U.S. Radiologists Dispute BMJ Mammography Study

Calling the results flawed, many in the radiology community are protesting a study released last week that suggested mammography hasn’t played a major role in the drop in breast cancer-related deaths.

The research, published in the July 28 British Medical Journal, compared the reduction in breast cancer deaths from 1989 to 2006 in several Northern European countries and concluded that improved disease management – not mammography – could most likely be credited with the decrease in deaths.

“Our study adds…to the evidence of studies that have used various designs and found that mammography screening by itself has little detectable impact on mortality due to breast cancer,” wrote the study authors, led by research director Phillippe Autier from the International Prevention Research Institute in Lyon, France.

Industry leaders in the United States, however, disagreed. According to them, these results are flawed and contradict proven research on the efficacy of using screenings to catch early-stage breast cancer.

“Improvements in therapy have, likely, played a role in the decrease of breast cancer deaths, but therapy cannot cure advanced cancer,” the American College of Radiology (ACR) and the Society of Breast Imaging (SBI) said in a joint statement. “Early detection via mammography is clearly the major reason for the decrease in deaths in the U.S.”

The ACR and SBI pointed to several errors in the study. Not only did the data analysis include deaths that occurred before the reviewed countries even introduced screenings, but it also didn’t account for the fact that not all countries began screenings in the same year. In addition, researchers did not adjust for the probability that not all women were invited to screenings, and among those who were, not all attended.

Memorial Sloan-Kettering Cancer Center diagnostic radiologist Carol Lee, MD, agreed the findings fly in the face of conventional practice.

“We know from direct observation in the United States that there has been a substantial decrease in breast cancer mortality since 1990,” said Lee, who is also chair of ACR’s Breast Imaging Communications Committee. “To suggest that improvements in treatment, and not early detection, are responsible defies common sense.”

Treatment advancements, she said, largely affect early-stage disease which can only be addressed when identified through mammography.

Dr. Stamatia Destounis, managing partner and a radiologist at Elizabeth Wende Breast Care in Rochester, N.Y., also questioned the data presented in this article, noting the countries compared had different screening programs instituted at different time intervals and varying incident and mortality rates.

“Their information is a bit contaminated, by lumping together patients that were added to the report having been found with cancer that were diagnosed prior to the screening interval and patients that were not invited to screening but were tested and found to have cancer,” she said.

Destounis said the U.S. can’t be compared to Europe. After implementing regular screening programs, the incidence of breast cancer in the U.S. increased as a result, and the breast cancer mortality decreased. Although treatments have improved and this may reduce mortality rates, screening is the most critical way to find tumors early, offering the greatest chances of survival, she said. “That’s the whole crux of what we do here… We want to offer early detection.”

Despite the controversial findings, Lee said she has no worries that U.S. physicians will change their perceptions about or their recommendations to undergo a mammography screening.
“I don’t believe this study will be one that will carry over into clinical practice in the United States,” Lee said. “Practitioners have seen the evidence and know the importance of screening mammography, and as a general rule, women will get a screening when their providers recommend it.”

According to Rachel Brem, MD, director of the Breast Imaging and Interventional Center and vice chair of the radiology department at George Washington University Medical Center, the design and underlying structure of the study is so poor, it should never have been published. 

"It's irresponsible to draw conclusions from data that should never be compared," she said, pointing out that in addition to previously mentioned errors, the researchers didn't account for differing cancer rates between countries. "Fortunately, I don't see any way this could have an impact on clinical practice."

The problem, she said, arises from how patients and government leaders interpret the study. In a tight economic climate, women could use this study as an excuse to forgo the expense associated with mammography screening. In addition, officials responsible for health care reform measures could cite the findings to support slicing funding for mammography services, making it harder for women to be screened even if they want to be.

"I fear that American women could potentially be significantly harmed by the results of this study, and we could step back from the advancements we've made in treating breast cancer," she said. "We could see an increase in breast cancer-related deaths."

Sara Michael contributed to this report.