Addition Of Screening Mammograms Adds No Benefit, But Causes Harm

Clamoring over screening protocols has started anew with the release of a Canadian study confirming earlier results that annual mammography screening did not reduce breast cancer deaths. Instead, it contributed to the diagnosis and treatment of conditions that would not have been life threatening in the first place. We’ve heard it all before, with varying degrees of belief, but there’s something different about this study and what it means for women. Not only did screening mammography fail to reduce breast cancer mortality, there is new evidence that overdiagnosis and overtreatment are major hazards for women of average risk.


Details of the Canadian Study

The Canadian study compared the incidence and mortality of breast cancer for nearly 90,000 women ages 40 to 59 who did, and did not, have an annual screening mammogram. The women were randomly assigned to two groups. The women in the mammography group had annual mammograms from 1980 to 1985. Those in the control group had no annual mammography. All women had routine health care with attention to breast health. The women in their 50s in both groups received annual clinical breast exams.

Then the researchers meticulously followed these women for 25 years to find out if they developed breast cancer, the characteristics of their tumors if they did develop the disease, and whether or not they died from breast cancer.

At the end of the screening period, there were more women in the mammography group diagnosed with an invasive (not stage zero) breast cancer, but the number of deaths between the groups was about the same – 500 women in the mammography group, compared to 505 women in the control. In other words, finding those additional breast cancers through screening did not translate to a reduction in the number of deaths.

This finding is, in itself, not that surprising.It confirms the results found after 13 years of follow-up. It also confirms more recent—and contentious evidence that as many as one in three of breast cancer cases are overtreated in the sense that the disease would not have affected the woman in her lifetime.

Some Screening History

Before mammograms became standard screening, a review in the Annals of Internal Medicine (1976) found little evidence on the long-term effects of mammography or the inherent risks associated with radiation. The authors cautioned that there was not enough evidence to warrant widespread promotion of mammography as a public health measure. Since then, there have been many studies of screening mammography, including statistical modeling to illustrate past trends and make projections about the relationship between screening and breast cancer mortality, and clinical trials that randomly assign people to receive (or not receive) the screening intervention.

Seven major clinical trials were conducted between 1963 and 1982 involving over half a million women. The Canadian study makes it eight, making screening mammography the most studied screening technique. Still, definitive answers are hard to come by in biomedicine. Finding meaning in data requires looking at all relevant studies, individually and then together.

The Lancet published the first systematic review of the trials in 2000, assessing their quality and evaluating their findings. When taking methodology into account, they concluded that there was no reliable evidence to justify mass screening because the studies had inconsistencies and some were highly flawed or had poor quality data. The best studies, like the Canadian one, showed zero mortality benefit. The studies were evaluated in greater detail in 2001, also in the Lancet, and there continued to be periodic updates with new data, concluding in 2013.

The most acceptable conclusion was that routine screening reduced breast cancer mortality for women ages 50 to 69 by about 15 percent. The remaining 85 percent died at the same rate as unscreened women who developed breast cancer. Two additional meta-analyses of these seven clinical trials (2002 Humphrey review and the 2006 Armstrong review) also concluded that reductions in mortality from mammography screening were variable, and fairly modest.

The Current Consensus from the Most Reliable Studies

The consensus on screening mammography’s benefit is now about 15 percent. But individual study results range from zero (like the Canadian study) to 35 percent and even higher, but from questionable observational methods. A Norwegian study published in the New England Journal of Medicine (2010) found that mammograms combined with modern treatment reduced the death rate by only 10 percent, and that the effect of mammograms alone could be as low as 2 percent, or even zero. Even with its limitations (it was not a randomized control trial), it offered a critical insight: improvements in treatment, which include less toxic radiotherapy and chemotherapies, are critical when it comes to reducing deaths from breast cancer. Quality treatment, not screening, is responsible for reducing the overall number of breast cancer deaths post-diagnosis.

The Canadian study is, once again, inviting pro-mammography groups like the American Cancer Society, the American College of Radiologists, and the Society of Breast Imaging, among others, to reconsider their screening recommendations. But this study is especially powerful because it has 25 years worth of high quality data.

Overdiagnosis and Overtreatment

Like the Norwegian study, the Canadian researchers speculate that since there was no reduction in deaths by adding mammography screening to the breast health repertoire, good health care that includes annual clinical breast exam by a professional examiner is just as good. What’s more, this strategy comes without the potential harms associated with routine mammography screening, radiation of course, but also most crucially overdiagnosis (the detection of conditions that, if left unattended, would not cause symptoms or death) and overtreatment (the treatment of indolent conditions with surgery, radiation, and chemotherapies).

Whereas under-diagnosis with both clinical breast exams and mammography screens is well known, with each missing a different 10 percent of cancers, when it comes to the problem of over-diagnosis, there has been considerable resistance. We can’t tell, exactly, which breast cancers might end up be life-threatening down the road, so all of them are treated like cancers.

However, an analysis of the effects of three decades of screening on breast cancer incidence strongly suggests that about 30 percent of women who get regular screening mammograms are over-diagnosed and overtreated. The Canadian study finds a similar magnitude of overdiagnosis, that about 22 percent of the screen-detected invasive cancers in the mammography group were over-diagnosed. An important caveat here is that the Canadian study counted only invasive breast cancers, not the stage zero breast cancer, Ductal Carcinoma In Situ (DCIS), which accounts for 1 in 5 breast cancers found from mammograms. The exclusion of DCIS may account for the 10 percent difference in overdiagnosis estimates. What they both show, is that screening averts fewer deaths than anticipated and causes more harm.

A Take-Home Message

Here’s where the Canadian study does a great service. By comparing the mammography group to a similar group of women who received regular care and annual clinical breast exams but not mammograms, we now know that the addition of mammography screening was not beneficial but, because all treatments have inherent risks, it did cause harm. With this data, individuals and physicians may become more comfortable in reassessing screening recommendations.

Be Sociable, Share!

Articles & Posts