Hard numbers put to screening mammography overdiagnosis

Over the years since the advent of widespread screening mammography, overdiagnosis and overtreatment have become accepted as unfortunate but tolerable collateral troubles. There’s no shortage of literature using mathematical modeling to suggest as much. Now comes a new analysis of hard data, published Oct. 13 in the New England Journal of Medicine, to quantify the problem.

Led by H. Gilbert Welch, MD, MPH, of Dartmouth and Barnett S. Kramer, MD, MPH, of the National Cancer Institute, researchers used SEER statistics spanning from 1975 through 2012 to calculate the tumor-size distribution and size-specific incidence of breast cancer among women 40 years of age or older.

The team then calculated the 10-year risk of death from breast cancer in two time periods: 1975 through 1979 (their baseline “pre-screening” period) and 2000 through 2002 (the most recent screening-era period in which 10 years of follow-up data were available).

They found that, after the advent of screening mammography, the proportion of detected breast tumors that were small (less than 2 centimeters or in situ carcinomas) increased from 36 percent to 68 percent. At the same time, the proportion of detected tumors that were large (2 centimeters or more) decreased from 64 percent to 32 percent.

However, this trend was less the result of a substantial decrease in the incidence of large tumors—30 fewer cases of cancer were observed per 100,000 women in the period after the advent of screening than in the period before screening—and more the result of a substantial increase in the detection of small tumors (with 162 more cases of cancer observed per 100,000 women), Welch and colleagues report.

Further, assuming that the underlying disease burden was stable, only 30 of the 162 additional small tumors per 100,000 women were expected to progress and grow.

This suggests that the remaining 132 cases of cancer per 100,000 women were overdiagnosed, the authors conclude.

“The potential of screening to lower breast cancer mortality is reflected in the declining incidence of larger tumors,” they note. “However, with respect to only these large tumors, the decline in the size-specific case fatality rate suggests that improved treatment was responsible for at least two thirds of the reduction in breast cancer mortality.”

In their discussion, the authors acknowledge that no perfect method exists to assess the population effects of cancer screening.

“We do not pretend to present a precise estimate of either the amount of overdiagnosis or the contribution of screening mammography to the reduction in breast-cancer mortality,” they write. “The data regarding size-specific incidence, however, make clear that the magnitude of overdiagnosis is larger than is generally recognized.”

In an accompanying NEJM opinion piece, Joann Elmore, MD, MPH, of the University of Washington notes that Welch et al. present “powerful data on a large number of women in a very clear fashion.” But it’s a problem, she states, that they “rely on data with extensive missing values, make assumptions about underlying disease burden that cannot be verified and acknowledge that their estimates are imprecise.”

“Rather than focusing on statistical issues and study design, we should move forward by agreeing that overdiagnosis does occur, even though the exact percentage of overdiagnosed cases remains unknown.”

Elmore says one way to reduce overdiagnosis is to target for screening only the individuals who are at elevated risk for breast cancer.

Welch has posted an 11-minute video on YouTube to further explain the new findings.