A headline in the newspaper proclaims, “Cancer screening expert to radiologists: Stop lying about mammograms.” The newspaper is not a supermarket tabloid, but the Los Angeles Times. The article1, which appeared following the publication of the CNBSS2, was written by the newspaper’s science editor.
Suggesting that the 90,000-patient study is “rather inconvenient for the die-hard fans of mammography,” the author observes that breast cancer was more likely to be diagnosed in those who underwent mammography. She thoroughly covers the commentary of Welch3, a critic of mammography. What’s a woman to think?
The latest findings of the controversial CNBSS have once again stirred the pot of public opinion about the viability of mammography and the value of having mammograms. The US media helped, providing widespread coverage of this latest study and fanning the flames of the battle over breast screening anew.
The effect of this ongoing debate on patient care is hard to assess. In 2009, the USPSTF revised its screening-mammography guidelines, recommending that women have their first mammograms at age 50 (instead of 40) and biennially (rather than annually) thereafter. Despite the fact that the American Cancer Society (ACS) continues to recommend mammograms beginning at age 40, a reduction in mammography volumes in women ages 40 to 49 has been noted.4
In November 2013, the USPSTF announced that it had started to review its 2009 guidelines for breast-cancer screening. Half of the key questions that it plans to review systematically include evaluations of women starting at age 40.5
Radiology Business Journal has discussed the CNBSS and its impact with four highly experienced breast-imaging specialists: one at an academic center; one at a women’s health clinic at a large, integrated multispecialty health-care organization; and two at urban radiology practices. Have they modified any recommendations to patients as a result of the CNBSS? What are patients asking about having mammograms as a result of learning about the study results? How many patients are aware of the study?
None of these specialists report changing how they and their colleagues conduct their mammography practices. On the contrary, the fact that the study’s results were published despite obvious weaknesses angers the radiologists and makes them more committed to protecting their patients by adhering to guidelines of the ACR®, the ACS, the Society of Breast Imaging, and others.
In sum, they advocate that all women start having annual screening mammograms at the age of 40 and continue them through their 70s—and possibly longer, for healthy elderly women. An exception would be made for women with serious comorbidities for whom a survival period of five or fewer years has been predicted.
Young women identified as being at high risk (having BRCA1 or BRCA2 genes, a family history of breast cancer, or a high level of radiation exposure from prior treatment) need some form of breast imaging to complement an annual clinical exam, generally starting in their 30s (potentially as early as age 25). Breast MRI exams are recommended in lieu of mammograms for women under 30, and mammograms alternating with breast ultrasound or breast MRI should start at 30.
In addition, a disturbing number of family-practice physicians, internists, and gynecologists are not taking 20 years of data about the value of annual breast screening at face value. These physicians either are not discussing mammograms with their patients or are recommending mammograms under the guidelines of the USPSTF (biennial screening beginning at age 50).
One radiologist notes a lemming-like effect, calling it the if-it’s-good-enough-for-the–New England Journal of Medicine, it’s-good-enough-for-me syndrome. Jeffry Lindenbaum, MD, PhD, of Billings Clinic in Montana, is more diplomatic.
“The USPSTF guidelines carry a lot of weight,” he says. “There have been primary-care physicians within the clinic who have modified their recommendations based on them.”
Matthew Gromet, MD, a breast imager with Charlotte Radiology in North Carolina, states, “There is no serious dispute about mammography saving lives. There are about nine worldwide randomized clinical trials that proved this. They showed a mortality reduction of about 25% to 31%. Women, however, should realize that the actual benefit of annual screening is much greater, for several reasons.”
He continues, “First, the studies, by their design, attribute the measured benefits to all those invited to be screened, even though some didn’t come for their mammograms. Second, every study has a limited number of screening rounds—often, four or five. A lifetime of screening would logically multiply the benefits significantly. In short, the large randomized trials proved that there is mortality reduction from screening, but they underestimated the magnitude of the benefit.”
The other major benefit of annual mammography is catching cancer early, allowing less severe treatments and a better prognosis. “The USPSTF was concerned about stress and anxiety from mammograms—very subjective criteria,” Gromet notes. “It didn’t comment about the stress and anxiety of a woman with a poor-prognosis advanced cancer who requires multiple surgeries or a total mastectomy, who has axillary-node dissection causing lymphedema, and who has severe side effects from chemotherapy treatments. The failure of the USPSTF to consider the other benefits of mammography is a serious shortcoming in its report.”
Mammograms are far from perfect. Every radiologist interviewed was quick to point out that between 10% and 15% of cancers are not identified. Some patients are recalled for additional studies (the majority of them unnecessarily).
