Agendia says AI breast cancer assay matched MammaPrint

Agendia said an investigational AI-driven digital pathology approach showed high concordance with its MammaPrint breast cancer recurrence test in a validation group of 372 early-stage breast cancer patients.

The company presented the research at the European Society for Medical Oncology Congress 2026 in Madrid. It said the deep-learning system was trained on 70,000 de-identified hematoxylin and eosin whole-slide images and then validated using de-identified data from the prospective RASTER study, with a median follow-up of 10.3 years.

Agendia said the investigational digital assay classified 49% of patients as digital high risk and 51% as digital low risk, with concordance to MammaPrint of r = 0.954.

The company said patients classified as digital low risk had a 10-year breast cancer-specific survival rate of 98.1%, compared with 86.7% for those classified as digital high risk. In an estrogen receptor-positive, HER2-negative subgroup of 274 patients, it said 10-year breast cancer-specific survival was 98.1% versus 84.7%.