Although numerous attempts have been made to forecast outcomes for prostate cancer after therapy using clinical and histological variables, the ability to accurately predict an individual's response to a specific treatment remains elusive. Recently, major advances in the field of genomics have made possible the near-comprehensive assessment of the genetic status of tumor genomes, with major concentration on predicting an individual's response to a specific treatment. Genomic approaches to treatment response include, but are not limited to, detection of gene rearrangements, DNA copy-number aberrations, single-nucleotide polymorphisms, epigenetic changes and differential gene-expression patterns. These approaches have been used to predict response to treatment for local and systemic disease in multiple small cohorts. Further study with larger cohorts and longer follow-up should result in more concordance among genomic approaches, and will enable physicians to gain insight into the heterogeneity of supposedly 'similar' cancers and help tailor treatments accordingly.