By replacing generic, age-determined breast cancer screening recommendations with personalized guidelines based on risk, a team of researchers says healthcare decision makers could save lives and reduce false-positives.
Oguzhan Alagoz, a professor of industrial and systems engineering at the University of Wisconsin-Madison, led a mathematical modeling study that showed the advantages of risk-based breast cancer screening strategies. The study appeared in the journal JAMA Network Open in January 2026. Alagoz’s collaborators included 15 researchers from across the United States and the Netherlands.
Alagoz, PhD student Yifan Lu and their collaborators tested an array of risk-based screening strategies using a pair of Cancer Intervention and Surveillance Modeling Network (CISNET) breast cancer models. They tested 50 screening strategies: three age-based and 47 risk-based—the latter determined by a five-year absolute invasive breast cancer risk calculator.
The researchers found that nine of the risk-based strategies, compared against age-based screenings for women ages 40-74, could yield comparable or greater numbers of averted deaths and reduce false-positives by 8-23%.
Traditionally, screening recommendations have been based on the average risk within a given age range, rather than an individualized approach.
“As our knowledge of individual risk factors grows, continuing with blanket recommendations becomes increasingly inefficient,” says Alagoz, who has spent the past two decades using modeling to improve breast cancer screening recommendations for various populations. “Our study demonstrates that by using available risk data, we can achieve a better balance of benefits and harms—maintaining or improving breast cancer survival rates while significantly reducing the burden of over-screening.”
Alagoz manages one of the models that’s part of the National Cancer Institute-funded CISNET, along with Amy Trentham-Dietz, a professor of population health sciences in the UW-Madison School of Medicine and Public Health.
He says that, in lieu of long and complex clinical trials, simulation modeling—in which researchers create a computational representation of a real system—offers a complementary method for assessing new screening strategies.
“While models cannot replace clinical trials, they allow us to evaluate a wide range of screening strategies and long-term outcomes that are difficult to study empirically,” he says. “So, they serve as a ‘virtual laboratory.’”
Oguzhan Alagoz is the Procter & Gamble Bascom Professor of industrial and systems engineering. UW-Madison alumni Natasha Stout, project scientist for CISNET, and Brian Sprague, a professor at the University of Vermont, are also authors on the study.
This research was supported by the National Cancer Institute under the National Institutes of Health (grants U01CA152958, U01CA253911, R01CA248068, R35CA283926, R35CA197289 and P30CA014520), as well as the American Cancer Society (CSDG-21-078-01-CPSH). Collection of Breast Cancer Surveillance Consortium data used in this study was supported by the National Cancer Institute (grant P01CA154292) and the National Institute of General Medical Sciences (grant U54GM115516).