External validation of a digital pathology-based multimodal artificial intelligence-derived prognostic model in patients with advanced prostate cancer starting long-term androgen deprivation therapy: a post-hoc ancillary biomarker study of four phase 3 randomised controlled trials of the STAMPEDE platform protocol
- PMID: 40467357
- DOI: 10.1016/j.landig.2025.100885
External validation of a digital pathology-based multimodal artificial intelligence-derived prognostic model in patients with advanced prostate cancer starting long-term androgen deprivation therapy: a post-hoc ancillary biomarker study of four phase 3 randomised controlled trials of the STAMPEDE platform protocol
Abstract
Background: Effective prognostication improves selection of patients with prostate cancer for treatment combinations. We aimed to evaluate whether a previously developed multimodal artificial intelligence (MMAI) algorithm was prognostic in very advanced prostate cancer using data from four phase 3 trials of the STAMPEDE platform protocol.
Methods: We included patients starting androgen-deprivation therapy in the docetaxel, docetaxel plus zoledronic acid, abiraterone, or abiraterone plus enzalutamide trials. Patients were recruited at 112 sites. We combined all standard-of-care control patients (including those allocated to standard of care [SOC-ADT] consisting of testosterone suppression with luteinising hormone-releasing hormone agonists or antagonists, and radiotherapy when indicated), and we combined the rest of the patients into docetaxel-treated or abiraterone-treated groups. Patients had either metastatic disease or were at very high-risk of metastatic disease, determined by node-positivity or, if node-negative, by T stage, serum prostate-specific antigen (PSA) level, and Gleason score. We used the locked ArteraAI Prostate MMAI algorithm that combined these clinical variables, age, and digitised prostate biopsy pathology images. We performed Fine-Gray and Cox regression adjusted for treatment allocation and cumulative incidence analyses at 5 years to evaluate associations with prostate cancer-specific mortality (PCSM) for continuous (per SD increase) and categorical (quartile-Q) scores. The STAMPEDE platform protocol is registered with ClinicalTrials.gov, NCT00268476.
Findings: Of 5213 eligible patients recruited from Oct 5, 2005, to March 31, 2016, 3167 were included in this analysis (1575 [49·7%] with non-metastatic disease, 1592 [50·3%] with metastatic disease; median follow-up 6·9 years [IQR 5·9-8·0]) with all datapoints available for score generation. The MMAI algorithm (per SD increase) was strongly associated with PCSM (hazard ratio [HR] 1·40, 95% CI 1·30-1·51, p<0·0001). On ad-hoc inspection, the highest scoring quartile of patients in each disease and treatment allocation group (MMAI Q4; vs the bottom three quartiles, Q1-3) had the highest PCSM risk in both patients with non-metastatic disease (HR 2·12, 1·61-2·81, p<0·0001) and those with metastatic disease (HR 1·62, 1·39-1·88, p<0·0001). MMAI quartile stratification split patients categorised by disease burden into groups with notably different risks of 5-year PCSM: patients with non-metastatic disease that were node-negative could be further stratified by MMAI score quartile Q1-3 (3%, 2-4) versus Q4 (11%, 7-15), those with non-metastatic disease that were node-positive could be stratified by Q1-3 (11%, 8-14) versus Q4 (20%, 13-26), those with metastatic disease with low-volume could be stratified by Q1-3 (27%, 23-31) versus Q4 (43%, 36-51), and those with metastatic disease with high-volume could be stratified by Q1-3 (48%, 44-52) versus Q4 (68%, 62-75).
Interpretation: Diagnostic prostate biopsy samples contain prognostic information in patients with, or at high-risk of, radiologically overt metastatic prostate cancer. MMAI algorithm combined with disease burden improves prognostication of advanced prostate cancer.
Funding: Prostate Cancer UK, UK Medical Research Council, Cancer Research UK, John Black Charitable Foundation, Prostate Cancer Foundation, Sanofi Aventis, Janssen, Astellas, Novartis, Artera.
Copyright © 2025 The Author(s). Published by Elsevier Ltd.. All rights reserved.
