Active surveillance (AS) allows selected men with localized prostate cancer to defer curative therapy and reduce treatment morbidity. Conversion from AS to treatment is commonly triggered by Gleason grade group (GGG) upgrading on confirmatory biopsy. We developed and validated a digital pathology artificial intelligence (DPAI)-derived risk score to predict GGG upgrading in AS-eligible patients.
The DPAI model was trained using histopathology image features from diagnostic biopsies of 998 patients and validated in an independent cohort of 296 patients meeting criteria for AS. Logistic regression estimated the probability of confirmatory-biopsy GGG increase, and feature selection identified the most predictive variables.
AI-GUR (Artificial Intelligence-Gleason Upgrade Risk) predicted GGG reclassification at confirmatory biopsy (OR 1.60; p = 0.0003) and provided information beyond conventional stratification (risk group, CAPRA) and cribriform morphology (all p < 0.01). Predicted risks were similar across time from diagnosis (~10-15% to ~85% at 1, 1.5, or 2 years; p for time = 0.50), consistent with initial biopsy mischaracterization rather than time-dependent progression.
AI-GUR provides individualized estimates of confirmatory-biopsy GGG upgrading for AS candidates. Using DPAI may improve shared decision-making by complementing standard clinicopathologic tools and molecular testing using the same biopsy specimen, while informing the likelihood of grade upgrade at confirmation.
This study focuses on improving care for men with early-stage prostate cancer who are eligible for active surveillance. Active surveillance is a management approach where cancer is closely monitored instead of treated right away, helping patients avoid or delay side effects from surgery or radiation. However, doctors often recommend follow-up biopsies because the initial biopsy may not fully capture how aggressive the cancer is and the disease may progress after initial diagnosis. The researchers developed a new artificial intelligence (AI) tool called AI-GUR (Artificial Intelligence–Gleason Upgrade Risk) to help predict whether a patient’s cancer may appear more aggressive on a follow-up biopsy. The tool uses digital images of biopsy tissue and analyzes patterns that may not be easily seen by the human eye as well as those patterns commonly used by pathologists to categorize cancer aggressiveness. The AI model was trained using data from 998 patients and tested in a separate group of 296 patients. Results showed that AI-GUR predicted which patients were more likely to have a higher cancer grade on first follow-up biopsy. This tool provided additional useful information beyond standard clinical measures, such as risk scores and common pathology features, and did not change based on how much time passed between biopsies. Overall, AI-GUR may help doctors and patients make more informed decisions related to personalizing their AS protocol. By providing personalized risk estimates, this approach could reduce uncertainty and improve confidence in care decisions.
Future oncology (London, England). 2026 Jun 16 [Epub ahead of print]
Brent Mabey, Lauren H Lenz, Matthew J Schiewer, Walter Rayford, Hassan Muhammad, Wei Huang, Robert Finch, Christina Nakamoto, Hosein Kouros-Mehr, Jeff Jasper, Hirak Basu, Chao Feng, Anurag Sharma, George Wilding, Rajat Roy, Dale Muzzey, Alexander Gutin
Myriad Genetics, Inc., Salt Lake City, UT, USA., Biostatistics, Myriad Genetics, Inc., Salt Lake City, UT, USA., Clinical Research, Myriad Genetics, Inc., Salt Lake City, UT, USA., The Urology Group, Memphis, TN, USA., PATHOMIQ, Inc., Cupertino, CA, USA., Medical Affairs - Germline Oncology, Myriad Genetics, Inc., Salt Lake City, UT, USA.