Radiology: Breast cancer overdiagnosis exaggeratedor not?

Twitter icon
Facebook icon
LinkedIn icon
e-mail icon
Google icon
Screening mammography with breast tomosynthesis in a 57-year-old woman.
Image source: RadioGraphics

A pair of editorials in the September issue of Radiology tackled the tough topic of breast cancer overdiagnosis by screening mammography. On one side, breast imagers argued that claims of overdiagnosis are inflated. On the other, a group of researchers suggested that screening mammography has a minor effect on mortality and breast imaging may cause more harm than benefit.

Although the editorials are unlikely to settle the decades-long debate, they do raise critical points. Overdiagnosis occurs when screening detects a tumor that would not have been diagnosed or progressed to potentially lethal disease in the absence of screening.

In “Mammographic Screening and ‘Overdiagnosis,’” Daniel B. Kopans, MD, director of breast imaging at Massachusetts General Hospital, Harvard Medical School, in Boston, acknowledged that overdiagnosis occurs.

However, breast cancer treatment has the greatest chance for success during the clinically occult phase of the disease. Kopans estimated that the likely extent of overdiagnosis is 10 percent or less, based on the Malmo trial, which indicated that mammography detected 10 percent more cancers than became clinically evident in 15 years of follow-up.

Co-editorialist Stephen W. Duffy, MSc, from the Centre for Cancer Prevention, Wolfson Institute of Preventive Medicine at Queen Mary University of London, examined the Swedish Two-County Trial data and calculated “that 2 to 2.5 lives were saved for every case that was estimated to have been ‘overdiagnosed.’”

Kopans and colleagues referred to increasing diagnosis of DCIS, noting that it accounts for 20 to 30 percent of all breast cancers, up from 2 to 5 percent prior to the inception of screening mammography. “There is still debate over the importance of the lesions … the fact remains there is no direct way to determine which cases of DCIS, if left alone, would not progress to become invasive,” he wrote.

Kopans also refuted the theory that overtreated screening-detected breast cancers would spontaneously regress, and wrote, “We have been unable to find any convincing evidence that a significant fraction of breast cancer regress spontaneously.”

Women should be provided information to make informed decisions regarding screening, concluded Kopans et al. This includes the possibility of overtreatment because nonprogressive and progressive cancers cannot yet be differentiated. Women also should be provided with information about the proven utility of screening in reducing the breast cancer mortality rate, according to Kopans and colleagues.

In “Is Mammographic Screening Justifiable Considering Its Substantial Overdiagnosis Rate and Minor Effect on Mortality?” Karsten Juhl Jorgensen, MD, of Nordic Cochrane Centre at University of Copenhagen, and colleagues wrote, “Overdiagnosis is substantial and certain and screening increases the number of mastectomies performed.”

Jorgensen credited advances in treatment with reducing breast cancer mortality and offered that treatment advances have curtailed the benefits of screening demonstrated in earlier trials such as the Swedish Two-County Trial.

He referred to a fall in breast cancer mortality in Europe between 1989 and 2005, which occurred prior to organized screening. In addition, “Despite the fact that Sweden has the longest running program, the widest age range and the shortest screening interval in Europe, the reductions in breast cancer mortality are lower than the European median.”

According to the Cochrane review, the Swedish Two-County Trial yielded 30 percent more cancers in the screening cohort, translating into 10 cases of overdiagnosis for every life prolonged. Jorgensen estimated that the overdiagnosis rate is 52 percent, based on analysis of countries with long-standing public screening programs.

“It is clear that screening has greatly increased the lifetime risk of becoming a breast cancer patient,” wrote Jorgensen and colleagues. They noted that autopsy studies suggest less than half of DCIS cases progress.

The editorialists cited a fairly staggering burden of overdiagnosis and minimal utility of screening. “A radiologist who reads 1,000 mammograms a year from women in their 40s would take 10 years to prolong one life yet burden one woman every year with overdiagnosis and overtreatment [based on an overdiagnosis rate of 30 percent and a 15 percent reduction in breast