Revisiting Breast Cancer Screening Guidelines
By Rachel Walden — October 26, 2009
A commentary in the current issue of the journal JAMA addresses breast and prostate cancer screening and the complexities and limitations of current screening approaches. The authors explain that while screening for the two diseases has increased, “the absolute numbers of more advanced disease have not decreased nearly as much as hoped for either cancer,” and that mortality has not decreased as much as expected.
It’s a complicated topic, even without getting into issues of access to screening and racial disparities in screening and treatment (which the JAMA piece does not). Essentially, the key limitations of screening are that widespread screening with current methods may sometimes detect slow-growing or inconsequential tumors, resulting in unnecessary treatment (because it’s not yet possible to predict which tumors present little risk), while in other cases, early detection doesn’t decrease mortality rates, because certain aggressive cancers, even when caught early, will not respond to treatment.
These messages are not easy message to convey or understand, and some worry that the related headlines may lead people to think screening is unwarranted. In a New York Times article on the commentary one biostatistician expressed concern that “the complex view of a changing landscape will be distilled by the public into yet another ‘screening does not work’ headline. The fact that population screening is no panacea does not mean that it is useless.”
The authors of the JAMA commentary don’t propose abandonment of breast and prostate cancer screening, however – they propose an improved approach consisting of determining validation tools that can identify and differentiate high and low risk cancers, reduction of treatment for minimal-risk disease, improved clinical and patient support tools to help guide decision-making, and identification of the the highest-risk patients for prevention initiatives. They suggest that there should be investment of significant funds – perhaps 10-20% of the $20 billion spent each year on screening – in these four areas “to improve screening, accelerate prevention research, and reduce harm from breast cancer and prostate cancer deaths.”
The New York Times article on the commentary also generated some controversy when it characterized the American Cancer’s Society’s response to the commentary as “saying that the benefits of detecting many cancers, especially breast and prostate, have been overstated” and that the organization was as a result “reconsider[in] its message about the risks as well as potential benefits of screening.”
The American Cancer Society has issued multiple responses to this characterization, explaining that it is standing by its screening guidelines. Organization representatives stated that they have long acknowledged that cancer screening isn’t perfect, but that “The bottom line is that mammography has helped avert deaths from breast cancer, and we can make more progress against the disease if more women age 40 and older get an annual mammogram.”
However, this research adds more fodder to the question of whether and how effective mammograms are. The Cochrane Collaboration, which creates meta-analysis of high quality systematic reviews, states in its review Screening for breast cancer with mammography that while screening likely reduces breast cancer mortality, ” it is not clear whether screening does more good than harm.”
I recently attended a workshop with the National Breast Cancer Coalition and they explained that the problem with pushing breast self exam and mammagrophy is that we’re wasting resources on promoting something that has been shown not to prevent breast cancer instead of putting our resources into finding what does prevent breast cancer.
A bombshell appeared in JAMA questioning the whole enterprise of cancer screening. We now have twenty years of data, and it’s not good. Dr Laura Esserman reviewed the data and has some harsh criticism. She concludes that screening the population for breast and prostate cancer has significant drawbacks, and the expected survival benefits have not materialized
Suppose that there is a one-in-1,000 chance that a woman in her 40s with no symptoms has breast cancer, and that 90 percent of the time a mammogram correctly classifies women as having cancer or not. If a woman in this group tests positive on her mammogram, what is the chance that she has cancer? The answer is not 90 percent. It is less than 1 percent, because of the large number of false positive results.