Prostate cancer is a major health problem with an incompletely understood pathogenesis and etiology. The advent of the prostate-specific antigen (PSA) test in the 1980s caused a revolution in how the disease was detected, but the evidence for PSA as a screening test is deficient. Biomarkers have been investigated, both for detection and discrimination of indolent from aggressive cancers. Refinements to the PSA test have been proposed but for practical and evidence-based reasons none have translated through to widespread clinical use. Of the novel biomarkers, the most promising is the prostate cancer antigen 3 (PCA3) test. New biomarkers to predict aggressive disease are even more contentious. The pathological grade of tumor remains the most powerful biomarker of prognosis. Other proven variables include tumor extent on biopsy and serum PSA. Tissue biomarkers have proven unhelpful due to a variable and biased literature with multiple methodological flaws, but Ki-67 probably shows more promise than any other current tissue biomarker. The recent discovery of a family of fusion genes in the prostate has led to considerable discussion on their prognostic role. Dissection of the genetic basis of the disease may lead to discoveries that will enhance our understanding and aid the search for prognostically valuable biomarkers.