(639g) Characterization and Modeling of Controlled Release From Silk Fibroin Based Drug Delivery Devices

The overall goal of the research project is to acquire a predictive model set for the controlled release of therapeutic agents from silk fibroin based drug delivery devices. Such a predictive model set would benefit the decision making process for drug delivery device formulation. When a new drug candidate for controlled release is presented, such a predictive model set would allow one to accurately approximate the in vitro and in vivo release profile from a silk based delivery device through simulation. This would greatly reduce the number of experiments required to develop delivery systems for novel drug candidates in controlled release. Model simulation of a robust predictive model will help one to consider how material properties and formulation steps may be manipulated as design variables to achieve a target release profile. A short list of material properties and design variables includes: mass loading percent, organic solvent treatment, device morphology/geometry, crystallinity, polymer molecular weight and molecular weight distribution, addition of excipients, porosity, and water solubility. For the majority of delivery systems optimal controlled release is slow, zero order, with good device scalability. In order to develop this predictive model set the mechanisms of controlled release and their kinetics must be identified and characterized through experimentation and mathematical modeling. Following this work the approach will focus on the establishment of mathematical relationships between the kinetic parameters of relevant release mechanisms and significant physiochemical properties of the chemical species being released from the device. The goal will be addressed in four steps: First, the mechanism of controlled release and relationships between the release kinetics and material properties must be identified through controlled released experimentation; second, confirm and quantify the kinetics with appropriate model selection and fitting; third, derive predictive models that define relationships between material properties and the kinetic parameters of relevant models used to approximate the kinetic parameters of new drugs; and fourth, test the predictive power of the developed model set through cross validation and simulation. The developed model set may be continually updated for material properties ? release kinetics relationships for more complex systems to expand the predictive capabilities of the model set.