Deep learning–based histologic classifiers enable molecular subtyping of metastatic prostate cancer
Open Access
- 6 August 2026
- journal article
- Published by American Society for Clinical Investigation in JCI Insight
- Vol. 11 (19)
Abstract
Metastatic prostate cancer is a clinically and molecularly heterogeneous disease. Under the selective pressure of androgen receptor–directed (AR-directed) therapies, resistant phenotypes frequently emerge, posing significant diagnostic and therapeutic challenges. Neuroendocrine prostate cancer (NEPC) is a clinically important phenotype characterized by lineage plasticity, neuroendocrine features, visceral metastases, and poor prognosis. Accurate diagnosis of NEPC remains difficult owing to its histologic and molecular complexity but has high clinical relevance. In this study, we developed a deep learning model that leverages interpretable cellular features to improve feature extraction from H&E-stained tissue sections (NEURAL-PC). By incorporating a multiple-instance learning framework, NEURAL-PC enables robust NEPC classification solely from H&E tumor images, achieving an area under the receiver operating characteristic curve of 0.921 in independent external validation. In addition to its diagnostic utility, NEURAL-PC provides prognostic information that enables further subclassification of advanced prostate cancer across diverse datasets, supporting its strong prognostic value and generalizability. Broadly, our work highlights a hybrid approach that integrates features across different domains, offering a promising strategy for developing reliable deep learning tools in pathology. Built on this framework, NEURAL-PC represents an extensively validated diagnostic and prognostic model for advanced prostate cancer.Keywords
Funding Information
- National Institutes of Health (1P01CA298991-01,R37CA286450,P30CA15704,P50CA097186,P01CA163227-06A1,R01CA234715-03,R01CA266452,R01CA280056,R50CA221836,R50CA274336,PO1CA163227,CA282978,S10OD028685)
- DOD Prostate Cancer Research Program (W81XWH-20-1-0111,W81XWH-21-1-0229,W81XWH-22-1-0278,W81XWH-18-1-0347,W81XWH-18-1-0689,W81XWH-21-1-0264,W81XWH-14-2-0183,W81XWH-17-2-0043,PC230420,PC230582)
- Doris Duke Charitable Foundation (2021184)
- V Foundation for Cancer Research
- Prostate Cancer Foundation (Felix Feng PC-SYNERGY award)
- Safeway Foundation
- Richard M. LUCAS Foundation
- Brotman Baty Institute for Precision Medicine
- UW/FHCC Institute for Prostate Cancer Research
- Pacific Northwest Prostate Cancer SPORE (CA097186)
- National Institutes of Health (ZIABC012163)
This publication has 65 references indexed in Scilit:
- Prostate CancerJAMA, 2025
- Annual Report to the Nation on the Status of Cancer, featuring state‐level statistics after the onset of the COVID‐19 pandemicCancer, 2025
- Characterization of HOXB13 expression patterns in localized and metastatic castration‐resistant prostate cancerThe Journal of Pathology, 2023
- Clinical, pathologic, and molecular features of amphicrine prostate cancerThe Prostate, 2023
- Combined TP53 and RB1 Loss Promotes Prostate Cancer Resistance to a Spectrum of Therapeutics and Confers Vulnerability to Replication StressCell Reports, 2020
- Clinical features of neuroendocrine prostate cancerEuropean Journal of Cancer, 2019
- Unsupervised pathology detection in medical images using conditional variational autoencodersInternational Journal of Computer Assisted Radiology and Surgery, 2018
- Classification of lung adenocarcinoma transcriptome subtypes from pathological images using deep convolutional networksInternational Journal of Computer Assisted Radiology and Surgery, 2018
- Androgen Receptor Pathway-Independent Prostate Cancer Is Sustained through FGF SignalingCancer Cell, 2017
- Characterizing the molecular features of ERG-positive tumors in primary and castration resistant prostate cancerThe Prostate, 2016