摘要：

Introduction: US breast cancer mortality rates have decreased from a 1990 level of 33.1 deaths per 100,000 to a 2001 level of 26.0 deaths per 100,000. The Healthy People 2010 (HP2010) goal for age-adjusted breast cancer mortality is 22.3 deaths per 100,000. The purpose of this study is to determine whether or not this HP2010 goal is attainable based on current dissemination patterns of screening mammography and adjuvant treatment. Method: A Monte Carlo simulation model was developed that reproduces the life history of women who were born in the US since 1890 and outputs population level incidence and mortality statistics from 1975 onward. The model relies on data from numerous national registries, such as the Surveillance, Epidemiology and End Results (SEER), Patterns of Care (POC) and Breast Cancer Surveillance Consortium (BCSC). The simulation model combines these databases by synthesizing well-accepted biological, epidemiological, clinical characteristics of breast cancer. The simulation model tracks the life history of individual women and records the following information for each breast cancer patient with invasive disease: her date of birth, her screening schedule, when and how she was detected, her tumor size and SEER historic stage at detection, ER status, what treatment she received, her survival time from diagnosis and her cause of death. While part of SEER was used to inform several components of our model, another part of SEER was left aside for model validation purposes. To predict future age-adjusted breast cancer mortality rates, trends in the secular breast cancer incidence, and screening and treatment remained at their current levels. Expert opinion was solicited for estimates of the dissemination patterns and the long-term efficacy of taxanes and aromatase inhibitors. All model assumptions were varied in a sensitivity analysis. Results: Our simulation model closely reproduces breast cancer population level trends observed in the SEER database that were not used in the model building process. It closely reproduces the tumor size and stage distribution as a function of calendar year by age group. We also found good agreement in terms of overall age-adjusted incidence and mortality trends. The model predicts that the HP2010 goals for breast cancer mortality will not be met. Instead breast cancer mortality is predicted to level off starting 2005. Even under optimal conditions (namely 100% annual screening for women ages 40-79 starting 2005 and optimistic estimates for treatment efficacy), it is not likely that the HP2010 breast cancer goal would be attained. These results were even more pronounced for the African American subpopulation, which has a lower overall incidence of breast cancer, higher rate of ER- disease and higher rate of breast cancer mortality. Conclusion: Breast cancer mortality rates are not likely to meet the HP2010 goal. We are currently working to identify what efforts would be needed to meet the HP2010 breast cancer goals.

展开