Abstract
Background
Increased body mass index (BMI) is associated with higher postmenopausal breast cancer risk and lower premenopausal breast cancer risk. Less is known about the central adiposity-breast cancer risk association, particularly for tumor subtypes.
Methods
We used prospective waist (WC) and hip circumference (HC) measures in the Nurses’ Health Studies. We examined associations of WC, HC, and waist-to-hip ratio (WHR) with breast cancer independent of BMI, by menopausal status. Cox proportional hazards models estimated the hazard ratios (HRs) and 95% confidence intervals (CIs) adjusting for breast cancer risk factors, with and without BMI.
Results
Adjusting for BMI, WC and HC were not associated, and WHR was positively associated with premenopausal breast cancer risk (WHR, quintile 5 vs 1: HRQ5vQ1, BMI-adjusted = 1.27, 95% CI = 1.04 to 1.54; Ptrend = .01), particularly for estrogen receptor-negative (ER-) and progesterone receptor-negative (PR-) and basal-like breast cancers. Premenopausal WC, HC, and WHR were not associated with postmenopausal breast cancer risk, with or without BMI adjustment. Postmenopausal WC, HC, and WHR were each positively associated with postmenopausal breast cancer (eg, WC HRQ5vsQ1 = 1.59, 95% CI = 1.36 to 1.86); after adjustment for BMI, only WC remained statistically significant (HRQ5vsQ1, BMI-adjusted = 1.38, 95% CI = 1.15 to 1.64; Ptrend = .002). In postmenopausal women, associations were stronger among never-users of hormone therapy and for ER+/PR+ breast cancers.
Conclusions
Central adiposity was positively associated with pre- and postmenopausal breast cancers independent of BMI. This suggests that mechanisms other than estrogen may also play a role in the relationship between central adiposity and breast cancer. Maintaining a healthy waist circumference may decrease pre- and postmenopausal breast cancer risk.
Because 1 in 8 US women develop breast cancer (1), prevention is important. Adiposity, a potentially modifiable risk factor, is commonly measured via body mass index (BMI), which represents adipose tissue and lean mass (2). Higher BMI is associated with increased postmenopausal breast cancer risk but lower premenopausal risk (3,4). Waist circumference (WC) and waist-to-hip ratio (WHR) are often used to measure central obesity, characterized by high levels of visceral adipose tissue (VAT) (5,6). Central adiposity may be particularly relevant to breast cancer development as VAT may be more metabolically active than subcutaneous adipose tissue (SAT) (3).
Central adiposity has generally been positively associated with postmenopausal breast cancer (4,7). Because WC is strongly correlated with BMI (r = 0.8–0.9) (8–13), adjusting for BMI is key to understanding central adiposity’s independent role. Among studies that did so (n = 7), the WC-postmenopausal breast cancer association was attenuated but still positive (hazard ratio [HR] per 10 cm = 1.05, 95% confidence interval [CI] = 1.02 to 1.08) in a meta-analysis (7).
Less is known for central adiposity and premenopausal breast cancer. In a prospective study meta-analysis, a positive association of WC with premenopausal breast cancer was stronger among studies adjusting for BMI (n = 5, HR per 10 cm = 1.09, 95% CI = 1.02 to 1.16) (7). However, sample sizes were limited, particularly for stratified analyses. Additionally, few studies have evaluated tumor heterogeneity, regardless of menopausal status. Therefore, we examined whether WC, hip circumference (HC), or WHR were associated with incident invasive breast cancer (overall and by tumor subtype), independent of BMI according to menopause status.
Methods
Study Population
The Nurses’ Health Study (NHS) began in 1976 with 121 700 female registered nurses aged 30–55 years, and the Nurses’ Health Study II (NHS2) began in 1989 with 116 429 female registered nurses aged 25–42 years. The participants complete a biennial questionnaire on lifestyle risk factors and illnesses. Among participants alive through 2016 (NHS) and 2017 (NHS2), follow-up rates were greater than 92%. The study protocol was approved by the institutional review boards of the Brigham and Women’s Hospital and Harvard T.H. Chan School of Public Health and those of participating cancer registries as required.
Anthropometric Measurement
Women measured their WC at the umbilicus and HC at the largest circumference around the hips in 1986 (NHS) and 1993 (NHS2) and again in 1996 and 2000 (NHS) or 2005 (NHS2), standing with a tape measure, to the nearest quarter inch. Women self-reported their weight and height at baseline (NHS = 1976; NHS2 = 1989), used to calculate BMI (kg/m2), and updated weight biennially thereafter. Weight at age 18 years was first reported in 1980 for NHS and 1986 for NHS2. Self-reported and technician anthropometric measurements are highly correlated (Supplementary Methods, available online) (14). In our analyses, a WC less than 15 or greater than 55 inches and an HC less than 20 or greater than 65 inches were excluded as outliers (n = 197) (10,11).
Breast Cancer Case Ascertainment
For participants who reported breast cancer diagnoses, medical records were reviewed to confirm and extract information on histopathology, invasiveness, and estrogen receptor (ER), progesterone receptor (PR), and HER2 results. Cases among deceased and nonrespondents were identified via death certificates and medical records. Because less than 1% of diagnoses were rejected after review, we included participant-confirmed incident breast cancers for whom records were not obtained. Molecular subtypes of breast cancer tumors based on immunohistochemistry were available for a subset of cases through 2006 (Supplementary Methods, available online) (15,16).
Statistical Analysis
Women contributed person-months from the return of the 1986 (NHS) and 1993 (NHS2) questionnaires through the earliest date of breast cancer diagnosis or other cancer diagnosis (except nonmelanoma skin cancers), date of death, or end of follow-up (NHS, June 2016; NHS2, June 2017). Time-varying Cox proportional hazards models, stratified by age in months and questionnaire cycle, were used to calculate hazard ratios and 95% confidence intervals for risk of invasive breast cancer with each central adiposity measure quintile, using the lowest quintile as the referent.
Because the BMI breast cancer association varies by menopause, we conducted analyses for premenopausal and postmenopausal adiposity measures separately and also updated menopause status every 2 years. For the premenopausal adiposity measure, we examined premenopausal (pre-pre) and postmenopausal (pre-post) breast cancer separately. For premenopausal anthropometric measure analyses, women who were postmenopausal or had unknown menopausal status at the initial WC and HC measurement were excluded. Premenopausal anthropometric measures were only updated if a woman was still premenopausal at the later assessment, otherwise the prior premenopausal measure was carried forward. Pre-pre and pre-post analyses were not mutually exclusive, with pre-pre analyses women only contributing time, whereas premenopausal and pre-post analyses contributing time until they were diagnosed with postmenopausal breast cancers or until other censoring events. Women did not contribute person-months in intervals when their menopause status was uncertain (<4%).
Age-adjusted and multivariable-adjusted models with time-varying covariates including age at menarche, height, parity and age at first birth, family history of breast cancer, benign breast disease diagnosis, alcohol intake, physical activity, smoking, menopausal status (pre-post models), hormone therapy (HT) use (pre-post, post-post models), and age at menopause (pre-post, post-post models) were run. Multivariable-adjusted models additionally controlling for BMI were run in 2 ways: BMI updated when WC and HC measures were updated and BMI updated every 2 years. Tests for trend across quintiles used the median of each quintile as a continuous variable.
Additional details on the methods are provided in the Supplementary Methods (available online). Statistical tests were 2-sided, and P values less than .05 were considered statistically significant. Analyses were conducted using SAS (SAS Institute Inc, Cary, NC).
Results
Over 24–30 years of follow-up, among 96 746 women who reported both WC and HC, 6129 (NHS = 4223, NHS2 = 1906) incident invasive breast cancer cases were identified (pre-pre = 1131, pre-post = 2089, post-post = 2909).
At the first WC measure, 70% of NHS and 5% of NHS2 women were postmenopausal (Table 1). Women with higher WC were older, had lower physical activity, and were more often prior smokers. Women with lower WC were more often current smokers, had a history of benign breast disease, and used HT. Supplementary Tables 1–2 (available online) show characteristics of women separately by menopause status at baseline.
