Predictive modelers must work together to create a more efficient drug pipeline.
By Stacey Tannenbaum, Astellas; Matthew Riggs, Metrum Research Group; and Marc Horner, ANSYS, Inc.
The pharmaceutical industry is rapidly increasing its reliance on predictive modeling to support study data and other evidence in the identification, development, and manufacturing of new drugs (Figure 1). These models, like the quintessentials of physics (F=ma) and chemistry (pV=nRT), describe physical events through mathematical principles. Predictive modeling thereby allows data and information to be synthesized and integrated to reduce uncertainty, mitigate risk, and support quantitative decisions, such as optimizing compound properties for rational formulation development, increasing manufacturing process efficiency, scaling preclinical and in vitro data to humans, and selecting doses in special populations. Modeling is also indispensable to the regulatory review process.