Abstract
Background
While obesity is an established risk factor for postmenopausal breast cancer, studies of weight loss and breast cancer provide are inconsistent. Therefore, we evaluated associations between weight change and breast cancer risk in postmenopausal women in the Women’s Health Initiative (WHI) Observational Study.
Methods
Postmenopausal women (n=61,335) with no prior breast cancer and normal mammogram had body weight and height measured and body mass index (BMI) calculated at baseline and year 3. Weight change at year 3 was categorized as: stable (change < 5%), loss (change ≥ 5%), or gain (change ≥ 5%) with further assessment of weight loss intentionality by self-report. Multi-variable Cox proportional hazards regression models were used to evaluate relationships between weight change and subsequent breast cancer incidence.
Results
During 11.4 years (mean) follow-up with 3,061 incident breast cancers, women with weight loss (n=8,175) had a significantly lower breast cancer risk compared with stable weight women (n=41,139) (hazard ratio [HR] 0.88 95% confidence interval [CI] 0.78-0.98, P= 0.02) with no interaction by BMI. Adjustment for mammography did not alter findings (HR 0.88 95% CI 0.78-0.99) with no significant difference by weight loss intentionality. Weight gain (≥ 5%) (n=12,021) was not associated with breast cancer risk (HR 1.02 95% CI 0.93-1.11) but was associated with higher triple negative breast cancer incidence (HR 1.54 95% CI 1.16-2.05).
Conclusions
Postmenopausal women who lost weight have lower breast cancer risk than those with stable weight. These findings suggest that postmenopausal women who lose weight may reduce their breast cancer risk.
Keywords: Breast cancer, obesity, weight loss, weight loss intentionality, Women’s Health Initiative
Introduction
Obesity is associated with increased risk of incident postmenopausal breast cancer1,2. As about one third of postmenopausal women in the US are obese, 3 obesity represents a common and potentially modifiable factor related to breast cancer outcome. However, it is not established that weight loss in postmenopausal women decreases breast cancer incidence or breast cancer mortality8.
Findings from observational studies in postmenopausal women examining weight loss and breast cancer incidence are inconsistent. Most recently, an analysis of 67,142 postmenopausal women in the Women’s Health Initiative (WHI) clinical trials found no association with subsequent breast cancer incidence in women with weight loss compared to women without weight loss5. These mixed findings preclude a strong public health message that postmenopausal women who are overweight or obese could reduce their breast cancer risk by losing weight.
The Take Off Pounds Sensibly (TOPS), a nationally available nonprofit weight loss program has demonstrated that weight loss above 5% of initial weight can be achieved and maintained for several years 6. Such modest magnitude of weight loss has been associated with health benefits including reduction in cardiovascular disease risk factors7, decreased progression of glucose intolerance to type II diabetes8 and decreased severity of obstructive sleep apnea9. Based on these findings, we evaluated the association of a similar magnitude of weight loss with breast cancer incidence in women participating in the WHI Observational Study.
Participants and Methods
Details of the design and conduct of the WHI Observational Study have been previously described10. Briefly, 93,676 women, ages 50–79 years with anticipated ≥ three year survival who were not eligible or interested in participating in the WHI clinical trials were recruited from 40 US clinical centers from 1993–1998. The study was approved by institutional review boards at the clinical centers and participants provided written consent.
Measures
Information on demographics and breast cancer risk were self-reported using questionnaires at baseline. Information on medication use was collected during interviews. Mammograms were not protocol mandated but information on mammography frequency was collected annually.
Measured height and weight were recorded at baseline and year 3 and were used to calculate body mass index (BMI, kg/m2). Percent weight change was calculated as measured weight at year 3 subtracted from measured weight at baseline divided by measured weight at baseline. The determined percent weight change variable categories were: weight gain (≥ 5% increase); weight loss (≥ 5% decrease); and weight stable (< ± 5% weight change). In addition, at the year 3 follow-up, participants were asked in a questionnaire, “In the past two years, did you gain or lose 5 or more pounds?”, “Was the change intentional or unintentional?” (yes/no). This information was used to categorize weight loss as intentional or unintentional,
Clinical outcomes were ascertained yearly and reported breast cancers were confirmed after medical record review by trained physician adjudicators at the clinical centers. Final adjudication and coding was performed at the WHI Clinical Coordinating Center11. Estrogen receptor (ER), progesterone receptor (PR) and Human Epidermal Growth Factor Receptor 2 (HER2) status was based on review of local laboratory reports.
Excluded from the Observational Study cohort (n=93,676) were: women with prior cancer (except non-melanoma skin cancer) (n=10,197); those with no follow-up information (n=422); or missing values for exposures of interest ; with BMI <18.5 kg/m2 and women who had breast cancer diagnosis, died or were lost to follow-up between baseline and year 3) (n=976). After exclusions, 61,335 women remained for the analysis (Figure 1).