Table 1.
Age-standardized characteristics (means [SD] or percentages) according to baseline waist circumference in the Nurses’ Health Study (1986) and Nurses’ Health Study II (1993)
| Characteristics | NHS | NHS2 | ||||
|---|---|---|---|---|---|---|
| Quintile 1 | Quintile 3 | Quintile 5 | Quintile 1 | Quintile 3 | Quintile 5 | |
| (n = 8282) | (n = 9424) | (n = 9517) | (n = 11 527) | (n = 8393) | (n = 9868) | |
| Mean waist circumference (SD), cm | 66.0 (2.6) | 76.7 (1.5) | 96.3 (8.2) | 65.7 (2.7) | 76.7 (1.4) | 98.6 (10.2) |
| Mean hip circumference (SD), cm | 91.9 (5.4) | 99.1 (5.7) | 113.1 (11.1) | 91.4 (5.0) | 98.9 (5.7) | 114.6 (13.3) |
| Mean waist-to-hip ratio (SD) | 0.72 (0.05) | 0.78 (0.05) | 0.86 (0.08) | 0.72 (0.04) | 0.78 (0.04) | 0.87 (0.10) |
| Mean body mass index (SD), kg/m2 | 20.7 (2.0) | 23.6 (2.2) | 30.6 (4.8) | 20.6 (1.9) | 23.5 (2.4) | 31.3 (6.0) |
| Mean height (SD), inches | 63.7 (2.3) | 64.6 (2.4) | 64.9 (2.5) | 64.2 (2.5) | 65.1 (2.6) | 65.3 (2.7) |
| Mean age (SD), ya | 50.9 (7.0) | 53.6 (7.0) | 55.0 (6.8) | 38.1 (4.6) | 38.8 (4.6) | 39.6 (4.6) |
| Mean age at menarche (SD), y | 12.7 (1.5) | 12.6 (1.4) | 12.4 (1.5) | 12.6 (1.5) | 12.5 (1.4) | 12.1 (1.4) |
| Mean parity (SD)b | 2.9 (1.4) | 3.2 (1.5) | 3.3 (1.6) | 2.1 (0.9) | 2.2 (0.9) | 2.2 (1.0) |
| Mean age at first birth (SD), y | 25.1 (3.3) | 25.1 (3.2) | 25.2 (3.4) | 25.9 (4.2) | 26.0 (4.2) | 25.7 (4.4) |
| Mean alcohol intake (SD), g/day | 6.6 (10.5) | 7.0 (11.5) | 5.1 (11.0) | 3.5 (6.1) | 3.4 (6.4) | 2.5 (6.1) |
| Mean physical activity, MET-h/wk | 19.0 (24.6) | 14.7 (20.1) | 10.6 (16.2) | 27.7 (33.5) | 20.6 (24.8) | 16.0 (22.2) |
| Family history of breast cancer, % | 8.1 | 8.5 | 8.3 | 6.2 | 6.2 | 5.9 |
| History of benign breast disease, % | 40.1 | 36.3 | 28.4 | 42.9 | 40.2 | 34.8 |
| Smoking status | ||||||
| Current smoker, % | 24.4 | 20.6 | 18.0 | 11.5 | 10.7 | 11.6 |
| Past smoker, % | 31.6 | 34.3 | 36.6 | 21.7 | 24.1 | 23.3 |
| Postmenopausal, %c | 68.4 | 68.4 | 69.9 | 4.2 | 4.8 | 5.7 |
| Mean age at menopause (SD), yc,d | 48.4 (5.6) | 48.7 (4.8) | 48.7 (4.7) | 39.2 (4.4) | 39.6 (3.9) | 39.5 (3.7) |
| Postmenopausal hormone usee | ||||||
| Current use, % | 34.2 | 31.0 | 20.8 | 52.5 | 53.6 | 49.7 |
| Past use, % | 21.0 | 22.1 | 23.2 | 10.1 | 7.8 | 12.1 |
Value is not age adjusted. MET = metabolic equivalent tasks; NHS = Nurses’ Health Study; NHS2 = Nurses’ Health Study II.
Number of children among parous women.
For NHS, at the last cycle of follow-up (2014–2016), 100.0% were postmenopausal, and age at menopause was 49.6 (4.5), 49.9 (4.4), and 49.7 (4.5) years for the first, third, and fifth quintile, respectively. For NHS2, at the last cycle of follow-up (2015–2017), 97.0%, 97.1%, and 96.9% were postmenopausal, and age at menopause was 49.9 (4.5), 50.2 (4.3), and 49.8 (4.5) years for the first, third, and fifth quintile, respectively.
Natural menopause or menopause due to bilateral oophorectomy.
Postmenopausal hormone use among postmenopausal women only.
WC, HC, and BMI were highly correlated (r = 0.71–0.81). WHR was highly correlated with WC (r = 0.73–0.76) but less correlated with HC and BMI (r = 0.18–0.39). Height was not correlated with WHR and weakly correlated with WC and HC (r = 0.13–0.23). Spearman correlations for anthropometric measures assessed 10–12 years apart ranged from 0.67 to 0.84 and were similar in pre-pre, pre-post, and post-post analyses.
Age-adjusted and multivariable-adjusted results (without BMI) were similar (data not shown), so only multivariable results are presented (Table 2). Multivariable models additionally adjusting for BMI updated biennially were similar, although slightly attenuated, to models adjusting for BMI updated only when WC and HC measurements were updated (data not shown). In the current pooled analysis, no evidence of study heterogeneity was found (all Pheterogeneity ≥.08; Supplementary Tables 3-4, available online). Estimates were similar when BMI at age 18 years was included (data not shown). Lastly, using cumulatively averaged or only initial central adiposity measures were similar to updated models shown.
Table 2.
