Prostate cancer is the most common non-cutaneous cancer and exhibits protean clinical manifestations from indolent and clinically insignificant to aggressive and lethal. I use genomic technologies and defined genetic and/or chemical perturbations to investigate prostate cancer biology and anticipate response to therapy using established prostate cancer cell lines, genetically-defined cell lines, prostate cancer mouse models, and clinical prostate cancer specimens. Some examples of discoveries from my work include being the first to publish a multi-gene expression-based predictor of prostate cancer recurrence after surgery, discovering decreased Hif pathway activity following mTOR inhibition, and identifying mechanisms of resistance associated with androgen ablation, docetaxel, imatinib, and mTOR inhibition. A unique aspect of my work has been a consistent focus on the incorporation of expression analysis into clinical trials. I have worked to create “learning loops” for prostate cancer clinical trials where the analysis of samples from one trial informs the design and implementation of subsequent clinical trials. Most recently, I am prospectively validating predictive, expression-based signatures for docetaxel, mTOR inhibition, and androgen receptor (AR) activity in metastatic prostate cancer. Our trial using a microarray-based signature of AR activity to determine therapy for men with metastatic prostate cancer will be the first trial to personalized care for men with prostate cancer based upon a molecular phenotype of their specific tumor and is emblematic of how we will improve and personalize care for the men diagnosed with this disease.