Figure 1.
Flow Diagram of Participants Included in the Analysis
Incident invasive breast cancer diagnosed after the year 3 visit was the primary study outcome. Participants were observed from the year 3 visit to the first of either a breast cancer diagnosis, date of death, loss to follow-up or end of follow-up (September 30, 2015). Follow-up continued per original protocol through the original trial completion date of March 31, 2005. Ongoing follow-up required two additional written consents at 2005 and 2010, obtained from 73% of surviving participants on the first occasion and 88% on the second.
Statistical Analyses
Baseline characteristics are described by post-baseline 3-year weight change category (stable weight [<5% weight change]), weight loss [≥ 5% decrease] and weight gain [≥5% increase]). To estimate associations between weight change category and breast cancer incidence, hazard ratios (HRs) and corresponding 95% confidence intervals (CI) were generated using Cox proportional hazards models adjusted for relevant covariates. Additional analyses evaluated weight loss intentionality and breast cancer incidence similarly. All statistical tests were two-sided.
In multi-variable models, we adjusted for results from the Breast Cancer Risk Assessment Tool (BCRAT) (which included age at enrollment, race/ethnicity, age of menarche, age of the mother at the birth of her first live child, number of first-degree relatives with breast cancer, and the number of previous breast biopsies), education, smoking pack-years, recreational physical activity, alcohol consumption, history of hormone therapy use, parity, and baseline BMI. Weight change associations with breast cancer incidence were also examined by subtypes defined by baseline BMI category (overweight [25 kg/m2 ≤ 30 kg/m2], obese [≥30 kg/m2] versus normal weight [<25 kg/m2]), race/ethnicity, age group (50 - <70, ≥ 70 years) and menopausal hormone therapy use. Because weight change associations were examined in four subgroups, less than one would be accepted to be positive by chance. Sensitivity analysis adjusted for: frequency of mammography and excluded breast cancers diagnosed within the first year of follow-up (after year 3 visit).
Results
Baseline characteristics by weight change categories from baseline to year 3 are listed in Table 1. Compared with the women with stable weight, women who had ≥ 5% weight gain were more likely to be younger, Black, to be heavier smokers and to be younger at their first child’s birth. Compared with women with stable weight, women who had intentional weight loss were more likely to have higher BMI, but less likely to be physically active or have used any menopausal hormone therapy (all P <.01).
Table 1.
Baseline Characteristics of Women’s Health Initiative Observational Study Participants by weight change category
| Weight change (%) | ||||||||
|---|---|---|---|---|---|---|---|---|
| Stable weight (Change within 5%) (n=41,139) | Weight gain (>=5%) (n=12,021) | Unintentional weight loss (>=5%) (n=3346) | Intentional weight loss (>=5%) (n=4829) | |||||
| N | % | N | % | N | % | N | % | |
| Age at screening | 12569 | 30.6 | 5058 | 42.1 | 690 | 20.6 | 1567 | 32.4 |
| 50–59 | ||||||||
| 60–69 | 18871 | 45.9 | 5131 | 42.7 | 1433 | 42.8 | 2149 | 44.5 |
| 70–79 | 9699 | 23.6 | 1832 | 15.2 | 1223 | 36.6 | 1113 | 23.0 |
| Ethnicity | 136 | 0.3 | 57 | 0.5 | 16 | 0.5 | 22 | 0.5 |
| American Indian or Alaskan Native | ||||||||
| Asian or Pacific Islander | 1330 | 3.2 | 297 | 2.5 | 89 | 2.7 | 97 | 2.0 |
| Black or African-American | 2583 | 6.3 | 917 | 7.6 | 272 | 8.1 | 337 | 7.0 |
| Hispanic/Latino | 1285 | 3.1 | 393 | 3.3 | 109 | 3.3 | 142 | 2.9 |
| White (not of Hispanic origin) | 35250 | 85.7 | 10202 | 84.9 | 2818 | 84.2 | 4182 | 86.6 |
| Other | 555 | 1.3 | 155 | 1.3 | 42 | 1.3 | 49 | 1.0 |
