Stratifying Underlying Risk in the Prostate cancer Active Surveillance Study (SURPASS)

/ABSTRACT Each year, over 100,000 patients in the US are diagnosed with favorable-risk prostate cancer for which clinical guidelines recommend active surveillance. Active surveillance (AS) involves cancer monitoring with clinical tests such as serum prostate specific antigen (PSA), prostate biopsies, and magnetic resonance imaging (MRI), such that definitive treatment (e.g. surgery or radiation) is delayed until signs of unfavorable cancer are evident, or treatment is avoided altogether. While the majority of men on AS will remain free of progression for more than 10 years, about 20% of cancers that appear low risk at diagnosis progress to aggressive phenotypes (grade group >3 cancer, recurrence after treatment, or metastasis). Early identification of these aggressive cancers will allow for earlier treatment, at a time when treatment may be curative. Although several biomarker tests have been developed and validated to improve prediction of higher risk disease after primary treatment, there are no biomarker tests that have been validated for clinical use in the large population of men on AS. The Canary Prostate Active Surveillance Study (PASS), established in 2008, is the only prospectively accrued, multi-institutional cohort of men on AS in North America. PASS has 2,300 participants with a median follow-up of 8 years. Extensive clinical, imaging (e.g. MRI), epidemiological data and longitudinal biospecimens are collected according to a standardized protocol. In this proposal, we will use the Canary PASS cohort to interrogate novel radiomic, pathomic, and liquid biomarkers for early identification of cancers that progress to aggressive disease. The overall goal of the proposed research is to develop a predictive tool that will accurately identify patients initially diagnosed with favorable-risk cancer that will progress to aggressive cancer better than routine clinical, pathologic, and imaging data. We will accomplish this goal using the following Aims: 1) Interrogate PSA isoform biomarkers for their ability to predict aggressive disease in patients using AS. We will leverage data from existing and emerging collaborations to evaluate if isoforms of PSA improve prediction of aggressive disease compared to the clinical standard of total PSA. 2) Interrogate MR imaging biomarkers for their ability to predict aggressive disease in patients using AS. We will A) evaluate whether computer-aided diagnostic (CAD) in a commercial, FDA cleared artificial intelligence (AI)-derived MRI platform detects aggressive cancer, B) develop and validate an AI-based radiomic biomarker for prediction of aggressive disease in patients on AS. 3) Interrogate histopathology signatures for improved prediction of aggressive disease in patients using AS. We will A) evaluate outputs from a commercial AI histopathology tool for prediction of aggressive cancer, B) develop and validate an AI-derived histopathology signature trained for prediction of aggressive disease in patients on AS. Secondary Aim: Develop a decision rule for incorporating multiple biomarkers and clinical data for optimal prediction of aggressive prostate cancer in patients using AS. Project Number: 1R01CA300695-01A1 | Fiscal Year: 2026 | NIH Institute/Center: National Cancer Institute (NCI) | Principal Investigator: DANIEL LIN | Institution: UNIVERSITY OF WASHINGTON, SEATTLE, WA | Award Amount: $688,959 | Activity Code: R01 | Study Section: Cancer and Hematologic Disorders Study Section[CHD] View on NIH RePORTER: https://reporter.nih.gov/project-details/11295658