By Clinical Research News Staff 

May 21, 2025 | A clinical research initiative has demonstrated that artificial intelligence could significantly enhance the identification of patients eligible for targeted breast cancer therapies, according to preliminary findings presented by Friends of Cancer Research. 

The Digital PATH Project, involving 31 contributing partners including pharmaceutical companies, academic centers, and government agencies, represents one of the most comprehensive evaluations of AI diagnostic technologies in oncology to date. The study utilized approximately 1,100 breast cancer tissue samples to assess how consistently these emerging tools can identify HER2 expression. 

The team assessed variability between different digital pathology tools, explained Jeff Allen, president and CEO of Friends of Cancer Research, who will present the study's results at the upcoming Next Generation Dx Summit. The intent wasn't to declare one technology better than another, but to learn if they could produce consistent and accurate results, he said.  

The research comes at a pivotal time in breast cancer treatment. The recent clinical recognition of "HER2-low" breast cancer has expanded treatment possibilities, with three antibody-drug conjugates now approved for patients previously classified as HER2-negative. This shift has heightened the importance of precise biomarker quantification. 

For many years, the diagnostic journey for patients with solid tumors begins with a biopsy where the sample tissue gets stained with H&E to enhance visualization, Allen explained. Certain types of cancer, including HER2-positive breast cancers, may be subsequently stained using different antibodies and immunofluorescence to further characterize the presence of molecular alterations. 

The study found that while AI tools consistently matched pathologists' assessments for high HER2 expression, they showed greater variability when evaluating low expression levels. This variability wasn't unexpected, as Allen noted these tools were trained before widespread recognition of the need to score HER2 at low levels because it was not yet an “actionable classification." 

The research methodology—using a standardized reference set across multiple platforms—demonstrates a potential pathway for efficient clinical validation of AI diagnostic tools. The speed with which the samples were evaluated—"in a matter of days and weeks"—highlights how this approach could accelerate regulatory approval processes. 

“We're very interested in exploring what the policy implications may be and how the use of independent reference sets could support the validation of these types of technologies for evaluating other biomarkers in the future," Allen said. 

Following this successful initiative, Friends of Cancer Research has already launched a similar project focused on AI-enabled radiographic imaging tools that measure tumor changes following treatment. 

Read Deborah Borfitz’s full story at Diagnostics World News.