| Education | 8063 | 19.6 | 2352 | 19.6 | 776 | 23.2 | 959 | 19.9 |
| HS diploma or less | ||||||||
| Some college/technical training | 14447 | 35.1 | 4517 | 37.6 | 1259 | 37.6 | 1847 | 38.2 |
| College graduate or more | 10350 | 25.2 | 2827 | 23.5 | 770 | 23.0 | 1095 | 22.7 |
| Master or more | 8279 | 20.1 | 2325 | 19.3 | 541 | 16.2 | 928 | 19.2 |
| Pack-year of cigarette smoking | 1180 | 2.9 | 375 | 3.1 | 87 | 2.6 | 147 | 3.0 |
| Never smoking | 21551 | 52.4 | 5800 | 48.2 | 1736 | 51.9 | 2376 | 49.2 |
| 8358 | 20.3 | 2393 | 19.9 | 576 | 17.2 | 959 | 19.9 | |
| 3523 | 8.6 | 1125 | 9.4 | 268 | 8.0 | 391 | 8.1 | |
| 2217 | 5.4 | 777 | 6.5 | 203 | 6.1 | 290 | 6.0 | |
| 1831 | 4.5 | 635 | 5.3 | 156 | 4.7 | 261 | 5.4 | |
| Alcohol intake | 4230 | 10.3 | 1222 | 10.2 | 443 | 13.2 | 527 | 10.9 |
| Non drinker | ||||||||
| Past drinker | 6740 | 16.4 | 2311 | 19.2 | 689 | 20.6 | 913 | 18.9 |
| Current drinker | 4528 | 11.0 | 1501 | 12.5 | 406 | 12.1 | 614 | 12.7 |
| History of hormone therapy use | 15545 | 37.8 | 4168 | 34.7 | 1413 | 42.2 | 1933 | 40.0 |
| Never hormone therapy use | ||||||||
| E-alone use | 12733 | 31.0 | 3765 | 31.3 | 1129 | 33.7 | 1480 | 30.6 |
| E+P use | 10091 | 24.5 | 3288 | 27.4 | 624 | 18.6 | 1120 | 23.2 |
| E-alone and E+P use | 2770 | 6.7 | 800 | 6.7 | 180 | 5.4 | 296 | 6.1 |
| Age at menarche, years (mean) | 6570 | 16.0 | 1942 | 16.2 | 544 | 16.3 | 806 | 16.7 |
| Number of Term Pregnancies | 4025 | 9.8 | 1211 | 10.1 | 354 | 10.6 | 518 | 10.7 |
| Never pregnant | ||||||||
| Never had term pregnancy | 1041 | 2.5 | 352 | 2.9 | 90 | 2.7 | 123 | 2.5 |
| 1 | 3507 | 8.5 | 1111 | 9.2 | 315 | 9.4 | 455 | 9.4 |
| 2 | 10863 | 26.4 | 3354 | 27.9 | 832 | 24.9 | 1224 | 25.3 |
| 3 or more | 10302 | 25.0 | 2856 | 23.8 | 749 | 22.4 | 1171 | 24.2 |
| Age at First Birth | 3615 | 8.8 | 1017 | 8.5 | 279 | 8.3 | 401 | 8.3 |
| Never had term pregnancy | 5066 | 12.3 | 1563 | 13.0 | 444 | 13.3 | 641 | 13.3 |
| < 20 | 3970 | 9.7 | 1523 | 12.7 | 416 | 12.4 | 532 | 11.0 |
| 20–29 | 25237 | 61.3 | 7115 | 59.2 | 1949 | 58.2 | 2873 | 59.5 |
| 30+ | 3251 | 7.9 | 803 | 6.7 | 258 | 7.7 | 382 | 7.9 |
| Hysterectomy ever | 42 | 0.1 | 6 | 0.0 | 2 | 0.1 | 3 | 0.1 |
| No | 24976 | 60.7 | 7207 | 60.0 | 1865 | 55.7 | 2912 | 60.3 |
| Yes | 16121 | 39.2 | 4808 | 40.0 | 1479 | 44.2 | 1914 | 39.6 |
| Bilateral oophorectomy | 29693 | 72.2 | 8647 | 71.9 | 2257 | 67.5 | 3463 | 71.7 |
| No | ||||||||
| Yes, both were taken out | 7675 | 18.7 | 2301 | 19.1 | 699 | 20.9 | 940 | 19.5 |
| Yes, one or part of an ovary removed | 3407 | 8.3 | 965 | 8.0 | 339 | 10.1 | 387 | 8.0 |
| Diabetes treated (pills or shots) | 51 | 0.1 | 12 | 0.1 | 5 | 0.1 | 1 | 0.0 |
| No | 39945 | 97.1 | 11606 | 96.5 | 3122 | 93.3 | 4626 | 95.8 |
| Yes | 1143 | 2.8 | 403 | 3.4 | 219 | 6.5 | 202 | 4.2 |
| Female relative had breast cancer | 22266 | 54.1 | 6470 | 53.8 | 1829 | 54.7 | 2540 | 52.6 |
| No | 11443 | 27.8 | 3353 | 27.9 | 912 | 27.3 | 1408 | 29.2 |
| Yes | 7430 | 18.1 | 2198 | 18.3 | 605 | 18.1 | 881 | 18.2 |
| Total energy expenditure (MET-hrs/wk) | 14.6 | 14.1 | 11.5 | 13.2 | ||||
| Body mass index (BMI, kg/m2) | 26.7 | 26.7 | 27.9 | 29.9 | ||||
Some categories do not add up to the total number of cases due to missing data
P values for all difference tests were <0.01 except for “Female relative had breast cancer“(p=0.4)
The mean weight loss from baseline to year 3 for the 4,829 women reporting intentional weight loss was 19.56 pounds (SD 27.12) while the mean weight loss for the 3,346 women reporting unintentional weight loss was 16.90 pounds (SD 18.69). Compared to weight at year 3, self-reported weight at year 6 for the intentional weight loss group was +2.55 pounds (SD13.68) and for unintentional weight loss group was +1.82 pounds (SD 12.03). Thus, most of early weight loss was maintained through year 6 (Appendix Table 1).
