Does Mammogram Risk Outweigh Benefit?

Value of Early Breast Cancer Screening Questioned

From the WebMD Archives

Oct. 14, 2003 -- For most women under 50, annual mammograms offer a small benefit. A growing number of researchers say this benefit may not outweigh the risks.

One of those researchers is Cornelia J. Baines, MD, professor emerita at the University of Toronto and co-leader of Canada's largest clinical trial of breast-cancer screening. Her editorial in the Oct. 15 issue of the Journal of the National Cancer Institute argues that women -- particularly those aged 40-49 -- are urged to get mammograms without being told of the risks.

One of those risks may be the "screening paradox" seen in women who start getting annual mammograms before the age of 50.

"There is a consistently observed increase in breast cancer deaths in the first 10 years of screening," Baines tells WebMD. "It is consistent, it occurs across time, and it occurs across countries."

There's a benefit to early screening. But it's much smaller than most women realize.

"If you truly look at the benefit to be obtained for women in the 40-49 age group, it is very small and takes a long time," Baines says. "It is a 9% decrease in cancer deaths after 16 years."

Meanwhile, she notes, the risk of death from breast cancer in the third year of screening is more than twice as high for women who get mammograms as for women who don't.

Is It Real?

The trouble with the screening paradox is that it might not be real, says Alfred O. Berg, MD, chairman of the department of family medicine at the University of Washington. Berg led the U.S. Preventive Services Task Force that rigorously evaluated the evidence for and against breast cancer screening.

"I don't agree with Dr. Baines that every study shows this," Berg says. "We are not sure the observation that she and others have made is actually there. ... We would probably judge the evidence as weak, because none of the studies were formally designed to look at this question."

Berg says it would be interesting to look into the questions raised by the screening paradox. And that's all Baines asks. But it's never been done.

Continued

"It is a shame this very clear message has been ignored for so long," she says. "We first reported this in the 1980s. If people had tried to get to the bottom of this, we might be further along than we are now. ... I want someone to tell me they will look into this paradox. To say, yes, that is a way we will get around to learning more about breast cancer."

How could mammograms cause breast cancer? Baines doesn't think it's the mammograms themselves. Women who get screened find small tumors sooner. This leads to earlier surgery. Some of these tumors might not really be dangerous. But removing them might be, animal studies suggest.

"Removal of the primary tumor may remove a negative growth factor that discourages metastases from growing," Baines suggests. "Not all women are vulnerable to this effect, or more would die sooner. Not all breast cancers are the same, and not all women are the same."

Berg notes that it's far from obvious that the early breast cancer known as ductal carcinoma in situ or DCIS is a clear and present danger.

"DCIS is commonly detected in early screening," he says. "It is not clear what is the natural history of DCIS. Some believe that by finding and treating DCIS aggressively, we may be overtreating. Some of those women may not need to be treated."

Other Risks From Early Screening

Berg may disagree with Baines about the screening paradox. But in most other things, they tend to agree. Berg, too, worries that women are being urged to seek early breast cancer screening without being told of the risks.

"Yes, there is evidence of a mortality benefit to mammography screening beginning at age 40 -- but there are risks," he says. "The ratio of harm to benefit gets smaller as women get older, but even for women in their 60s, the odds that a woman will benefit are quite small. When a woman considers screening, the odds ought to be explicit."

So should a woman get screened for breast cancer? The American Cancer Society says yes, without question, most women should start annual screening at age 40 -- and some should start sooner. But Berg says that based on the available evidence, a woman who says no to screening is not wrong.

Continued

"Our view is if you work through the numbers with an individual woman, it is perfectly legitimate for that woman to decide or not to decide to have a mammogram at any one time. The decision is a subjective one," he says.

Even if there were no screening paradox, the risks of screening are very real.

"There are always risks for any screening program," Berg says. "There are false positives, which lead to unnecessary testing, and unnecessary anxiety. Some women with positive mammograms, even after getting a negative biopsy, have heightened anxiety for a period of time. Studies in the last five years show that with regular screening, a large proportion of women -- 25-30 percent -- will have a biopsy sometime in life that is negative. Results can also falsely reassure: They can be negative when something really is there."

But screening is good for most women, right? Not so, says Berg.

"The number of women you need to screen to prevent one death from breast cancer is about 1,000," he says. "Out of every 1,000 women screened, only one will get the benefit. And the others won't. Most women feel that mammograms are an absolute test, that it will detect breast cancer early and if they are treated early they will benefit. The data is not encouraging that this is so."

WebMD Health News

Sources

SOURCES: Baines, C.J. Journal of the National Cancer Institute, Oct. 15, 2003; vol 95: pp 1508-1511, 1512-1513. Berg. A.O. Journal of the National Cancer Institute, Oct. 15, 2003; vol 95: pp 1511-1512. Cornelia J. Baines, MD, professor emerita, University of Toronto, Ontario, Canada. Alfred O. Berg, MD, chairman, department of family medicine, University of Washington; and chairman, U.S. Preventive Services Task Force.
© 2003 WebMD, Inc. All rights reserved.

Pagination