Multivariable-adjusted hazard ratios and 95% confidence intervals for invasive breast cancer by menopausal status at adiposity measure and diagnosis: Nurses’ Health Study (1986-2016) and Nurses’ Health Study II (1993-2017)
| Measure, quintiles | Quintile 1 | Quintile 2 | Quintile 3 | Quintile 4 | Quintile 5 | P trend |
|---|---|---|---|---|---|---|
| HR (95% CI) | HR (95% CI) | HR (95% CI) | HR (95% CI) | HR (95% CI) | ||
| Premenopausal at baseline measure and diagnosis | ||||||
| Waist circumference, cm | ||||||
| Quintile range | <69.2 | 69.2–74.2 | 74.3–79.3 | 79.4–86.9 | ≥87.0 | |
| Cases/person-years | 237/ 149 734 | 237/ 145 315 | 214/ 116 159 | 234/ 112 303 | 209/ 133 813 | |
| MV-adjusteda | 1.00 (Referent) | 0.94 (0.78 to 1.13) | 1.02 (0.84 to 1.23) | 1.15 (0.95 to 1.39) | 0.83 (0.68 to 1.01) | .18 |
| MV-adjusted and BMI adjustedb | 1.00 (Referent) | 0.98 (0.82 to 1.19) | 1.11 (0.91 to 1.36) | 1.34 (1.09 to 1.64) | 1.12 (0.86 to 1.45) | .12 |
| Hip circumference, cm | ||||||
| Quintile range | <92.7 | 92.7–97.1 | 97.2–101.5 | 101.6–107.2 | ≥107.3 | |
| Cases/person-years | 233/ 150 965 | 283/ 156 082 | 180/ 99 943 | 227/ 120 740 | 208/ 129 593 | |
| MV-adjusteda | 1.00 (Referent) | 1.05 (0.88 to 1.25) | 0.96 (0.78 to 1.17) | 0.98 (0.81 to 1.19) | 0.80 (0.65 to 0.98) | .01 |
| MV-adjusted and BMI adjustedb | 1.00 (Referent) | 1.09 (0.91 to 1.30) | 1.03 (0.83 to 1.27) | 1.11 (0.89 to 1.37) | 1.03 (0.78 to 1.36) | .81 |
| Waist-to-hip ratio | ||||||
| Quintile range | <0.73 | 0.73–0.75 | 0.76–0.78 | 0.79–0.82 | ≥0.83 | |
| Cases/person-years | 223/ 135 199 | 203/ 128 677 | 228/ 127 744 | 228/ 131 390 | 249/ 134 313 | |
| MV-adjusteda | 1.00 (Referent) | 0.96 (0.79 to 1.16) | 1.09 (0.91 to 1.32) | 1.06 (0.87 to 1.28) | 1.12 (0.93 to 1.36) | .15 |
| MV-adjusted and BMI adjustedb | 1.00 (Referent) | 0.97 (0.80 to 1.18) | 1.13 (0.94 to 1.37) | 1.13 (0.93 to 1.37) | 1.27 (1.04 to 1.54) | .01 |
| Body mass index, kg/m2c | ||||||
| Quintile range | <20.9 | 20.9–22.4 | 22.5–24.3 | 24.4–27.4 | ≥27.5 | |
| Cases/person-years | 260/ 148 535 | 233/ 135 538 | 232/ 124 984 | 206/ 116 261 | 200/ 132 005 | |
| MV-adjusteda | 1.00 (Referent) | 0.90 (0.75 to 1.08) | 0.95 (0.80 to 1.14) | 0.88 (0.73 to 1.06) | 0.75 (0.61 to 0.91) | .004 |
| Premenopausal at baseline measure and postmenopausal at diagnosis | ||||||
| Waist circumference, cm | ||||||
| Quintile range | <69.2 | 69.2–74.2 | 74.3–79.3 | 79.4–86.9 | ≥87.0 | |
| Cases/person-years | 392/ 288 766 | 464/ 293 148 | 426/ 237 668 | 390/ 236 290 | 417/ 278 650 | |
| MV-adjusteda | 1.00 (Referent) | 1.09 (0.95 to 1.25) | 1.22 (1.06 to 1.40) | 1.15 (0.99 to 1.33) | 1.16 (1.00 to 1.34) | .09 |
| MV-adjusted and BMI adjustedb | 1.00 (Referent) | 1.08 (0.94 to 1.24) | 1.19 (1.03 to 1.38) | 1.10 (0.94 to 1.29) | 1.08 (0.90 to 1.30) | .56 |
| Hip circumference, cm | ||||||
| Quintile range | <92.7 | 92.7–97.1 | 97.2–101.5 | 101.6–107.2 | ≥107.3 | |
| Cases/person-years | 356/ 279 133 | 477/ 312 050 | 355/ 206 856 | 462/ 262 415 | 439/ 274 067 | |
| MV-adjusteda | 1.00 (Referent) | 1.06 (0.92 to 1.22) | 1.13 (0.97 to 1.31) | 1.14 (0.99 to 1.32) | 1.17 (1.00 to 1.36) | .04 |
| MV-adjusted and BMI adjustedb | 1.00 (Referent) | 1.05 (0.91 to 1.21) | 1.11 (0.95 to 1.30) | 1.11 (0.95 to 1.30) | 1.10 (0.91 to 1.35) | .31 |
| Waist-to-hip ratio | ||||||
| Quintile range | <0.73 | 0.73–0.75 | 0.76–0.78 | 0.79–0.82 | ≥0.83 | |
| Cases/person-years | 454/ 278 828 | 416/ 262 453 | 421/ 258 057 | 437/ 266 396 | 361/ 268 787 | |
| MV-adjusteda | 1.00 (Referent) | 1.01 (0.88 to 1.15) | 1.07 (0.93 to 1.22) | 1.09 (0.95 to 1.25) | 0.97 (0.84 to 1.12) | .99 |
| MV-adjusted and BMI adjustedb | 1.00 (Referent) | 1.00 (0.88 to 1.15) | 1.06 (0.92 to 1.21) | 1.06 (0.93 to 1.22) | 0.93 (0.80 to 1.08) | .55 |
| Body mass index, kg/m2c | ||||||
| Quintile range | <20.9 | 20.9–22.4 | 22.5–24.3 | 24.4–27.4 | ≥27.5 | |
| Cases/person-years | 356/ 274 317 | 431/ 269 939 | 441/ 260 439 | 429/ 250 455 | 432/ 279 371 | |
| MV-adjusteda | 1.00 (Referent) | 1.10 (0.95 to 1.27) | 1.14 (0.99 to 1.31) | 1.15 (0.99 to 1.33) | 1.18 (1.02 to 1.37) | .05 |
| Postmenopausal at both baseline measure and diagnosis | ||||||
| Waist circumference, cm | ||||||
| Quintile range | <69.2 | 69.2–74.2 | 74.3–79.3 | 79.4–86.9 | ≥87.0 | |
| Cases/person-years | 209/ 85 541 | 413/ 123 305 | 430/ 131 930 | 698/ 18 9409 | 1159/ 280 518 | |
| MV-adjusteda | 1.00 (Referent) | 1.31 (1.10 to 1.55) | 1.22 (1.03 to 1.44) | 1.39 (1.19 to 1.63) | 1.59 (1.36 to 1.86) | <.001 |
| MV-adjusted and BMI adjustedb | 1.00 (Referent) | 1.27 (1.07 to 1.51) | 1.16 (0.98 to 1.37) | 1.28 (1.09 to 1.51) | 1.38 (1.15 to 1.64) | .002 |
| Hip circumference, cm | ||||||
| Quintile range | <92.7 | 92.7–97.1 | 97.2–101.5 | 101.6–107.2 | ≥107.3 | |
| Cases/person-years | 342/ 117 976 | 531/ 151 002 | 412/ 120 332 | 757/ 207 785 | 867/ 213 607 | |
| MV-adjusteda | 1.00 (Referent) | 1.18 (1.03 to 1.36) | 1.11 (0.96 to 1.29) | 1.19 (1.04 to 1.36) | 1.38 (1.21 to 1.57) | <.001 |
| MV-adjusted and BMI adjustedb | 1.00 (Referent) | 1.14 (0.99 to 1.31) | 1.04 (0.89 to 1.20) | 1.05 (0.91 to 1.21) | 1.10 (0.93 to 1.30) | .60 |
| Waist-to-hip ratio | ||||||
| Quintile range | <0.73 | 0.73–0.75 | 0.76–0.78 | 0.79–0.82 | ≥0.83 | |
| Cases/person-years | 291/ 95 782 | 363/ 113 608 | 438/ 134 033 | 645/ 173 040 | 1172 /294 239 | |
| MV-adjusteda | 1.00 (Referent) | 1.03 (0.88 to 1.20) | 1.02 (0.88 to 1.19) | 1.17 (1.02 to 1.35) | 1.24 (1.08 to 1.42) | <.001 |
| MV-adjusted and BMI adjustedb | 1.00 (Referent) | 1.02 (0.87 to 1.19) | 0.99 (0.85 to 1.15) | 1.11 (0.97 to 1.29) | 1.15 (1.00 to 1.32) | .006 |
| Body mass index, kg/m2c | ||||||
| Quintile range | <20.9 | 20.9–22.4 | 22.5–24.3 | 24.4–27.4 | ≥27.5 | |
| Cases/person-years | 325/ 107 181 | 411/ 126 527 | 544/ 157 367 | 722/ 199 186 | 907/ 220 440 | |
| MV-adjusteda | 1.00 (Referent) | 1.06 (0.91 to 1.22) | 1.11 (0.96 to 1.27) | 1.17 (1.02 to 1.34) | 1.44 (1.26 to 1.65) | <.001 |
Multivariable model adjusted for age at menarche (years, continuous), height (inches, continuous), parity/age at first birth (nulliparous, 1-2 children aged younger than 25 years, 1-2 children aged 25 years and older, 3+ children aged younger than 25 years, 3+ children aged 25 years and older), family history of breast cancer (yes/no), benign breast disease diagnosis (yes/no), alcohol intake (none, 1-4, 5-14, 15+ g/day), physical activity (MET-hours/week, quartiles), and smoking (never, past, current). In models that included both premenopausal and postmenopausal, we additionally adjusted for menopausal status (premenopausal or postmenopausal), and for postmenopausal women, we additionally adjusted for postmenopausal hormone use (never, past, current) and age at menopause (younger than 45, 45-49, 50-54, 55 years or older). BMI = body mass index; CI = confidence interval; HR = hazard ratio; MET = metabolic equivalent tasks; MV = multivariable.
