Application of the Dynamic Gastric Model to evaluate the effect of food on the drug release characteristics of a hydrophilic matrix formulation.

Characterisation of the effect of food on the bio-performance of modified and extended release dosage forms can be very challenging due to the need to replicate the dynamic biochemical conditions of the human gut as well as the complex physical processing modalities under fed state. Classical compendial methods are useful for testing the quality of pharmaceutical dosage forms but typically have limitations in the accurate prediction of food-effect in-vivo. Preliminary evaluation of the Dynamic Gastric Model (DGM) shows that it can provide substantially more detailed mechanistic information on dosage form properties compared to conventional compendial testing. The potential effect of food on the drug release and physical properties of a hydrophilic matrix formulation containing a model drug, hydrochlorothiazide, was studied using compendial methods, bio-relevant media and the DGM (in combination with an off-line intestinal model). Whilst the compendial methods with biorelevant media provided good correlation with the dissolution rates observed using the DGM/intestinal model under simulated fasted state, the quantification of simulated fed state performance changes was much more challenging using the compendial methods. Classical compendial studies using biorelevant FeSSIF and FaSSIF media could not readily discern differences in dissolution performance under fasted and fed states; however, the DGM could detect significant changes in both physical properties as well as drug release performance under fed state processing.

Preparation and characterization of poly(epsilon-caprolactone) polymer blends for the delivery of proteins.

A series of ternary blend matrices, based on high- and low-molecular-weight poly(epsilon-caprolactone) (PCL) and a poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide) block copolymer (Synperonic L61), has been developed for the delivery of proteins. The inclusion of Synperonic L61 served to enhance the water content of the matrix available for protein diffusion. Blends comprising high-molecular-weight PCL (PCLH) and Synperonic L61 alone were found by scanning electron microscopy to be incompatible over a wide composition range. The addition of a low-molecular-weight PCL (PCLL) enhanced compatibility of the polymeric components. Analysis of the dynamic mechanical and thermal properties of these blends showed a significant shift in the glass transition temperature of the material (-38 to -55 degrees C), as the weight fraction of Synperonic L61 increased to 30 wt%, indicative of limited miscibility of the components in the non-crystalline phase. All ternary blended matrices showed a higher degree of hydration relative to homogeneous PCLH. The maximum water content could be modified by adjusting the weight fraction of Synperonic L61 in the blend. The rate of release of a model protein, bovine serum albumin (BSA), from matrices containing Synperonic L61 was considerably higher than that from PCLH and PCLH/PCLL alone. Analysis of the release mechanisms of BSA from these matrices suggested that, whilst diffusion was the predominant mode of protein transport, the elution of Synperonic L61 from blended matrices during the dissolution caused a notable deviation from Fickian control.

The microencapsulation of protein using a novel ternary blend based on poly(epsilon-caprolactone).

Microspheres with an entrapped protein were prepared from poly(epsilon-caprolactone) (PCL), and a novel ternary blend, comprising of high and low molecular weight PCL in combination with poloxamer 181, a triblock copolymer of poly(ethylene oxide)-poly(propylene oxide)-poly(ethylene oxide). The inclusion of low molecular weight PCL served to enhance phase mixing by a reduction in the molecular weight of the polymeric components. Encapsulation of the protein, bovine serum albumin, was possible using a water-in-oil-in-water multiple emulsion solvent evaporation technique. Microspheres prepared from unblended PCL were irregular and porous in nature. The presence of surface imperfections and macroscopic pores was attributable to the high rate of crystallization of the PCL polymer from solution. The inclusion of poloxamer 181 into the matrix retarded the rate of crystallization of the PCL, thereby enhancing particulate sphericity and regularity. Manipulation of the process parameters of blended microspheres provided a means of controlling the particle size and the entrapment efficiency of the protein. The influence of variables such as protein to polymer ratio, internal phase volume and emulsifier concentration in both the internal and external aqueous phases, on the properties of the microspheres was investigated. A mean particle size ranging from 10 to 42 microns could be achieved by altering the internal phase volume of the primary emulsion, whilst a high protein entrapment (11% w/w) was possible with a protein to polymer ratio of 1:4. Native-PAGE analysis of the entrapped protein indicated a maintenance of bulk structural integrity.