Computational Methods for Breast Cancer Molecular Profiling through Routine Histopathology: A Review

TL;DR

This paper investigates how AI can extract molecular biomarkers from routine H&E breast cancer histopathology to support precision oncology. It surveys non-omic biomarker prediction and omic inference across genomics, transcriptomics, proteomics, and metabolomics, detailing methods such as CNNs, MIL, CLAM, transformers, and graph networks applied to datasets like TCGA, HEROHE, and BCI. It shows that non-omic biomarkers (e.g., ER, PR, HER2, Ki-67, PD-L1) are increasingly detectable from H&E, while omics predictions are advancing but constrained by limited multi-omics datasets and domain-specific pretraining, with AUCs often ranging from 0.7 to 0.9 for various targets. The discussion highlights challenges in data availability, annotation, stain variability, generalizability, and explainability, arguing for standardized validation to translate AI-based digital pathology into clinical practice.

Abstract

Precision medicine has become a central focus in breast cancer management, advancing beyond conventional methods to deliver more precise and individualized therapies. Traditionally, histopathology images have been used primarily for diagnostic purposes; however, they are now recognized for their potential in molecular profiling, which provides deeper insights into cancer prognosis and treatment response. Recent advancements in artificial intelligence (AI) have enabled digital pathology to analyze histopathologic images for both targeted molecular and broader omic biomarkers, marking a pivotal step in personalized cancer care. These technologies offer the capability to extract various biomarkers such as genomic, transcriptomic, proteomic, and metabolomic markers directly from the routine hematoxylin and eosin (H&E) stained images, which can support treatment decisions without the need for costly molecular assays. In this work, we provide a comprehensive review of AI-driven techniques for biomarker detection, with a focus on diverse omic biomarkers that allow novel biomarker discovery. Additionally, we analyze the major challenges faced in this field for robust algorithm development. These challenges highlight areas where further research is essential to bridge the gap between AI research and clinical application.

Figures (2)

• Figure 1: Conventional workflow in breast cancer management from screening to personalized treatment
• Figure 2: Overview of molecular profiling approaches in breast cancer, each contributing to personalized treatment insights