The relationship between dietary fat and the risk of breast cancer has been controversial for decades. Early international correlation studies ( 1 ), buttressed by later case–control studies ( 2 ), suggested a positive relationship. However, the substantial confounding in the international correlation studies by many factors—especially reproductive variables—and the potential for recall and selection biases in retrospective studies cast doubt on those results. Indeed, most prospective studies ( 2 ), including two large pooled analyses of prospective studies ( 3 , 4 ), have not demonstrated a link between fat intake and breast cancer risk.
However, several recent cohort studies have reported results that suggest a modest positive association between fat intake and the risk of breast cancer. For example, in a meta-analysis of 31 case–control studies and 14 cohort studies ( 5 ), the summary risk estimates for the two study designs were similar to each other (and neither was statistically significant), whereas the overall summary relative risk estimate from all studies for the highest versus the lowest fat intakes was 1.13 (95% confidence interval [CI] = 1.03 to 1.25).
More recently, the Women's Health Initiative Dietary Modification Trial ( 6 ) reported a marginally statistically significant 9% reduction in breast cancer incidence among women in the low-fat dietary pattern group compared with women in the control group for a period of 8.1 years of follow-up (hazard ratio [HR] = 0.91, 95% CI = 0.83 to 1.01). However, this result cannot be attributed solely to a decrease in fat intake because, in addition to statistically significantly lowering their fat intake, the intervention group also lowered their energy intake and body weight and increased their fruit, vegetable, fiber, and folate intakes compared with the control group. It has been suggested that the notable reduction in body weight probably contributed to the reduction in breast cancer incidence ( 7 , 8 ).
In this issue of the Journal, Thiébaut et al. ( 9 ) provide further high-quality data from the prospective National Institutes of Health (NIH)–AARP Diet and Health Study that bear on this issue. They report that among the postmenopausal women in this cohort, those with a median fat intake of 90 g/day (or 40% cal from fat) had, depending on the statistical model, breast cancer that were 11%–22% higher than the rates of women with a median fat intake of 24.2g/day (or 20% cal from fat). This result is similar to the summary relative risk estimate from the meta-analysis ( 5 ). Thiébaut et al. corrected for measurement error in fat and energy intakes, which strengthened the observed association; for a doubling in total fat intake, the uncorrected HR was 1.15 (95% CI = 1.05 to 1.26) and the corrected HR was 1.32 (95% CI = 1.11 to 1.58). Thus, even with correction for measurement error, this study provides further evidence that the association between total fat intake and the risk of breast cancer among postmenopausal women is likely to be modest. Analyses of specific subtypes of fat yielded increased risks that ranged from 10% to 13% for a doubling of percent energy from fat, an unrealistic scenario except, perhaps, for the small minority of women with the lowest intakes. When all types of fat were considered simultaneously, only the association for saturated fat remained statistically significant.
Thiébaut et al. found a stronger association between saturated and monounsaturated fat intakes and breast cancer risk among current users of menopausal hormone therapy than among never or former users, which suggests that dietary fat intake may have more influence on breast cancer risk when it occurs in the context of an estrogen-rich environment (i.e., menopausal hormone therapy users or premenopausal women). Indeed, the pooled analysis of prospective studies ( 4 ) reported positive associations for polyunsaturated fat and vegetable fat among current users of menopausal hormones but not among never or former users. Furthermore, in the Women's Health Initiative Dietary Modification Trial ( 6 ), women who were randomly assigned to the low-fat group and who were either using postmenopausal hormones at baseline or who were also randomly assigned to the estrogen plus progestin arm of the trial had a 17% (95% CI = 1% to 31%) reduction in breast cancer incidence compared with the control group; by contrast, there was no difference in breast cancer incidence between the low-fat dietary pattern group and the control group among women not using postmenopausal hormones ( P for interaction = .06). Similarly, in the predominantly premenopausal Nurses’ Health Study II cohort, women in the highest quintile of animal (but not vegetable) fat intake had a statistically significantly increased risk of breast cancer compared with those in the lowest quintile of intake ( 10 ). Thus, perhaps the association between dietary fat intake and the risk of breast cancer varies by menopausal status or by menopausal hormone therapy use, as does the association between adiposity and breast cancer risk ( 11 , 12 ).
On the basis of the main findings from the NIH–AARP Diet and Health Study ( 9 ) and other studies described above, it appears that dietary fat intake during postmenopausal years has little, if any, impact on breast cancer risk, although intake may be associated with higher risks of breast cancer for specific population subgroups, such as menopausal hormone users or younger women. This potential interaction of dietary fat with the hormonal milieu is of interest and can be addressed using a variety of approaches, including mechanistic studies, studies of younger age groups, and pooled analyses of multiple studies, to increase statistical power. Examining a wide range of populations with different dietary patterns may help to disentangle potentially different effects of specific subtypes of fat. For example, in the NIH–AARP Diet and Health Study ( 9 ), despite the wide range in total fat intakes, the high correlation between saturated and monounsaturated fat intakes (Pearson correlation coefficient = .81) limited the ability to evaluate their independent associations with risk. Also, additional investigation of particular contributors to fat intake, such as red meat and dairy products, would be valuable.
The modest associations that have been observed for dietary fat and breast cancer risk in observational studies and clinical trials stand in sharp contrast to the robust evidence for a strong link between adiposity and the risk of postmenopausal breast cancer ( 11 , 13 ). Thus, from a prevention perspective, interventions to control the amount of body fat (e.g., promotion of exercise and caloric restraint) are likely to have a greater impact on breast cancer incidence than a reduction in fat intake.