Initial Development of the Risk Adapted Active Surveillance Monitoring in MUSIC Stratification Scheme: RETRO MUSIC

Document Type

Conference Proceeding

Publication Date

5-2025

Publication Title

Journal of Urology

Abstract

INTRODUCTION AND OBJECTIVE: Few risk stratification schemes have been developed to tailor the intensity of active surveillance (AS) monitoring to promote resource utilization stewardship and minimize the burden of surveillance testing. We developed a novel classification system to stratify patients based on their risk of upgrading to ≥GG3 disease utilizing factors present at enrollment into AS. METHODS: We identified men undergoing AS in the Michigan Urological Surgery Improvement Collaborative (MUSIC) from 2012 to 2024. The cohort was randomly split 80%/20% into development and validation cohorts. An adaptive elastic net Cox regression model was fit to the development cohort to identify factors associated with upgrading to GG3 disease. The model included biopsy GG (1, 2), number of positive cores (≤2, 3-5, ≥6), PSA density (< 0.15, ≥0.15), pre-biopsy MRI (none, PIRADS ≤3, ≥4), post-biopsy MRI (none, PIRADS ≤3, ≥4), and genomic classifier (none, reassuring, non-reassuring). The model was used to then create two RETRO MUSIC levels: normal and elevated. RESULTS: We identified 6,909 patients that underwent at least one biopsy on AS. The median age at the start of AS was 66, 80% of men had GG1 disease, and PSA was < 10 ng/ml in 92% of patients. Factors associated with an increased hazard of reclassification from the multivariable model were: 1) GG2, 2) PIRADS ≥4 lesion on pre-biopsy MRI or 3) up to 6 months post-biopsy, 4) PSA density ≥0.15, and 5) a non-reassuring genomic classifier. Notably, number of positive cores (≤2, 3-5, and ≥6) was not associated with an increased risk of reclassification. Patients were classified as RETRO MUSIC normal risk if none of the 5 concerning features were present, and as RETRO MUSIC elevated risk if any of these features were present. In the validation cohort, the 3-year risk of upgrading to GG3 in RETRO MUSIC normal and elevated risk groups was 10% and 24%, respectively (Figure 1, p< 0.001). CONCLUSIONS: We identified an easy to implement, two tiered classification scheme to identify which patients are at an elevated risk for reclassification to ≥GG3 disease. These data may help inform efforts to de-intensify surveillance testing in patients at low risk for reclassification; patients at high risk for reclassification should undergo more intensive monitoring on AS. These initial criteria and recommended follow-up strategies will be further refined for optimal clinical implementation.

Volume

213

Issue

5S

First Page

e347

Last Page

e348

Comments

American Urological Association Annual Meeting, April 26-29, 2025, Las Vegas, NV

DOI

10.1097/01.JU.0001109828.87037.24.09

Link to Full Text

Share

COinS