SEARCH

SEARCH BY CITATION

When the British Medical Journal (BMJ) published results in February from a 25-year Canadian study comparing screening mammography with clinical breast examination or usual care, it led to strong reaction from medical professionals on both sides of the screening mammography debate.[1] It also set off a flurry of media coverage, some of which likely confused the public, says Otis Brawley, MD, chief medical officer of the American Cancer Society (ACS).

“My problem with the media is there are not enough people who want to go in-depth into the issue and understand that one tagline, ‘go get screened,’ is too simplistic,” says Dr. Brawley.

Nevertheless, even among medical professionals, the debate continues. The results of the Canadian National Breast Screening Study (CNBSS) concluded that annual screening of women aged 40 to 59 years does not reduce breast cancer beyond that of physical examination or usual care. They also found that 22% of screen-detected breast cancers were overdiagnosed, representing 1 of every 424 women in the trial who underwent screening.

The researchers, based in Toronto, compared breast cancer incidence and mortality over a span of 25 years in more than 89,000 women aged 40 to 59 years who either underwent mammography screening or did not. Women in the mammography arm aged 40 to 49 years and all women aged 50 to 59 years in both arms also received annual clinical breast examinations. Women aged 40 to 49 years in the control arm received a single clinical breast examination followed by usual care in the community.

During the study period, 3250 women in the mammography arm and 3133 women in the control arm were diagnosed with breast cancer whereas 500 women and 505 women, respectively, died of the disease. Thus, the authors found that cumulative mortality was similar in both arms. They concluded that although education, early diagnosis, and excellent clinical care should continue, annual mammography does not result in a reduction in breast cancer-specific mortality for women aged 40 to 59 years beyond clinical breast examination or usual care.

In an accompanying editorial, Mette Kalager, MD, PhD, an epidemiologist and screening researcher at the University of Oslo in Norway and the Harvard School of Public Health in Boston, Massachusetts, and colleagues concluded that long-term follow-up did not support mammography screening for women aged younger than 60 years.2 They and the authors of the CNBSS study recommended that the rationale for the practice “be urgently reassessed by policy makers.”

Criticism of the Study

  1. Top of page
  2. Criticism of the Study
  3. Support for the Study Results
  4. The Debate Continues
  5. References

The study results were interpreted differently by various groups. For example, the American College of Radiology and the Society of Breast Imaging called the study highly flawed, citing the use of outdated equipment and poor methodology. Elizabeth Thompson, MD, MPH, a New York-based radiation oncologist, concurs. “The study has many, many problems,” she says. “First, it was not blindly randomized. Patients were examined and then put into mammography or no mammography arms after the practitioner found a mass. That biased the screening group to find cancer, and more cancers were found in the screening group.”

She adds that the mammography machines and techniques were outdated and not applicable to US practices, in which digital mammography and other sophisticated techniques are applied. Furthermore, she says, in the United States, radiologists and not technologists read the scans, which was not always the case in the Canadian study.

In a response published later in BMJ, the study's lead author, Anthony Miller, MD, of the University of Toronto, answered such criticisms, noting that the centers in the study did not use secondhand mammography machines. Many were newly purchased for the trial, and all were carefully assessed by the trial's reference physicist, who ensured that the radiation dose was low and the study films were of high quality, he notes.

Responding to questions concerning the soundness of the trial's randomization, Dr. Miller added that the National Cancer Institute of Canada arranged for it to be assessed by 2 internationally recognized epidemiologists. “After full evaluation, they determined the randomization fully complied with accepted standards. In brief, the nurses had no role in the randomization. They were trained to recognize abnormalities, not make a diagnosis, and could have no opinion on whether the woman had a cancer or not and certainly could not judge the curability of any cancers found,” he wrote.

In her assessment of current data, however, Dr. Thompson stands by the current ACS-recommended guidelines that specify annual screening mammograms for women starting at age 40 years, and she notes that women should continue to have them as long as they are in good health. The ACS also states that women should be told of the benefits and harms of mammography. The organization also is currently reviewing its screening guidelines.

