Verghese, G. | Search bioRxiv

bioRxiv2025

SMART: A Spatio-Molecular Atlas of Response Trajectories in Triple-Negative Breast Cancer

A major challenge in treating Triple-Negative Breast Cancer (TNBC) lies in its molecular, morphological and clinical heterogeneity, which hampers accurate prediction of responses to neoadjuvant treatment. To address this, we introduce SMART: Spatio-Molecular Atlas of Response Trajectories, a comprehensive, multimodal resource compiled from 129 TNBC samples across 89 patients, obtained before, during, and after neoadjuvant chemotherapy (NACT). SMART comprises of 5,096 high quality manually selected spatial transcriptomic profiles enriched for epithelial, immune, or stromal compartments; paralleled with histological annotations, imagebased network analysis and protein expression. Seven novel spatial epithelial archetypes (EAs), seven tumour-immune microenvironments (TIMEs) and their co-localisation patterns were defined, revealing an opposing prevalence of functionally divergent EAs between response groups and the prognostic significance of B-cell enriched TIMEs, in particular those surrounding histologically normal epithelium adjacent to the tumour. The SMART dataset and analytical tools are publicly available via the PharosAI platform, providing the research community with the most comprehensive, manually annotated spatio-molecular transcriptomics atlas of NACT-treated TNBC to date.

cancer biology↗

bioRxiv2025

Normal breast tissue classifiers assess large-scale tissue compartments with high accuracy

Cancer research emphasises early detection, yet quantitative methods for normal tissue analysis remain limited. Digitised haematoxylin and eosin (H&E)-stained slides enable computational histopathology, but artificial intelligence (AI)-based analysis of normal breast tissue (NBT) in whole slide images (WSIs) remains scarce. We curated 70 WSIs of NBTs from multiple sources and cohorts with pathologist-guided manual annotations of epithelium, stroma, and adipocytes (https://github.com/cancerbioinformatics/OASIS). We developed robust convolutional neural network (CNN)-based, patch-level classification models, named NBT-Classifiers, to tessellate and classify NBTs at different scales. Across three external cohorts, NBT-Classifiers trained on 128{square}x{square}128{square}{micro}m and 256{square}x{square}256{square}{micro}m patches achieved AUCs of 0.98-1.00. The model learned independent normal features different from those of precancerous and cancerous epithelium, which were further visualised using two explainable AI techniques. When integrated into an end-to-end preprocessing pipeline, NBT-Classifiers facilitate efficient downstream analysis within peri-lobular regions. NBT-Classifiers provide robust compartment-specific analytical tools and enhance our understanding of NBT appearances, which serve as valuable reference points for identifying premalignant changes and guiding early breast cancer prevention strategies.

pathology↗