Value of Mammography Questioned by Canadian Trial

Roxanne Nelson

February 11, 2014

The value of screening mammograms for detecting breast cancer and reducing mortality has been fiercely debated. Studies have come to conflicting conclusions, and a new study showing that mammography has no effect on breast cancer deaths will undoubtedly fan the fires once again.

The Canadian researchers conclude that annual screening mammography in women 40 to 59 years of age does not reduce mortality from breast cancer beyond that of a physical examination or usual care. Results from the trial were published online February 11 in BMJ.

"Thus, the cumulative mortality from breast cancer was similar between women in the mammography arm and in the control arm," write Anthony B. Miller, MD, from the Dalla Lana School of Public Health at the University of Toronto, and colleagues.

In addition, almost one quarter of cancers detected during screening were overdiagnosed, they point out.

However, the trial has been heavily criticized, and a leading mammography proponent noted that "it's been known for years that the study was compromised from the start."

In an accompanying editorial, Mette Kalager, MD, PhD, from the University of Oslo, Norway, and the Harvard School of Public Health in Boston, and colleagues, agree with the researchers that the "rationale for screening by mammography be urgently reassessed by policy."

The editorialists note that as "time goes by," more efficient mechanisms are needed to "reconsider priorities and recommendations for mammography screening and other interventions."

They point out that this is not as easy as it sounds, because "governments, research funders, scientists, and medical practitioners may have vested interests in continuing activities that are well established."

Trial Was "Compromised From the Start"

A leading proponent of mammography, Daniel Kopans, MD, professor of radiology at Harvard Medical School and senior radiologist in the Department of Radiology at Massachusetts General Hospital in Boston, disagrees with the study results. "It would be an outrage for women if access to screening was curtailed because of the poor results in the Canadian National Breast Screening Study [CNBSS] when it has been known for years that the trial was compromised from the start," he told Medscape Medical News.

Dr. Kopans said that he reviewed the mammograms in the trial in 1990, and "I can personally attest to the fact that the quality was poor." He added that even the trial's own reference physicist noted that the quality "was far below the state of the art," even for the early 1980s.

"The documented poor quality of the mammography is sufficient to explain the study results, and all of the above disqualifies the CNBSS as a scientific study of mammography screening. But it is even worse than that," Dr. Kopans said. The CNBSS also violated the fundamental rules of randomization by conducting a clinical examination of women prior to randomization and then placing women with suspicious findings in the mammography group.

The fundamental compromise of the allocation process, which is "indisputable," along with the documented poor quality of the mammography should have, long ago, disqualified the CNBSS as a legitimate trial of screening mammography, Dr. Kopans explained.

A joint statement from the American College of Radiology and the Society of Breast Imaging echoes these concerns. It asserts that the study is "an incredibly misleading analysis based on the deeply flawed and widely discredited" CNBSS.

Thus, these results, along with others from the CNBSS trial, "should not be used to create breast cancer screening policy as this would place a great many women at increased risk of dying unnecessarily from breast cancer," it states.

Will Not Change Practice

Two other mammography experts approach by Medscape Medical News also have reservations about the trial.

Catherine M. Dang, MD, associate director of the Wasserman Breast Cancer Risk Reduction Program and surgeon at the Saul and Joyce Brandman Breast Center at Cedars-Sinai Medical Center in Los Angeles, said the study will add to the controversy over the value of mammography.

"It has the benefit of a large number of participants, which will yield significant results because of the large number alone, but lacks much of the detail needed to really make this a sound study on which to radically alter our screening recommendations," she told Medscape Medical News. "Most studies to date actually state that screening mammography is more reliable than clinical breast exam alone."

Dr. Dang noted that this study differs from others, in that women not undergoing mammography were very closely followed with clinical breast examination on an annual basis and taught self-breast-exam. "In clinical practice, the reality is that many women are not getting a breast examination annually by a physician or trained healthcare provider and are not being taught breast self-exam," she said. "I would argue that the 'usual care' given to these research study participants is not the clinical care for many women in the United States."

"I don't think this study alone will change national guidelines in the United States regarding screening mammography because there are a number of studies that have shown a mortality benefit," she added.

Helen Cappuccino, MD, a surgical oncologist from the Roswell Park Cancer Institute in Buffalo, New York, pointed out that when looking at a population, "one can often make an argument that screening is not cost-effective, or that the population of patients in question isn't benefited by reduced mortality from breast cancer mammographic screening."

However, "we are in the business of caring for patients, not populations," Dr. Cappuccino explained. Mammography can help to detect earlier, smaller breast cancers, which are hopefully more readily treatable for cure," she said.

Study Limitations

Dr. Cappuccino noted that there was crossover in the study population. "Some of the patients who were not supposed to receive screening mammograms did, in fact, receive 1 or more mammogram."

