MRI biomarkers to detect cribriform growth and support selection for active surveillance in prostate cancer

Abstract

MRI is increasingly used to characterize prostate cancer, but its ability to capture the biological features that drive disease aggressiveness remains incompletely understood. In this thesis, we demonstrated that MRI-based hypoxia is associated with cribriform growth, a histopathological pattern associated with poor prognosis, which exhibits a significantly higher levels of MRI-based hypoxia than other patterns of intermediate risk prostate patients. We further showed that a diffusion-weighted MRI feature (90th Percentile ADC) is associated with the presence of cribriform growth, suggesting that MRI can provide non-invasive information about underlying tumor architecture and biology. These findings contribute to the development of imaging biomarkers for improved prostate cancer risk stratification and treatment decision-making.