During 11.4 years (mean) of follow-up from the year 3 visit, 3,061 women developed invasive breast cancer. The mean time from year 3 weight determination to breast cancer diagnosis was 6.47 years (range 0.005 −17.0 years). In multi-variable analyses, compared with women with stable weight (n=41,139), women with weight loss (≥ 5%) (n=8,175) had a significantly lower breast cancer risk (hazard ratio [HR] 0.88 95% confidence interval [CI] 0.78–0.98, p=0.02) (Table 2). Adjustment for mammography did not alter findings (HR 0.88 95% CI 0.78–0.99). Statistical test between intentional and unintentional weight loss and breast cancer incidence found no significant difference (P= 0.2) with HRs < 1 for both intentional and unintentional weight loss and breast cancer risk. Results were also similar after excluding breast cancers diagnosed within the first year of follow-up (after the year 3 visit) and in analyses adjusting for mammogram frequency. Weight gain (≥ 5%) (n=12,021) was not associated with higher overall breast cancer risk but was associated with higher risk of triple negative breast cancer (HR 1.54 95% CI 1.16–2.05).
Table 2:
Hazard ratios (HR) and 95% CIs for the association between weight change and risk of breast cancer
| Cases | Age-adjusted HR (95CI) | Multivariable-adjusteda HR (95% CI) | |
|---|---|---|---|
| Weight change between baseline and year 3 (percentage change) | |||
| Stable weight (within ±5% change) | 2092 | Reference | Reference |
| Weight gain (≥5%) | 620 | 1.03 (0.94 1.13) | 1.02 (0.93 1.11) |
| Weight loss (≥5%) | 349 | 0.92 (0.82 1.03) | 0.88 (0.78 0.98) |
| Intentional | 229 | 0.97 (0.85 1.11) | 0.91 (0.79 1.04) |
| Unintentional | 120 | 0.84 (0.70 1.01) | 0.82 (0.68 0.99) |
In multivariable models, we adjusted for Gail score, education, smoking pack-year, recreational physical activity, alcohol, history of hormone therapy use, parity, and BMI.
Associations reflect findings for overall weight change.
Breast cancer characteristics are shown in appendix Table 2e with associations (HR 95% CI) shown in Table 3. The risk of triple negative breast cancers was highest in the weight gain group (HR 1.54 95% CI 1.16–2.05). Weight change association with breast cancer incidence was examined in four subgroups: baseline BMI (normal, overweight, obese), race/ethnicity, age group (50≤70, 70+ years) and hormone therapy use and findings are shown in Table 4. No interaction term test was statistically significant.
Table 3.
Characteristics of invasive breast cancer (n=3,061) by weight change category*
| Weight Loss | ||||||
|---|---|---|---|---|---|---|
| Characteristic | Stable Weight N=2092 (%) | Weight Gain N=620 (%) | Intentional N=229 (%) | Unintentional N=120 (%) | Weight Loss (overall) N=349 (%) | P value |
| Histology | 0.7 | |||||
| Ductal | 62.9 | 64.5 | 62.4 | 72.5 | 65.9 | |
| Lobular | 11.8 | 10.0 | 11.4 | 8.3 | 10.3 | |
| Ductal and lobular | 14.1 | 14.2 | 15.3 | 10.8 | 13.8 | |
| Other | 11.3 | 11.3 | 10.9 | 8.3 | 10.0 | |
| Estrogen receptor status | 0.3 | |||||
| Positive | 82.6 | 81.3 | 80.3 | 85.8 | 82.2 | |
| Negative | 12.8 | 15.3 | 16.2 | 10.0 | 14.0 | |
| Progesterone receptor status | 0.4 | |||||
| Positive | 69.5 | 69.8 | 62.9 | 70.8 | 65.6 | |
| Negative | 24.7 | 25.6 | 30.6 | 25.0 | 28.7 | |
| Estrogen/progesterone receptor status | 0.07 | |||||
| ER+PR+ | 68.2 | 69.2 | 61.6 | 70.8 | 64.8 | |
| ER+PR− | 13.3 | 11.1 | 16.2 | 15.0 | 15.8 | |
| ER-PR− | 11.3 | 14.5 | 14.4 | 10.0 | 12.9 | |
| HER2 overexpression | 10.9 | 11.0 | 10.9 | 10.8 | 10.9 | 0.4 |
| Triple-negative tumor | 7.2 | 11.0 | 9.6 | 6.7 | 8.6 | 0.02 |
| Stage | 0.2 | |||||
| Local | 76.4 | 74.7 | 70.7 | 75.0 | 72.2 | |
| Regional or distant | 22.7 | 24.8 | 28.4 | 22.5 | 26.4 | |
| Grading | 0.5 | |||||
| Well differentiated | 27.4 | 27.1 | 27.5 | 23.3 | 26.1 | |
| Moderately differentiated | 44.3 | 43.7 | 43.2 | 51.7 | 46.1 | |
| Poorly differentiated | 20.2 | 23.4 | 21.8 | 17.5 | 20.3 | |
Some categories do not add up to 100% due to missing data
Association (HR 95% CI) between weight change category and subtypes of breast cancer are outlined in Appendix Table 1e.
