The issue of the safety of drug metabolites in humans is a complex one. In this commentary, a proposal is made regarding how to deal with drug metabolites observed in humans such that the safety of these molecules can be assured. The human radiolabeled ADME study, in which metabolites are identified and quantified in circulation and excreta, is proposed as the primary source of information on human metabolites from which decisions can be made regarding the need for further risk assessment. Although radiolabel ADME studies yield quantitative metabolite profiles that are commonly reported as a percentage of the total drug related material (for circulating metabolites) and a percentage of total dose (for excretory metabolites), it is essential to convert these values into absolute abundances. The structure of a metabolite, its abundance, the biofluid in which it is observed (circulation or excreta), and the toxicity mechanism of concern serve as the four most important characteristics for determination as to whether further safety consideration is warranted. Metabolites in circulation require consideration for toxicity that can arise by effects on specific receptors and/or enzymes (either target or off-target). Metabolites in excreta require consideration for their potential to indicate a body-burden to chemically reactive intermediary metabolites, which can yield toxicities of nonspecific mechanisms commonly associated with covalent binding (e.g., carcinogenicity, immunoallergic response, etc.). Through an analysis of 24 drugs removed from the market because of human toxicity, it was concluded that further testing of human metabolites would not have yielded any additional information that could have predicted human safety findings because human metabolites would have been present in the animal species routinely used in toxicology testing after the administration of the parent compound.