Breast Tumors Grow More Slowly with Age

By Crystal Phend
OSLO, Norway,09 may 2008 -- In women between 50 and 70, breast tumor growth is faster in those at the younger end of the spectrum, which may have implications for sensitivity of breast cancer screening programs, researchers said.The estimated time for a tumor to double in diameter from 10 to 20 mm was 1.4 years in women ages 50 to 59 compared with 2.1 years for women 60 to 69, reported Harald Weedon-Fekjær, Ph.D., of the Cancer Registry of Norway and University of Oslo, and colleagues, online in Breast Cancer Research.The sensitivity of mammographic screening would rise with tumor size from 26% at 5 mm to 91% at 10 mm, according to a mathematical model developed in the large registry study.
Tumor growth rates are important for determining intervals between mammographic screening and ages at which mammography is recommended for women, the researchers said.
Previous studies have found decreasing tumor progression with older age, but this measure of growth rates was indirect, typically used a model without "explicit relation to the biological process of tumor growth," and was often difficult to compare between different countries, they said.
To develop a model directly connected to tumor measurements, the researchers examined data from the Cancer Registry of Norway. The analysis included first mammograms for 395,188 women ages 50 to 69 and clinical data from tumors found through screening.
The mean time for tumors to grow from 10 to 20 mm was estimated at 1.7 years, but growth varied widely, with a standard deviation of 2.2 years.
Time for a tumor to double in diameter from 15 to 30 mm ranged from an average 41 to 234 days for the first and last quartiles, respectively.
The mammography screening test sensitivity estimates increased sharply with tumor size, reaching 26% at 5 mm and 91% at 10 mm.
However, estimated sensitivity did not differ by age (P=0.83 at 5 mm), which Dr. Weedon-Fekjær and colleagues said was surprising.
According to simulations using the model and data for another group of women with clinically detected breast cancer, almost all cancers among women in the screening population would have been visible at mammography before reaching clinical detection.
The mean time tumors would be visible at screening before clinical detection was 2.9 years overall and significantly longer in older women.
The model developed in the study appeared to be a better fit to the registry data than the classical Markov model (P<0.0001) and increased the predictive power for observed versus predicted number of cases by 85%.
Estimates in the new model did not appear to be biased by missing tumor measurements in the registry, the researchers said.
However, the model appeared to be less effective at predicting cancer incidence on subsequent rounds of screening.
It predicted a 71% drop in detected cancers from the first to second screening whereas as decline of 46% was observed.
"In addition to possible problems with the model itself," the investigators said, "this can be an effect of changes in hormone replacement therapy use in the study period, of increased sensitivity in the second round due to use of earlier mammograms, of better training of staff with time, or of an overrepresentation of communities with high cancer risk in the second screening round."
The relatively high rate of hormone therapy use, which increases breast density, during the study period may have affected tumor growth as well as mammographic screening sensitivity, they noted.
The project received financial support from the Norwegian Cancer Society. The researchers reported no conflicts of interest.
Additional source: Breast Cancer ResearchSource reference: Weedon-Fekjær H, et al "Breast cancer tumor growth estimated through mammography screening data" Breast Cancer Res 2008; DOI: 10.1186/bcr2092.