Pathologists who used an artificial intelligence (AI)-assisted training platform for assessing immunohistochemistry (IHC) showed a greater degree of accuracy at classifying cases of low or ultralow HER2 expression compared with reference scores that did not use AI assistance, according to recent research presented at the 2025 ASCO Annual Meeting. 

“Roughly 65% of breast tumors once called HER2‑negative actually demonstrate some level of HER2 expression and belong to subgroups now classified as HER2-low or HER2-ultralow breast cancers. Some of these tumors could be treated with HER2-targeted drugs, but only if we detect their HER2 expression levels. Our study provides the first multinational evidence that AI can help close a critical diagnostic gap and open the door to new therapies like antibody-drug conjugates for a majority of patients who, until recently, had not been offered these options.” 

- Marina De Brot, MD, PhD, of A.C. Camargo Cancer Center in São Paolo, Brazil, stated in a press release. 

De Brot and colleagues analyzed digital HER2 IHC assessment of breast cancer samples from 105 pathologists who attended a masterclass and were asked to conduct an assessment of 20 IHC-stained HER2 cases using AI and without AI. In total, the pathologists performed 1940 readings, with AI-assisted readings having an average of 89.6% agreement compared with 76.3% agreement without AI. 

Classification of HER2 clinical categories such as positive, low, ultralow, and null was 88.5% with AI assistance compared with 66.7% without AI assistance, while misclassification of cases as HER2 null when they were HER2-ultralow was 4.0% with AI assistance compared with 29.5% without the use of AI. 

The researchers said they intend to use the AI tool for routine diagnostics across multiple centers and will assess the use of AI for clinical changes such as treatment and time-to-therapy for HER2-low and HER2-ultralow cases. 

“Accurate HER2 scoring is important to ensure that patients receive the best treatment for their breast cancer. This international study shows that an AI-assisted approach improved HER2 scoring, including in situations that would affect treatment decisions. These findings shed light on the promising role for AI in oncology, not as a replacement for the physician, but as a powerful tool to help us work smarter and faster to deliver high-quality, more personalized care.” 

- Julian Hong, MD, MS, medical director of Radiation Oncology Informatics at the University of California, San Francisco, who was not involved with the study, said in a press release. 

-- 

References:  

https://meetings.asco.org/abstracts-presentations/253080 

https://www.asco.org/about-asco/press-center/news-releases/artificial-intelligence-assistance-improve-accuracy-identifying-her2-low-ultralow-breast-cancers  

Disclosures: Some authors declared financial ties to drugmakers. See full study for details. This study was supported by AstraZeneca. 

Photo Credit: Getty Images 

By Jeff Craven