According to the 2002 CNBSS report, more than 20% of the patients who weren't supposed to receive a mammogram actually did so outside the study, she explained.

"The impact of that is unclear," she said. "Furthermore, the unscreened patients were carefully educated about breast care and had regular annual follow-ups, so the results of this can't be translated to a general population, where such education and regular follow-up might not actually occur," Dr. Cappuccino noted.

Another limitation of the study is the similarity in tumor size and proportion of patients with nodal involvement at diagnosis in the 2 study groups. Dr. Dang explained that, in reality, breast cancers detected with modern digital screening (used in majority of facilities in United States) are usually about half the size of those detected using "film-screen mammography," and are less likely to present with nodal involvement.

Also, because of the design, there was no information about tumor stage, tumor characteristics, or characteristics of the study participants. Although the participants were randomized, it is unclear whether there was any stratification for age, family history, or other risk factors, Dr. Dang pointed out.

"I would also argue that mortality should not be the only end point, although it is perhaps the easiest to determine based on registries of death; it is very difficult to follow patients for that length of time," she explained. "Mortality from breast cancer can be predicted by stage, age, and tumor characteristics, but we are not given that information in the study for those with breast cancer, so it is unclear whether the 2 groups were really equal."

Morality is also affected by things like treatment with systemic adjuvant therapy, and again, there is no information on that in the study, Dr. Dang added.

Study Details

In the CNBSS, 44,925 women from 40 to 59 years of age were randomized to 5 annual mammography screens and 44,910 were randomized to no mammography (control group). All women 50 to 59 years received annual physical breast examinations, as did women 40 to 49 years in the mammography group. Women 40 to 49 years in the control group received a single exam followed by usual care.

The participants were recruited from 15 screening centers in 6 Canadian provinces from 1980 to 1985 (Alberta, British Columbia, Manitoba, Nova Scotia, Ontario, and Quebec).

The primary outcome measure was breast-cancer-specific mortality.

During the 5-year screening period, 666 invasive breast cancers were diagnosed in the mammography group and 524 were diagnosed in the control group. During the 25-year follow-up period, 180 women in the mammography group and 171 in the control group died of breast cancer (hazard ratio [HR], 1.05).

The findings for women 40 to 49 years and 50 to 59 years were nearly identical.

For the entire study period, the cumulative death rate from breast cancer was similar in the mammography and control groups (500 vs 505; HR, 0.99).

The 25-year survival for women with palpable tumors was similar in the mammography and control groups (66.3% vs 62.8%), but was higher for women diagnosed with a nonpalpable tumor on mammography (79.6%).

Overdiagnosis With Mammography

At the end of the 5-year screening period, there was an excess of 142 breast cancers in the mammography group. At 15 years, the excess became constant at 106 cancers.

This indicates that 22% of the screen-detected invasive cancers in the mammography group were overdiagnosed, the researchers note. Expressed another way, there was 1 overdiagnosed breast cancer for every 424 women who received mammography screening in the trial.

But Dr. Dang noted that "most would argue that detection of cancer at all is not considered a 'false positive' because most cancers are excised and there is no way to know whether or not, at some point, the progression of cancer that has not been excised will affect a person's mortality."

Dr. Cappuccino agrees. "Currently, we are doing our best to detect and treat any cancer within breast tissue," she said. "At this point, most of my patients would prefer to err on the side of early diagnosis and intervention, rather than wait to see the implications of a breast abnormality."

With time and additional study, a series of known risk parameters (such as family history, genetic predisposition, other medical and pharmaceutical considerations, and gynecologic and obstetric history) might be useful in assessing which patients are most likely to derive a benefit from screening tests like mammography, she noted.

"Like much of evolving cancer care, in the future it is more likely to be a personalized recommendation. For now, there is enough of a benefit that I am most comfortable continuing to screen according to the current national recommendations," Dr. Cappuccino concluded.

This study was supported by the Canadian Breast Cancer Research Alliance, the Canadian Breast Cancer Research Initiative, the Canadian Cancer Society, Health and Welfare Canada, the National Cancer Institute of Canada, the Alberta Heritage Fund for Cancer Research, the Manitoba Health Services Commission, the Medical Research Council of Canada, le Ministère de la Santé et des Services Soçiaux du Québec, the Nova Scotia Department of Health, and the Ontario Ministry of Health. The authors and editorialists have disclosed no relevant financial relationships.

BMJ. Published online February 11, 2014. Abstract, Editorial


Comments on Medscape are moderated and should be professional in tone and on topic. You must declare any conflicts of interest related to your comments and responses. Please see our Commenting Guide for further information. We reserve the right to remove posts at our sole discretion.