Study finds high rate of overdiagnosis from mammograms

Better treatment and not earlier detection with screening mammography may be the reason rates of death from breast cancer have declined in recent years, according to controversial results from a large U.S. database study.

In addition, the research team reports in The New England Journal of Medicine, "women were more likely to have breast cancer that was overdiagnosed than to have earlier detection of a tumor that was destined to become large."

"This new article will not change our guidelines," said Dr. Richard Wender, chief cancer control officer at the American Cancer Society, who was not involved in the new research. It is simply showing a trend, which is "the weakest form of evidence," he said.

The American Cancer Society's guidelines call for women to consider having a mammogram at age 40, to actually have one each year beginning at age 45 and then, at age 55, to consider having one every other year. Wender said other medical groups have made similar recommendations.

The chief author of the new study, Dr. Gilbert Welch of the Dartmouth Institute for Health Policy and Clinical Research in Hanover, New Hampshire, and his coauthors found that mammography resulted in the discovery of 162 more cases of breast cancer for every 100,000 women, but only 30 of those small tumors were expected to grow and become a danger.

That suggests more than four times as many cases - 132 in all - were what is known as overdiagnosed.

"Everyone should understand that there are tradeoffs in screening. It's not a simple it-can-only-help-you kind of story," he told Reuters Health.

"If you have a new breast lump you should have a mammogram. There's no doubt about that," he said. But otherwise, "women should understand it's a choice."

The debate is part of ongoing discussion about the effectiveness and timing of breast cancer screening, whether better detection technologies are identifying cancers that don't really pose a threat, and how to treat the abnormalities that are found.

"We've always been trained to think that the best test is the one that finds the most cancer," Welch said. "That's way too easy. The question is, who can find the cancers that matter? Who can help sort out what cancers are really important?"

Women who get breast cancer screening already have to deal with false positives and paying for evaluations that may not be covered by insurance, writes Dr. Joann Elmore in an accompanying editorial. Now it's becoming clear that screening is causing overdiagnosis of diseases that would never cause symptoms or death, she continues.

"We need better methods of distinguishing biologically self-limited tumors from harmful tumors that progress," writes Elmore, a professor of medicine at the University of Washington School of Medicine in Seattle.

Wender disputes the conclusion in the new research that improved treatment, and not mammography, is responsible for the decline in breast cancer deaths.

Other studies have shown that breast screening cuts the death rate, he said. In addition, in Denmark, which embraced screening 10 years later than most Western countries, a decline in breast cancer deaths was delayed by a decade.

The data in the new study came from the U.S. national Surveillance, Epidemiology, and End Results (SEER) program.

Over the period from 1975 to 2012, mammograms did shift the balance in how big tumors were when they were first discovered, researchers found.

The proportion of large tumors (that is, at least 2 cm in diameter) went from 64 percent to 32 percent. Over time, smaller tumors became the norm, representing 36 percent of the cancers at the start of the study period and 68 percent by the end.

Welch's team writes that the modest decrease in the number of large tumors detected over time "suggests that screening has had the desired effect of advancing the time of diagnosis of some tumors that were destined to become large." At the same time, a much larger increase in the incidence of small tumors suggests that screening has led to women being "considerably more likely to have tumors that were overdiagnosed than to have earlier detection of a tumor that was destined to become large."

The researchers estimate that at least two thirds of the reduction in deaths may be due to better treatment. But the results presume that the risk of developing breast cancer has not increased over time.

"Virtually every other expert group has concluded that, in fact, the incidence of breast cancer has been increasing," which would shift the benefit in favor of mammography, Wender noted.