Predicting biological effects of drugs, chemicals and therapeutic interventions is fraught with hazard. Approximately 80 percent of candidate drugs fail in human clinical trials because they are found to be unsafe or ineffective. More than 30 percent of promising medications have failed in clinical trials because they are found to be harmful to human health (i.e., they have high toxicity), despite costly pre-clinical studies in animal and cell models. Because these models often do not adequately represent human biology, they do not always accurately reflect how patients will react to an experimental compound.
A major area of emphasis at NCATS is the development of model systems for drug and toxicity testing that more closely resemble human physiology. Such advances could save enormous amounts of time and expense by preventing patients from being exposed to potentially harmful or ineffective candidate drugs in clinical studies. In addition, these models have the potential to provide useful information about the basic biology of disease and serve as improved testing platforms for predicting toxicity or other physiological processes as well as evaluating environmental chemicals.