Pharmacometricists have increasingly drawn attention to nonlinear mixed effects modeling methods as a more efficient approach to the design and conduct of clinical trials in oncology. The crudeness of current widely used methods of disease response assessment is well recognized in the oncology community. However, there is significant conflict and controversy on how best to improve treatment response evaluation and thereby the success rate of new therapeutics development. The potential benefits of quantitative methods are apparent to most statisticians. But the necessary change in clinical trial operations and data collection to implement quantitative methods faces significant obstacles. These include: 1)the absence of examples of successful implementation and impact, 2)a focus in the statistical and regulatory community on outcomes closely associated with survival benefit such as progression-free survival or endpoints based on underdeveloped technology, 3) a hyperattention to new technologies, especially molecular markers and functional imaging, and 4)typical resistance to change among staff, investigators, sponsors, and regulators. To overcome the first obstacle, our group has modeled available data and worked to analyze previously collected data with frustrating results. The details will be reviewed and put into perspective to inspire greater contributions from the statistical and regulatory community in effecting the change necessary to improve cancer therapeutics development.