BALTIMORE--Breast cancer is preventable, says Johns Hopkins epidemiologist Jonathan Samet, MD. "But we don't yet know how to do it."
BALTIMORE--Breast cancer is preventable, says Johns Hopkins epidemiologistJonathan Samet, MD. "But we don't yet know how to do it."
This paradox arises, in part, from the rapid changes seen in breastcancer patterns worldwide that may point to an important rolefor environmental risk factors and suggest the potential for primarydisease prevention, Dr. Samet said at a conference on neoplasticdisorders sponsored by Johns Hopkins Medical Institutions.
Studies offer clues as to the predictors of breast cancer risk,Dr. Samet said, but even when risk factors are identified, weas a society and women as individuals may not be ready to takethe necessary steps to modify them significantly. And some importantrisk factors such as family history are not subject to alteration.
A woman's reproductive history can be used to help predict herrisk of breast cancer, he said. Early menarche is associated withincreased risk, and the age of menarche has been falling overthe last century.
Other factors include later menopause, later age at first pregnancy,nulliparity (especially for women over 40), high parity for womenbefore menopause, and low parity for women after menopause. (Lactationmay confer some protection.)
These reproductive risk factors suggest that exposure to estrogen(and possibly progesterone) plays some role in the onset of breastcancer. Thus, factors that reduce the lifetime exposure to estrogenmay be helpful. Exercise during adolescence, for example, maydelay menarche and result in fewer overall menstrual cycles.
"However, these factors are not easily modified," Dr.Samet said. "It just isn't socially acceptable to proposeearly childbearing, for example, as a form of breast cancer protection."
The suspect role of estrogen prompts questions regarding otherhormonal sources, he said. But there is no clear-cut pattern forany danger from either long-term contraceptive use or postmenopausalestrogen replacement therapy.
While women who smoke have lower levels of estrogen, Dr. Sametsaid, there is no good evidence that smoking affects breast cancerrisk. "The irony is that lung cancer continues to rise inwomen," he said, "even though lung cancer is relativelyeasy to prevent, while breast cancer is so hard to prevent."
Obesity--possibly due to hormonal differences--appears to reducebreast cancer risk in premenopausal women but increase it postmenopausally.Even as few as one or two alcoholic drinks a day may elevate risk,but any effects of selenium, antioxidants, or caffeine are unclear,he stated.
Some of the best evidence about dietary factors comes from transnationalcomparisons. As fat consumption rises in a country, it seems,so does breast cancer incidence. But the evidence, Dr. Samet said,is "confusing," and epidemiologic studies are not likelyto settle the issue.
Clinical trials are now underway to study whether consuming morefiber, less fat, and less alcohol will affect risk. The Women'sHealth Initiative has enrolled 70,000 women to test whether limitinga diet to 20% of calories from fat will help.
Other studies are looking at ways of reducing hormonal exposure,either by avoiding oral contraceptives and estrogen replacementtherapy, or by using hormonal antagonists like tamoxifen (Nolvadex)in younger women.
"We should look for substantial progress over the next decadeas we learn how to use the emerging information on breast cancergenetics to guide primary prevention," Dr. Samet concluded.