Prostate cancer (PCa) is the most common non-skin cancer type among men and one of the leading causes of cancer deaths worldwide [1]. The current diagnostic options of biopsy in combination with PSA-based tests are not conclusive to the aggressiveness of the disease. Novel diagnostic tests for a reliable non-invasive identification of aggressive PCa with high sensitivity and specificity are urgently needed and would be a great advance in clinical routine. At the moment, no single biomarker could be identified to be specific to prostate cancer. In this project, we attempt the identification and validation of urinary metabolite biomarkers and biomarker networks, i.e. high-dimensional classifiers, associated with aggressive prostate cancer.