Development and applications of a material library for pharmaceutical continuous manufacturing of solid dosage forms

The purpose of this study is to establish a material library and discuss its potential application to the development and lifecycle management of a continuous manufacturing process for solid dosage forms. Particularly, this study addresses the importance of selecting process-relevant testing conditions for material characterization, proposes a methodology to capture relevant information with a reduced set of measurements, and correlates material properties with process performance. This study included 20 pharmaceutical materials, and each material was characterized by 44 properties, capturing 880 data points. The stress conditions of a commonly used feeder hopper were calculated using the Janssen model for six selected materials. Multivariate analysis, such as principal component analysis and clustering analysis, was used to explore the knowledge space of the material library. Statistically similar and dissimilar material properties were evaluated for material feeding performance from a loss-in-weight feeder to test utility of the material library. 20 materials included in this study show a wide range of material properties. Consolidation stress during testing significantly impacts obtained material properties. Based on a material similarly metric, a reduced set of characterization tests that captures > 95% of the relevant information was identified. Materials were then grouped into six clusters. The material loss-in-weight feeding results show that the materials within the same cluster show similar feeding performance, while selected materials from different cluster have different feeding performance. Additional material understanding regarding flow properties may be needed to implement a continuous manufacturing process. Characterization using multiple tests under process-relevant conditions can be helpful to establish the correlation between material properties and process and product performance using multivariate analysis tools.