12: 4759.
  • 18
    Ritenbaugh, C., Patterson, R., Chlebowski, R., et al (2003) The Women's Health Initiative Dietary Modification trial: overview and baseline characteristics of participants. Ann Epidemiol. 13: S87S97.
  • 19
    Patterson, RE, Kristal, AR, Tinker, LF, Carter, RA, Bolton, MP, Agurs-Collins, T. (1999) Measurement characteristics of the Women's Health Initiative food frequency questionnaire. Ann Epidemiol. 9: 178187.
  • 20
    Langer, R., White, E., Lewis, C., Kotchen, J., Hendrix, S., Trevisan, M. (2003) The Women's Health Initiative observational study: baseline characteristics of participants and reliability of baseline measures. Ann Epidemiol. 13: S107S121.
  • 21
    McTiernan, A., Kooperberg, C., White, E., et al (2003) Recreational physical activity and the risk of breast cancer in postmenopausal women: the Women's Health Initiative Cohort Study. JAMA. 290: 13311336.
  • 22
    Vermeulen, A., Verdonck, L., Kaufman, JM (1999) A critical evaluation of simple methods for the estimation of free testosterone in serum. J Clin Endocrinol Metab. 84: 36663672.
  • 23
    Rinaldi, S., Geay, A., Dechaud, H., et al (2002) Validity of free testosterone and free estradiol determinations in serum samples from postmenopausal women by theoretical calculations. Cancer Epidemiol Biomarkers Prev. 11: 10651071.
  • 24
    Hays, J., Hunt, J., Hubbell, F., et al (2003) The Women's Health Initiative recruitment methods and results. Ann Epidemiol. 13 (Suppl 9): S18S77.
  • 25
    McTernan, P., Anderson, L., Anwar, A., et al (2002) Glucocorticoid regulation of p450 aromatase activity in human adipose tissue: gender and site differences. J Clin Endocrinol Metab. 87: 13271336.
  • 26
    McTiernan, A., Tworoger, SS, Ulrich, CM, et al (2004) Effect of exercise on serum estrogen in postmenopausal women: a 12-month randomized clinical trial. Cancer Res. 64: 29232928.
  • 27
    Kwa, H., Cleton, F., Bulbrook, R., Wang, D., Hayward, J. (1981) Plasma prolactin levels and breast cancer: relation to parity, weight and height, and age at first birth. Int J Cancer. 28: 3134.
  • 28
    Unzer, S., dos Santos, J., Moreira, A., Vilanova, M., de Sa, M. (1995) Alternatives in plasma gonadotropin and sex steroid levels in obese ovulatory and chronically anovulatory women. J Reprod Med. 40: 516520.
  • 29
    Flint, D., Binart, N., Kopchick, J., Kelly, P. (2003) Effects of growth hormone and prolactin on adipose tissue development and function: review. Pituitary. 6: 97102.
  • 30
    Ling, C., Svensson, L., Oden, B., et al (2003) Identification of functional prolactin (PRL) receptor gene expression: PRL inhibits lipoprotein lipase activity in human white adipose tissue. J Clin Endocrinol Metab. 88: 18041808.
  • 31
    Baker, E., Mathur, R., Kirk, R., Landgrebe, S., Moody, L., Williamson, H. (1982) Plasma gonadotropins, prolactin, and steroid hormone concentrations in female runners immediately after a long-distance run. Fertil Steril. 38: 3841.
  • 32
    Boyden, T., Pamenter, R., Grosso, D., Stanforth, P., Rotkis, T., Wilmore, J. (1982) Prolactin responses, menstrual cycles, and body composition of women runners. J Clin Endocrinol Metab. 54: 711714.
  • 33
    Shangold, M., Gatz, M., Thysen, B. (1981) Acute effects of exercise on plasma concentrations of prolactin and testosterone in recreational women runners. Fertil Steril. 35: 699702.
  • 34
    Kraemer, R., Blair, M., McCaferty, R., Castracane, V. (1993) Running-induced alterations in growth hormone, prolactin, triiodothyronine, and thyroxine concentrations in trained and untrained men and women. Res Q Exerc Sport. 64: 6974.
  • 35
    Wang, CY, Anderson, G., Prentice, RL (1999) Estimation of the correlation between nutrient intake measures under restricted sampling. Biometrics. 55: 711717.
  • 36
    Tworoger, S., Chubak, J., Aiello, E., et al (2004) The effect of CYP19and COMT polymorphisms on exercise-induced fat loss inpostmenopausal women. Obes Res. 12: 972981.
  • 37
    Bingham, S., Luben, R., Welch, A., Wareham, N., Khaw, K., Day, N. (2003) Are imprecise methods obscuring a relation between fat and breast cancer? Lancet. 362: 212214.
  • 38
    Lichtman, SW, Pisarska, K., Berman, ER, et al (1992) Discrepancy between self-reported and actual caloric intake and exercise in obese subjects. N Engl J Med. 327: 18931898.
  • 39
    Hebert, J., Patterson, R., Gorfine, M., Ebbeling, C., St Jeor, S., Chlebowski, R. (2003) Differences between estimated caloric requirements and self-reported caloric intake in the Women's Health Initiative. Ann Epidemiol. 13: 629637.
  • 40
    Toniolo, P., Koenig, K., Pasternack, B., et al (1994) Reliability of measurements of total, protein-bound, and unbound estradiol in serum. Cancer Epidemiol Biomarkers Prev. 3: 4750.
  • 41
    McTiernan, A., Tworoger, S., Rajan, B., et al (2004) Effect of exercise on serum androgens in postmenopausal women: a 12-month randomized clinical trial. Cancer Epidemiol Biomarkers Prev. 13: 17.