Adjusted for same variables as in multivariable model (a) plus body mass index updated every 2 years (continuous).
Updated only when waist-and-hip circumference measured.
In pre-pre analyses (Table 2), WC was not associated with invasive breast cancer. HC was inversely associated with breast cancer; however, adjusting for BMI nullified results. WHR was positively associated after adjustment for BMI (quintile 5 vs 1: HRQ5vsQ1 = 1.27, 95% CI = 1.04 to 1.54; Ptrend = .01). Conversely, BMI was inversely associated (HRQ5vsQ1 = 0.75, 95% CI = 0.61 to 0.91; Ptrend = .004), even with WHR adjustment (data not shown).
In the pre-post analysis (Table 2), WC and HC associations with breast cancer were attenuated after adjusting for BMI. WHR was not associated with breast cancer, however, BMI was associated with a statistically significant 18% increase in breast cancer risk. In a sensitivity analysis, censoring women if they were postmenopausal at additional WC and HC measurements, BMI was no longer statistically significant, but other results remained consistent.
In post-post analyses (Table 2), all measures had positive associations (eg, WC HRQ5vsQ1 = 1.59, 95% CI = 1.36 to 1.86); however, after adjustment for BMI, only the WC and breast cancer relationship remained statistically significant (HRQ5vsQ1 = 1.38, 95% CI = 1.15 to 1.64; Ptrend = .002). BMI was similarly positively associated with risk, although this association was attenuated with adjustment for WC (HRQ5vsQ1 = 1.23, 95% CI = 1.03 to 1.45; data not otherwise shown).
In pre-pre analyses (Table 3), associations for WC and WHR varied by ER/PR status and for WC by molecular subtype (Pheterogeneity ≤ .05 for all). WC was inversely associated with ER+/PR- breast tumors (90th vs 10th percentile: HRP90vsP10 = 0.46, 95% CI = 0.22 to 0.96) and positively associated with ER-/PR- breast tumors (HRP90vsP10 = 1.99, 95% CI = 1.35 to 2.94). WC was positively associated with basal-like cancer (HRP90vsP10 = 3.58, 95% CI = 1.94 to 6.63). Similar associations were seen for WHR. Although heterogeneity was observed for ER/PR status (Pheterogeneity = .03), no strata were statistically significant.
Table 3.
Multivariable-adjusted hazard ratios and 95% confidence intervals for breast cancer subtypes by menopausal status at both anthropometric measure and diagnosis: Nurses’ Health Study (1986-2016) and Nurses’ Health Study II (1993-2017)
| Measure, tertiles | Pre-pre |
Pre-post |
Post-post |
||||||
|---|---|---|---|---|---|---|---|---|---|
| No. of cases | Continuous, 90th vs 10th percentilea | Continuous, 90th vs 10th percentilea | No. of cases | Continuous, 90th vs 10th percentilea | Continuous, 90th vs 10th percentilea | No. of cases | Continuous, 90th vs 10th percentilea | Continuous, 90th vs 10th percentilea | |
| MV-adjustedb | MV-adjusted and BMI adjustedc | MV-adjustedb | MV-adjusted and BMI adjustedc | MV-adjustedb | MV-adjusted and BMI adjustedc | ||||
| HR (95% CI) | HR (95% CI) | HR (95% CI) | HR (95% CI) | HR (95% CI) | HR (95% CI) | ||||
| Waist circumference, cm | |||||||||
| ER+, PR+ | 699 | 0.86 (0.72 to 1.04) | 1.13 (0.87 to 1.47) | 1074 | 1.08 (0.95 to 1.24) | 0.99 (0.83 to 1.18) | 1647 | 1.44 (1.29 to 1.61) | 1.27 (1.09 to 1.46) |
| ER+, PR- | 66 | 0.35 (0.17 to 0.73) | 0.46 (0.22 to 0.96) | 203 | 0.93 (0.68 to 1.26) | 0.85 (0.61 to 1.18) | 383 | 0.93 (0.72 to 1.19) | 0.81 (0.62 to 1.06) |
| ER-, PR- | 162 | 1.50 (1.07 to 2.10) | 1.99 (1.35 to 2.94) | 254 | 1.09 (0.83 to 1.43) | 1.00 (0.75 to 1.34) | 416 | 1.00 (0.79 to 1.26) | 0.88 (0.68 to 1.13) |
| Pheterogeneityd | <.001 | <.001 | .63 | .64 | <.001 | <.001 | |||
| Luminal A | 436 | 0.73 (0.57 to 0.93) | 0.98 (0.71 to 1.35) | 817 | 1.00 (0.86 to 1.17) | 0.89 (0.73 to 1.09) | 1089 | 1.28 (1.11 to 1.48) | 1.14 (0.95 to 1.36) |
| Luminal B | 190 | 0.84 (0.59 to 1.19) | 1.12 (0.75 to 1.69) | 324 | 1.23 (0.97 to 1.54) | 1.09 (0.84 to 1.42) | 415 | 1.40 (1.13 to 1.73) | 1.24 (0.98 to 1.58) |
| HER2-type | 33 | 1.43 (0.69 to 2.98) | 1.95 (0.91 to 4.21) | 76 | 0.90 (0.54 to 1.48) | 0.80 (0.48 to 1.33) | 99 | 0.88 (0.55 to 1.42) | 0.79 (0.48 to 1.27) |
| Basal-like | 47 | 2.60 (1.47 to 4.58) | 3.58 (1.94 to 6.63) | 25 | 2.31 (1.01 to 5.31) | 2.05 (0.88 to 4.74) | 91 | 1.04 (0.62 to 1.74) | 0.90 (0.53 to 1.53) |
| Pheterogeneityd | .001 | .001 | .13 | .13 | .29 | .28 | |||
| Hip circumference, cm | |||||||||
| ER+, PR+ | 699 | 0.80 (0.66 to 0.96) | 0.97 (0.73 to 1.29) | 1074 | 1.15 (1.00 to 1.32) | 1.07 (0.88 to 1.30) | 1647 | 1.36 (1.21 to 1.52) | 1.11 (0.94 to 1.31) |
| ER+, PR- | 66 | 0.54 (0.29 to 1.00) | 0.65 (0.33 to 1.25) | 203 | 0.85 (0.62 to 1.18) | 0.79 (0.56 to 1.13) | 383 | 1.01 (0.79 to 1.29) | 0.82 (0.63 to 1.08) |
| ER-, PR- | 162 | 1.23 (0.87 to 1.72) | 1.50 (1.00 to 2.26) | 254 | 1.10 (0.83 to 1.44) | 1.02 (0.75 to 1.39) | 416 | 0.98 (0.77 to 1.25) | 0.80 (0.62 to 1.04) |
| Pheterogeneityd | .03 | .03 | .23 | .23 | .01 | .01 | |||
| Luminal A | 436 | 0.66 (0.51 to 0.84) | 0.76 (0.54 to 1.08) | 817 | 1.05 (0.90 to 1.23) | 0.98 (0.78 to 1.22) | 1089 | 1.26 (1.09 to 1.46) | 1.01 (0.83 to 1.23) |
| Luminal B | 190 | 0.82 (0.58 to 1.16) | 0.95 (0.62 to 1.45) | 324 | 1.37 (1.09 to 1.72) | 1.27 (0.96 to 1.68) | 415 | 1.21 (0.96 to 1.51) | 0.96 (0.74 to 1.25) |
| HER2-type | 33 | 1.19 (0.56 to 2.51) | 1.39 (0.63 to 3.08) | 76 | 1.01 (0.61 to 1.68) | 0.94 (0.55 to 1.60) | 99 | 0.93 (0.57 to 1.51) | 0.75 (0.45 to 1.23) |
| Basal-like | 47 | 1.56 (0.85 to 2.86) | 1.83 (0.94 to 3.54) | 25 | 1.65 (0.68 to 4.00) | 1.54 (0.63 to 3.77) | 91 | 0.87 (0.51 to 1.48) | 0.69 (0.40 to 1.18) |
| Pheterogeneityd | .05 | .05 | .22 | .22 | .38 | .37 | |||
| Waist-to-hip ratio | |||||||||
| ER+, PR+ | 699 | 1.04 (0.88 to 1.23) | 1.11 (0.93 to 1.32) | 1074 | 0.97 (0.86 to 1.10) | 0.95 (0.83 to 1.08) | 1647 | 1.17 (1.07 to 1.29) | 1.13 (1.03 to 1.24) |
| ER+, PR- | 66 | 0.45 (0.24 to 0.87) | 0.48 (0.25 to 0.92) | 203 | 1.04 (0.78 to 1.38) | 1.01 (0.76 to 1.35) | 383 | 0.91 (0.73 to 1.12) | 0.86 (0.69 to 1.07) |
| ER-, PR- | 162 | 1.43 (1.07 to 1.93) | 1.50 (1.13 to 2.01) | 254 | 1.07 (0.84 to 1.38) | 1.05 (0.81 to 1.34) | 416 | 0.98 (0.81 to 1.20) | 0.94 (0.77 to 1.15) |