Mary Mahoney, MD, vice chair of radiology at the University of Cincinnati Medical Center in Ohio and director of breast imaging at the UC Barrett Cancer Center, comments that even with well-trained mammographers and technologists, there will always be technical and positioning issues. The center offers digital tomosynthesis (at no additional cost) to any woman with dense breasts who is having routine screening. This has helped reduce the number of recalls.
“When we need to recall a patient, we call the patient, tell her that we need to take more images to rule out something we could not see clearly, and advise her that for the overwhelming majority of women who come back, the repeat mammogram will positively clear things up,” she says. “Women are anxious about returning for an additional work-up, but the inconvenience is minor, compared with the greater harm of missing a small, easily curable cancer.”
The imaging department at the Women’s Health Clinic of Billings Clinic also has added digital tomosynthesis for use with dense-breasted women. “This isn’t perfect, either,” Lindenbaum says. “We use ultrasound and breast MRI as problem solvers to supplement inconclusive mammograms, and these exams are very useful, but no study exists today with 100% positive predictive value. Our recall rates fall within the Mammography Quality Standards Act guidelines, but it would be nice if we were perfect.”
Critics of mammograms might focus on imperfections, but the women who come to Billings Clinic for screening do not. Neither do the patients of X-Ray Associates of New Mexico (Albuquerque). At that practice, according to Michael Linver, MD, he is preaching to the choir. “The women who have mammograms are true believers,” he says. “The ones I would really like to talk to are the ones who don’t come to our imaging center—or any other facility offering breast screening.”
He does try: Linver says that for years, he has routinely done a lot of public speaking, mostly arranged through religious groups and community organizations. “In my screening-mammogram presentation, I present the basic facts and the statistics from clinical studies,” he says. “I tell the people who attend what resources are available in the community to help them, if they do not have insurance or cannot afford the copayment. Some people tell me afterward that nobody, not even their physicians, has talked to them about what I presented.”
Linver is openly critical of physicians—especially family physicians and internists—who fail to talk to their patients about mammograms or who say that their patients don’t need them. He will call physicians whose patients—a large number—show up at the practice and tell the technologists that their physicians have said that they don’t need mammograms—but that they decided to have screening exams anyway.
Lindenbaum makes a point of talking to his colleagues at Billings Clinic, taking advantage of formal opportunities as well. He spoke with the clinic’s chairs of family practice, internal medicine, and obstetrics/gynecology immediately following the publication of the latest CNBSS findings. Their reaction was positive and supportive.
Mahoney says that there are many opportunities—including grand rounds, multidisciplinary tumor boards, and newsletters—at the University of Cincinnati Medical Center to discuss breast imaging. The CNBSS recently was a topic for residents and fellows to discuss. Physicians affiliated with the medical center understand the importance of identifying young patients at risk and of making sure that their patients have breast screening, she says.
Linver suggests that federal and state licensing requirements for mammogram interpretation should be tightened. “The key weakness of the mammography exam is its inability to show cancers buried in dense glandular or fibrous tissue,” he says, “but an equal weakness is the fact that there are a lot of radiologists reading mammograms who are not truly qualified to do so. In Europe, it is necessary to read 5,000 mammograms each year to qualify to interpret screening mammograms. In British Columbia, you need to pass an exam. In the United States, you only need to read 480 mammograms a year—and you don’t even have to read them accurately.”
Every radiologist interviewed welcomes the idea of using telemammography to send mammograms from low-volume facilities to centers of excellence. Linver predicts that this could increase national sensitivity statistics by at least 5%. Mahoney points out that the technology, used with some mobile mammography vans, clearly demonstrates the viability of telemammography services.
Likewise, all interviewees recommend that women who are at high risk, as defined by the ACS, should begin breast screening earlier. Mahoney recommends that young women who received radiation therapy in their teens start breast-cancer screening eight years following this treatment or at the age of 25, whichever is later. For very young women, the radiologists recommend ultrasound or MRI breast exams, and—starting at age 30—an annual mammography screening. Alternating use of screening methods is recommended, so that a young at-risk woman will be screened every six months.
Linver says that when he identifies a patient who meets high-risk criteria, he advises her referring physician, and he also puts in his report a recommendation that the patient be independently assessed using breast-cancer–risk models or be tested for the BRCA mutations. If the patient is determined to be at high risk, many insurance companies will cover a breast MRI exam.
Billings Clinic has a genetic counselor on staff. Charlotte Radiology has an information-packed website that its patients (and other women) rely on, according to Michelle Russell, the practice’s breast-services marketing manager. She says, “We want women to be informed and to make intelligent decisions. Therefore, we were very proactive in our education. We immediately posted information about the CNBSS. We heard from many women in our community that the clarity is not only appreciated, but life saving.”
