Mammography outcomes in a practice setting by age: prognostic factors, sensitivity, and positive biopsy rate

The separate unplanned analysis of women ages 40-49 in population-based randomized controlled trials has resulted in demonstration of statistically significant breast cancer mortality reduction due to screening mammography in only two of the individual trials, and in all such trials only through meta-analysis. Therefore, many researchers have utilized the surrogate endpoints of tumor size and axillary lymph node status to evaluate screening efficacy. For the present study, these endpoints were evaluated in an audit of 854 screen-detected cancers found in 147,125 mammographic examinations performed in women over 40 between 1988 and 1994 in a community practice setting. The concerns that mammography in the 40-49 group has a lower sensitivity and higher biopsy rate were also addressed. Median invasive tumor size and lymph node positivity were found to be equally small (1.0-1.1 cm and 13.5-12.2%, respectively), and the sensitivity and overall biopsy rate were found to be constant over all ages 40 and above. Positive biopsy rate (PBR) varied directly with increasing age, paralleling the measured cancer detection rate in each decade, with no abrupt change at age 50. We conclude that modern mammography in a community practice setting can successfully detect breast cancers with favorable prognostic factors and achieve constant sensitivity and acceptable PBRs in all women over 40. Our data also suggest that many of the large differences seen by inappropriately dividing data at age 50 decrease or disappear when analysis is performed by decade.