Development and Validation of a Breast Cancer Polygenic Risk Score on the Basis of Genetic Ancestry Composition

Area of Science:

• Genetics
• Oncology
• Population Health

Background:

• Polygenic risk scores (PRSs) for breast cancer (BC) risk stratification are primarily developed in European ancestry populations.
• Their application to non-European ancestries is limited due to challenges in incorporating genetic ancestry and ancestry-dependent variant data.

Purpose of the Study:

• To propose and evaluate a multiple-ancestry PRS (MA-PRS) for equitable breast cancer risk prediction across diverse ancestries.
• To address limitations in current PRSs by formally incorporating genetic ancestry and ancestry-dependent variant information.

Main Methods:

• Developed and validated a MA-PRS combining ancestry-specific PRSs based on individual genetic ancestral composition (African, East Asian, European).
• Utilized large cohorts (development n=189,230; validation n=89,126) referred for hereditary cancer testing.
• Assessed risk stratification using multivariable logistic regression and goodness-of-fit analyses.

Main Results:

• The MA-PRS demonstrated significant association with BC risk in the full cohort (OR, 1.43) and within major ancestries.
• The top MA-PRS decile identified individuals with a two-fold increased risk of developing BC.
• The MA-PRS showed good calibration and accurate prediction in both European and non-European women.

Conclusions:

• The MA-PRS effectively expands polygenic risk prediction utility to non-European women by using genetic ancestral composition.
• Incorporating genetic ancestry into PRSs offers potential for more personalized treatment decisions for women of diverse and mixed ancestries.