Did patients know about the study? Did its findings rattle them? About 60% of the patients who’ve had screening mammograms since the study’s results were publicized ask questions about them, technologists at Billings Clinic report, and some additional patients say that they have inquired elsewhere about the CNBSS findings.
Technologists are given relevant information to share with patients, and (depending on the individual situation) can ask clinical colleagues to be available to answer questions as well. At Charlotte Radiology, any staff member who interacts with patients is trained to discuss the value of mammograms and to know at what point he or she might need to involve a colleague.
A breast MRI exam, even with insurance coverage, is expensive. The design for the Women’s Health Clinic, built in 2013, includes a room for whole-breast ultrasound equipment that Mahoney expects to be purchased during the next round of capital-equipment upgrades. Gromet and Linver say that their practices are carefully watching as this technology improves. They also have hope for low-cost breast-MRI capabilities that are being developed.
Has the CNBSS had an impact on the number of women presenting for breast screening? It’s too early to tell, but every radiologist interviewed cites a drop in volume following issuance of the USPSTF guidelines.
What is sad, they add, is the case of the patient who presents with a lump in her breast and who receives the diagnosis of a cancer that could have been caught much earlier. They agree that a woman should not be asking whether an annual mammogram will increase her life expectancy. Better questions, they say, are what her benefit would be from a lifetime of screening—and if she does eventually develop breast cancer, whether catching it at an early stage would make her quality of life better than it would be if the cancer went undiagnosed for a longer period.
Nobody is forcing women to have mammograms. The radiologists interviewed hope that they will.
- Kaplan K. Cancer screening expert to radiologists: stop lying about mammograms. Los Angeles Times. Published February 19, 2014. Accessed April 8, 2014.
- Miller AB, Wall C, Baines CJ, Sun P, To T, Narod SA. Twenty five year follow-up for breast cancer incidence and mortality of the Canadian National Breast Screening Study: randomised screening trial. BMJ. 2014;348:g366.
- Welch HG. Don’t slam Canada for mammogram study. CNN Opinion. Published February 19, 2014. Accessed April 8, 2014.
- Sprague BL, Bolton KC, Mace JL, et al. Registry-based study of trends in breast cancer screening mammography before and after the 2009 U.S. Preventive Services Task Force recommendations. Radiology. 2014;270(2):354-361.
- US Preventive Services Task Force. Draft research plan: screening for breast cancer. Published December 2013. Accessed April 8, 2014.
Mammography’s Ongoing Ordeal
The recent publication of the Canadian National Breast Screening Study1 (CNBSS) was not the first time that mammography has been challenged—and it won’t be the last. An earlier media frenzy erupted after the 2013 publication of an article by Bleyer and Welch.2
The authors conclude that “screening mammography has only marginally reduced the rate at which women present with advanced cancer”; that there is substantial overdiagnosis; and “that screening is having, at best, only a small effect on the rate of death from breast cancer.”2 Kopans3 dissected their research and has formally called for the withdrawal of the article by the publisher.
The brouhaha caused by the United States Preventive Services Task Force (USPSTF) 2009 revision of mammography-screening guidelines has receded, but the controversial decision to recommend the first mammogram at age 50 (as opposed to 40) will soon be revisited. In November 2013, the USPSTF announced that it had started to review its 2009 guidelines for breast-cancer screening.
The proper frequency for breast screening continues to be debated, even among supporters of mammography. In addition to questioning the value of an annual mammogram, regardless of the patient’s age, some critics question its cost. Research reported by O’Donoghue et al4 in February 2014 used 2010 data and economic models to show that billions of dollars could be saved, every year, by switching to biennial screening. The authors found that the simulated cost of annual screening for 85% of US candidates was $10.1 billion (compared with $2.6 billion for screening every two years).
Why the CNBSS team decided to revisit and analyze data from a clinical trial that was profoundly and repeated criticized for numerous study-design flaws is unknown. Even one of its own consultants has decried the study.
Yaffe5,6 has pointed out that the quality levels of images acquired using today’s mammographic technology and the equipment of 1980 are so dissimilar that findings from that time are simply not applicable today. He also has reported that while involved with the study when it was being conducted, he and outside radiologist consultants complained frequently to the investigators about quality issues: subpar image quality, exams performed by inexperienced technologists, and mammograms interpreted by underqualified radiologists.
He has reiterated another frequently made criticism: Because the patient cohort consisted only of volunteers who had undergone clinical exams prior to being assigned to a group, this compromised randomization. Yaffe believes that the nurses who examined patients subsequently assigned women with large, palpable cancers to the screening arm of the study so that these women would be assured prompt mammograms and immediate treatment. By itself, this might have distorted the study, Yaffe and others assert. Yaffe also has been critical of the media for presenting the data as a new study.