Inferring Clinically Relevant Molecular Subtypes of Pancreatic Cancer from Routine Histopathology Using Deep Learning

2026-01-06

medRxiv

AI Summary

This study addresses the clinical challenge of molecular subtyping in pancreatic ductal adenocarcinoma (PDAC), which is hindered by the high cost and tissue requirements of sequencing-based approaches. To overcome this, we propose PanSubNet—an interpretable deep learning framework that, for the first time, accurately predicts therapeutically relevant basal-like and classical PDAC subtypes using only routine hematoxylin and eosin (H&E)-stained whole-slide images. Our method integrates cellular- and tissue-level features through a multi-scale representation learning architecture enhanced with attention mechanisms for transparent attribution. Ground-truth labels are refined using the Moffitt 50-gene signature and GATA6 expression. PanSubNet achieves an AUC of 88.5% in internal five-fold cross-validation and 84.0% on an external TCGA cohort, demonstrating superior prognostic stratification in metastatic cases compared to conventional RNA-seq–based subtyping.

Technology Category

• Machine Learning: Multi-instance/Multi-view Learning
• Computer Vision: Other Foundations of Computer Vision
• Search and Optimization: Learning to Search

Application Category

• Web Mining and Content Analysis: Large pretrained models with web data
• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking

Abstract

Background and aims: Molecular subtyping of pancreatic ductal adenocarcinoma (PDAC) into basal-like and classical has established prognostic and predictive value. However, its use in clinical practice is limited by cost, turnaround time, and tissue requirements, thereby restricting its application in the management of PDAC. We introduce PanSubNet (PANcreatic SUBtyping NETwork), an interpretable deep learning framework that predicts therapy-relevant molecular subtypes directly from standard hematoxylin and eosin (H&E)-stained whole-slide images. Methods: PanSubNet was developed using data from 1,055 patients across two multi-institutional cohorts (PANCAN, n=846; TCGA, n=209) with paired histology and RNA sequencing data. Ground-truth labels were derived using the validated Moffitt 50-gene signature refined by GATA6 expression. The model employs dual-scale architecture that fuses cellular-level morphology with tissue-level architecture, leveraging attention mechanisms for multi-scale representation learning and transparent feature attribution. Results: On internal validation within PANCAN using five-fold cross-validation, PanSubNet achieved mean area under the receiver operating characteristic curve (AUC) of 88.5% in high-confidence cases, with balanced sensitivity and specificity. External validation on the independent TCGA cohort without fine-tuning demonstrated robust generalizability (AUC 84.0%). PanSubNet preserved and, in metastatic disease, strengthened prognostic stratification compared to RNA-seq based labels. Prediction uncertainty linked to intermediate transcriptional states, not classification noise. Model predictions are aligned with established transcriptomic programs, differentiation markers, and DNA damage repair signatures. Conclusions: By enabling rapid, cost-effective molecular stratification from routine H&E-stained slides, PanSubNet offers a clinically deployable and interpretable tool for genetic subtyping. We are gathering data from two institutions to validate and assess real-world performance, supporting integration into digital pathology workflows and advancing precision oncology for PDAC

Problem

• pancreatic cancer
• molecular subtyping
• clinical application
• histopathology
• prognostic stratification

Innovation

• deep learning
• molecular subtyping
• histopathology
• interpretable AI
• pancreatic cancer