Purpose: As part of the NCI Cancer Intervention and Surveillance Network (CISNET) consortium, we developed a simulation model that predicts U.S. breast cancer mortality trends from 1975 to 2000 and quantifies the impact of screening mammography and adjuvant therapy on the overall age-adjusted trends. The primary objective of this work was to evaluate the impact of screening and adjuvant therapy on the age-specific trends in breast cancer mortality.

Methods: We modified our existing Monte-Carlo simulation model wherein individual women undergo varying screening schedules and treatment regimes consistent with national trends and long-term outcomes are predicted. We extended the model to incorporate age-specific mammographic detection thresholds and tumor doubling times. We estimated the tumor doubling times by calibrating to the Surveillance, Epidemiology, and End Results (SEER) incidence trends, assuming a linear relationship between tumor doubling time and age of diagnosis in the absence of screening. We estimated the age-specific mammographic detection thresholds by calibrating to both SEER incidence trends and data on screened tumor sizes provided by the Breast Cancer Surveillance Consortium (BCSC), assuming a quadratic relationship between mammographic detection threshold and age of diagnosis in the absence of screening.

Results: We computed the contribution from screening mammography and adjuvant therapy on breast cancer mortality in the year 2000 for different age groups. The contribution of screening relative to adjuvant therapy increased with age. If women ages 40-49 years were not screened, there would have been a 5.5% and 4.9% increase in mortality among women ages 40-49 years and ages 50-59 years at death, respectively. If women ages 50-59 years were not screened, there would have been a 10.2% and 8.1% increase in mortality among women ages 50-59 years and ages 60-69 years at death, respectively. If women ages 60-69 years were not screened, there would have been a 9.4% and 9.1% increase in mortality among women ages 60-69 years and ages 70-79 years at death, respectively.

Conclusions: Screening women above age 50 provided the largest mortality benefit. This result was expected since younger women have a lower rate of screening, lowered sensitivity to mammography and faster tumor volume doubling times. An analysis based on life years saved may provide a different result for the magnitude of the benefit attributable to screening varying age groups.