Conflict of interest statement
Declaration of interests Artera could gain commercially from clinical implementation of this study’s results and UCL could receive a share of commercial revenue for its contribution to this study. CTAP, LMe, EG, MAP, LMu, SL, SSV, MG, ST, AW, LZ, DW, CLA, NBA, MRS, MKBP, LCB, and GA are employees of UCL. VYTL, SJ, RY, AM, EC, H-CH, JG, DvdW, TT, and AE are employees of Artera, and hold stock options in Artera, and FYF and AE are company founders. AE sits on the Artera board. JG also receives travel support from Artera. AE has patents pending for Artera intellectual property. AS reports having received grants from the Prostate Cancer Foundation, John Black Charitable Foundation, and Cancer Research UK; and personal fees or travel support from Veracyte, Ipsen, and AIRA Matrix, outside the submitted work. CCP reports advisory roles at Blue Earth Therapeutics and Janssen, outside the submitted work. DES reports having received personal fees from Boston Scientific; and having taken advisory board roles for Astellas, AstraZeneca, Bayer, GlaxoSmithKline, Janssen, Novartis, and Pfizer, and leadership roles for the National Comprehensive Cancer Network prostate cancer guidelines outside the submitted work. DMB reports having received research funding from the charity Orchid during the conduct of the study. FYF reports being a co-founder of, having stock options in, and being on the advisory board of Artera, during the conduct of the study; having received personal fees from Janssen Oncology, Astellas Pharma, Foundation Medicine, Exact Sciences, Bristol Myers Squibb, Varian Medical Systems, Novartis, Tempus, Amgen, Myovant Sciences, Roivant, Genentech, and Bayer; being a co-founder having stock options in PFS Genomics; and having stock options and serving on the scientific advisory board of Bluestar Genomics and SerImmune, all outside the submitted work. GA reports having received personal fees, grants, and travel support from Janssen, Novartis, and Astellas Pharma during the conduct of the study; personal fees or travel support from Pfizer, Bayer Healthcare Pharmaceuticals, AstraZeneca, Arvinas, Propella, Orion, and Sanofi-Aventis, outside the submitted work. In addition, GA’s former employer, The Institute of Cancer Research, receives royalty income from abiraterone and GA receives a share of this income through the Institute’s Rewards to Discoverers Scheme. GA’s current employer (UCL) receives clinical trial costs from Novartis, Janssen, Astellas, and Pfizer; receives licensing fees from Artera and Veracyte; receives equipment from Agilent; and holds the patent for “Blood-based methylation markers” (GB1915469.9). LMu reports having received grants from Cancer Research UK, and from Janssen, Astellas, Novartis, Sanofi, and Clovis as part of the wider STAMPEDE trial for the duration of this study. MKBP reports educational grants and non-financial support to the Clinical Trials Unit from Novartis, Pfizer, ROCHE, ROCHE Products, Sanofi, Serum Institute of India, Shionogi, SUMVAX, Synteny Biotechnology, Takeda, Tibotec, Transgene, ViiV Healthcare, Virco Xenothera, CSL Behring, Eli-Lilly, Emergent Biosolutions, Gilead Sciences, GlaxoSmithKline, Grifols, ICON, Janssen Products LP, Janssen-Cilag, Janssen Pharmaceutica, Johnson & Johnson, Merck Serono, Micronoma, Modus Theraputics, MSD, Mylan, Abcodia Pvt Ltd, Advanced Accelerator – Applications International SA, Akagera, Amgen, Aspirin Foundation, Astellas, AstraZeneca, AoA, Baxter, Bayer, Bristol Myers Squibb US, Brii Biosciences, B&C Group, Cepheid, Cipla, and Clovis, outside the submitted work. MRS reports having received grants and non-financial support from Sanofi-Aventis, Janssen, and Astellas, during the conduct of the study; personal fees from Eisai, Janssen, and Eli-Lilly; travel support from Health Research Board Ireland, Trials Research Methodology Network Ireland and National Cancer Grid India; and independent data monitoring committee or trial steering committee roles for Institute of Cancer Research, Imperial College London, Kings College London, University of Oxford, University of Leeds, and UCL (all UK) outside the submitted work. NDJ reports having received direct support from the Institute of Cancer Research during the conduct of the study; having received research grants and personal fees from AstraZeneca, Janssen, Astellas and Novartis; personal fees from Bayer, Clovis, Merck Sharp & Dohme, Sanofi; and advisory roles for AstraZeneca, Bayer, Clovis, Janssen, Merck, Novartis, Sanofi, and Gilead, and for AAA Accelerator Systems as part of the VISION trial Independent Data Monitoring Committee, outside the submitted work. NWC reports leadership roles for National Prostate Cancer Audit England outside the submitted work, and participation in the STAMPEDE trial management group and safety committee. PTT reports having received grants from the US National Institutes of Health and the US Department of Defense; personal fees and travel assistance from RefleXion Medical, Natsar Pharmaceuticals, Bayer Healthcare, Janssen, Lantheus, Pfizer, and Regeneron; and royalties from Nastar Pharmaceuticals for patent “Compounds and Methods of Use in Ablative Radiotherapy” (patent#: 9114158) outside of the submitted work; and holds leadership roles at NRG Oncology, the American Society for Radiation Oncology, and the American Association for Cancer Research. RJJ reports having received grants and personal fees from Astellas, Bayer, and Roche; grants from Clovis and Exelixis; personal fees from Janssen, Novartis, Pfizer, Merck Serono, MSD, Ipsen, and Bristol-Myers Squibb; travel assistance from Bayer and Janssen; and advisory board role for Roche; outside the submitted work. SG reports personal fees from Tolremo, Ipsen, and Avalere Health; and speakers fees and travel support from AstraZeneca, Bayer, Intellisphere LLC, Gilead, Silvio Grasso Consulting, WebMD-Medscape, Peer Voice, European Society for Medical Oncology, American Society of Clinical Oncology Genitourinary Cancers Symposium, Deutschsprachingen Europäischen Schule für Onkologie (DESO), AdMeTech Foundation, EPG Health, Meister Concept GmbH, and Swiss Group for Clinical Cancer Research (SAKK); advisory or leadership roles at Telixpharma, Novartis, Modra Pharmaceuticals, Orion, Bayer, AstraZeneca, Myriad Genetic, Amgen, MSD, Bristol-Myers Squibb, Daiichi Sankyo, Boehringer Ingelheim, Innomedica, Macrogenics, Astells, and Pfizer; and co-inventor on patent application (WO 2009138392 A1) for “a method for biomarker discovery” (granted in China, Europe, Japan, and USA) outside the submitted work. WC reports having received personal fees and travel support from Janssen, Bayer, Astellas, Ipsen, and AAA Novartis outside the submitted work. All other authors declare no competing interests.
Publication types
MeSH terms
Substances
Associated data
LinkOut - more resources
Full Text Sources
Medical
Research Materials
Miscellaneous