| Pheterogeneityd | .004 | .01 | .76 | .75 | .04 | .03 | |||
| Luminal A | 436 | 0.99 (0.80 to 1.24) | 1.08 (0.86 to 1.35) | 817 | 0.95 (0.82 to 1.10) | 0.92 (0.79 to 1.07) | 1089 | 1.09 (0.97 to 1.23) | 1.05 (0.93 to 1.18) |
| Luminal B | 190 | 0.97 (0.71 to 1.34) | 1.05 (0.76 to 1.44) | 324 | 0.96 (0.77 to 1.21) | 0.93 (0.74 to 1.17) | 415 | 1.25 (1.06 to 1.47) | 1.21 (1.02 to 1.43) |
| HER2-type | 33 | 1.25 (0.64 to 2.42) | 1.33 (0.70 to 2.53) | 76 | 0.89 (0.55 to 1.43) | 0.85 (0.53 to 1.38) | 99 | 0.85 (0.56 to 1.27) | 0.81 (0.54 to 1.22) |
| Basal-like | 47 | 2.07 (1.37 to 3.13) | 2.14 (1.43 to 3.20) | 25 | 1.68 (0.93 to 3.03) | 1.65 (0.90 to 2.99) | 91 | 1.15 (0.74 to 1.77) | 1.08 (0.69 to 1.68) |
| Pheterogeneityd | .04 | .05 | .42 | .40 | .29 | .27 | |||
The incremental units were based on the difference between the 90th percentile and 10th percentile. The increments are 28.6 cm for waist circumference, 24.8 cm for hip circumference, and 0.17 for waist-to-hip ratio. BMI = body mass index; CI = confidence interval; ER = estrogen receptor; HR = hazard ratio; MV = multivariable; PR = progesterone receptor.
Multivariable model adjusted for age at menarche (years, continuous), height (inches, continuous), parity/age at first birth (nulliparous, 1-2 children <25 years, 1-2 children 25+ years, 3+ children <25 years, 3+ children 25+ years), family history of breast cancer (yes/no), benign breast disease diagnosis (yes/no), alcohol intake (none, 1–4, 5–14, 15+ g/day), physical activity (MET-hours/week, quartiles), and smoking (never, past, current).
Adjusted for same variables as in multivariable model (a) plus BMI updated every 2 years (continuous).
Pheterogeneity was assessed using likelihood ratio tests comparing models that assumed a common association between each anthropometric measure and breast cancer subtypes by ER/PR status and by molecular subtype to models that allow for separate associations for each breast cancer subtype.
For pre-post breast cancer (Table 3), the associations for WC, HC, and WHR did not vary by ER/PR status (Pheterogeneity > .23 for all) or by tumor molecular subtype (Pheterogeneity > .13 for all). Hazard ratios were elevated for basal-like tumors, although only the association with WC was statistically significant; no associations were statistically significant in BMI-adjusted models.
For post-post breast cancer (Table 3), WC associations varied by ER/PR status (Pheterogeneity < .001), with a positive association among ER+/PR+ tumors (HRP90vsP10 = 1.27, 95% CI = 1.09 to 1.46) but no association with other hormone subtypes (eg, ER-/PR- HRP90vsP10 = 0.88, 95% CI = 0.68 to 1.13). Although no heterogeneity was observed by molecular subtypes, WC was positively associated with both luminal A and luminal B tumors but attenuated after BMI adjustment. For HC, there was statistically significant heterogeneity by ER/PR status (Pheterogeneity = .01) but no strata were statistically significant after BMI adjustment. Lastly, WHR associations varied by ER/PR status, with positive associations with ER+/PR+ (HRP90vsP10 = 1.13, 95% CI = 1.03 to 1.24) and luminal B tumor subtypes (HRP90vsP10 = 1.21, 95% CI = 1.02 to 1.43).
Associations of WC, HC, or WHR with breast cancer (pre-pre and post-post) did not vary by BMI or age. There was no interaction for WC and WHR with a family history of breast cancer regardless of menopausal status. In pre-pre analyses, there was an interaction between HC and family history of breast cancer (Pinteraction = .01); however, neither strata were statistically significant. For post-post, associations did not statistically significantly vary by family history or HT use. However, among never-HT users, higher WC was positively associated with breast cancer (HRT3vsT1 = 1.49, 95% CI = 1.12 to 1.97), whereas among ever-HT users, higher WC was not associated with risk (HRT3vsT1 = 1.14, 95% CI = 0.98 to 1.33; Pinteraction = .06). Associations with WHR were stronger among never-HT users, although the interaction was statistically nonsignificant (Pinteraction = .22).
Discussion
In this study, higher WHR was associated with increased premenopausal breast cancer risk, and larger postmenopausal WC was associated with increased postmenopausal breast cancer risk, independent of BMI. Specifically, higher WC and WHR were associated with a higher risk of ER-/PR- and basal-like premenopausal breast cancers. Higher WC and WHR were primarily associated with ER+/PR+ and possibly luminal B postmenopausal breast cancers.
Previously in NHS, HC was inversely associated with risk (10), whereas in NHS2, HC was null (11). In the current, much larger pooled analysis, with 4477 additional cases (pre-pre = 314) (10,11), no evidence of statistically significant study heterogeneity was found. Among postmenopausal women, there continued to be no association with HC, the positive trend for WHR was not statistically significant (10), and the association with WC was statistically significantly positive.
Most prospective cohorts that examined WC or WHR and premenopausal breast cancer controlling for BMI found similar positive associations (17–20). Among cohorts that did not account for BMI (21,22) or additionally reported associations unadjusted for BMI (17,20,23), statistically nonsignificant inverse associations were generally observed. In the 2 cohorts that examined HC, 1 did not control for BMI and found no association (22), whereas the second reported a positive association when including BMI (17), which we did not see.
No other study has examined premenopausal central adiposity measures and postmenopausal breast cancer. Prior studies have assessed menopause at either measurement or diagnosis but not at both. As the hazard ratio for basal-like breast cancers in the pre-post analysis were similar in magnitude to the pre-pre, albeit statistically nonsignificant, menopause status at exposure may be the more important feature for this subtype. The association between premenopausal BMI and postmenopausal breast cancer appeared intermediary between pre-pre and post-post associations.