Table 4.
Association (HR 95% CI) between weight change category and breast cancer risk by stratifying variable*
| Weight Loss | ||||||
|---|---|---|---|---|---|---|
| Stable Weight | Weight gain | Intentional | Unintentional | Weight loss (overall) | Interaction P value | |
| Baseline BMI category | 0.4 | |||||
| Normal weight | 1.0 (ref) | 0.92 (0.80 1.06) | 0.94 (0.72 1.24) | 0.70 (0.50 0.96) | 0.82 (0.66 1.02) | |
| Over weight | 1.0 (ref) | 1.10 (0.95 1.28) | 1.00 (0.80 1.25) | 0.97 (0.72 1.30) | 0.99 (0.82 1.19) | |
| Obese | 1.0 (ref) | 1.07 (0.88 1.29) | 0.86 (0.68 1.08) | 0.82 (0.58 1.16) | 0.85 (0.69 1.03) | |
| Race/Ethnicity | 0.9 | |||||
| White | 1.0 (ref) | 1.02 (0.93 1.13) | 0.92 (0.80 1.07) | 0.83 (0.68 1.00) | 0.89 (0.79 1.00) | |
| African-American | 1.0 (ref) | 1.16 (0.78 1.74) | 0.68 (0.33 1.42) | 0.86 (0.37 1.99) | 0.75 (1.42 1.33) | |
| Hispanic/Latina | 1.0 (ref) | 0.89 (0.44 1.80) | 1.06 (0.39 2.85) | - | 0.56 (0.21 1.50) | |
| Asian-American | 1.0 (ref) | 0.47 (0.19 1.17) | 0.48 (0.14 1.70) | 1.00 (0.31 3.27) | 0.67 (0.27 1.65) | |
| American Indian/Alaska Native | 1.0 (ref) | 3.13 (0.45 21.86) | - | 9.93 (0.45 218.74) | 0.15 (0.007 2.99) | |
| Other | 1.0 (ref) | 1.25 (0.48 3.24) | 1.52 (0.42 5.46) | 2.86 (0.60 13.58) | 1.88 (0.66 5.29) | |
| Age group | 0.8 | |||||
| 50–<70 | 1.0 (ref) | 1.03 (0.93 1.13) | 0.89 (0.76 1.04) | 0.77 (0.61 0.97) | 0.85 (0.74 0.97) | |
| 70+ | 1.0 (ref) | 0.97 (0.77 1.23) | 0.99 (0.73 1.35) | 0.91 (0.67 1.25) | 0.95 (0.76 1.20) | |
| Hormone use | 0.6 | |||||
| Never | 1.0 (ref) | 0.99 (0.85 1.16) | 0.80 (0.63 1.02) | 0.78 (0.58 1.06) | 0.79 (0.65 0.96) | |
| E-alone | 1.0 (ref) | 1.14 (0.96 1.34) | 1.02 (0.79 1.32) | 0.92 (0.66 1.27) | 0.98 (0.79 1.21) | |
| E+P | 1.0 (ref) | 0.95 (0.81 1.12) | 0.97 (0.75 1.24) | 0.79 (0.54 1.15) | 0.91 (0.73 1.13) | |
Subgroup analyses of hazard ratios (HRs and 95% confidence interval CI) by weight change category for breast cancer incidence by subgroups of participants
Discussion
In this large prospective study of postmenopausal women, compared to women with stable weight, women with weight loss of ≥ 5% were at a lower risk for invasive breast cancer. There was no significant interaction between weight loss and intentionality of weight loss with respect to breast cancer risk.
