Predicting response to intravesical BCG in high-risk non-muscle invasive bladder cancer using an artificial intelligence-powered pathology assay: development and validation in an international 12 center cohort

Yair Lotan; Viswesh Krishna; Waleed M Abuzeid; Bryn Launer; Sam S Chang; Vrishab Krishna; Siddhant Shingi; Jennifer B Gordetsky; Thomas Gerald; Solomon Woldu; Eugene Shkolyar; Dickon Hayne; Andrew Redfern; Lisa Spalding; Courtney Stewart; Eduardo Eyzaguirre; Shamsunnahar Imtiaz; Vikram M Narayan; Vignesh T Packiam; Michael A O'Donnell; Roger Li; Loic Baekelandt; Steven Joniau; Tahlita Zuiverloon; Mario I Fernandez; Marcela Schultz; Patrick J Hensley; Derek Allison; John A Taylor; Ameer Hamza; Ashish Kamat; Vivek Nimgaonkar; Snehal Sonawane; Daniel L Miller; Drew Watson; Damir Vrabac; Anirudh Joshi; Jay B Shah; Stephen B Williams

doi:10.1097/JU.0000000000004278

Back

Predicting response to intravesical BCG in high-risk non-muscle invasive bladder cancer using an artificial intelligence-powered pathology assay: development and validation in an international 12 center cohort

Journal article

Open access

Peer reviewed

Predicting response to intravesical BCG in high-risk non-muscle invasive bladder cancer using an artificial intelligence-powered pathology assay: development and validation in an international 12 center cohort

Yair Lotan, Viswesh Krishna, Waleed M Abuzeid, Bryn Launer, Sam S Chang, Vrishab Krishna, Siddhant Shingi, Jennifer B Gordetsky, Thomas Gerald, Solomon Woldu, …

The Journal of urology, Vol.212(1), pp.192-204

02/2025

DOI: 10.1097/JU.0000000000004278

PMCID: PMC12674634

PMID: 39383345

Files and links (1)

url

https://doi.org/10.1097/JU.0000000000004278View

Published (Version of record) Open Access

Abstract

There are few markers to identify those likely to recur or progress after treatment with intravesical BCG. We developed and validated artificial intelligence-based histologic assays that extract interpretable features from transurethral resection of bladder tumor digitized pathology images to predict risk of recurrence, progression, development of BCG unresponsive disease, and cystectomy. Pre-BCG resection-derived whole-slide images and clinical data were obtained for high-risk non-muscle invasive bladder cancer cases treated with BCG from 12 centers and were analyzed through a segmentation and feature extraction pipeline. Features associated with clinical outcomes were defined and tested on independent development and validation cohorts. Cases were classified into high or low risk for recurrence, progression, BCG unresponsive disease, and cystectomy. 944 cases (development:303, validation:641, median follow-up:36 months) representative of the intended use population were included (high-grade Ta:34.1%, high-grade T1:54.8%; carcinoma-in-situ only:11.1%, any carcinoma-in-situ:31.4%). In the validation cohort, "High recurrence risk" cases had inferior high-grade recurrence-free survival versus "Low recurrence risk" cases (HR 2.08, p<0.0001). "High progression risk" patients had poorer progression-free survival (HR 3.87, p<0.001) and higher risk of cystectomy (HR 3.35, p<0.001) than "Low progression risk". Cases harboring the BCG unresponsive disease signature had a shorter time to development of BCG unresponsive disease than cases without the signature (HR 2.31, p<0.0001). AI assays provided predictive information beyond clinicopathologic factors. We developed and validated AI-based histologic assays that identify high-risk non-muscle invasive bladder cancer cases at higher risk of recurrence, progression, BCG unresponsive disease, and cystectomy, potentially aiding clinical decision-making.

Bladder Cancer

Progression

Recurrence

Artificial intelligence

Details

Title: Subtitle: Predicting response to intravesical BCG in high-risk non-muscle invasive bladder cancer using an artificial intelligence-powered pathology assay: development and validation in an international 12 center cohort
Creators: Yair Lotan - The University of Texas Southwestern Medical Center
Viswesh Krishna - Valar Labs, Palo Alto, CA, USA
Waleed M Abuzeid - Valar Labs, Palo Alto, CA, USA
Bryn Launer - Vanderbilt University Medical Center
Sam S Chang - Vanderbilt University Medical Center
Vrishab Krishna - Valar Labs, Palo Alto, CA, USA
Siddhant Shingi - Valar Labs, Palo Alto, CA, USA
Jennifer B Gordetsky - Vanderbilt University Medical Center
Thomas Gerald - The University of Texas Southwestern Medical Center
Solomon Woldu - The University of Texas Southwestern Medical Center
Eugene Shkolyar - Stanford Medicine
Dickon Hayne - The University of Western Australia
Andrew Redfern - The University of Western Australia
Lisa Spalding - The University of Western Australia
Courtney Stewart - The University of Texas Medical Branch at Galveston
Eduardo Eyzaguirre - The University of Texas Medical Branch at Galveston
Shamsunnahar Imtiaz - Emory University School of Medicine
Vikram M Narayan - Emory University
Vignesh T Packiam - University of Iowa
Michael A O'Donnell - University of Iowa
Roger Li - Moffitt Cancer Center
Loic Baekelandt - Universitair Ziekenhuis Leuven
Steven Joniau - Universitair Ziekenhuis Leuven
Tahlita Zuiverloon - Erasmus MC
Mario I Fernandez - Universidad del Desarrollo
Marcela Schultz - Universidad del Desarrollo
Patrick J Hensley - University of Kentucky
Derek Allison - University of Kentucky
John A Taylor - University of Kansas Medical Center
Ameer Hamza - University of Kansas Medical Center
Ashish Kamat - The University of Texas MD Anderson Cancer Center
Vivek Nimgaonkar - Valar Labs, Palo Alto, CA, USA
Snehal Sonawane - Valar Labs, Palo Alto, CA, USA
Daniel L Miller - Valar Labs, Palo Alto, CA, USA
Drew Watson - Valar Labs, Palo Alto, CA, USA
Damir Vrabac - Valar Labs, Palo Alto, CA, USA
Anirudh Joshi - Valar Labs, Palo Alto, CA, USA
Jay B Shah - Stanford Medicine
Stephen B Williams - The University of Texas Medical Branch at Galveston
Resource Type: Journal article
Publication Details: The Journal of urology, Vol.212(1), pp.192-204
DOI: 10.1097/JU.0000000000004278
PMID: 39383345
PMCID: PMC12674634
NLM abbreviation: J Urol
ISSN: 1527-3792
eISSN: 1527-3792
Publisher: LIPPINCOTT WILLIAMS & WILKINS
Language: English
Electronic publication date: 10/09/2024
Date published: 02/2025
Academic Unit: Urology
Record Identifier: 9984722944302771

Metrics

10 Record Views

22 Times Cited - Web of Science