When conducting a study during an early phase of testing a nonvalidated procedure or intervention, such as a Phase I oncology study, investigators know little about the risks of participation and the probability of individual benefit is essentially nonexistent. The procedure or intervention must be considered nontherapeutic (administered solely for scientific purposes), and the risk-benefit assessment must be conducted accordingly. Investigators and IRB members must first ensure that all the risks associated with the procedure or intervention are minimized, and that the investigator has provided justification for the study of the procedure or intervention in humans by supplying sufficient preclinical data, and must then assess whether the risks are reasonable in relation to the knowledge to be gained. Once these criteria are met, subjects may assume significant research risks, as long as these risks have been properly disclosed during the informed-consent process. Risks are minimized in these studies by close monitoring of subjects during the course of the study and early identification of adverse effects, enrollment of the minimum number of subjects (sample size) to obtain the necessary pharmacokinetic and safety endpoints, and design of the subject inclusion and exclusion criteria so that individuals least susceptible to harm (to the degree possible) are included in the study.