Mammography is no more sensitive at detecting breast cancer in women with a
family history of the disease than in women without it, according to a new
study by University of California, San Francisco and San Francisco Veterans
Affairs Medical Center researchers and their colleagues.
The new findings call into question a recommendation included in most screening
guidelines - that women whose immediate relatives have had the disease should
be screened at a younger age than women without such a family history.
All guidelines for mammography screening recommend it for women between ages 50
and 69 (some guidelines, especially in the US, recommend screening should begin
at age 40). Most also recommend earlier screening for women whose mother,
sister or daughter had the disease, said Karla Kerlikowske, MD, chief of the
Women Veterans Comprehensive Health Center at the SFVAMC and UCSF assistant
professor of medicine and epidemiology and biostatistics. Women with such a
family history are nearly twice as likely to get breast cancer in their
“People have thought that because women with a family history are at higher
risk than women of a similar age without a family history, maybe mammography
would perform better in these women. But there’s no data to support that
hypothesis,” she said.
The study by Kerlikowske and her colleagues, which is published in the latest
issue of the Annals of Internal Medicine, assessed the accuracy of mammography
by analyzing data from more than 389,000 women in seven mammography
They found that mammography it missed a similar percentage of cancers whether
women had a family history or not. “We hypothesized that the chance that
mammography misses a cancer case would be lower in these women at higher risk,
but it’s not. The sensitivity of mammography is primarily influenced by the
age of the woman, and is not affected by family history ,” she said.
Previous studies have shown that mammography is more sensitive at detecting
cancer in older women. Kerlikowske’s new study confirms that mammography
detects nearly 85 percent of cancer cases in women ages 50 to 69, but only
about 75 percent of cases for women 30 to 49.
The study also showed that mammography found more cases of breast cancer among
the women with a family history, an expected result given that breast cancer is
more common in these women. Mammography detected six cases per 1000 women with
a family history, versus four cases per 1000 women without a family history.
Detecting a few more cases of cancer might sound like a good reason to
recommend mammography at an earlier age for women with a family history, but
Kerlikowske argues that the justification for such a recommendation should not
be based simply on detecting more cancers, but on whether the benefits of the
test outweigh the harms.
The overall benefit of screening women under 50 is comparatively small,
Kerlikowske said. To prevent one breast cancer death among women 40 to 49 years
of age, 2500 women would have to be regularly screened, whereas only 270 women
in the 50- to 69-year old group need to be regularly screened to prevent one
breast cancer death.
Since mammography is no more sensitive in women with a family history,
recommending earlier screening for these women would subject thousands of
younger women to the test for a very small benefit, Kerlikowske said.
The risks of mammography include the extra testing and anxiety that accompany
a positive result, which in a great majority of cases turns out to be a false
positive (the test suggested cancer where none existed). The new study found
that women with a family history of breast cancer had a greater chance of a
Kerlikowske said she hopes this study will convince some doctors to consider
their decision more carefully before automatically advising mammography at an
earlier age for women with a family history, and to inform women of the
potential benefits and harms of screening. Co-investigators on this study
included Virginia Ernster, PhD, UCSF vice chair and professor of epidemiology
and biostatistics; and researchers at seven other universities and
The study was supported by grants from the National Cancer Institute and the
Department of Defense.