Among postmenopausal women, many cohorts (22,24–28), but not all (18,19,21,23), observed similar positive associations for WC and/or WHR. A statistically significant inverse association for WC was observed using insurance data; however, they had limited control for breast cancer risk factors (20). Among cohorts where HC was examined, 2 reported a positive association (25,26) and 1 found no association (22), similar to our study; however, prior studies did not control for BMI.
Inconsistencies among cohorts may be due to variation in defining menopause status, study sizes and power to detect associations particularly among premenopausal women, and population characteristics (eg, age distributions, HT use).
Associations tended to be stronger for WC and WHR among never-HT users, although Pheterogeneity was not statistically significant. Similar associations have been observed in prior studies that observed statistically significant heterogeneity with WC (17,29,30), HC (17,29,30), or WHR (30), although other studies did not observe statistically significant heterogeneity for WC (21,23,24,27) or WHR (17,23,24,27,29). Because HT users generally have higher estradiol levels than nonusers, estrogen from adipose tissue might be less impactful (31), diluting potential associations.
Few prospective cohorts have examined tumor heterogeneity; those that have examined ER status (25,29,32) or ER/PR status (24,33,34) and 1 compared triple-negative breast cancer to ER+ cancers (29). Of these, fewer have examined subtypes in premenopausal women (24,33). The Sister Study found no association of WC or WHR with premenopausal ER+/PR+ cancers and had insufficient numbers to examine ER-PR- tumors (24). In the E3N cohort, HC was positively associated with premenopausal ER+/PR+ and ER-/PR- cancers adjusting for BMI, whereas WC and WHR were unrelated to risk, regardless of subtype (33). Conversely, potentially because of more premenopausal cases (927 in NHS cohorts with ER/PR status vs 277 in E3N), we observed statistically significant heterogeneity by ER/PR status for WC and WHR, with the strongest positive associations for ER-/PR-.
For postmenopausal breast cancers, WC (24,34) and WHR (24) have generally been positively associated with ER+/PR+ but null for ER-/PR- tumors (24,33,34), similar to our findings; conversely, in the E3N cohort, WC was null for both tumor subtypes (33). For HC, no statistically significant heterogeneity was observed by ER/PR status (33), and no association was noted among triple-negative postmenopausal breast cancers (29).
Two case-only analyses have examined whether measures of central adiposity vary by tumor molecular subtypes (35,36). In one, a larger WC was positively associated with HER2- and HER2+ luminal b tumors, but not triple-negative tumors or HER2+ nonluminal tumors, compared with luminal A tumors among premenopausal women (n = 596). No associations were seen among postmenopausal women (n = 1100) (35). A smaller case-only study found no association for WC and premenopausal tumors (n = 382) and no association among triple-positive or HER2-enriched tumors compared with luminal A tumors (36). They did report an association for larger WC and lower risk of triple-positive tumors (ie, ER+, HER2+, any Ki-67, any PR) compared with luminal B tumors (36). When we conducted a similar case-only analysis (reference = luminal A), we observed a stronger association of WC with basal-like tumors among premenopausal women and no difference by subtype in postmenopausal women (data not shown).
Contrasting BMI associations with pre- and postmenopausal breast cancer are hypothesized to be due to differing primary endogenous estrogen sources during these periods of life (ie, ovary vs adipose tissue) (37,38). Heavier postmenopausal women have higher estrogen and testosterone levels and lower sex hormone–binding globulin levels than leaner women, a pattern associated with higher postmenopausal breast cancer risk (39–42). Attenuation of the postmenopausal BMI–breast cancer association after controlling for endogenous estrogen levels further supports estrogen as a mediator (43,44). Similar results have been observed for WC and WHR (42,43). However, central obesity may be a better predictor of VAT than BMI (12), with 1 study observing associations between estrogen concentration and production in VAT and BMI and WC (45). The stronger association of WC and WHR among never-HT users and with ER+ and luminal breast cancers in this study suggests that central adiposity may increase postmenopausal breast cancer partly via estrogen levels.
Conversely, a higher premenopausal BMI is often associated with increased anovulation and lower sex hormone–binding globulin and estradiol levels compared with normal-weight women (37,46–48). Similar hormonal profiles have been observed for WC (49,50). However, our finding of an increased premenopausal breast cancer risk with greater central adiposity and stronger associations with ER-/PR- and basal-like subtypes suggests that the increased risk among premenopausal women is potentially estrogen independent.
Although WC does not differentiate between SAT and VAT (51,52) and correlates more strongly with SAT than VAT (51–54), WC (and WHR less so) is suggested to be a better surrogate for VAT than BMI (12,52–55). VAT contains more inflammatory cells, androgen receptors, and glucocorticoid receptors; is more metabolically active and insulin resistant; and has greater glucose uptake capacity than SAT (56). Thus, multiple mechanisms implicated in breast cancer (eg, inflammatory or insulin pathways) (57–62) may link WC with premenopausal breast cancer, and, to a lesser extent, postmenopausal breast cancer.
Our study had many strengths, including a large number of premenopausal women, the ability to comprehensively examine central adiposity and breast cancer risk by menopausal status and tumor subtypes, control for many breast cancer risk factors updated over follow-up, and multiple measures of BMI, WC, and HC. However, the small number of less common tumor subtypes is a limitation. We cannot exclude the possibility that residual or unknown confounders may have affected our results, nor can we exclude the possibility of chance findings because of multiple testing. Also, adjusting for BMI may be an overadjustment because of the high WC-BMI correlation; however, associations were still seen with WHR, which is not as correlated, and in sensitivity analyses using WC residuals (Supplementary Table 5, available online). Although self-reported, anthropometric measures correlated highly with technician measurements (14) and were collected prospectively, minimizing differential misclassification. Lastly, because both cohorts are predominately White (98%) and associations may vary by ethnicity (7), our results may not be generalizable to other ethnicities.
Our findings add to evidence suggesting that aspects of central adiposity, measured by WC or WHR, increase the risk of pre- and postmenopausal breast cancer independent of general adiposity, measured via BMI. Further, associations may vary by breast cancer subtype, although replication is needed. Additional studies in other racial and ethnic groups are needed, as are studies to elucidate the mechanism(s) underlying the central adiposity-breast cancer association, particularly among premenopausal women.
Funding
This work was supported by the National Cancer Institute at the National Institutes of Health (UM1 CA186107, U01 CA176726, P01 CA87969) and the Breast Cancer Research Foundation. SCH was supported through National Research Service Awards F32 CA224677 by the National Cancer Institute.
Notes
Role of the funder: The study sponsors had no role in the design of the study; the collection, analysis, and interpretation of the data; the writing of the manuscript; or the decision to submit the manuscript for publication.
Author contributions: Conceptualization: SCH, SEH. Formal analysis: SCH, SEH. Funding acquisition: SCH, HE, RMT, WCW, SEH. Investigation: SCH, SEH. Methodology: SCH, BAR, SEH. Resources/Data-curation: SCH, HE, RMT, WCW, SEH. Writing - original draft: SCH, SEH. Writing - review & editing: SCH, HE, RMT, WCW, BAR, SEH.
Disclosures: The authors have no conflicts of interest to disclose.
Prior presentations: Presented at the 2020 Virtual American Society for Preventative Oncology Conference, Virtual Poster Showcase, March 23-27, 2020.
Acknowledgments: We would like to thank the participants and staff of the Nurses’ Health Study and the Nurses’ Health Study II for their valuable contributions as well as the following state cancer registries for their help: AL, AZ, AR, CA, CO, CT, DE, FL, GA, ID, IL, IN, IA, KY, LA, ME, MD, MA, MI, NE, NH, NJ, NY, NC, ND, OH, OK, OR, PA, RI, SC, TN, TX, VA, WA, WY. The authors assume full responsibility for analyses and interpretation of these data.
Data Availability
Further information including the procedures to obtain and access data from the Nurses’ Health Studies and Health Professionals Follow-up Study is described at https://www.nurseshealthstudy.org/researchers (contact email: nhsaccess@channing.harvard.edu) and https://sites.sph.harvard.edu/hpfs/for-collaborators/.
