Why Are Breast-Cancer Rates Dropping?

"You have breast cancer." Many of us either know someone who has received that diagnosis or have heard those scary words ourselves. Although breast cancer is still the second most common cancer among American women (behind skin cancer), rates have been declining by about 2 percent a year since 1999, after having increased for the previous 20 years. That's good news—but it's also a puzzle that researchers have been trying to solve. If we can find out why fewer women are being diagnosed with this terrible disease, maybe we can find ways to make it even less common and perhaps apply those lessons to other forms of cancer as well.

The latest clue comes in a study to be published next month in the American Journal of Public Health. (The study is currently available onlineat the journal's website.) Harvard researchers analyzed data from more than 350,000 women with breast cancer and concluded that the drop is linked to decreased use of menopausal hormone therapy in the wake of the July 2002 release of data from the Women’s Health Initiative, the massive federal study that was halted early because women on a combination of estrogen and progesterone had an increased risk of breast cancer and heart disease.

The Harvard researchers came to that conclusion because much of the decline in incidence is among white, affluent women over 50. Those are the same women who are most likely to take hormones at menopause. "Our study can't settle the question," says Nancy Krieger, an epidemiologist at Harvard's School of Public Health who is the lead author. But, she adds, her work "does show trends that make it hard to imagine any other explanation for the fall in these rates."

For several decades before 2002, hormone therapy was routinely prescribed to women at menopause (in the industrialized world, the average age is 51) because doctors believed that it would prevent heart disease. The reasoning behind that theory was simple. Before menopause, women are much less likely than men to suffer from heart attacks or strokes. After their periods stop, women's vulnerability to heart disease escalates for about a decade. At that point, the rates for men and women are roughly equal. That led doctors to conclude that estrogen had a protective effect that disappeared after menopause. Hormones seemed to be the solution.

The Women's Health Initiative (WHI), one of the largest federal research studies ever conducted, was meant to definitively prove whether or not hormones lowered the risk of heart disease. The study turned out to be a bombshell for doctors and patients. Major medical groups, like the American College of Obstetricians and Gynecologists and the North American Menopause Society, revised their guidelines to advise against using hormones for disease protection. Hormones are now recommended only to ease symptoms like hot flashes.

In the first year after the WHI results were released, more than a third of all menopausal women who were taking hormones threw out their pills. That's a move Krieger would endorse. "I think the increased risk of breast cancer is very clear," she says, "and not remotely outweighed by any reported protection against cardiac disease or cognitive decline."

Krieger's study is not the only one to link the decline in hormone therapy with the drop in breast-cancer rates, although it's one of the largest. In an article published in January in the journal Breast Cancer Research, scientists looked at the data from the California Teachers Study and came to a similar conclusion: the drop in hormone-therapy use was related to the decline in cancer diagnoses.

But not everyone agrees that the drop in hormone use explains everything. Because breast-cancer rates started declining before 2002, some researchers think other factors are at work. Susan Gapstur, an epidemiologist at the American Cancer Society, points out that the decline began around the time that mammography use peaked. She and others believe that changes in frequency of screening may account for some of the drop. And other scientists say that because cancers can take years to develop, it's hard to pin the decline in rates to the sudden plunge in hormone use.

Krieger's study may answer at least some of those concerns. A decline in mammography use can't explain a dip in diagnoses, she says, because the drop in screening was much smaller than the decline in cancer incidence. And the study's authors say that rates may have begun to decline before 2002 because of smaller studies around that time that raised doubts about the safety of hormone therapy.

There are still many more questions about the overall effects of taking estrogen. "We'd like to understand hormone therapy's impact on the levels of breast cancer," Krieger says. "Does it vary by age? Does it depend on duration or timing of use? What happens if you only take it for a short time?" Designing a definitive study would be ethically challenging, Krieger admits. "How can you ask women to take a drug that you know causes them harm?"

Krieger's work raises another issue that is currently on the research agenda: how breast-cancer risk varies by a woman's race and socioeconomic status. Her study shows that in the case of hormone therapy, wealthier women didn't necessarily benefit from their greater access to health care.

Beyond health care, researchers are finding real differences in the kinds of breast cancer that affect various populations. "We now know that breast cancer is not one size fits all," says Dr. Worta McCaskill-Stevens, a medical oncologist at the National Cancer Institute. "There are different subtypes, and they are distributed differently among various racial and ethnic populations." African-American women are more likely to get an aggressive form of the disease at a younger age, for example.

The reason for these differences is still unclear, McCaskill-Stevens says, but the research is ongoing. "We are learning a lot about the differences in the risk of developing it," she says. One critical step would be developing biological markers that could help doctors spot disease at its earliest stages. McCaskill-Stevens also says that it's important to understand how environmental factors (like taking hormone therapy or having children early in life) interact with an individual's genetic risk. And once scientists know more about individual risk, they can create more personalized treatments.

All that is in the future, of course. In the meantime, studies like Krieger's help unravel the mysteries of breast cancer and make sure that the number of women with this disease continues to decline.

Join the Discussion