Agendia Unveils Clinical Validation of AI-Driven Digital Pathology Approach

Agendia, Inc., a leader in precision oncology for breast cancer, recently presented research introducing its investigational artificial intelligence (AI)-driven digital pathology approach for the MammaPrint®70-gene recurrence signature at the European Society for Medical Oncology (ESMO) Congress 2026 in Madrid. The podium presentation, delivered by co-founder and Chief Research Officer, Laura van ’t Veer, Ph.D., detailed the development and clinical validation of a deep learning pipeline that can infer MammaPrint (MP) risk categories directly from hematoxylin and eosin (H&E)-stained, whole-slide images (WSIs). The company also presented two additional posters highlighting data from the prospective, real-world FLEX Study (NCT03053193).

“AI has been part of Agendia’s scientific foundation from the beginning and this study represents its continued evolution,” said Dr. Laura van ‘t Veer. "For the first time, we are presenting our work exploring whether the biological information represented by MammaPrint can be recognized through AI applied to routinely generated pathology images.”

The investigational AI-driven digital pathology model was trained on 70,000 de-identified H&E WSIs with MP-derived supervision. The deep learning pipeline included tissue detection, classification and feature extraction using self-supervised and supervised foundation models. The model then predicted a continuous recurrence score that was dichotomized into High- (D-High) or Low- (D-Low) risk groups. Results were validated using de-identified data from 372 early-stage breast cancer (EBC) patients from the prospective, real-world RASTER study with a median follow-up of 10.3 years.

• The investigational digital assay demonstrated a high concordance with MP (r = 0.954), classifying 49% of patients as D-High and 51% as D-Low.
• Patients classified as D-Low had a 10-year breast cancer-specific survival (BCSS) rate of 98.1%, compared to 86.7% for those classified as D-High (HR 0.12, p < 0.0001).
  • In the estrogen receptor-positive (ER+)/HER2-negative subgroup of 274 patients, 10-year BCSS was 98.1% versus 84.7%, respectively (HR 0.10).
• In multivariable analysis, digital risk classification was the only independent predictor of BCSS (HR 0.084, p = 0.0002), outperforming clinical factors. Similar findings were observed for distant recurrence-free intervals.

“AI-driven digital pathology approaches have the potential to expand access to genomic information for breast cancer patients globally, particularly in scenarios where access to tissue-based genomics is limited,” said William Audeh, M.D., M.S., Chief Medical Officer at Agendia. “This research brings together decades of genomic science with emerging digital pathology capabilities and reflects our commitment to developing innovative approaches to understanding tumor biology.”

To see other data presented by Agendia at ESMO, refer to the ESMO 2026 landing page.

About Agendia

Agendia is a global leader in precision oncology focused on early-stage breast cancer. The company’s genomic assays, MammaPrint + BluePrint, deliver essential biological insights to inform personalized treatment decisions for patients and their care teams. With operations in Amsterdam and Irvine, Agendia partners with academic and community oncology centers worldwide to generate real-world evidence through the landmark FLEX Study (NCT03053193), the largest whole-transcriptome registry of early-stage breast cancer.

View source version on businesswire.com: https://www.businesswire.com/news/home/20261026703849/en/