Supplementary Material
References
- 1.American Cancer Society. Breast Cancer Facts & Figures 2017-2018. Atlanta, GA: American Cancer Society, Inc; 2017. [Google Scholar]
- 2.Rothman KJ.BMI-related errors in the measurement of obesity. Int J Obes. 2008;32(S3):S56–S59. [DOI] [PubMed] [Google Scholar]
- 3.Amadou A, Hainaut P, Romieu I.. Role of obesity in the risk of breast cancer: lessons from anthropometry. J Oncol. 2013;2013:906495. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.World Cancer Research Fund and American Institute for Cancer Research. Continuous Update Project Report: Diet, nutrition, physical activity and breast cancer; 2017:124.
- 5.Shuster A, Patlas M, Pinthus JH, Mourtzakis M.. The clinical importance of visceral adiposity: a critical review of methods for visceral adipose tissue analysis. Br J Radiol. 2012;85(1009):1–10. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Tchernof A, Després J-P.. Pathophysiology of human visceral obesity: an update. Physiol Rev. 2013;93(1):359–404. [DOI] [PubMed] [Google Scholar]
- 7.Chen G-C, Chen S-J, Zhang R, et al. Central obesity and risks of pre- and postmenopausal breast cancer: a dose-response meta-analysis of prospective studies: central obesity and breast cancer. Obes Rev. 2016;17(11):1167–1177. [DOI] [PubMed] [Google Scholar]
- 8.Freedman DS, Ford ES.. Are the recent secular increases in the waist circumference of adults independent of changes in BMI? Am J Clin Nutr. 2015;101(3):425–431. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Ryan MC, Fenster Farin HM, Abbasi F, Reaven GM.. Comparison of waist circumference versus body mass index in diagnosing metabolic syndrome and identifying apparently healthy subjects at increased risk of cardiovascular disease. Am J Cardiol. 2008;102(1):40–46. [DOI] [PubMed] [Google Scholar]
- 10.Huang Z, Willett WC, Colditz GA, et al. Waist circumference, waist: hip ratio, and risk of breast cancer in the Nurses’ Health Study. Am J Epidemiol. 1999;150(12):1316–1324. [DOI] [PubMed] [Google Scholar]
- 11.Harris HR, Willett WC, Terry KL, Michels KB.. Body fat distribution and risk of premenopausal breast cancer in the Nurses’ Health Study II. J Natl Cancer Inst. 2011;103(3):273–278. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Janssen I, Heymsfield SB, Allison DB, Kotler DP, Ross R.. Body mass index and waist circumference independently contribute to the prediction of nonabdominal, abdominal subcutaneous, and visceral fat. Am J Clin Nutr. 2002;75(4):683–688. [DOI] [PubMed] [Google Scholar]
- 13.Ngueta G, Laouan-Sidi EA, Lucas M.. Does waist circumference uncorrelated with BMI add valuable information? J Epidemiol Community Health. 2014;68(9):849–855. [DOI] [PubMed] [Google Scholar]
- 14.Rimm EB, Stampfer MJ, Colditz GA, Chute CG, Litin LB, Willett WC.. Validity of self-reported waist and hip circumferences in men and women. Epidemiology. 1990;1(6):466–473. [DOI] [PubMed] [Google Scholar]
- 15.Tamimi RM, Baer HJ, Marotti J, et al. Comparison of molecular phenotypes of ductal carcinoma in situ and invasive breast cancer. Breast Cancer Res. 2008;10(4):R67. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Collins LC, Marotti JD, Baer HJ, Tamimi RM.. Comparison of estrogen receptor results from pathology reports with results from central laboratory testing. J Natl Cancer Inst. 2008;100(3):218–221. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Lahmann PH, Hoffmann K, Allen N, et al. Body size and breast cancer risk: findings from the European Prospective Investigation into Cancer And Nutrition (EPIC). Int J Cancer. 2004;111(5):762–771. [DOI] [PubMed] [Google Scholar]
- 18.Sonnenschein E, Toniolo P, Terry MB, et al. Body fat distribution and obesity in pre- and postmenopausal breast cancer. Int J Epidemiol. 1999;28(6):1026–1031. [DOI] [PubMed] [Google Scholar]
- 19.Muti P, Stanulla M, Micheli A, et al. Markers of insulin resistance and sex steroid hormone activity in relation to breast cancer risk: a prospective analysis of abdominal adiposity, sebum production, and hirsutism (Italy). Cancer Causes Control CCC. 2000;11(8):721–730. [DOI] [PubMed] [Google Scholar]
- 20.Lee KR, Hwang IC, Han KD, Jung J, Seo MH.. Waist circumference and risk of breast cancer in Korean women: a nationwide cohort study. Int J Cancer. 2018;142(8):1554–1559. [DOI] [PubMed] [Google Scholar]
- 21.Catsburg C, Kirsh VA, Soskolne CL, et al. Associations between anthropometric characteristics, physical activity, and breast cancer risk in a Canadian cohort. Breast Cancer Res Treat. 2014;145(2):545–552. [DOI] [PubMed] [Google Scholar]
- 22.Kaaks R, Van Noord PA, Den Tonkelaar I, Peeters PH, Riboli E, Grobbee DE.. Breast-cancer incidence in relation to height, weight and body-fat distribution in the Dutch ‘DOM’ cohort. Int J Cancer. 1998;76(5):647–651. [DOI] [PubMed] [Google Scholar]
- 23.Palmer JR, Adams-Campbell LL, Boggs DA, Wise LA, Rosenberg LA.. Prospective study of body size and breast cancer in Black women. Cancer Epidemiol Biomark Prev. 2007;16(9):1795–1802. [DOI] [PubMed] [Google Scholar]
- 24.White AJ, Nichols HB, Bradshaw PT, Sandler DP.. Overall and central adiposity and breast cancer risk in the Sister Study. Cancer. 2015;121(20):3700–3708. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 25.Macinnis RJ, English DR, Gertig DM, Hopper JL, Giles GG.. Body size and composition and risk of postmenopausal breast cancer. Cancer Epidemiol Biomark Prev. 2004;13(12):2117–2125. [PubMed] [Google Scholar]
- 26.Guo W, Key TJ, Reeves GK.. Adiposity and breast cancer risk in postmenopausal women: results from the UK Biobank prospective cohort. Int J Cancer. 2018;143(5):1037–1046. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Krebs EE, Taylor BC, Cauley JA, Stone KL, Bowman PJ, Ensrud KE.. Measures of adiposity and risk of breast cancer in older postmenopausal women. J Am Geriatr Soc. 2006;54(1):63–69. [DOI] [PubMed] [Google Scholar]
- 28.Gaudet MM, Patel AV, Teras LR, et al. Obesity-related markers and breast cancer in CPS-II Nutrition Cohort. Int J Mol Epidemiol Genet. 2013;4(3):156–166. [PMC free article] [PubMed] [Google Scholar]
- 29.Phipps AI, Chlebowski RT, Prentice R, et al. Body size, physical activity, and risk of triple-negative and estrogen receptor-positive breast cancer. Cancer Epidemiol Biomarkers Prev. 2011;20(3):454–463. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30.Ahn J, Schatzkin A, Lacey JV, et al. Adiposity, adult weight change, and postmenopausal breast cancer risk. Arch Intern Med. 2007;167(19):2091–2102. [DOI] [PubMed] [Google Scholar]
- 31.Morimoto LM, White E, Chen Z, et al. Obesity, body size, and risk of postmenopausal breast cancer: the Women’s Health Initiative (United States). Cancer Causes Control CCC. 2002;13(8):741–751. [DOI] [PubMed] [Google Scholar]
- 32.Sellers TA, Davis J, Cerhan JR, et al. Interaction of waist/hip ratio and family history on the risk of hormone receptor-defined breast cancer in a prospective study of postmenopausal women. Am J Epidemiol. 2002;155(3):225–233. [DOI] [PubMed] [Google Scholar]
