Colorectal cancer is the second most common malignancy in the U.S. In the past 15 years we witnessed the development of new anticancer agents for patients with colorectal cancer. A multitude of therapeutic options creates a challenge for a physician to select the most efficacious and least toxic agent for an individual patient. To this end, pharmacogenomic variability in metabolic pathways for fluorouracil, capecitabine, irinotecan and oxaliplatin has been characterized. Pharmacogenomic investigations of dihydropyrimidine dehydrogenase, thymidylate synthase and thymidine phosphorylase identified a panel of clinically relevant polymorphisms for fluoropyrimidine drugs. Polymorphisms in nucleotide excision repair genes such as ERCC1 can modulate oxaliplatin exposure. For irinotecan, a polymorphism of the uridine diphosphate glucuronosyltransferase increases the risk for neutropenia. The pharmacogenomic test for this variation (UGT1A1*28) is approved by the FDA for patients who are candidates for irinotecan therapy. Disposition and activity of targeted agents, such as therapeutic antibodies bevacizumab and cetuximab, are influenced by additional genomic variations. Deficient nucleotide mismatch repair system, a hallmark of hereditary colon cancer syndrome (identified as high frequency microsatellite instability), emerged as a predictive marker for 5-fluorouracil in the adjuvant setting and the KRAS mutation as a determinant of sensitivity to anti-EGFR antibodies. Collectively, these advances establish a firm basis towards future development of an integrated pharmacogenomics array for prime time applications in the clinic to customize pharmacotherapy for colorectal cancer. We also emphasize that individualizing anticancer chemotherapy is an arduous task because drug efficacy and toxicity represent multifactorial complex traits mediated by multiple DNA sequence variants, temporal variations in gene expression, environmental factors and gene-environment interactions.