The current findings are consistent with those from bariatric surgery where the substantial weight loss associated with the procedure in severely obese women has been associated with lower breast cancer risk, supporting the benefits of substantial (usually > 20 kg) weight loss12. Our current findings suggest benefit for smaller degrees of weight loss achievable without surgery.
The association of weight loss with reduction in breast cancer risk in postmenopausal women is supported by findings from the WHI Dietary Modification Trial13, where 48,835 women were randomized to a dietary intervention based on a low-fat eating plan or to a usual diet control in a breast cancer primary prevention trial which were recently updated14, 15. During the 8.5 years (mean) dietary intervention period, women in the dietary intervention group, compared to women in the usual diet control group, significantly reduced their fat intake which was associated with a statistically significant, 3% weight loss. Now, after 16.1 years (mean) cumulative follow-up with 3,030 incident breast cancers, deaths after breast cancer were significantly reduced in the dietary intervention group, measured from randomization14 or from breast cancer diagnosis15.
The current finding that weight loss determined over a relatively short 3 year period was associated with a significantly lower breast cancer incidence differs somewhat from findings previously reported from analyses in the separate WHI Clinical Trials population where weight loss was not associated with lower breast cancer incidence 5. The demographic characteristics of the WHI Clinical Trial and Observational Study participants are comparable. However, findings in the WHI Clinical Trial report are based on serial body weight measures obtained annually for about 12 years from clinical trial initiation in 1993 through 2005. As a group, women in the US lose weight beginning in the seventh decade of life16. As 23% of women in the WHI Clinical Trials were between 70–79 years of age at entry between 1993–1998, the report by Neuhouser and colleagues16 included a much larger contribution of body weight determinations in older women who more commonly experience involuntary weight loss, for both medical and social reasons (such as death of spouse, change in economic status, poor oral health, and activities of daily living disability 17). Thus, the finding from the WHI Clinical Trial which incorporated cumulative weight change over two decades are not directly comparable to the current report on relatively short term weight change in a younger population.
The current analyses examined weight loss by intentionality since unintentional weight loss, especially in older individuals, is correlated with increased morbidity and mortality and may contribute to reverse causality17. In previous analyses in the same WHI OS study cohort 18, overall weight loss ≥ 5% was associated with significantly lower endometrial cancer risk (HR 0.71, 95% CI 0.54–0.95). However, for endometrial cancer, the association was stronger in women reporting intentional (0.60 95% CI 0.42–0.86) compared to unintentional weight loss (0.94 95% CI 0.62–1.41)18. The findings regarding weight loss intentionality appears different from that seen for breast cancer in the current report. However, in the prior endometrial cancer report and the current breast cancer report, the HRs for both cancers were < 1 for intentional as well as unintentional weight loss and no significant interaction was seen. Taken together, these findings suggest additional larger studies are needed to provide more definitive answer to role of intentional, as compared to unintentional, weight loss on cancer risk.
To our review, weight loss intentionality and breast cancer risk has only been addressed in only one previous prospective study, the Iowa Women’s Health Study19. That study used recalled weight change for both prior and current weight over a 35-year period and observed that women with intentional weight loss ≥20 lbs. had 19% lower breast cancer risk (RR 0.81, 95% CI 0.66–1.00) while unintentional weight loss was not associated with lower breast cancer risk19. In contrast to recalled weight change in that report, in the current study, we evaluated serial measured weights with intentionality over a relatively short, defined 3 year period.
Study strengths include the large study population of postmenopausal women, adjustment for breast cancer risk factors, serial measured body weights, adjudicated breast cancers, weight loss intentionality linked to body weight change, long follow-up and large number of breast cancer cases. A limitation is having measured weight at baselines and year 3 only, but by self-report at year 6, most of the weight loss was sustained.
Our study has limitations. The observational study design precludes causal inference. As we have no strong biological rationale for the association between weight gain and increasing only triple negative breast cancer, we now suggest this finding should be interpreted with caution. While measured weight loss was available for a relatively short period prior to breast cancer diagnosis, there is evidence that short term interventions may affect undiagnosed cancer which influence subsequent health outcomes 5 to 10 years later. Results from a WHI randomized trial evaluating menopausal hormone therapy found that 5.6 years (mean) use of combined estrogen plus progestin significantly increased breast cancer risk which persisted through 13 years follow-up (HR 1.37 95% CI 1.06–1.77 20. Also, in a randomized, placebo-controlled breast cancer primary prevention trial (MAP.3), after median follow-up of only 36 months, compared to placebo, the aromatase inhibitor exemestane reduced breast cancer incidence by 65% (P=0.002) 21.
