(UroToday.com) The 2025 Society of Urologic Oncology (SUO) annual meeting held in Phoenix between December 2 and December 5, 2025, was host to the Poster Session. Dr. Roger Li presented poster #197: Development and Validation of a Computational Histology AI (CHAI)-based biomarker to Prognosticate High Grade Progression risk in Low-Grade Non-muscle Invasive Bladder Cancer.
Dr Li opened by noting that current guidelines allow low-grade NMIBC to be managed with active surveillance, intravesical chemotherapy, or BCG.1,2 A clearer understanding of an individual patient’s risk of grade or stage progression could meaningfully influence decisions to escalate or prolong intravesical therapy.
The Computational Histology AI (CHAI) platform has previously generated prognostic biomarkers in high-grade and intermediate-risk NMIBC. Building on that groundwork, the investigators applied CHAI to develop and validate a biomarker aimed at predicting progression risk specifically in low-grade NMIBC.3
For this analysis, a pathologist selected representative diagnostic slides from 425 cases of low-grade NMIBC treated between 2007 and 2024. These slides were digitized at 40x magnification and analyzed using the CHAI platform.3 In the development cohort of 123 cases, image-derived features most strongly associated with progression-free survival were used to create the LG-prog biomarker, which generates a continuous score and categorizes tumors into risk groups using a 15 percent threshold. The validation cohort included 302 cases, and the association between the LG-prog score and progression-free survival was evaluated using a Cox proportional hazards model.
Dr Li noted that tumors classified as CHAI LG-prog high risk demonstrated a significantly greater likelihood of progression than those labeled low risk. In the validation cohort, the high-risk group had an HR of 4.75 (95 percent CI 2.48 to 9.09, p<0.001). Importantly, this association remained robust even after adjusting for established IBCG intermediate-risk factors, including tumor size, multifocality, prior intravesical therapy failure, recurrence history, and receipt of first-line induction therapy, with the adjusted HR rising to 5.66 (95 percent CI 2.81 to 11.44, p<0.001).
Progression curves separated early and continued to diverge over time. By 12 months, progression occurred in 17% of high-risk patients compared with 4.6% of the low-risk group. This gap widened further at 36 months (50% versus 12%t) and again at 60 months (73% versus 15%). These findings highlight substantial and durable risk stratification achieved with the CHAI LG-prog biomarker in low-grade NMIBC.
Dr Li concluded the poster presentation with the following key points:
- Leveraging pre-treatment H&E slides, the CHAI LG-prog biomarker identifies a subset of low-grade NMIBC patients with a substantially higher likelihood of progression, with long-term risk surpassing 70% at 60 months, independent of conventional clinical and pathologic features.
- Individuals flagged as high risk by this model may benefit from earlier therapeutic intensification rather than standard surveillance-focused approaches.
- By distinguishing patients with truly elevated progression risk, the CHAI platform brings a precision-medicine framework to the management of low-grade NMIBC.
Presented by: Roger Li, MD, Genitourinary Medical Oncologist, Moffitt Cancer Center, Tampa, FL
Written by: Julian Chavarriaga, MD, Urologic Oncologist at Penn State Health, @chavarriagaj on Twitter during the 2025 Society of Urologic Oncology (SUO) annual meeting held in Phoenix, AZ, between the 2nd and 5th of December 2025.
References:- Holzbeierlein JM, Bixler BR, Buckley DI, Chang SS, Holmes R, James AC, Kirkby E, McKiernan JM, Schuckman AK. Diagnosis and Treatment of Non-Muscle Invasive Bladder Cancer: AUA/SUO Guideline: 2024 Amendment. J Urol. 2024 Apr;211(4):533-538. doi: 10.1097/JU.0000000000003846. Epub 2024 Jan 24. Erratum in: J Urol. 2024 Dec;212(6):936. doi: 10.1097/JU.0000000000004251. PMID: 38265030.
- NCCN Bladder Cancer Guidelines Version 2.2025.
- Lotan Y, Krishna V, Abuzeid WM, Launer B, Chang SS, Krishna V, Shingi S, Gordetsky JB, Gerald T, Woldu S, Shkolyar E, Hayne D, Redfern A, Spalding L, Stewart C, Eyzaguirre E, Imtiaz S, Narayan VM, Packiam VT, O'Donnell MA, Li R, Baekelandt L, Joniau S, Zuiverloon T, Fernandez MI, Schultz M, Hensley PJ, Allison D, Taylor JA, Hamza A, Kamat A, Nimgaonkar V, Sonawane S, Miller DL, Watson D, Vrabac D, Joshi A, Shah JB, Williams SB. Predicting Response to Intravesical Bacillus Calmette-Guérin in High-Risk Nonmuscle-Invasive Bladder Cancer Using an Artificial Intelligence-Powered Pathology Assay: Development and Validation in an International 12-Center Cohort. J Urol. 2025 Feb;213(2):192-204. doi: 10.1097/JU.0000000000004278. Epub 2024 Oct 9. PMID: 39383345.