Enhancing the OncoSim-Breast model using Canadian breast density information

Area of Science:

• Oncology
• Biostatistics
• Public Health

Background:

• Breast cancer is a leading diagnosis for Canadian women.
• Breast density significantly impacts cancer risk and mammogram effectiveness.
• The OncoSim-Breast model previously lacked explicit breast density parameters.

Purpose of the Study:

• To integrate breast density-specific data into the OncoSim-Breast microsimulation model.
• To enhance the model's accuracy in representing breast cancer dynamics.
• To support evidence-based breast cancer screening policies in Canada.

Main Methods:

• Integrated breast density parameters (prevalence, relative risk, mammography sensitivity/specificity) into OncoSim-Breast.
• Utilized data from five Canadian provinces, categorized by Breast Imaging Reporting and Data System (BI-RADS) A-D.
• Calibrated and validated the updated model against Canadian Cancer Registry data (2010-2019).

Main Results:

• Dense breasts (BI-RADS C/D) were more prevalent in younger women (58% <50 yrs) and decreased with age (26% ≥70 yrs).
• Mammography sensitivity decreased with increasing breast density (e.g., 88% for Category A vs. 69% for Category D in women <50 yrs).
• The updated model accurately reproduced historical age-specific incidence and stage distribution.

Conclusions:

• Incorporating breast density significantly improved OncoSim-Breast's accuracy and policy relevance.
• The enhanced model serves as a validated tool for informing breast cancer screening policies.
• The model now accounts for breast density variations among women, crucial for policy development.