Impact of Population-Based Pathogenic Variant Testing on Risk-Based Breast Screening Recommendations: A Secondary Analysis of the WISDOM Study

Why This Study Matters

Risk-based breast cancer screening is designed to match screening intensity to a person’s estimated chance of developing breast cancer. Instead of giving every woman the same screening schedule, this approach uses information such as age, family history, breast density, prior biopsies, and genetic factors to decide who may benefit from earlier or more intensive screening, including breast MRI.

This study asks an important question: if a woman carries a pathogenic variant, or PV, in a breast cancer susceptibility gene, would her clinical risk profile alone—or her clinical risk plus a polygenic risk score—have identified her as someone who should receive high-risk screening? The answer matters because PV testing may detect people who would otherwise be missed by standard risk models.

Background: What Are Pathogenic Variants and Polygenic Risk Scores?

Pathogenic variants are inherited changes in genes that increase the risk of disease. In breast cancer, well-known genes include BRCA1 and BRCA2, but there are many others, such as PALB2, TP53, ATM, CHEK2, and RAD51C. Some variants are high penetrance, meaning they confer a large increase in cancer risk, while others are moderate or lower penetrance.

A polygenic risk score is different. Rather than focusing on a single gene, it combines the small effects of many common genetic variants across the genome to estimate inherited risk. In principle, adding a polygenic risk score to clinical factors may improve screening decisions. However, it is not yet clear whether this approach captures the same women who would be flagged by direct PV testing.

Study Design

This cohort study used retrospective data from the WISDOM Study, a national clinical trial comparing risk-based breast cancer screening with annual mammography. Women aged 40 to 74 years who tested positive for a pathogenic variant as part of the trial were included.

Researchers compared two kinds of screening assignments:
1. The actual screening assignment, which incorporated the participant’s PV status.
2. Hypothetical screening assignments based on either a clinical risk model alone or a clinical risk model combined with a polygenic risk score.

The clinical model used was based on the Breast Cancer Surveillance Consortium risk tool, which estimates breast cancer risk using clinical and demographic factors. The study included data collected from September 2016 through February 2023, with follow-up through September 2025.

Who Was Included?

A total of 712 women with a pathogenic variant were analyzed. The median age was 53 years, with an interquartile range of 46 to 62 years.

The genetic findings were divided into three broad groups:
– 232 women (33%) carried high-penetrance variants
– 278 women (39%) carried moderate-penetrance variants
– 202 women (28%) carried CHEK2 low-penetrance variants

This mix is important because not all genetic findings lead to the same cancer risk or the same screening recommendations.

Main Findings

The study found little overlap between the screening recommendations based on actual PV status and the hypothetical recommendations based on clinical risk plus polygenic risk score.

Among women with high-penetrance pathogenic variants, only 2 of 232 (0.9%) would have been assigned the same high-risk screening recommendation using the clinical model plus polygenic risk score. These women were in a group that would typically be recommended breast MRI alternating with mammography every 6 months, reflecting their very high inherited risk.

The mismatch was especially notable in women aged 40 to 49 years. Of 279 PV carriers in this age group, 178 (63.8%) would otherwise have been advised to delay screening until age 50 based on clinical plus polygenic risk assessment. In other words, more than half of the women carrying a cancer-predisposing variant would not have been identified as needing earlier screening if PV testing had not been performed.

Among the 433 carriers aged 50 to 74 years, 385 (88.9%) would have been recommended to receive biennial mammography rather than intensified high-risk screening.

The results were similar when the researchers compared PV-informed screening with recommendations based on the clinical model alone, showing that the addition of polygenic risk did not substantially close the gap.

What the Results Mean

The key message is that pathogenic variant testing identifies a different group of women than clinical risk models and polygenic risk scores do.

This matters for several reasons. First, women with high-penetrance variants may need earlier and more intensive surveillance to detect breast cancer at a curable stage. Second, if a risk model misses these women, they may be placed on routine screening pathways that are not appropriate for their inherited risk. Third, relying only on clinical features and polygenic risk could give a false sense of reassurance.

The study suggests that population-based PV testing may add unique value in risk-based screening programs. In other words, genetic testing for actionable breast cancer genes may uncover women who would not be recognized as high risk by other methods.

Clinical Implications

These findings support the idea that screening programs should consider including population-based pathogenic variant testing, especially when the goal is to personalize breast cancer screening and prevention.

For women with high-penetrance variants, high-risk screening often includes annual breast MRI and mammography, sometimes alternating every six months. For moderate-risk genes, management may vary depending on the exact gene, age, family history, and local guidelines. Women with CHEK2 variants may also benefit from tailored surveillance, although risk and recommendations can differ by variant type and family context.

The study does not mean that clinical risk models are unhelpful. On the contrary, they remain important tools for many women, especially when genetic testing is unavailable or negative. But the findings show that clinical models and polygenic scores alone are not enough to identify all women who carry clinically important pathogenic variants.

Strengths and Limitations

A major strength of the study is that it comes from a large, national trial with real-world screening data. It directly compares actual PV-informed decisions with what would have happened under risk-model-based strategies.

There are also limitations. This was a retrospective secondary analysis, so it was not originally designed to answer every question about screening outcomes. The study focused on women who already had a pathogenic variant and were enrolled in a specific trial population, which may limit generalizability. In addition, a screening recommendation is not the same as a final clinical outcome; whether a woman actually benefits depends on access, adherence, and follow-up care.

Polygenic risk scores are also still evolving. Their performance can vary by ancestry and by the populations used to develop the score. That means the results may not apply equally across all racial and ethnic groups.

Bottom Line

In this study, most women carrying pathogenic variants in breast cancer genes would not have been recommended for high-risk screening based on clinical risk alone or clinical risk plus polygenic risk score. Population-based pathogenic variant testing therefore appears to identify additional high-risk women who might otherwise be missed.

For risk-based breast cancer screening to work well, it may need to combine clinical tools, family history, and genetic testing rather than rely on any single method. This approach could improve the precision of screening and help ensure that women at the highest inherited risk receive appropriate care.

Study Details

Trial registration: ClinicalTrials.gov Identifier NCT02620852.

Published in JAMA Oncology on 2026-05-31.

Citation: Shieh Y, Heise RS, Madlensky L, Sabacan LP, Soto IA, Fiscalini AS, Ross K, Goodman D, Blanco A, Brain S, Heditsian DM, Moya J, Fergus KB, Olopade OI, Scheuner MT, Eklund M, Ziv E, Tice JA, van ’t Veer L, Esserman LJ, and collaborators from the Athena Breast Health Network and the WISDOM Study.