Center News

Life after breast-cancer diagnosis

Aug. 22, 2003
toddler in parent's arms

A new PHS Division study reveals that women with breast cancer who give birth 10 or more months after diagnosis experience no increased risk of mortality compared to patients who do not subsequently bear children.

Photo by Todd McNaught

Breast-cancer patients under age 45 who subsequently bear children are no more likely to die of their disease than patients who do not, according to a new study led by researchers in the Public Health Sciences Division. In a large analysis of women from Seattle, Los Angeles and Detroit, scientists found no increase in the mortality rate of breast-cancer survivors who gave birth 10 or more months after their diagnosis regardless of their race or ethnic background.

The results may reassure younger cancer patients who are evaluating their reproductive options, said Dr. Beth Mueller, a PHS investigator and lead author of the study.

"Breast-cancer incidence in women under 45 has increased, and this is an age group that is having kids or thinking about having kids," she said "We felt that this was an important study to conduct for breast-cancer patients who are making these decisions."

First of its kind

The study, which is consistent with previous analyses conducted in Scandinavia, is the first to examine the effect of childbirth on survival after breast-cancer diagnosis in an ethnically and racially diverse population of young women.

The results appear in the online version of the journal Cancer and will be published in print in the September 15 issue. Co-investigators of the study, which was funded by the Department of Defense, included PHS collaborators Aruna Kamineni, Dr. Kathi Malone and Dr. Janet Daling, as well as colleagues at the Karmanos Cancer Institute at Wayne State University and the University of Southern California.

The researchers investigated the effects of childbirth on mortality rates among 438 breast-cancer survivors under age 45 who had children after their diagnosis and 2775 women of the same age group who did not have children after their diagnosis. Study subjects were identified from three cancer registries that are part of a nationwide cancer-registry system run by the National Cancer Institute, including the Cancer Surveillance System housed in the PHS Division. Identification of women with births was conducted by linking cancer-registry data with birth-certificate records. The median follow-up time for women after delivery ranged from about four to nine years.

Healthy-mother bias

Overall, women who had children 10 or more months after their diagnosis actually had nearly half the risk of death compared to women who did not have subsequent children. However, Mueller cautioned that this result should not be interpreted to mean that pregnancy confers a protective effect on survival due to what has been termed the "healthy-mother" bias.

"Although we tried to control statistically for disease stage, only relatively healthier women are likely to decide to become pregnant," she said. "Cancer patients-even those with the same stage disease-can experience their disease differently. Our study could not measure aspects of health status at the time the pregnancy occurred, so we can't directly address this issue."

When researchers looked at mortality rates by subgroup, African-American breast-cancer patients, while also not at increased risk of death after childbirth 10 or more months following diagnosis, did not exhibit the same decrease in mortality incidence compared to other racial or ethnic groups. Although the reason for this disparity is uncertain, the researchers speculate that it could reflect differences in access to medical care.

Increased risk of mortality was observed, however, among subgroups of women who were pregnant at the time of their diagnosis, including those who delivered within three months of their diagnosis, those 35 years of age or older at diagnosis and those with certain disease characteristics. For example, women who had had regional disease (cancer that has spread from the breast to nearby sites in the body) were nearly twice as likely to die of their disease than women who did not give birth.

Multiple factors could contribute to the increased mortality in these women, Mueller said.

"It's possible that this is due to a pregnancy-related hormonal effect on the breast tissue," she said. "But it's also possible that women who are pregnant when they are diagnosed with breast cancer may not be treated as vigorously as women who are not pregnant."

Fred Hutchinson Cancer Research Center is a world leader in research to prevent, detect and treat cancer and other life-threatening diseases.