An extended analysis of cancer rates reinforces a strong association between use of hormone replacement therapy (HRT) and increased breast cancer incidence, according to research led by scientists at The University of Texas M. D. Anderson Cancer Center and published in the New England Journal of Medicine.
In the published study, the investigators say that plummeting use of HRT in mid-2002, after results of the Women's Health Initiative study were announced, correlated with a steep decline in new breast cancer diagnoses that started shortly thereafter and continued through 2003. Incidence in 2004 leveled maintained the same low level of incidence, the lowest rate seen since about 1987, the researchers say.
The decline occurred primarily in women age 50-69, the researchers find, and was predominantly seen in estrogen-receptor(ER)-positive cancer - the type of tumors is fueled by estrogen, a hormone that is supplemented in HRT. Such cancers declined 14.7 percent in this time period, compared to a non-significant decline of 1.7 percent in ER-negative tumors.
"For our new data set, 2004, the drop in breast cancer incidence leveled off and remained low in that year, "showing that the decreased rates seen in 2003 were also present in 2004, meaning that the decline was not a one-year wonder, a short-lived anomaly," says Ravdin, the study's lead investigator.
"This kind of study can't prove causality, but the data present a very compelling link between hormone replacement therapy and breast cancer," says Berry, the study's senior investigator.
Using data derived from National Cancer Institute (NCI) cancer registries that report on 9 percent of the U.S. population, they found that the total decrease in breast cancer incidence was 6.7 percent between 2002 and 2003. They also calculated that by the end of 2002, about 20 million fewer prescriptions for HRT were written in the United States- a decrease of 38 percent. Interest in HRT use dropped after the 16,608-participant federal Women's Health Initiative study results were announced in July 2002 and showed that the risks of taking these agents outweighed the benefits for many post-menopausal women.
Ravdin and Berry strongly stress, however, that their study is not suggesting that all women stop their use of HRT. "This study is not saying that an individual woman will reduce her absolute risk of developing breast cancer by 15 percent by immediately discontinuing use of HRT," Berry says.
While it may be true that stopping use of HRT may have prevented as many as 14,000 breast cancers in 2003 compared with 2002, the percentage of decline is based on an entire population, he explains. "At best, based on this analysis, an individual woman could reduce her individual risk of developing breast cancer by one in 60, or about 1.7 percent, if she stopped using hormones," Berry says.
As a physician, Ravdin tells his patients to follow currently accepted guidelines for HRT use: to use the drug at the lowest dose and for the shortest time period to control hot flashes and other debilitating symptoms caused by the onset of menopause.
"The risk of developing breast cancer from use of these hormones is relatively small and for some women with postmenopausal symptoms, the benefits of HRT are well worth that risk," he says. "This is just another small piece of the puzzle to help women gauge the risks and benefits of using HRT."
The researchers also say that their study cannot answer three key questions: whether stopping the use of HRT leads to a permanent or a temporary decline in breast cancer incidence; if this effect is seen for stopping all types of HRT; and how much of a contributing role other factors may have played in the decline.
"There are several possibilities as to what effect stopping HRT has. Possibly, it slows the growth of tumors that are there but aren't big enough yet to be detected on a mammogram. Or it could be removing the hormone fuel stops the growth completely or even causes tumor regression," Berry says. "We don't know which is correct."
While Berry, an expert in statistics, adds that he was initially surprised that stopping HRT use could have such an immediate impact on breast cancer growth, Ravdin, the clinician, says he was not.
"We know that if you treat ER-positive breast cancer with anti-hormone treatment, you can see shrinkage within weeks, so why wouldn't withdrawing hormones have the same kind of effect on smaller cancers that have not yet been detected?" says Ravdin. "My thought is that these tumors don't completely disappear, but they have stopped growing - hopefully, for many of them, forever."
As to the impact of other factors on breast cancer decline, the researchers say that one contributing factor could be declining use of mammography by women who have stopped using HRT. NCI data has reported a 3.2 percent decline in screening mammography in 2003 for women 50-65 years old, compared to 2000, Berry says, but adds, "such a change would seem insufficient to explain the decline in breast cancer incidence." A large drop in screening would have been seen in breast tumors that are both ER-positive and ER-negative, and that wasn't the case.
Finally, the researchers say that this study may lead to new insights into both the etiology of breast cancer and its prevention. "We will continue modeling incidence rates to try to understand whether what we are seeing is a slowing or a regressing of tumors, or a mix between these two things," Berry says.
The study was funded by grants from the National Cancer Institute and from M. D. Anderson. Co-authors also include Kathy Cronin, Ph.D., Nadia Howlander, M.S., Christine Berg, M.D., Eric Feuer, Ph.D., and Brenda Edwards, Ph.D., from the National Cancer Institute, and Rowan Chlebowski, M.D., Ph.D., from Harbor UCLA Medical Center.
M. D. Anderson's Peter Ravdin M.D., Ph.D., professor in the Department of Biostatistics, and Donald Berry, Ph.D., head of the Division of Quantitative Sciences, presented the first analysis of the 2002-2003 falling breast cancer rates at the December 2006 meeting of the San Antonio Breast Cancer Symposium. They have since examined breast cancer incidence in 2004.
Cite This Page: