At the recent Breast Cancer Research Foundation scientific symposium held in New York City and Memorial Sloan-Kettering on October 25, I summarized evidence on lifestyle approaches for breast cancer prevention. Let me briefly review the evidence presented in this talk.
You might ask how can we assume that breast cancer is preventable?
Evidence comes from many sources. Among the most convincing are migrant studies which show that the rate of breast cancer in offspring of women who migrate are far higher than the rate of breast cancer in the nieces of these women, daughter of their sisters who did not migrate, born in Asian countries. In addition, within countries we have seen substantial changes in the rates for beast cancer in short periods of time. Such changes must reflect changing exposures to factors promoting cancer rather than a change in the underlying genetic predisposition to breast cancer. For example after the results of the Women’s Health Initiative were released in 2002, there was a marked decrease in the use of combination estrogen plus progestin. The rates of breast cancer fell by approximately 10% among postmenopausal women in many countries (1). In addition, with rapidly changing reproductive patterns in Korea, the rates of breast cancer have risen rapidly among premenopausal women to almost parallel those of US women of the same age.
Further strong evidence comes from randomized controlled trials of selective estrogen receptor modulators that block the action of estrogen at the level of the receptor and lead to a 50% or greater reduction in breast cancer incidence in the space of five years or so. Tamoxifen and Raloxifene have both been tested using this rigorous study design (2, 3). Finally, among women who have undergone bilateral removal of the ovaries due to elevated genetic inherited risk of breast cancer, a 50% reduction in incidence has been reported (4).
The importance of adolescent and early life exposures driving lifetime risk of breast cancer is clearly documented from the follow-up of the atomic bomb survivors (5). Now with over 40 years of follow-up we have clear evidence that the greatest risk for dollars of radiation overture is for those who were under 15 at the time of the atomic bomb. The continuing international variation in postmenopausal breast cancer further illustrates the variation in cancer incidence.
One must still ask “How will we prevent breast cancer”?
At what age should we intervene? What do we need to change? By how much and for how long? Accumulating evidence shows that exposures from birth to late adolescence drive breast cancer risk. Growth from birth to adolescence has a major impact on lifetime breast cancer risk. Evidence in support of this conclusion comes from a wide range of studies addressing height and the risk of cancer. A notable example is presented by Sung and colleagues who followed 339,000 Korean women for 10 years and observed a significant increase in risk with taller height (6). In that analysis the authors controlled for breast cancer risk factors. To achieve one’s height growth proceeds from birth at different paces quickening prior to the menarche (onset of menstrual periods) and concluding shortly thereafter. By this time a women has reached her adult height. With my colleague Catherine Berkey, we have analyzed growth velocity and risk of breast cancer showing in the Nurses’ Health Study that young women with the highest rates of growth have significantly higher risk of both premenopausal and subsequent postmenopausal breast cancer (7). Recently, we have carried this analysis to risk of benign breast disease, a precursor lesion for breast cancer (8). Again, we see that the women who have the highest peak growth velocity, have significantly elevated risk of benign breast disease through their premenopausal years. These data point to childhood and adolescence as an important period for risk accumulation that may drive breast cancer risk for the rest of life. What then may modify the rate of risk accumulation, slowing this down, and in essence protect women against risk of breast cancer.
To answer these questions we have conducted a number of studies using biopsy confirmed diagnosis of benign breast disease as a breast cancer risk marker. This allows us to collect diet activity, weight, height, and other risk factor data from girls as they grow through adolescence and early adult years. We have followed them through their 20s to confirm their diagnosis of benign breast disease. We focus on a number of modifiable risk factors. We observe that those women who drank 3 to 5 drinks of alcohol per week more than doubled the risk of benign breast disease during 10 years of follow-up into their 20s (9). This finding has been replicated with the use of recall of alcohol intake during high school and early adult years in a separate study, the Nurses’ Health Study II. Other adolescent exposures that deserve mention include higher fiber intake through high school years. Again in a prospective study we observed that those with higher fiber intake in high school had lower risk of benign breast disease. In fact, women consuming 27 g of fiber per day compared to those consuming less than 15 grams per day, had a 25% reduction in their risk of premalignant breast lesions (10).
We have also evaluated physical activity in relation to both premenopausal breast cancer and to benign breast disease (11-13). We recorded details of walking, moderate and vigorous activity, through high school years and adult years. Women who had the highest sustained levels of physical activity from ages 12 to 22 had a substantial reduction in their risk of breast cancer through to the age of menopause. Compared to women who had the lowest levels of activity the most active women experienced 25% fewer cases of breast cancer. Similar results have been observed for benign breast disease.
Based on these data we conclude that higher levels of activity will contribute substantially to re-juiced risk of premenopausal breast cancer. Likewise limiting alcohol intake and increasing fiber intake should also reduce risk of premalignant breast lesions and subsequent risk of breast cancer.
Together, these data point to the importance of understanding the timing of lifestyle habits in relation to lifetime risk of breast cancer. To maximize benefits of prevention, we must focus on the most biologically relevant periods. For breast cancer this increasingly looks like the interval between menarche and the age at which a woman has her first full term pregnancy. There is much potential to improve lifestyle through these years. Reduce alcohol intake, increase levels of physical activity, and get to higher fiber intake. These relatively modest changes in lifestyle could alter the trajectory for breast cancer risk throughout life. We have the tools to make these changes. As a society we will all benefit if we implement them now.
Related CNiC posts
Literature cited