Mammograms are far from strangers to the paparazzi. It was only recently that a television news reporter from a prominent broadcasting company reluctantly agreed to have a mammogram performed on live television to promote breast cancer screening and help “save lives” . That same study diagnosed her with breast cancer, which was followed by the impressive statement from her physicians: “that mammogram just saved your life”. But over the last day or two several major newspapers are presenting a very different story. The titles of the articles from two prominent newspapers say it all: “Vast Study Casts Doubt on Value of Mammograms” and “Mammogram Screenings Don’t Reduce Cancer Death Rates, Study Finds” [2,3]. So which one is it? Do they save lives, or don’t they? And the obligate follow up question: should we continue to subject patients to them, or not?
The answer to that remains controversial. One can start by looking at the study referenced by the above-mentioned newspaper articles, a Canadian study that was recently published in the British Medical Journal . The study was a randomized screening trial that included 89,835 women aged 40-59, with a primary outcome of death from breast cancer. Participants were randomly assigned to mammography or control. More specifically, all participants received a preliminary clinical breast exam (including being taught self-examination techniques), followed by:
• Four rounds of yearly two-film mammography AND physical examinations for women in the mammography group.
• Instructions to remain under the care of their family doctor and receive “usual care” for women aged 40-49 in the control group (assuming their preliminary breast exam was negative)
• Annual physical examinations without mammography for the next four years for women aged 50-59 in the control group
The screening period (during which the above tests were performed) lasted a total of five years, including the preliminary clinical breast exam; a follow-up period of 25 years ensued (mean follow-up was 22 years). It should be noted that the primary analysis only included deaths that were secondary to invasive breast cancers diagnosed during the five-year screening period. But participants were followed from the date of randomization to the end of the follow-up period, and secondary analysis data were also presented. The results may surprise more than a few of us.
A total of 1190 invasive breast cancers were diagnosed during the screening period: 666 (1.48%) in the mammography group (n = 44,925) and 524 (1.16%) in the control group (n = 44,910). Of these, 180 (27%) women from the mammography group and 171 (32%) from the control group died secondary to the cancer during the 25-year follow-up period. Furthermore, during the entire study period (including screening and follow-up periods) 3250 (7.23%) women in the mammography and 3133 (6.97%) women in the control group received a diagnosis of breast cancer. Of these, 500 and 505, respectively, died from breast cancer.
Given the rather striking similarities between the primary outcome in both groups, the study’s conclusions are as one would expect: “…annual mammography does not result in a reduction in breast cancer specific mortality… The data suggest that the value of mammography screening should be reassessed”.
Before continuing with our analysis, and in the interest of rational, unbiased scientific understanding, a few flaws in the study are worth pointing out: even though the overall follow up time (25 years) was appropriately long, patients in the mammogram group only received four years worth of screening. Furthermore, the paper provides no data as to what type of screening (if any) was pursued and/or received by patients in any of the groups after the four year screening period was over, let alone what effect this unaccounted-for screening may have had –for example, the lack of difference between both groups at 25 years could potentially be explained by large numbers of participants in both groups deciding to pursue mammography after the screening period was over.
That being said, the study’s results are, without a doubt, discouraging for mammograms. Moreover, they are not entirely new. A 2012 study published in the New England Journal of Medicine specifically looked at mammograms as a way to decrease late-stage presentation of breast cancer and thus improve survival. The conclusions are similar: while there appears to be an increase in the detection of early-stage cases, mammography “has only marginally reduced the rate at which women present with advanced cancer” . Other authors have even suggested that the mortality benefits that were first reported during the 1980’s and 90’s may be more a result of subsequent advances in therapy than the mammographic screening itself –ie, therapies that became available in the late 1990’s and 2000’s, after the original mammogram data was published .
So what is the solution then? The Canadian study indirectly suggests that mammography is no better than regular clinical breast exams. Is the solution to encourage patients to seek out yearly clinical breast exams and/or encourage them to perform regular self-examinations? At least for regular self-examinations, it depends on where you look: a 2002 study performed in Shanghai showed no reduction in mortality from breast cancer even after intensive instruction and self-examinations practiced under medical supervision .
But back to the topic of mammograms. It can be argued that even though the final word has yet to be spoken, the evidence appears to be piling against them. In the interim, they remain a common tool in the arsenal of breast cancer detection. So what should we tell patients? When to start? How often? Guidelines vary from institution to institution, and even large organizations still disagree on the best approach –e.g., the United States Preventive Services Task Force recommends starting at age 50 and continuing every other year, while the American Cancer Society recommends starting at age 40 and continuing annually . It will be interesting to see how the medical institution reacts to this latest paparazzi frenzy, and ultimately, what better options will become available.
Dr. Miguel A. Saldivar is a first year resident at NYU Langone Medical Center
Peer reviewed by Susan Talbot, MD, Attending Physician, Dept. of Medicine Division of Hematology/Oncology, NYU Langone Medical Center
1. ABC News’ Amy Robach reveals breast cancer diagnosis. http://abcnews.go.com/blogs/health/2013/11/11/abc-news-amy-robach-reveals-breast-cancer-diagnosis/
2. Vast Study Casts Doubt on Value of Mammograms. http://www.nytimes.com/2014/02/12/health/study-adds-new-doubts-about-value-of-mammograms.html
3. Mammogram Screenings Don’t Reduce Cancer Death Rates, study finds. http://www.latimes.com/science/la-sci-breast-cancer-mammography-20140212,0,6053567.story#axzz2t7PjgvSH
4. Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial. Miller, et al. BMJ 2014;348:g366 doi: 10.1136/bmj.g366. http://www.bmj.com/content/348/bmj.g366
5. Effect of Three Decades of Screening Mammography on Breast-Cancer Incidence. Bleyer, et al. N Engl J Med 2012; 367:1998-2005. November 22, 2012. DOI: 10.1056/NEJMoa1206809. http://www.nejm.org.ezproxy.med.nyu.edu/doi/full/10.1056/NEJMoa1206809
6. Screening for Breast Cancer: Evidence for Effectiveness. Fletcher, et al. July 26, 2013. Uptodate online database. http://www-uptodate-com.ezproxy.med.nyu.edu/contents/screening-for-breast-cancer-evidence-for-effectiveness?source=see_link&anchor=H3757815#H3757815
7. Randomized trial of breast self-examination in Shanghai: final results.. Thomas, et al. J Natl Cancer Inst. 2002 Oct 2;94(19):1445-57. PMID: 12359854. http://www.ncbi.nlm.nih.gov/pubmed/12359854
8. With differing mammogram guidelines, I’m not sure when to begin mammogram screening. What does Mayo Clinic recommend? http://www.mayoclinic.org/tests-procedures/mammogram/expert-answers/mammogram-guidelines/FAQ-20057759