No abstract

No abstract

No abstract

Roller compaction (RC) is a continuous process for solid dosage form manufacturing within the pharmaceutical industry achieving similar goals as wet granulation while avoiding liquid exposure. From a quality by design perspective, the aim of the present study was to demonstrate the applicability of statistical design of experiments (DoE) and multivariate modeling principles to identify the Design Space of a roller compaction process using a predictive risk-based approach. For this purpose, a reduced central composite face-centered (CCF) design was used to evaluate the influence of roll compaction process variables (roll force, roll speed, gap width, and screen size) on the different intermediate and final products (ribbons, granules, and tablets) obtained after roll compaction, milling, and tableting. After developing a regression model for each response, optimal settings were found which comply with the response criteria. Finally, a predictive risk based approach using Monte Carlo simulation of the factor variability and its influence on the responses was applied which fulfill the criteria for the responses in a space where there is a low risk for failure. Responses were as follows: granule throughput, ribbon porosity, granules particle size, and tablets tensile strength. The multivariate method orthogonal partial least-squares (OPLS) was used to model product dependencies between process steps e.g. granule properties with tablet properties. Those results confirmed that the tensile strength reduction, known to affect plastic materials when roll compacted, was not prominent when using brittle materials. While direct compression qualities are frequently used for roll compacted drug products because of their excellent flowability and good compaction properties, this study confirmed earlier findings that granules from these qualities were more poor flowing than the corresponding powder blend.