# Risk models for breast cancer and their validation

**Authors:** Adam R Brentnall, Jack Cuzick  
**arXiv:** [1907.02829](https://arxiv.org/abs/1907.02829)  
**Date:** 2019-07-05  
**Summary page:** https://www.summarizepaper.com/en/arxiv-id/1907.02829v2/

## Key points

- The article focuses on the development and evaluation of the IBIS (Tyrer-Cuzick) breast cancer risk model
- The model combines various risk factors into a hybrid genetic segregation and regression model
- It assesses an individual's risk of developing breast cancer at different follow-up times
- The IBIS model is recommended by several organizations, including the American Cancer Society, to guide preventive and early detection strategies
- It has been used to guide entry criteria into prevention trials and determine eligibility for supplemental breast magnetic resonance imaging
- New graphical methods for assessing calibration over follow-up time are introduced
- Different methods for assessing calibration of cumulative absolute risk are reviewed
- Providing a detailed description of the method for expected risk is emphasized when presenting calibration findings
- Opportunities for refinement and improvement of the current risk model exist
- Other factors that affect breast cancer risk have difficulties being included in a relatively simple model
- The research applies the breast cancer risk model and calibration assessment methods to data from a large cohort of women attending mammographic screening in Washington, USA
- Overall, this research contributes to ongoing efforts to prevent cancer and diagnose it early when it is most treatable.

## Summary

This article focuses on the development and evaluation of a breast cancer risk model known as the IBIS (Tyrer-Cuzick) model. The model combines various risk factors into a hybrid genetic segregation and regression model to assess an individual's risk of developing breast cancer at different follow-up times. The authors review the statistical foundations of the model and describe recent updates. The IBIS model has been used to guide entry criteria into prevention trials and is recommended by several organizations, including the American Cancer Society, to guide preventive and early detection strategies. It has also been used to determine eligibility for supplemental breast magnetic resonance imaging. The article introduces new graphical methods for assessing calibration over follow-up time which can be combined with Poisson regression analysis. The impact of using different methods to assess calibration of cumulative absolute risk is reviewed. The authors emphasize the importance of providing a detailed description of the method for expected risk when presenting calibration findings. While the current risk model is effective, there are opportunities for refinement and improvement. Other factors that affect breast cancer risk exist but have difficulties being included in a relatively simple model. The article concludes by applying the breast cancer risk model and calibration assessment methods to data from a large cohort of women attending mammographic screening in Washington, USA. Overall, this research contributes to ongoing efforts to prevent cancer and diagnose it early when it is most treatable. By accurately assessing an individual's breast cancer risk, targeted interventions can be implemented to reduce the public health burden associated with rising disease incidence.