MRI has a wide variety of applications in the clinical trials process. MR has shown particular utility in the early phases of clinical development, when trial sponsors are interested in demonstrating proof of concept and must make decisions about allocation of resources to a particular compound based on the results from a small number of experimental subjects. This utility is largely due to the many different imaging endpoints that can be measured using MR, ranging from structural (tumor burden, hippocampal volume) to functional (blood flow, vascular permeability) to molecular (hepatic fat fraction, glycosaminoglycan content). The unique flexibility of these systems has proven to be both a blessing and a curse to those attempting to deploy MR in multi-center clinical trials, however, as differences among scanner manufacturers and models in pulse sequence implementation, hardware capabilities, and even terminology make it increasingly difficult to ensure that results obtained at one center are comparable to those at another. These problems are compounded by the differences between the procedures used in clinical trials and those used in routine clinical practice, which make trial-specific training for site technologists and radiologists a necessity in many cases. This article will briefly review the benefits of including quantitative MR imaging in clinical trials, then explore in detail the challenges presented by the need to develop and deploy a detailed MR protocol that is both effective and implementable across many different MR systems and software versions.