“The role of mammography not only gets patients to have a mammogram, but it also reinforces their relationship with a physician and allows them to get information about their risk factors,” Dr. Thompson says. Although she agrees that overdiagnosis is a problem, she adds that experts are continually working to improve technology and readings to reduce the number of unnecessary biopsies and call-backs.

Support for the Study Results

  1. Top of page
  2. Criticism of the Study
  3. Support for the Study Results
  4. The Debate Continues
  5. References

Meanwhile, Russell Harris, MD, MPH, a professor of medicine at the University of North Carolina at Chapel Hill and a screening expert, believes the CNBSS findings as well as those from other studies indicate that the benefit of mammography screening for women is very small. Among women in their 40s, a total of 2000 women would need to be screened for a period of 10 years for 1 woman to live longer, he notes.

Dr. Harris served as a member of the US Preventive Services Task Force until 2008 and participated in the screening recommendations they issued in 2009. Those recommendations, which also generated controversy, concluded that women aged 50 to 74 years should undergo biennial screening mammograms. They also recommended that the decision to initiate regular biennial screening mammography before age 50 years should be an individual decision that takes into account the patient's values regarding specific benefits and harms, and the knowledge that there is likely to be only a small benefit from such screening.

“We have been too focused in putting all our eggs into the basket of screening to reduce the problem of breast cancer in our society, and we can't pin all our hopes on it,” says Dr. Harris.

Researchers need to begin concentrating more on lifestyle issues such as obesity and lack of physical activity and their ties to the disease, he adds. “We need to get our minds off of screening as the final answer. We should screen some people but in a limited kind of way.” That may mean screening women only every other year in their 60s, he says.

The Debate Continues

  1. Top of page
  2. Criticism of the Study
  3. Support for the Study Results
  4. The Debate Continues
  5. References

The disagreement over the CNBSS is an indication of the “conundrum” the cancer community faces regarding the value of screening mammography, Dr. Brawley says. “I don't think we will ever get an answer,” he adds. “We can't, as a community, even come to grips about whether we are talking about 8 or 12 studies, but we're all talking about the same data.”

At the same time, he says, every one of the studies, whether it is favorable or less favorable to mammography screening, has flaws and inconsistencies. Mammography proponents tend to emphasize the problems with studies that conflict with their position, and vice versa, he says.

Although Dr. Brawley believes that mammography screening helps to reduce breast cancer mortality, he also thinks that its benefits have been exaggerated. “We always quote the best study and tend to exclude others,” he says.

The most mammogram-favorable Swedish Two-County Trial showed a 35% relative risk reduction in death for women in their 40s who are screened every 2 to 3 years, which translates to 8 women instead of 12 dying of breast cancer, Dr. Brawley says. “That's the best we can do with the current science, but I think 35% is a little high; I think it's more like 10 women dying instead of 12,” he adds, noting that in contrast, the US Preventive Services Task Force concluded a 15% percent relative risk reduction of death for women screened in their 40s.

Because these studies are continually evolving, all the major organizations that issue guidelines review the data every 5 to 7 years and update their results. The US Agency for Healthcare Research and Quality lists these guidelines on its Web site (guideline.gov). The ACS is currently in the midst of an extensive review of its breast cancer screening guidelines, which were last updated in 2004. The process of its review is new, based on recommendations from a statement by the Institute of Medicine. As a result, the ACS has commissioned a structured literature review by experts from the schools of public health at Duke University and the University of North Carolina at Chapel Hill. This extensive review will then be read by 15 experts who understand the issues surrounding the disease but have “as few financial and emotional conflicts as possible,” Dr. Brawley says. This group will then develop proposed guidelines, which will then be open for a public comment period.

We have been too focused on putting all our eggs into the basket of screening to reduce the problem of breast cancer in our society, and we can't pin all our hopes on it. —Russell Harris, MD, MPH

References

  1. Top of page
  2. Criticism of the Study
  3. Support for the Study Results
  4. The Debate Continues
  5. References
  • 1
    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.