In terms of randomized weight reduction trials in populations without cancer, a recent meta-analysis of 45 trials found that behavioral interventions that incorporate both diet and physical activity can achieve significant, maintained weight loss up to 24 months22. Taken together with current study findings, these results suggest that a short-term lifestyle intervention may similarly reduce the progression of undiagnosed breast cancers. The similar findings of lower endometrial cancer risk with weight loss determined in the same 3 year period18 also supports the current results associating weight loss in postmenopausal women with lower breast cancer risk.
Conclusions
Weight loss in postmenopausal women is associated with lower breast cancer risk. These findings suggest that interventions in postmenopausal women designed to generate weight loss may result in a reduction in breast cancer risk.
Women’s Health Initiative Investigators
Program Office
(National Heart, Lung, and Blood Institute, Bethesda, MD) Jacques Roscoe, Shari Ludlum, Dale Burden, Joan McGowan, Leslie Ford, and Nancy Geller
Clinical Coordinating Center
(Fred Hutchinson Cancer Research Center, Seattle, WA) Garnet Anderson, Ross Prentice, Andrea LaCroix, and Charles Kooperberg
Investigators and Academic Centers
(Brigham and Women’s Hospital, Harvard Medical School, Boston, MA) JoAnn E Manson; (MedStar Health Research Institute/Howard University, Washington, DC) Barbara V Howard; (Stanford Prevention Research Center, Stanford, CA) Marcia L. Stefanick; (The Ohio State University, Columbus, OH) Rebecca Jackson; (University of Arizona, Tucson/Phoenix, AZ) Cynthia A. Thompson; (University at Buffalo, Buffalo, NY) Jean Wactawski-Wende; (University of Florida, Gainesville/Jacksonville, FL) Marian Limacher; (University of Iowa, Iowa City/Davenport, IA) Robert Wallace; (University of Pittsburgh, Pittsburgh, PA) Lewis Kuller; (City of Hope National Medical Center, Duarte, CA) Rowan T. Chlebowski; (Wake Forest University School of Medicine, Winston-Salem, NC) Sally Shumaker
Supplementary Material
Funding/Support
The WHI program is supported by the National Heart, Lung and Blood Institute, National Institutes of Health, Department of Health and Human Services through contracts N01WH22110, 24152, 32100-2, 32105-6, 32108-9, 32111-13, 32115, 32118-32119, 32122, 42107-26, 42129-32, and 44221. This report was additionally funded by American Institute for Cancer Research Grant 30210-01 (RTC).
Footnotes
Conflict of Interest Disclosures
Dr. Chlebowski reported being a consultant for AstraZeneca, Novartis and Pfizer. No other author reported conflicts.
Findings were presented, in part, at the San Antonio Breast Cancer Symposium on December 8, 2017
Anatomic site: Breast cancer
Condensed abstract: In a large prospective study of postmenopausal women, women with weight loss of ≥5% were at a lower risk for invasive breast cancer compared to women with stable weight.
Contributor Information
Rowan T. Chlebowski, City of Hope National Medical Center, Duarte, CA.
Juhua Luo, Indiana University Bloomington, Bloomington, IN
Garnet L. Anderson, Fred Hutchinson Cancer Research Center, Seattle, WA
Wendy Barrington, University of Washington, Seattle WA
Kerryn Reding, University of Washington, Seattle, WA
Michael S. Simon, Karmanos Cancer Institute, Detroit, MI.
JoAnn E. Manson, Brigham and Women’s Health Hospital, Harvard Medical School, Boston
Thomas E. Rohan, Albert Einstein College of Medicine, Bronx, NY
Jean Wactawski-Wende, University at Buffalo, SUNY, Buffalo, NY
Dorothy Lane, Stony Brook University School of Medicine, Stony Brook, NY
Howard Strickler, Albert Einstein College of Medicine, Bronx, NY
Yasmin Mosaver-Rahmani, Albert Einstein College of Medicine, Bronx, NY
Jo L. Freudenheim, University at Buffalo, SUNY, Buffalo, NY
Nazmus Saquib, Sulaimaon Al Rajhi College School of Medicine, Al Bukayri ah, Saudi Arabia
Marcia L. Stefanick, Stanford University School of Medicine, Stanford, CA
References
- 1.Chlebowski RT, Aiello E, McTiernan A. Weight loss in breast cancer patient management. J Clin Oncol. 2002; 20:1128–43. [DOI] [PubMed] [Google Scholar]
- 2.Protani M, Coory M, Martin JH. Effect of obesity on survival of women with breast cancer: systematic review and meta-analysis. Breast Cancer Res Treat. 2010; 123:627–35. [DOI] [PubMed] [Google Scholar]
