Genetic polymorphisms and variable expression of drug receptors, metabolizing enzymes, and transporters have been linked to interindividual differences in efficacy and toxicity of many Food and Drug Administration-approved therapeutic agents. In multiple sclerosis, the combination of heterogeneity of disease pathology and significant variation in clinical response to disease-modifying agents necessitates the definition of biomarkers that can a priori predict therapeutic response and define appropriate therapeutic regimens. Pharmacogenomic studies will directly address the question of heterogeneity by analysis of the correlation between different genomic variants and clinical responses to therapy. These studies will include longitudinal designs, maximize clinical response variables, include whole-genome technologies, use large patient cohorts, and require the development of novel mathematical algorithms designed to integrate the wealth of disparate data to identify modest genetic effects and interactions.