ABSTRACT Prostate cancer is a curable disease when treated early before metastases occur. Current diagnostic methods struggle to accurately predict which patients will develop metastases, resulting in the development of advanced, incurable disease in some patients who initially present with localized disease. We propose to tailor the development of our 3D spatial biology platform towards predicting which prostate cancer patients will develop metastases. Our 3D spatial biology platform is the basis for multiple prominent publications, has an international user base, and has demonstrated compatibility with current laboratory workflows during our Phase I award studies. The specific aims of this proposal are: 1) Develop an accelerated and automated sample preparation process, 2) Develop machine learning models to segment tissue structures and extract 3D features for use in predictive models, and 3) Create a multiparameter model to predict metastatic risk from prostate biopsy samples. The studies in this proposal will accelerate the workflow to a clinically feasible turnaround time, build a predictive model, and demonstrate the benefits of our technology in a defined prostate cancer patient subpopulation. Our partners in this proposal (CorePlus) can rapidly deploy the developed technologies in a clinical laboratory with regulatory accreditation and an existing referral base. Successful completion of this project will position Alpenglow Biosciences towards an advanced stage in commercialization with high potential for success. This project addresses the IMAT goal of developing substantially improved cancer detection and risk assessment technologies.