Published On Sep 30, 2015
Second Opinion–Fall 2015
A Proto feature about changing screening protocols draws a vigorous response.
Miss on Mammography
The readers of Proto need to know that the article covering breast cancer screening (“Screening Under Scrutiny,” Spring 2015) does not reflect the scientific evidence.
1. There are absolutely NO data (none) that show that any of the parameters of screening change abruptly at the age of 50. There is NO scientific or biological support for using the age of 50 as a threshold for screening or analyzing the data as the article has suggested. The age of 40 is the threshold simply because the scientific studies did not include anyone younger than age 40. Since we base recommendations on the scientific evidence, the age of 40 is the threshold.
2. The most rigorous scientific studies, called randomized controlled trials (RCTs), show that screening and early detection saves lives for women ages 40–74. No older women were included in the trials, but there is no reason to believe that healthy older women will not benefit from screening.
3. When screening has been introduced for the general public, for women ages 40 and over, the death rate declines even when all women have access to the same therapy. Therapy has improved, but therapy saves the most lives when breast cancers are treated earlier.
4. A major study at Massachusetts General Hospital and Brigham and Women’s Hospital (two of Harvard’s teaching hospitals) showed that more than 70% of the women who died from breast cancer were among the 20% who were not participating in screening.
5. The National Cancer Institute’s Cancer Intervention and Surveillance Modeling Network (CISNET) computer models all agree that the most lives are saved by annual mammography beginning at the age of 40.
6. Ductal carcinoma in situ (DCIS) are a series of lesions that look like breast cancer, under the microscope, to the pathologist. Breast cancers begin in the inside of the milk ducts, and DCIS lesions are thought to be the earliest form of breast cancer. No one knows how many or which of these lesions will progress to become invasive breast cancers. When DCIS become “invasive” breast cancers, having broken out of the milk ducts, they can spread through the lymphatics or blood vessels around the ducts to other parts of the body (metastatic disease). When cancer spreads to other organs, it becomes lethal. The article is incorrect in suggesting that the “vast majority...won’t cause harm.” We actually do not know which of these lesions, if left untreated, will not cause harm.
7. Unfortunately, these “harms” of screening have been exaggerated by calling them, as is repeated in the article, “false positives.” These have been misunderstood as suggesting that women are told that they have breast cancer when they do not. In fact, most “false positives” are simply women recalled from screening for a few extra pictures or an ultrasound, and the vast majority are reassured that everything is all right. The recall rate for mammography is approximately 10%, which is the same as for cervical cancer screening (Pap testing). Some inconvenience and unfortunate anxiety is associated with being recalled, but for most women it is not “huge emotional distress” as suggested by this article.
8. No one has ever suggested that mammography is the ultimate answer to breast cancer. It does not find all cancers, and does not find all cancers early enough to result in a cure. We all hope for a universal cure, or a safe way to prevent breast cancer, but neither is on the horizon. Mammography and early detection are available today, and thousands of lives are being saved because of screening.
9. The Proto article is simply wrong in suggesting that improvements in technology meant finding “smaller and smaller lesions, many of which weren’t dangerous.” There is no scientific evidence that “invasive” cancers aren’t dangerous. The few papers that have suggested this have been shown to be based on flawed analyses. Finding smaller invasive cancers leads to curing more women.
10. The article’s suggestion that “the USPSTF committee had ample evidence that earlier, more frequent mammography wasn’t desirable” is also false. In fact, the analysis done for the USPSTF using the CISNET models showed that the most lives are saved by annual mammography starting at the age of 40. The models showed that if women waited to start screening until the age of 50, and then every two years, then tens of thousands of lives would be lost that could be saved by annual screening starting at the age of 40.
11: The article incorrectly categorized the Canadian National Breast Screening Studies (CNBSS) as “landmark.” The fact is that they were notoriously compromised trials. They failed to follow fundamental rules for these trials. They did not have a blinded assignment of women (you cannot know anything about the women before assigning them to the screening arm or the control arm to avoid imbalancing the trial). “Blinded” assignment is critical, but was violated by the CNBSS. This fundamental failure casts doubt on their results, which differed from all the other trials. In addition, it is astonishing that a trial to test the value of mammography not only did not have good mammography, but two other radiologists and I were asked, by the investigators, to review the quality of the mammograms, and we judged them to be “poor to unacceptable” for much of the trial. Their own physicist is still highly critical of the mammograms. How can the Proto article suggest this was a “landmark” trial unless the author and editors meant that it showed how to not do a trial? The recent summary of the more modern UK Age Trial shows that screening women ages 40–49 saves lives. The CNBSS was so flawed that it should not be used to advise women.
12. The article claimed that mammography was not as important because modern treatment for breast cancer “has become increasingly effective even when tumors aren’t found and treated at the earliest stages of disease.” Therapy has improved, but the author and editors seem to have ignored Dr. Daniel Haber, a leading oncologist, who was quoted as stating with regard to screening and early detection, “That’s critical for curing cancer before it spreads.” Therapy has improved, but the most lives are saved when breast cancer is treated earlier.
13. The author and editors did not check their arithmetic with regard to claims of “overdiagnosis” in the Canadian NBSS. The article quoted a rate of 22%, but had they read the paper they would have found that there were 3,250 cancers in the mammography arm and 3,133 in the control arm, a difference of only 117, which is 3.7%. There was little if any “overdiagnosis” even in the CNBSS, which was compromised from the start. The scientific evidence shows that there is little if any “overdiagnosis” of invasive breast cancers.
14. The reduction in deaths for women screened is 30% or more and not the 10%–15% suggested by the article.
15. The article suggests that only women at high risk should be screened. This sounds like it makes sense until you realize that women at high risk (those with a genetic predisposition with BRCA1 or 2 gene mutations, and women with a family history and other risk factors) account for only 25% of women diagnosed each year with breast cancer. The vast majority of women (75%) who are diagnosed each year have no known elevated risk. Why would you exclude them from screening?
It is bad enough that scientifically flawed papers have been published due to poor peer review at the medical journals, but it is extremely disappointing that a Massachusetts General Hospital publication would provide its readers with so much misinformation. All women should be encouraged to participate in breast cancer screening every year starting at the age of 40.
Daniel Kopans // Senior Radiologist, Dept. of Radiology, Massachusetts General Hospital, Boston
In the weeks since Proto discussed skepticism of cancer screening (“Screening Under Scrutiny,” Spring 2015), new developments from my research and others’ have highlighted the risks of overdiagnosis and questioned the lasting benefits of screening mammography. Although these studies advance research, they add to the public’s confusion and frustration with the medical community, which seems to be incapable of agreeing on the true value of screening mammograms.
We know that women have different risks of developing breast cancer, and it is natural to suspect that women also have different risks of overdiagnosis. But when deciding whether to screen, and how often, the same blanket mammography recommendations are given to almost everyone.
We should strive for more individualized recommendations for breast cancer screening. For those who believe there isn’t enough screening, individualization has the potential to direct more screening at higher-risk women. And for those who, like me, believe there is too much screening, individualizing should be able to reduce screens that lead to overdiagnosis.
Charles Harding // Data Scientist, Seattle, Wash.