In 2010, a year after the US Preventive Services Task Force (USPSTF) revised its recommendations in favor of less screening mammography, NEJM published findings from a major cohort study suggesting that screening mammography offers only a modest mortality benefit.
This week in NEJM, Bleyer and Welch provide evidence that many women are being overdiagnosed with breast cancer due to screening mammography.
The authors examined breast cancer incidence trends over thirty years (from 1976-78 to 2006-08) using data from the Surveillance, Epidemiology, and End Results (SEER) database. The goal was to determine whether screening mammography, by identifying small lesions that on subsequent biopsy are found to be malignancies, leads to earlier detection of disease that otherwise would have progressed to late-stage cancer. If it does, then the increase in women undergoing screening mammography should be associated with a rise in early-stage cancer diagnoses and concurrent fall in late-stage cancer diagnoses.
Over the study period, the rate of early-stage cancer detection more than doubled (from 112 to 234 per 100,000 women), yielding an absolute increase of 122 additional early-stage diagnoses per 100,000 women. Meanwhile, the rate of late-stage cancer detection decreased by only 8 (from 102 to 94) cancers per 100,000 women. The authors posit that this means of the 122 additional early-stage cancers detected due to screening, only 8 would have otherwise progressed to late-stage cancer.
The remaining 114 per 100,000, in turn, represent “overdiagnoses.” The authors estimate that up to 31% of all breast cancer diagnoses in 2008 were overdiagnoses — and that in the last thirty years, more than a million women in the U.S. have been overdiagnosed with breast cancer. The implications of overdiagnosis can be profound: women undergo surgery, radiation, and chemotherapy that they arguably don’t need. Screening mammography may be saving the lives of some women, but it does so at a cost to many other women.
And just how many lives does screening save? Over the study period, the observed reduction in breast cancer mortality was greater than the reduction in late-stage cancer detection (20 vs. 8 per 100,000). Moreover, the reduction in mortality was actually greater among women under age 40 — i.e., women not exposed to screening mammography — as compared to women age 40 and over (42% vs. 28% reduction, respectively). These observations suggest the predominant driver of mortality reduction has been better treatment rather than increased screening.
NEJM Deputy Editor Dan Longo, M.D., states: “This interesting analysis points to an important gap in our knowledge: we cannot tell whether an individual cancer will follow a benign course or instead represents a threat to the patient’s life. This is the challenge if overdiagnosis is to be curtailed and mammography is to have maximal benefit.”
The findings of this study deepen the uncertainty surrounding screening mammography. We now have further data to illustrate the nuances of screening and the full range of potential consequences. But the more data we have, it seems, the less we can depend on guidelines to indicate the right course of action. Going forward, patients likely will need to play a more active role in decision-making. Meanwhile, physicians may need to dedicate even greater time to conversations about screening to ensure each decision is well-informed and tailored to the needs and wishes of the individual patient.
In your practice, has the use of screening mammography changed since the revision of USPSTF guidelines in 2009? How do you approach the subject of screening mammography with your patients? Will these study findings change your approach?
For more, see our latest Clinical Decisions:
What is the best strategy for breast cancer screening: start at age 40, at age 50, or not at all? This Clinical Decisions provides options based on new data on the risks and benefits of mammography. Vote and comment at NEJM.org.