Weight and Weight Change during Adulthood
Attained weight and weight change in adults provide sensitive measures of the balance between long-term energy intake and expenditure. Although the relationship between these variables and breast cancer risk has been complex and confusing, recent findings provide a coherent picture and indicate a major contribution of weight gain to risk of postmenopausal breast cancer risk. Two reproducible findings have been particularly enigmatic: (a) In affluent Western populations with high rates of breast cancer, measures of body fatness have been inversely related to risk of premenopausal breast cancer, and (b) body fatness has been only weakly related to postmenopausal breast cancer risk despite strong associations between body fat and endogenous estrogen levels.

The inverse relationship between body weight [typically considered as body mass index (BMI), calculated as weight in kilograms divided by the square of height in meters, to account for variation in height] and incidence of premenopausal breast cancer has been consistently seen inprospective studies and in a meta-analysis of case control and cohort studies.

Little relationship of BMI to breast cancer mortality has been observed in premenopausal women, probably because delayed detection and diagnosis in heavier women counterbalances the lower incidence among heavier women. Heavier premenopausal women, even at the upper limits of what are considered to be healthy weights, have more irregular menstrual cycles and increased rates of anovulatory infertility, findings suggesting that their lower risk may be due to fewer ovulatory cycles and less exposure to ovarian hormones. Increased rates of menstrual irregularity and anovulatory infertility are also seen among very lean women, but such women are uncommon in Western populations. In case control studies, a consistent relationship between menstrual cycle regularity and breast cancer risk has not emerged, which could cast doubt on this explanation, but this may be due to the indirect relationship between menstrual regularity and ovulation and to difficulties in remote recall.

In a prospective study among younger women, both shorter and longer or irregular cycles, compared with regular cycles of approximately 28 days, were associated with reduced risk of breast cancer41; this finding lends support to irregular anovulation as the explanation for the lower risk in heavier women.

In case control and prospective studies conducted in affluent Western countries, the association between BMI and risk of breast cancer among postmenopausal women has been only weakly positive or nonexistent. The lack of a stronger association has been surprising, because obese postmenopausal women have plasma levels of endogenous estrogens nearly twice those of lean women, due to conversion of androstenedione to estrogens in adipose tissue and also to lower levels of sex hormone-binding globulin. The lack of a stronger positive association now appears to be due to two factors. First, as with the protective effect of early pregnancy, the reduction in breast cancer risk associated with being overweight in early adult life appears to persist throughout later life.

Thus, an elevated BMI in a postmenopausal woman represents two opposing risks: a protective effect due to the correlation between early weight and postmenopausal weight, and an adverse effect due to elevated estrogen levels after menopause. For this reason, weight gain from early adult life to after menopause should be more strongly related to postmenopausal breast cancer risk than attained weight. Indeed, a positive association between weight gain and increased risk of postmenopausal breast cancer has been consistently supported by both case control and prospective studies.

A second reason for the failure to appreciate a greater adverse effect of excessive weight or weight gain on risk of postmenopausal breast cancer is that the use of postmenopausal hormones obscures the variation in endogenous estrogens due to adiposity and elevates breast cancer risk regardless of body weight. To appreciate fully the impact of weight or weight gain, an analysis should be limited to women who never used postmenopausal hormones. Thus, among women who never used postmenopausal hormones in the Nurses’ Health Study, those who gained 25 kg or more after age 18 had double the risk of breast cancer compared with women who maintained their weight within 2 kg. In this population, the combination of using postmenopausal hormones or gaining weight after age 18 years accounted for one-third of postmenopausal breast cancer cases.

The relationship between body weight and breast cancer risk in lower-risk, mainly non-Western, countries has beenobserved to be somewhat different than that in higher-risk countries.233 In general, the inverse relationship between weight and premenopausal breast cancer risk has not been observed, and the association between weight and postmenopausal risk has been stronger. This difference is likely to be due to the lower prevalence of overweight among premenopausal women in these low-risk countries; few women are likely to be sufficiently overweight to cause anovulation and a reduction in premenopausal breast cancer risk. As a result, BMI after menopause would reflect only the adverse effects of high levels of endogenous estrogens, unopposed by a residual protective effect due to correlation with overweight in early adult life.

In summary, as in animal studies, energy balance appears to play an important but complex role in the causation of breast cancer in humans. During childhood, rapid growth rates accelerate the occurrence of menarche, an established risk factor, and result in greater attained stature, which has been consistently associated with increased risk. During early adult life, overweight is associated with a lower incidence of breast cancer before menopause, but no reduction in breast cancer mortality. Weight gain after age 18 years, however, is associated with a graded and substantial increase in postmenopausal breast cancer that is seen most clearly in the absence of hormone replacement therapy.

Walter C. Willett, Beverly Rockhill, Susan E. Hankinson, David J. Hunter and Graham A. Colditz

W. C. Willett: Harvard Medical School, Boston, Massachusetts; Department of Nutrition, Harvard School of Public Health, Boston, Massachusetts
B. Rockhill: Harvard Medical School, Brigham and Women’s Hospital, Boston, Massachusetts
S. E. Hankinson: Departments of Medicine and Epidemiology, Harvard Medical School and Harvard School of Public Health, Boston Massachusetts
D. J. Hunter: Departments of Epidemiology and Nutrition, Harvard School of Public Health, Boston, Massachussetts
G. A. Colditz: Department of Medicine, Harvard Medical School, Boston, Massachussetts

References