RESUMEN
In recent years, continuous manufacturing techniques, such as twin screw wet granulation, have gained significant momentum. Due to the large diversity of the (model) formulations and equipment, it is often difficult to generalize conclusions about the importance of process settings. As only limited knowledge is available on the importance of formulation variables, this study focused on the systematic quantification of both process as formulation effects on critical quality attributes of granules from several model formulations. Apart from conventional process and formulation variables, also non-conventional process factors such as nozzle diameter, nozzle orientation and inclusion of a new type of size control elements were evaluated using a Plackett-Burman screening design. Although effects were often formulation-dependent, liquid-to-solid ratio proved the most influential variable for all formulations. Furthermore, binder concentration had a clear effect on granule characteristics, whereas barrel fill level and barrel temperature were less influential. The novel type of size control elements improved granule size distribution and density. The impact of nozzle diameter and wet binder addition proved negligible towards granule properties. Overall it was apparent that lactose/MCC-based formulations correlated better than lactose-based formulations, indicating the possible process robustness of the first filler combination to accommodate API and excipient variability and to handle APIs with different characteristics.
Asunto(s)
Química Farmacéutica/métodos , Composición de Medicamentos/métodos , Tornillos Óseos , Excipientes/química , Lactosa/química , Tamaño de la Partícula , Solubilidad , Comprimidos/química , Temperatura , Resistencia a la TracciónRESUMEN
Due to the numerous advantages over batch manufacturing, continuous manufacturing techniques such as twin screw wet granulation are rapidly gaining importance in pharmaceutical production. Since a large knowledge gap on the importance of formulation variables exists, this study systematically assessed the impact of different screw configurations and process settings on eight model formulations, varying in filler type, active pharmaceutical ingredient (API) characteristics and drug load. Although liquid to solid (L/S) ratio was the most influential variable for all formulations, also a large effect of the kneading element thickness was observed. Narrow kneading elements with a length to diameter ratio (L/D) of 1/6 had a significant detrimental effect on granule size, flow and strength compared to 1/4 L/D elements. The effects of kneading element distribution and barrel fill level were less pronounced. At low drug load, both filler types could be used to obtain granules with acceptable critical quality attributes (CQAs) for both APIs. Granulation at high drug load of the poorly soluble, poorly wettable API mebendazole proved challenging as it could not be processed using lactose as filler, in contrast to lactose/MCC. As formulations containing lactose/MCC as filler were less influenced by different screw configurations, process settings and API characteristics than formulations without MCC, lactose/MCC/HPMC was considered a promising platform formulation.
Asunto(s)
Mebendazol/química , Metformina/química , Tecnología Farmacéutica/métodos , Celulosa/química , Composición de Medicamentos , Excipientes/química , Lactosa/química , Tamaño de la Partícula , Solubilidad , HumectabilidadRESUMEN
In current study a holistic material characterization approach was proposed and an extensive raw material property database was developed including a wide variety of APIs and excipients with different functionalities. In total 55 different materials were characterized and described by over 100 raw material descriptors related to particle size and shape distribution, specific surface area, bulk, tapped and true density, compressibility, electrostatic charge, moisture content, hygroscopicity, permeability, flowability and wall friction. Principal component analysis (PCA) was applied to reveal similarities and dissimilarities between materials and to identify overarching properties. The developed PCA model allows to rationalize the number of critical characterization techniques in routine characterization and to identify surrogates for use during early drug product development stages when limited amounts of active pharmaceutical ingredients are available. Additionally, the developed database will be the basis to build predictive models for in silico process and formulation development based on (a selection of) property descriptors.