In the 1970s and early 1980s, population-based screening mammography was studied in numerous randomized control trials (RCTs), with the primary outcome of reduced breast cancer mortality. Although technology and the sensitivity of mammography in the 1980s was somewhat rudimentary compared with current screening, a meta-analysis of these RCTs demonstrated a clear mortality benefit for screening mammography.1 As a result, widespread population-based mammography was introduced in the mid-1980s in the United States and has become a standard for breast cancer screening.
Since that time, few RCTs of screening mammography versus observation have been conducted because of the ethical challenges of entering women into such studies as well as the difficulty and expense of long-term follow-up to measure the effect of screening on breast cancer mortality. Without ongoing RCTs of mammography, retrospective, observational, and computer simulation trials of the efficacy and harms of screening mammography have been conducted using proxy measures of mortality (such as stage at diagnosis), and some have questioned the overall benefit of screening mammography.2,3
To further complicate this controversy, some national guidelines have recommended against routinely recommending screening mammography for women aged 40 to 49 based on concerns that the harms (callbacks, benign breast biopsies, overdiagnosis) exceed the potential benefits (earlier diagnosis, possible decrease in needed treatments, reduced breast cancer mortality).4 This has resulted in a confusing morass of national recommendations with uncertainty regarding the question of whether to routinely offer screening mammography for women in their 40s at average risk for breast cancer.4-6
Recently, to address this question Duffy and colleagues conducted a large RCT of women in their 40s to evaluate the long-term effect of mammography on breast cancer mortality.7 Here, I review the study in depth and offer some guidance to clinicians and women struggling with screening decisions.
Breast cancer mortality significantly lower in the screening group
The RCT, known as the UK Age trial, was conducted in England, Wales, and Scotland and enrolled 160,921 women from 1990 through 1997.7 Women were randomly assigned in a 2:1 ratio to observation or annual screening mammogram beginning at age 39–41 until age 48. (In the United Kingdom, all women are screened starting at age 50.) Study enrollees were followed for a median of 22.8 years, and the primary outcome was breast cancer mortality.
The study results showed a 25% relative risk (RR) reduction in breast cancer mortality at 10 years of follow-up in the mammography group compared with the unscreened women (83 breast cancer deaths in the mammography group vs 219 in the observation group [RR, 0.75; 95% confidence interval (CI), 0.58–0.97; P = .029]). Based on the prevalence of breast cancer in women in their 40s, this 25% relative risk reduction translates into approximately 1 less death per 1,000 women who undergo routine screening in their 40s.
While there was no additional significant mortality reduction beyond 10 years of follow-up, as noted mammography is offered routinely starting at age 50 to all women in the United Kingdom. The authors concluded that “reducing the lower age limit for screening from 50 to 40 years [of age] could potentially reduce breast cancer mortality.”
Was overdiagnosis a concern? Another finding in this trial was related to overdiagnosis of breast cancer in the screened group. Overdiagnosis refers to mammographic-only diagnosis (that is, no clinical findings) of nonaggressive breast cancer, which would remain indolent and not harm the patient. The study results demonstrated essentially no overdiagnosis in women screened at age 40 compared with the unscreened group.
Continue to: Large trial, long follow-up are key strengths...