Geometry, growth rates, and duration of cancer and carcinoma in situ of the breast before detection by screening.

The purpose of this study is to report the time elapsing between the moment a breast cancer reaches threshold size that would permit detection and the size at actual detection defined as sojourn time (STt) using data from the Breast Cancer Detection and Demonstration Project (Louisville data and reported composite data from 27 centers) by dividing prevalence rates by incidence rates. The number of cellular generations (n) required to produce cancers of different volumes was calculated at threshold (nt), at detection (nd), and the difference between the two (nd - nt). By dividing the difference (nd - nt) into STt, the average actual or net tumor volume doubling time (DTact) in this interval have been estimated. The STt value for carcinoma in situ was 557 and for cancer it was 538. At ages 35-39, STt ranged from 365-456 days, by ages 70-74, 942-1383 days. The average DTact similarly varied with age, carcinoma in situ, cancer with negative axillary nodes, and cancer with positive axillary nodes over a range of 28-732 days (95% confidence). The estimated DTact in the predetectable period was manyfold less than the DTact measured for mammographically visible cancers, in keeping with the predictions of decelerating growth. Models of the cytokinetic and clinical behavior of breast cancer are discussed.