Breast Cancer Not Overdiagnosed by Mammography

The idea that breast cancer is overdiagnosed is being perpetuated in an effort to reduce access to breast cancer screening, according to Daniel B. Kopans, MD, professor in the department of radiology at Harvard Medical School.

“Overdiagnosis and overtreatment are used to create fear and confusion,” Kopans said during his presentation “Overdiagnosis and Overtreatment of Breast Cancer: What is the Reality?” at the 32nd Annual Miami Breast Cancer Conference held February 26–28 in Miami Beach, Florida. “However, mammography is not responsible for either one. Pathologists make breast cancer diagnoses and oncologists determine treatment.”

Instead, any overdiagnosis or overtreatment that exists is just part of modern medical practice, he said. To draw a comparison, every patient diagnosed with bacterial pneumonia is treated with antibiotics even though it is possible that they may recover without them and despite the fact that it is recognized that antibiotics could lead to a superinfection or adverse reactions.

Early treatment of breast cancer consisted solely of mastectomy. Studies from the early 20th century show that there was a 30% 30-year survival among women with breast cancer treated with mastectomy alone, Kopans said. However, if these women were treated today, many of them would have been given systemic therapy, even though the data from the earlier study show that in many of these women systemic therapy would have been “overtreatment” because they were not going to die from their cancers.

“Overdiagnosis is a myth that has been created by a handful of individuals who provide no care for women with breast cancer,” Kopans told Cancer Network. “Many physicians and women have been misled by analyses by individuals who do not provide care for women with breast cancer who, using scientifically, flawed analyses advise waiting until the age of 50 and then screening every 2 years.”

During his presentation, Kopans discussed two major published studies that are contributing to this myth.

The first study was published in the New England Journal of Medicine by Bleyer and Welch in 2012, and used data from SEER to conclude that breast cancer was overdiagnosed in 1.3 million US women in the past 30 years. However, according to Kopans, this paper has no scientific validity.

“One of the authors admitted that their analysis was based on ‘guesses,’” Kopans told Cancer Network. “They actually had no data on who was having mammograms or which cancers were found by mammography.”

In addition, the study combined ductal carcinoma in situ with small invasive breast cancers into a category called early breast cancer, which Kopans said is very misleading.

“It turns out that their guesses were incorrect,” Kopans said. “Using actual data, not guesses, there is little if any overdiagnosis of invasive breast cancers, but the media have not brought this to the attention of the public and their physicians.”

Kopans also pointed to the results of another study responsible for some of the misinformation out there: the Canadian National Breast Screening Study (CNBSS). The CNBSS concluded that annual mammography in women aged 40 to 59 years did not reduce mortality from breast cancer compared with a physical examination or usual care. Although this study was largely discredited in the 1990s, Kopans said that the 25-year results have recently gained attention again and its compromised results are being used to deny women access to screening.

“To do a successful randomization, it has to be blinded,” Kopans said. “There should be two identical groups with the same number of women dying of breast cancer in both groups if nothing at all was done, no intervention.”

Unfortunately, this was not the case with the CNBSS, Kopans said. Instead, all women in the study had clinical breast exams before allocation so it was known who had palpable lesions and who had large axillary lymph nodes. Women were allocated using open lists, meaning that study coordinators knew which patients had problems. Data show that more women who had advanced breast cancer were placed into the screening arm, and subsequently, there were more deaths in the screening group in the first 10 years with a greater than 90% 5-year survival in the control women.

In addition, Kopans said that the mammography used in the CNBSS was poor. Examination of the study showed that screening was done with old devices, without grids, and with little training for imaging technicians or radiologists.

“Results of CNBSS have been ignored in Canada with provinces continuing to support screening,” Kopans said. “Yet it is being promoted, again by those who do not take care of women with breast cancer, and, unfortunately, may well guide the next, inexpert, USPSTF panel that is meeting to review their guidelines.”