- 33.Fagherazzi G, Chabbert-Buffet N, Fabre A, et al. Hip circumference is associated with the risk of premenopausal ER-/PR- breast cancer. Int J Obes. 2012;36(3):431–439. [DOI] [PubMed] [Google Scholar]
- 34.Canchola AJ, Anton-Culver H, Bernstein L, et al. Body size and the risk of postmenopausal breast cancer subtypes in the California Teachers Study cohort. Cancer Causes Control. 2012;23(3):473–485. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 35.Agresti R, Meneghini E, Baili P, et al. Association of adiposity, dysmetabolisms, and inflammation with aggressive breast cancer subtypes: a cross-sectional study. Breast Cancer Res Treat. 2016;157(1):179–189. [DOI] [PubMed] [Google Scholar]
- 36.Crispo A, Montella M, Buono G, et al. Body weight and risk of molecular breast cancer subtypes among postmenopausal Mediterranean women. Curr Res Transl Med. 2016;64(1):15–20. [DOI] [PubMed] [Google Scholar]
- 37.Potischman N, Swanson CA, Siiteri P, Hoover RN.. Reversal of relation between body mass and endogenous estrogen concentrations with menopausal status. J Natl Cancer Inst. 1996;88(11):756–758. [DOI] [PubMed] [Google Scholar]
- 38.Tchernof A, Després JP.. Sex steroid hormones, sex hormone-binding globulin, and obesity in men and women. Horm Metab Res Horm Res. 2000;32(11/12):526–536. [DOI] [PubMed] [Google Scholar]
- 39.Key TJ, Appleby PN, Barnes I, et al. Endogenous sex hormones and breast cancer in postmenopausal women: reanalysis of nine prospective studies. J Natl Cancer Inst. 2002;94(8):606–616. [DOI] [PubMed] [Google Scholar]
- 40.Key TJ, Appleby PN, Reeves GK, et al. Body mass index, serum sex hormones, and breast cancer risk in postmenopausal women. J Natl Cancer Inst. 2003;95(16):1218–1226. [DOI] [PubMed] [Google Scholar]
- 41.McTiernan A, Wu L, Chen C, et al. Relation of BMI and physical activity to sex hormones in postmenopausal women. Obesity. 2006;14(9):1662–1677. [DOI] [PubMed] [Google Scholar]
- 42.Liedtke S, Schmidt ME, Vrieling A, et al. Postmenopausal sex hormones in relation to body fat distribution. Obesity. 2012;20(5):1088–1095. [DOI] [PubMed] [Google Scholar]
- 43.Rinaldi S, Key TJ, Peeters PHM, et al. Anthropometric measures, endogenous sex steroids and breast cancer risk in postmenopausal women: a study within the EPIC cohort. Int J Cancer. 2006;118(11):2832–2839. [DOI] [PubMed] [Google Scholar]
- 44.Dashti SG, Simpson JA, Karahalios A, et al. Adiposity and estrogen receptor-positive, postmenopausal breast cancer risk: quantification of the mediating effects of fasting insulin and free estradiol. Int J Cancer. 2020;146(6):1541–1552. [DOI] [PubMed] [Google Scholar]
- 45.Hetemäki N, Savolainen-Peltonen H, Tikkanen MJ, et al. Estrogen metabolism in abdominal subcutaneous and visceral adipose tissue in postmenopausal women. J Clin Endocrinol Metab. 2017;102(12):4588–4595. [DOI] [PubMed] [Google Scholar]
- 46.Zain MM, Norman RJ.. Impact of obesity on female fertility and fertility treatment. Womens Health. 2008;4(2):183–194. [DOI] [PubMed] [Google Scholar]
- 47.Tworoger SS, Eliassen AH, Missmer SA, et al. Birthweight and body size throughout life in relation to sex hormones and prolactin concentrations in premenopausal women. Cancer Epidemiol Biomark Prev Publ. 2006;15(12):2494–2501. [DOI] [PubMed] [Google Scholar]
- 48.Giviziez CR, Sanchez EGM, Approbato MS, Maia MCS, Fleury EAB, Sasaki RSA.. Obesity and anovulatory infertility: a review. JBRA Assist Reprod. 2016;20(4):240–245. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 49.Evans DJ, Hoffmann RG, Kalkhoff RK, Kissebah AH.. Relationship of androgenic activity to body fat topography, fat cell morphology, and metabolic aberrations in premenopausal women. J Clin Endocrinol Metab. 1983;57(2):304–310. [DOI] [PubMed] [Google Scholar]
- 50.Freeman EW, Sammel MD, Lin H, Gracia CR.. Obesity and reproductive hormone levels in the transition to menopause. Menopause. 2010;17(4):718–726. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 51.Bosy-Westphal A, Booke C-A, Blöcker T, et al. Measurement site for waist circumference affects its accuracy as an index of visceral and abdominal subcutaneous fat in a Caucasian population. J Nutr. 2010;140(5):954–961. [DOI] [PubMed] [Google Scholar]
- 52.Camhi SM, Bray GA, Bouchard C, et al. The relationship of waist circumference and BMI to visceral, subcutaneous, and total body fat: sex and race differences. Obesity. 2011;19(2):402–408. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 53.Gradmark AMI, Rydh A, Renström F, et al. Computed tomography-based validation of abdominal adiposity measurements from ultrasonography, dual-energy X-ray absorptiometry and anthropometry. Br J Nutr. 2010;104(4):582–588. [DOI] [PubMed] [Google Scholar]
- 54.Neamat-Allah J, Wald D, Hüsing A, et al. Validation of anthropometric indices of adiposity against whole-body magnetic resonance imaging–a study within the German European Prospective Investigation into Cancer and Nutrition (EPIC) cohorts. PloS One. 2014;9(3):e91586. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 55.Pouliot MC, Després JP, Lemieux S, et al. Waist circumference and abdominal sagittal diameter: best simple anthropometric indexes of abdominal visceral adipose tissue accumulation and related cardiovascular risk in men and women. Am J Cardiol. 1994;73(7):460–468. [DOI] [PubMed] [Google Scholar]
- 56.Ibrahim MM.Subcutaneous and visceral adipose tissue: structural and functional differences. Obes Rev. 2010;11(1):11–18. [DOI] [PubMed] [Google Scholar]
- 57.Renehan AG, Zwahlen M, Egger M.. Adiposity and cancer risk: new mechanistic insights from epidemiology. Nat Rev Cancer. 2015;15(8):484–498. [DOI] [PubMed] [Google Scholar]
- 58.Agurs-Collins T, Ross SA, Dunn BK.. The many faces of obesity and its influence on breast cancer risk. Front Oncol. 2019;9:765. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 59.Laudisio D, Muscogiuri G, Barrea L, Savastano S, Colao A.. Obesity and breast cancer in premenopausal women: current evidence and future perspectives. Eur J Obstet Gynecol Reprod Biol. 2018;230:217–221. [DOI] [PubMed] [Google Scholar]
- 60.Crespi E, Bottai G, Santarpia L.. Role of inflammation in obesity-related breast cancer. Curr Opin Pharmacol. 2016;31:114–122. [DOI] [PubMed] [Google Scholar]
- 61.Andò S, Gelsomino L, Panza S, et al. Obesity, leptin and breast cancer: epidemiological evidence and proposed mechanisms. Cancers. 2019;11(1):62. doi:10.3390/cancers11010062. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 62.Rose DP, Gracheck PJ, Vona-Davis L.. The interactions of obesity, inflammation and insulin resistance in breast cancer. Cancers. 2015;7(4):2147–2168. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
Further information including the procedures to obtain and access data from the Nurses’ Health Studies and Health Professionals Follow-up Study is described at https://www.nurseshealthstudy.org/researchers (contact email: nhsaccess@channing.harvard.edu) and https://sites.sph.harvard.edu/hpfs/for-collaborators/.