Abstract
This breast cancer adjuvant systemic therapy dissemination simulation generates individual-level breast cancer adjuvant therapy usage that is representative of the US population over the years 1975-2000. The model assigns women diagnosed with breast cancer between 1975 to 2000 to one of each treatment modality: chemotherapy, 2 years of tamoxifen, 5 years of tamoxifen, chemotherapy and 2 years of tamoxifen, chemotherapy and 5 years of tamoxifen or, no adjuvant therapy. These individual-level assignments of adjuvant therapy represent the US population over target time-period when taken on aggregate. The generated assignments were subsequently used in larger modeling efforts by CISNET consortium members to model the impact of mammography and adjuvant therapy on incidence and mortality over the same time period. More information on this effort is available on the CISNET website.
Clinical trials have shown adjuvant tamoxifen and multi-agent chemotherapy to be effective in the treatment of breast cancer for some group of women. Trends in the utilization of adjuvant therapy were modeled (Mariotto JNCI, 2002) as a function of age, stage of the disease, estrogen receptor (ER) status and year of diagnosis using population data on treatment use from cancer registry (Surveillance, and Epidemiology and End Results (SEER) program).
Summary
This section includes a brief description of the purpose and methods behind this simulator. An example is also included.
Description of simulation
This simulation generates individual-level breast cancer adjuvant therapy usage that is representative of the US population over the years 1975-2000. Given a year of birth, stage of the tumor, ER status and year of diagnosis the program will assign each women in the cohort to one of six modalities of adjuvant systemic therapy: chemotherapy, 2 years of tamoxifen, 5 years of tamoxifen, chemotherapy and 2 years of tamoxifen, chemotherapy and 5 years of tamoxifen and no adjuvant therapy.
The details breast cancer adjuvant systemic therapy dissemination mode can be found in Mariotto et al. (JNCI, 2002). In brief, data on treatment of patients diagnosed with breast cancer from 1975 through 1999 were obtained from eight registries of Surveillance, and Epidemiology and End Results (SEER) program. Because the SEER information on the use of adjuvant systemic therapy is incomplete, supplementary information from a series of population-based patterns of care (POC) studies that randomly selected cases from the SEER registries was also included to verify and adjust the SEER information on treatment. These population-based data sources are combined and use to estimate the proportion of women by age, stage at diagnosis, ER status and year of diagnosis using one of the 6 modalities of adjuvant therapy as described above. These proportions are then used to assign each women to one treatment.
Example
To aid in further understanding of this model, the following hypothetical example is provided. If the model is provided with the following parameters:
| Classification | AJCC stage/node status with ER Status |
|---|---|
| Age DX | Ages 50 to <70 |
| Stage | Regional |
| ER Status | ER positive |
| Year DX | 1990 |
| Treatment | Cumulative Range | Probability |
|---|---|---|
| Chemotherapy | 0 to <= 0.034196 | 0.03420 |
| Tamoxifen* | > 0.034196 to <= 0.617043 | 0.58285 |
| Both Chemotherapy and Tamoxifen (2-year) | > 0.617043 to <= 0.72640 | 0.10936 |
| No Treatment | > 0.726405 | 0.27360 |
Graphically, these treatment distributions may be displayed as follows:
Next, for each simulated woman, two independent uniformly distributed pseudo random numbers are generated. The first number determines which treatment is received and the second determines the duration of tamoxifen treatment, should it apply the woman. For example, if a simulated woman receives the following pseudo random numbers:
| Treatment Random Number: | 0.32 |
|---|---|
| Tamoxifen Treatment Random Number: | 0.16 |
She would be assigned Tamoxifen for 5 years, as depicted grapically below.
Contact Information
If you have questions on the modeling used in the simulation or if you would like access to the simulation source code, please send an email to mariotta@mail.nih.gov.
References
- Trends in use of Adjuvant Multi-Agent Chemotherapy and Tamoxifen for Breast Cancer in the United States: 1975-1999. Mariotto A, Feuer EJ, Harlan LC, Wun L, Johnson K, Abrams J. Journal of the National Cancer Institute 2002;94:1626-34.
- The Dissemination of Adjuvant Multi-Agent Chemotherapy and Tamoxifen for Breast Cancer in the US using Estrogen receptor information: 1975-1999.Angela Mariotto, Eric J Feuer, Linda C Harlan, Jeffrey Abrams. Submitted to JNCI
Interface
Important: These models not designed for use by or for cancer patients or clinicians seeking treatment guidance for individual cases. Rather, they are largely based on retrospective surveillance data and have been built to aid our understanding of the impact of cancer control interventions (e.g., prevention, screening treatment) on population trends in incidence and mortality. Furthermore, the models here represent a small part of the larger CISNET effort and are used as population level inputs to cancer natural history and prognosis models. Please see the CISNET home page for more details.