# Predicting MammaPrint Recurrence Risk from Breast Cancer Pathological Images Using a Weakly Supervised Transformer

## TL;DR

A new AI model called CPMP predicts breast cancer recurrence risk from histopathological images, offering a cost-effective alternative to genomic tests.

## Contribution

CPMP is a weakly supervised transformer model that enables spatial and morphological analysis of tumor images to predict MammaPrint risk groups.

## Key findings

- CPMP achieves an AUROC of 0.824 ± 0.03 in predicting MammaPrint recurrence risk groups.
- The model reveals distinct spatial and morphological patterns associated with high and low MammaPrint risk groups.
- Prognostic evaluation shows CPMP significantly stratifies distant metastasis risk in an external cohort.

## Abstract

Recurrence related to poor prognosis is a leading cause of mortality in patients with breast cancer (BC). The MammaPrint (MP) genomic assay is designed to stratify recurrence risk and evaluate chemotherapy benefits for early‐stage HR+/HER2‐ BC patients. However, MP fails to reveal spatial tumor morphology and is limited by high costs. In this study, a BC MP cohort is established and CPMP is developed, a weakly supervised agent‐attention transformer model, to predict MP recurrence risk from annotation‐free BC histopathological slides. CPMP achieves an AUROC of 0.824 ± 0.03 in predicting MP risk groups. CPMP is further leveraged for spatial and morphological analyses to explore histological patterns associated with MP risk groups. The model reveals tumor spatial localization at the whole‐slide level and highlights distinct intercellular interaction patterns of MP groups. It also characterizes the diversity in tumor morphology and uncovers MP high‐specific, low‐specific, and colocalized morphological phenotypes that differ in quantitative cellular composition. Prognostic evaluation in the external cohort exhibits significant stratification of distant metastasis risk (HR: 3.14, p‐value = 0.0014), underscoring the prognostic power of CPMP. These findings demonstrate the capability of CPMP in MP risk prediction, offering a flexible supplement to genomic risk assessment in early‐stage BC.

This study presents CPMP, a weakly supervised transformer model that predicts MammaPrint recurrence risk directly from routine histopathological images of early‐stage HR+/HER2− breast cancer patients. CPMP enables spatial heatmap visualization, analysis of cellular‐level interaction patterns, and an in‐depth characterization of morphological phenotypes associated with MP risk groups, providing an AI‐driven and cost‐efficient tool for breast cancer recurrence risk assessment.