Prostate cancer is the most common urologic neoplasm and the second leading cause of cancer-related death among men in many developed countries. Given the highly heterogeneous behaviour of the disease, there is a great need for prognostic factors, in order to stratify the clinical risk and give the best treatment options to the patient. Clinical factors, such as prostate-specific antigen value and derivatives, and pathological factors, such as stage and Gleason grading, are well kown prognostic factors. Nomograms can provide useful prediction in each clinical sceario. The field of molecular biomarkers is briskly evolving towards personalized medicine. TMPRSS2-ERG fusion, deletion of PTEN ed and gene panels are some of the more extensively explored molecular features in prostate cancer outcome prediction. In the near future, circulating tumour cells, exosomes and microRNAs could give us further, not invasive important tools.