A Pilot Study on Patient-specific Computational Forecasting of Prostate Cancer Growth during Active Surveillance Using an Imaging-informed Biomechanistic Model

• DOI: 10.1158/2767-9764.crc-23-0449
• Publication Date: 2024-03-01T14:11:26Z
• Authors: Guillermo Lorenzo; Jon S. Heiselman; Michael A. Liss; Michael I. Miga; Hector Gomez; Thomas E. Yankeelov; Alessandro Reali; Thomas J.R. Hughes
• Keywords: Concordance

Abstract Active surveillance (AS) is a suitable management option for newly diagnosed prostate cancer, which usually presents low to intermediate clinical risk. Patients enrolled in AS have their tumor monitored via longitudinal multiparametric MRI (mpMRI), PSA tests, and biopsies. Hence, treatment prescribed when these tests identify progression higher-risk cancer. However, current protocols rely on detecting through direct observation according population-based monitoring strategies. This approach limits the design of patient-specific plans may delay detection progression. Here, we present pilot study address issues by leveraging personalized computational predictions cancer growth. Our forecasts are obtained with spatiotemporal biomechanistic model informed mpMRI data (T2-weighted apparent diffusion coefficient maps from diffusion-weighted MRI). results show that our technology can represent forecast global burden individual patients, achieving concordance correlation coefficients 0.93 0.99 across cohort (n = 7). In addition, model-based biomarker cancer: mean proliferation activity (P 0.041). Using logistic regression, construct risk classifier based this achieves an area under ROC curve 0.83. We further coupling enables early identification disease more than 1 year. Thus, posit predictive constitutes promising decision-making tool patients Significance: Personalization prediction progression, thereby showing promise guide during each patient.