Skip to main content

Researchers publish first-ever breast cancer risk model for Hispanic women

December 22, 2016

By Anne Harding

NEW YORK (Reuters Health) - The first-ever breast cancer risk model specifically for Hispanic women was published on December 20, and will soon be incorporated into the National Cancer Institute's Breast Cancer Risk Assessment Tool (BCRAT).

"We hope this new tool is not only more accurate for that population but that health care providers would use this to not only estimate but also inform their Hispanic female patients about their risk for breast cancer," Dr. Matthew Banegas, a cancer-disparities researcher at the Center for Health Research at Kaiser Permanente Northwest in Portland, Oregon, and the lead author of the new study, told Reuters Health by phone.

Hispanics account for nearly 17% of the United States population, with about two-thirds being born in this country, Dr. Banegas and his team note in their report, published in the Journal of the National Cancer Institute. Presently, the BCRAT is accurate for non-Hispanic whites, African-Americans, and Asian and Pacific Islander Americans, they add, but underestimates risks for Hispanic women.

To develop a nativity-specific model for Hispanic women (meaning whether or not a woman was born in the US would be included as a risk factor), the researchers used data from the San Francisco Bay Area Breast Cancer Study, the California Cancer Registry and the NCI's Surveillance, Epidemiology, and End Results program.

Their analysis included 533 U.S.-born and 553 foreign-born breast cancer patients and 464 U.S.-born and 947 foreign-born controls.

Other risk factors used in the model along with nativity included age at first full-term pregnancy, biopsy for benign breast disease, family history of breast cancer, and age at menarche (for the foreign-born group only).

Using Women's Health Initiative data on 6,220 postmenopausal Hispanic women, the researchers found their model was calibrated well for U.S.-born Hispanic women, with an observed/expected ratio of 1.07, but the model appeared to overestimate risk for foreign-born Hispanic women (O/E ratio, 0.66). The area under the curve was 0.564 for women born in the U.S. and 0.625 for foreign-born women.

Given that the model is based on data from California residents, the new model will be appropriate for Hispanic women living in the western U.S., who are mainly of Mexican and Central American descent, the authors note.

"Studies highlighting heterogeneity in breast cancer risk between Hispanic women underscore the importance of differences in country of origin, duration of residence in the United States, and acculturation in estimating the risk of breast cancer," they add.

"We as researchers and providers and patients need to keep in mind when we're using this tool we haven't evaluated the performance of this new model in specific Hispanic subgroups, yet we do hope to do so, and I have been contacted by individuals across the U.S. expressing interest in the model and saying we would like to test it in our group of patients," Dr. Banegas said.

"We do understand that this is not a perfect model but we agree that it's a great first step, and we do intend to work with others to continue to improve this model so that the risk estimates that we get from this tool only get better and better over time," he added.

The study did not have commercial funding.


J Natl Cancer Inst 2016.

(c) Copyright Thomson Reuters 2016. Click For Restrictions -
Back to Top