Abstract 5905: A non-invasive, methylation-based NGS assay for sensitive detection and monitoring of bladder cancer in urine

Abstract Background: Bladder cancer, the sixth most common cancer worldwide, has a high recurrence rate, necessitating frequent and reliable disease monitoring. Current methods, such as cystoscopy, are invasive and can hinder patient compliance. However, detecting therapy response and disease progression on a molecular level still remains challenging. Urinary tumor DNA (utDNA) analysis offers a promising, non-invasive, and highly sensitive diagnostic tool for bladder cancer diagnosis and monitoring. By detecting utDNA methylation profiles, this analysis could enable personalized treatment decisions and improve clinical outcomes. Methods: We developed PredicineALERT, a methylation-based next-generation sequencing (NGS) assay designed for bladder cancer monitoring using urine. Methylation markers were identified through comparative analysis of profiles from bladder cancer patients and healthy donors. These markers were used to train a model capable of distinguishing bladder cancer from non-cancer samples. The assay's sensitivity and specificity were validated using serially diluted methylated DNA controls to determine its detection limit. To assess the performance of the PredicineALERT assay in detecting tumor fractions, serial urine samples were collected from muscle-invasive bladder cancer (MIBC) patients undergoing neoadjuvant therapies and analyzed using both PredicineALERT and a tumor-informed minimal residual disease (MRD) assay. Results: A panel of methylation markers was identified from urine samples, enabling the model to accurately differentiate bladder cancer patients from healthy individuals. PredicineALERT demonstrated the ability to detect tumor signals at levels as low as 0.01%, underscoring its potential for minimal residual disease detection and early recurrence monitoring. Comparison with the tumor-informed MRD assay revealed excellent concordance in tumor fraction detection, further highlighting the assay's robustness and reliability. Conclusion: PredicineALERT is a non-invasive and highly sensitive assay for bladder cancer monitoring. Its design facilitates early recurrence detection while enhancing patient compliance, presenting a promising advancement in bladder cancer management. Citation Format: Feng Xie, Hang Dong, Haoran Tang, Yue Zhang. A non-invasive, methylation-based NGS assay for sensitive detection and monitoring of bladder cancer in urine [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2025; Part 1 (Regular Abstracts); 2025 Apr 25-30; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2025;85(8_Suppl_1):Abstract nr 5905.