For decades, genetic counselors have helped patients understand inherited breast cancer risk by evaluating family history and pathogenic variants in genes such as BRCA1, BRCA2, PALB2, ATM, and CHEK2. Yet these pathogenic variants account for only a fraction of genetic breast cancer susceptibility; another substantial source is the combined effect of many low-penetrance variants.^1-3 Polygenic risk scores (PRS) aggregate these effects into a single estimate of genetic risk.

A major milestone for breast cancer PRS came in 2015 when the Breast Cancer Association Consortium published a 77-variant score,² later expanded to 313 variants,³ that identified approximately 10% of females in the general population as having inherited risk high enough to influence screening and prevention decisions. Importantly, multigene panel testing for pathogenic variants in known susceptibility genes does not detect this source of inherited risk.

Accuracy and Equity Across Ancestries

As polygenic risk assessment began to move from research into clinical practice, some questioned whether scores built on predominantly European populations would be accurate for non-European ancestries. Within a year of the landmark 77-variant PRS publication by Mavaddat et al., researchers launched large-scale efforts to investigate this concern.⁴ In 2019, Martin et al. highlighted the broader challenge of PRS transferability across ancestries, and subsequent breast cancer studies showed reduced stratification of European-derived PRS in people of African ancestry.^5,6

In parallel, studies in large, diverse cohorts demonstrated a problem beyond reduced stratification: poor calibration led to substantial inaccuracies when a European-derived PRS was applied to patients who self-identified as Black, Hispanic, Asian, or of mixed non-European ancestry.⁴ These studies further demonstrated that polygenic risk scores based on self-reported ancestry are likely to be inaccurate for people of non-European ancestry, particularly among admixed populations.⁴

To address these limitations, researchers developed a multiple-ancestry PRS (MA-PRS) framework. It characterizes genetic ancestry using ancestry-informative markers rather than self-reported ancestry and applies ancestry-specific allele frequencies and risk estimates to each variant. The MA-PRS has been independently validated in nearly 700,000 individuals across multiple cohorts, including ancestry-specific studies of Ashkenazi Jewish, Black/African, East Asian, Hispanic, mixed non-European, South Asian, and White populations.⁴^,7-¹⁰

Although developed for breast cancer, the principles underlying MA-PRS are broader and may be useful in the development of equitable polygenic risk scores for other cancers and common diseases.⁴

Analytical Validity Matters, Too

Most PRS discussions have focused on clinical validity, but analytical validity (validity of the clinical assay) deserves equal attention. Some PRS rely heavily on imputation, a statistical process that infers unmeasured variants from nearby genetic markers. Imputation is widely used in research, but its implications for clinical testing, especially in historically underrepresented populations, warrant careful consideration.^11,12

Recently, Ho et al. found that PRS from different genotyping arrays and low-coverage sequencing could differ materially for the same individual, with much of the observed discordance attributed to imputation.¹³ All platforms were highly concordant when analyses were restricted to directly genotyped variants.

These results do not imply that imputation is unsuitable for clinical testing. Rather, they demonstrate the importance of validating the clinical assay: research initiatives such as the UK Biobank can establish important aspects of clinical validity, but not the validity of the assay used for patient testing. Regardless of the technical approach used, laboratories offering PRS for clinical decision-making should demonstrate concordance, reproducibility, and population-specific performance using the clinical assay and pipeline.

Clinical Validity: Why Prospective Validation Matters

Many PRS studies rely on case-control analyses, which can show that scores tend to be higher in affected than unaffected individuals but generally cannot demonstrate the accuracy of absolute risk estimates. To determine whether a model accurately predicts risk over time, prospective longitudinal validation is needed.

For breast cancer prevention and screening, accuracy of absolute risk estimates matters: clinical decisions frequently depend on thresholds such as a 20% lifetime risk. Genetic counselors should therefore consider whether a PRS has been validated in prospective longitudinal cohorts that include diverse populations.

Evaluating PRS for Clinical Use

Polygenic risk scores intended for clinical use should demonstrate ancestry-specific accuracy, prospective longitudinal validation, and analytical validity of the clinical assay. When appropriately validated, PRS provide substantial genetic risk information that is complementary to family history and independent of other established risk factors.^4,7-10 Current professional guidelines increasingly acknowledge the emerging role of polygenic risk assessment, while emphasizing the importance of careful validation before clinical implementation. Genetic counselors have an important role in critically evaluating PRS, distinguishing tools supported by rigorous evidence from those with more limited data, and helping patients and providers interpret results in the context of overall risk.

Questions remain about when and how PRS should be incorporated into routine clinical care. As the evidence continues to evolve, it will be essential for genetic counselors to communicate the strengths and limitations of these tools and to help ensure that advances in risk prediction are evaluated and implemented in ways that benefit all patients, regardless of ancestry.

References

1. Torabi Dalivandan S, et al. Risks and function of breast cancer susceptibility alleles. Cancers. 2021;13(16):3953.
2. Mavaddat N, et al. Prediction of breast cancer risk based on profiling with common genetic variants. J Natl Cancer Inst. 2015;107(5):djv036.
3. Mavaddat N, et al. Polygenic risk scores for prediction of breast cancer and breast cancer subtypes. Am J Hum Genet. 2019;104(1):21-34.
4. Hughes E, et al. Development and validation of a breast cancer polygenic risk score on the basis of genetic ancestry composition. JCO Precis Oncol. 2022;6:e2200084.
5. Martin AR, et al. Clinical use of current polygenic risk scores may exacerbate health disparities. Nat Genet. 2019;51(4):584-591.
6. Du Z, et al. Evaluating polygenic risk scores for breast cancer in women of African ancestry. J Natl Cancer Inst. 2021;113(9):1168-1176.
7. Mabey B, et al. Validation of a clinical breast cancer risk assessment tool combining a polygenic score for all ancestries with traditional risk factors. Genet Med. 2024;26(7):101128.
8. Simmons, T, et al. Longitudinal validation in UK Biobank of a breast cancer risk model combining a multiple-ancestry polygenic score with traditional risk factors. Genetics in Medicine Open. 2025.
9. Simmons, T, et al. "Improving a polygenic risk score (PRS) for breast cancer (BC) risk assessment in diverse ancestries." 2024: 10533-10533.2025.
10. Simmons T, et al. A breast cancer risk model incorporating a polygenic score for all ancestries: clinical validation in the prospective All of Us cohort. Presented at: San Antonio Breast Cancer Symposium; 2026; San Antonio, TX.
11. Cahoon JL, et al. Imputation accuracy across global human populations. Am J Hum Genet. 2024; 111(5), 979-989.
12. Sengupta D, et al. Performance and accuracy evaluation of reference panels for genotype imputation in sub-Saharan African populations. Cell Genom. 2023;3(6):100332.
13. Ho PJ, et al. Genomic platform-specific polygenic risk scores impact breast cancer risk stratification. Commun Med. 2026;6:41.

Elisha Hughes, PhD

is a mathematician and Director of Biostatistics at Myriad Genetics. She leads a team that investigates how genetic variation affects disease risk, prognosis, and therapeutic response, and translates genetic insights into clinically useful tools for personalized disease prevention and treatment. She has led efforts to improve the accuracy of polygenic risk assessment across diverse populations, and her research has been featured in numerous scientific publications and presentations. Dr. Hughes is passionate about advancing equity in medical genetics and expanding access to personalized care.