Genomics - having quickly emerged as the central discipline in basic science and biomedical research - is poised to take the center stage in clinical medicine as well over the next few decades. Although there is no specific regulatory guideline on the application of pharmacogenetics to drug development, some recommendations are already included in several published guidelines on drug development. The patients more likely to provide the most valuable information on the specific contribution of a given gene or its variant are those who fail to respond to a drug ('therapeutic failures') and those who develop toxicity to the drug. However, before drawing definite conclusions on subgroups following pharmacogenomic analyses, one must be aware of disease classification, data collection, and how much is known about the disease process. It seems reasonable to collect genomic DNA from all patients enrolled in clinical drug trials (along with appropriate consent to permit pharmacogenetic studies) for the purpose of post hoc analyses. One exception to post hoc genomic analysis is when patients with a specific genotype are excluded from randomization into a clinical trial. Physicians will need to understand the concept of genetic variability, its interactions with the environment (e.g. drug-drug or drug-disease interactions), and its implication for patient care.