By Kathleen Berger, Executive Producer of Science and Technology

AI-based breast cancer prediction technology applied to mammograms may catch breast cancer very early for women at high risk of developing breast cancer. It may also reduce the need for additional imaging for all women at different times in their lives for different reasons, to include women at high risk or low risk for developing breast cancer.

“If you can give somebody the gift of saying, ‘You’ve got five more years,’” said Debbie Bennett, MD, Breast Radiologist at Siteman Cancer Center and the Ronald and Hanna Evens Professor of Women’s Health and Section Chief of Breast Imaging at WashU Medicine in St. Louis. “The anxiety that is caused by knowing that ‘Okay, at some point in the years remaining to me there’s a high chance I am going to develop breast cancer,’ but not knowing what year that’s going to be is really difficult. People come in here and every time they’re here they’re waiting for the other shoe to drop. Every year they think, ‘This is the year I am going to get breast cancer.’”

Bennett is leading a new clinical trial to help advance groundbreaking breast cancer risk prediction technology developed at WashU Medicine by Graham A. Colditz, MD, DrPH, and Shu (Joy) Jiang, PhD. The new technology that harnesses AI to analyze mammograms and improve the accuracy of predicting a woman’s personalized five-year risk of developing breast cancer has received Breakthrough Device designation from the Food and Drug Administration (FDA).

Colditz is the Niess-Gain Professor of Surgery at WashU Medicine and associate director of prevention and control at Siteman Cancer Center, based at Barnes-Jewish Hospital and WashU Medicine. Jiang is an associate professor of surgery in the Division of Public Health Sciences in the Department of Surgery at WashU Medicine.

“A woman with a family history may be at higher risk for lifetime risk (of developing breast cancer), which is a metric being used right now in clinic. We don’t know how long we are going to live, so looking at lifetime risk doesn’t make much sense. But looking at a five-year risk, you can actually see that horizon. That’s five years from today,” said Jiang. “Currently in clinic right now, women can get their five-year breast cancer risk. The way of getting that is by filling out a bunch of questionnaires, but the accuracy is not very good.”

So, the researchers developed new technology that harnesses AI to identify individuals at high risk of developing breast cancer more accurately than the standard, questionnaire-based method.

“We’ve done several validations on a global scale,” said Jiang. “Our tool can achieve up to 85% accuracy. It’s over twice as more accurate, compared to questionnaires alone.”

“I’ll call it a giant leap forward towards personalized prevention strategies,” said Colditz. “With millions of bits of data from the mammogram, we get a much richer summary of the breast tissue than a four-point scale on density.”

The FDA Breakthrough Device designation provides an expedited review process for full market approval.