Adding SNPs To Breast Cancer Risk Model Does Not Increase Accuracy Meaningfully

Previous research had recommended a comparison of BCRAT, commonly known as the Gail Model, and BCRATplus7, which includes the seven SNPs associated with breast cancer.

Based on that recommendation, Mitchell H. Gail, of the Division of Cancer Epidemiology and Genetics, National Cancer Institute in Bethesda, Md., investigated four medically important applications that are based on risks and benefits to compare the performance of BCRAT with BCRATplus7. The applications were used to decide which women could benefit from tamoxifen to prevent breast cancer; which should have screening mammography; assessing the extent of reclassification of breast cancer risk; and allocating access to screening mammography under cost constraints.

Gail found that for all applications, the value added with BCRATplus7 compared with BCRAT was very small.

"In view of these uncertainties and the small improvements from BCRATplus7 in these applications, further studies are needed to validate models with SNPs and to assess how much they improve performance over simpler models," the author writes.

This research was published in the Journal of the National Cancer Institute on June 17, 2009.