- 3.Ogden CL, Carroll MD, Kit BK, et al. Global, regional and national prevalence of overweight and obesity in children and adults during 1980–2013: a systematic analysis for the Global Burden of Disease Study 2013. Lancet. 2014; 384:766–81. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Lauby-Secretan B, Scoccianti C, Loomis D, et al. Body fatness and cancer – viewpoint of the IARC Working Group. N Engl J Med. 2016; 375:794–8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Neuhouser ML, Aragaki AK, Prentice RL, et al. Overweight, obesity and postmenopausal invasive breast cancer risk. A secondary analysis of the Women’s Health Initiative randomized clinical trials. JAMA Oncol. 2015; 1:611–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.Mitchell NS, Polsky S, Catenacci VA, Furniss AL, Prochazka AV. Up to 7 years sustained weight loss for weight-loss program completers. Am J Prev Med. 2015; 49(2):248–58. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 7.Pi-Sunyer X, Blackburn G, Brancati FL, et al. Look AHEAD Research Group. Reduction in weight and cardiovascular disease risk factors in individuals with type 2 diabetes: one-year results of the look AHEAD trial. Diabetes Care. 2007; 30(6):1374–1383. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8.Haw JS, Galaviz KI, Straus AN, et al. Long-term sustainability of diabetes prevention approaches: A systematic review and meta-analysis of randomized clinical trials. JAMA Intern Med. 2017. [Epub ahead of print]. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9.Foster GD, Borrandaile KE, Sanders MH, et al. Sleep AHEAD Research Group of Look AHEAD Research Group. A randomized study on the effect of weight loss on obstructive sleep apnea among obese patients with type 2 diabetes: the Sleep AHEAD study. Arch Intern Med. 2009; 169:1619–1626. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.The Women’s Health Initiative Study Group. Design of the Women’s Health Initiative clinical trial and observational study. Control Clin Trials. 1998; 19:61–109. [DOI] [PubMed] [Google Scholar]
- 11.National Cancer Institute. Surveillance Epidemiology and End Results. http://www.seer.cancer.gov (assessed July 2, 2017).
- 12.McCauley GM, Ferriss JS, Geffel D, et al. Cancer in obese women: potential protective impact of bariatric surgery. J Am Coll Surg. 2009; 208:1093–8. [DOI] [PubMed] [Google Scholar]
- 13.Prentice RL, Caan B, Chlebowski RT, et al. Low-fat dietary pattern and risk of invasive breast cancer: the Women’s Health Initiative randomized controlled dietary modification trial. JAMA 2006; 295:629–42. [DOI] [PubMed] [Google Scholar]
- 14.Chlebowski RT, Aragaki AK, Anderson GL, et al. Low-fat dietary pattern and breast cancer mortality in the Women’s Health Initiative (WHI) randomized controlled trial. J Clin Oncol. 2017; 35(25):2919–2926. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Chlebowski RT, Aragaki AK, Anderson GL, et al. Association of low-fat dietary pattern with breast cancer overall survival: A secondary analysis of the Women’s Health Initiative randomized clinical trial. JAMA Oncology 2018. March 26 (in press). [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Newman AD, Yanez D, Harris T, et al. Weight change in older age and its association with mortality. J Am Geriatr Soc. 2001; 49(10):1309. [DOI] [PubMed] [Google Scholar]
- 17.Wijnhoven HA, van Zon SK, Twisk J, Visser M. Attribution of causes of weight loss and weight gain to three-year mortality in older adults: results from the Longitudinal Aging Study Amsterdam. J Gerontol A Biol Med Sci. 2014; 69:1236–43. [DOI] [PubMed] [Google Scholar]
- 18.Luo J, Chlebowski RT, Hendryx M, et al. Voluntary weight loss and endometrial cancer risk. J Clin Oncol. 2017; 35:1189–1193Reeves GK, Pirie K, Beral V, et al. Cancer incidence and mortality in relation to body mass index in the Million Women Study: cohort study. BMJ. 2007; 335:1134. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Parker ED, Folsom AR. Intentional weight loss and incidence of obesity-related cancers: the Iowa Women’s Health Study. Int J Obes Relat Metab Disord. 2003; 27:1447–52. [DOI] [PubMed] [Google Scholar]
- 20.Chlebowski RT, Rohan TE, Manson JE, et al. Breast cancer after use of estrogen plus progestin and estrogen alone: analyses of data from 2 Women’s Health Initiative Randomized clinical trials. JAMA Oncol. 2015; 1:296–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Chlebowski RT, Rohan TE, Manson JE, et al. Breast cancer after use of estrogen plus progestin and estrogen alone: analyses of data from 2 Women’s Health Initiative Randomized clinical trials. JAMA Oncol. 2015; 1:296–305. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Dombrowski SU, Knittle K, Avenell A, Araujo-Soares V, Sniehotta FF. Long term maintenance of weight loss with non-surgical interventions in obese adults: systematic review and meta-analyses of randomized controlled trials. BMJ. 2014; 348:g2646. [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.