Computational Mapping of Reactive Stroma in Prostate Cancer Yields Interpretable, Prognostic Biomarkers

2026-01-10

AI Summary

Current prostate cancer grading relies predominantly on glandular architecture, overlooking the prognostic value of reactive stroma (RS). To address this gap, we propose PROTAS, a deep learning framework that, for the first time, enables automated and interpretable quantification of RS in routine hematoxylin and eosin (H&E)-stained slides. By integrating domain-adversarial training with spatial feature analysis, PROTAS links RS morphological characteristics to underlying biological mechanisms, such as contraction pathways, and demonstrates robust generalization to diagnostic biopsies in the PLCO dataset. Notably, PROTAS independently predicts biochemical recurrence beyond established clinical and pathological metrics, achieving a concordance index (c-index) of 0.80—significantly outperforming pathologist-based assessments—and thereby introduces a novel dimension for prostate cancer prognosis.

Technology Category

• Machine Learning: Feature Construction/Reformulation
• Computer Vision: Adversarial Attacks & Robustness
• Intelligent Robots: Learning & Optimization for ROB

Application Category

• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Economics, Online Markets and Human Computation: Data quality aspects of human-annotated datasets
• User Modeling, Personalization and Recommendation: On-Device user modeling, personalization, and recommendation

Abstract

Current histopathological grading of prostate cancer relies primarily on glandular architecture, largely overlooking the tumor microenvironment. Here, we present PROTAS, a deep learning framework that quantifies reactive stroma (RS) in routine hematoxylin and eosin (H&E) slides and links stromal morphology to underlying biology. PROTAS-defined RS is characterized by nuclear enlargement, collagen disorganization, and transcriptomic enrichment of contractile pathways. PROTAS detects RS robustly in the external Prostate, Lung, Colorectal, and Ovarian (PLCO) dataset and, using domain-adversarial training, generalizes to diagnostic biopsies. In head-to-head comparisons, PROTAS outperforms pathologists for RS detection, and spatial RS features predict biochemical recurrence independently of established prognostic variables (c-index 0.80). By capturing subtle stromal phenotypes associated with tumor progression, PROTAS provides an interpretable, scalable biomarker to refine risk stratification.

Problem

• reactive stroma
• prostate cancer
• tumor microenvironment
• histopathological grading
• biomarker

Innovation

• reactive stroma
• deep learning
• prostate cancer
• interpretable biomarker
• domain-adversarial training