Background: Over the past 30 years, eight major randomized controlled trials of breast cancer screening--with mammography and/or clinical breast examination--have been conducted. Results from several trials have been updated during the past year, and initial results of three other trials have been reported.
Purpose: The National Cancer Institute held an International Workshop on Screening for Breast Cancer in February 1993 to conduct a thorough and objective critical review of the world's most recent clinical trial data on breast cancer screening, consider the new evidence, assess the current state of knowledge, and identify issues needing further research.
Methods: Investigators representing the eight randomized controlled trials of breast cancer screening in women aged 40-74 presented published and unpublished data. Evidence relating to the effectiveness of breast cancer screening in different age groups, especially women aged 40-49, was presented.
Results: For women aged 40-49, randomized controlled trials consistently demonstrated no benefit from screening in the first 5-7 years after study entry. A meta-analysis of six trials found a relative risk of 1.08 (95% confidence interval = 0.85-1.39) after 7 years' follow-up. After 10-12 years of follow-up, none of four trials have found a statistically significant benefit in mortality; a combined analysis of Swedish studies showed a statistically insignificant 13% decrease in mortality at 12 years. Only one trial (Health Insurance Plan) has data beyond 12 years of follow-up, and results show a 25% decrease in mortality at 10-18 years. Statistical significance of this result is disputed, however. In women aged 50-69, all studies show mortality reductions; three of four studies show reductions of about 30% at 10-12 years after study entry. Results from two of these trials were statistically significant. Too few women over age 70 have been included in studies for adequate analysis.
Conclusions: For women aged 40-49, randomized controlled trials of breast cancer screening show no benefit 5-7 years after entry. At 10-12 years, benefit is uncertain and, if present, marginal; thereafter, it is unknown. For women aged 50-69, screening reduces breast cancer mortality by about a third. Currently available data for women age 70 or older are inadequate to judge the effectiveness of screening.
Implications: Randomized trials have provided stronger scientific evidence regarding the effectiveness of screening for breast cancer than for any other cancer. However, much still needs to be learned. Periodic gatherings of scientists in the field should speed the process.