Despite major advances in genetic testing for breast cancer, women of African ancestry continue to experience disproportionately high death rates from the disease. Researchers have long attributed this gap to multiple factors, including shortcomings in existing genetic risk models, a higher prevalence of aggressive tumor subtypes, and diagnoses that often occur at later stages.
A new study from the University of Chicago Medicine may mark a turning point. Scientists have developed advanced polygenic risk score (PRS) models that significantly improve breast cancer risk prediction for women of African ancestry, addressing a long-standing inequity in precision medicine. Drawing on genetic data from more than 36,000 women, the research represents the most comprehensive effort to date to tailor breast cancer risk assessment for this historically underserved population. The findings were published today in Nature Genetics.
Why existing models fall short
Most widely used genetic tools for breast cancer risk prediction were developed using data from women of European ancestry. While these models perform well in that population, they are far less accurate for African American women, particularly when predicting aggressive forms of the disease such as triple-negative breast cancer (TNBC).
Genetic diversity plays a central role in this disparity. Populations of African ancestry have greater genetic variation, and differences in how genetic variants are distributed can influence disease risk. Models trained primarily on European genetic data often fail to capture critical risk signals present in African genomes.
Polygenic risk scores estimate disease risk by analyzing thousands of single nucleotide polymorphisms, or SNPs—small variations in DNA that, when considered collectively, can meaningfully alter cancer risk. While PRS approaches have been effective in European populations, limited sample sizes and greater genetic diversity have historically reduced their accuracy for women of African ancestry.
“Polygenic risk scores worked well for European-Americans but weren’t accurate for African American women due to smaller sample sizes and greater genetic diversity,” said Dezheng Huo, PhD, professor of public health sciences at the University of Chicago and senior author of the study. “By forming a large consortium and combining data from investigators across 20 institutions, we were able to substantially improve prediction accuracy.”
Tailoring models to the right population
The research team used data from the African Ancestry Breast Cancer Genetics Consortium, which includes participants from the United States, the Caribbean, and sub-Saharan Africa. Women in the study either had a breast cancer diagnosis or served as healthy controls.
Using this dataset, the researchers developed PRS models specifically designed for women of African ancestry, covering four categories: overall breast cancer, estrogen receptor–positive (ER+), estrogen receptor–negative (ER–), and triple-negative breast cancer.
Model performance was measured using the area under the curve (AUC), a standard metric that reflects how well a model distinguishes between individuals who will develop breast cancer and those who will not. The new models achieved AUC scores ranging from 0.61 to 0.64, compared with 0.56 to 0.58 for earlier tools—representing a meaningful improvement in predictive power.
To enhance clinical practicality and reduce costs, the team also created streamlined versions of the models. One simplified TNBC model relied on just 162 genetic markers while maintaining comparable performance, achieving an AUC of 0.626.
“With improved risk prediction, clinicians can begin screening earlier for women at higher risk, personalize care, and detect cancers at more treatable stages,” Huo said.
Earlier screening and higher-risk identification
The study found that women in the top 1% of overall PRS scores had a 25.7% lifetime risk of developing breast cancer. For TNBC, women in the top 1% faced a lifetime risk of 7.4%—notable given the aggressive nature of the disease.
Based on these findings, researchers suggest that women at very high genetic risk could benefit from screening as early as age 32, rather than waiting until ages 40 or 45, as current guidelines often recommend.
Family history strengthens predictive power
The predictive value of the new models increased further when combined with family history, a well-established breast cancer risk factor. Women in the top 1% of PRS scores who also had a first-degree relative with breast cancer faced a lifetime risk exceeding 50%.
Such risk levels could justify earlier and more frequent screening, as well as preventive strategies such as risk-reducing medications or genetic counseling, the researchers noted.
To confirm the reliability of the models, the team validated them across multiple independent datasets, including the National Institutes of Health’s All of Us Research Program and three additional studies involving women of African ancestry. In one validation cohort, the TNBC model achieved an AUC of 0.652, demonstrating consistent performance across diverse populations.
Toward more equitable precision medicine
While the current research focused primarily on African American women and those of West African ancestry, the authors emphasized the need for further studies. Genetic differences among West, East, North, and South African populations—as well as across the global African diaspora—underscore the importance of continued research and model refinement.
“These advanced testing models bring us closer to a future where everyone, regardless of ancestry, has an equal opportunity for early detection, effective treatment, and survival,” Huo said.
The study, titled “Improved Polygenic Risk Prediction Models for Breast Cancer Subtypes in Women of African Ancestry,” was supported by grants from the National Institutes of Health, the Breast Cancer Research Foundation, and the Susan G. Komen Foundation.
