From benchtop to pilot scale-experimental study and computational assessment of a hot-melt extrusion scale-up of a solid dispersion of dipyridamole and copovidone.

The aim of our work was to study and define a computationally-based adiabatic scale-up methodology for a hot-melt extrusion (HME) process to produce an amorphous solid dispersion (ASD). As a drug product becomes commercially viable, there is a need for scaling up the manufacturing process. In the case of HME used for the formation of ASDs, scale-up can be challenging due to the fundamental differences in how heat is generated in extruders of differing scale, i.e. conduction vs. viscous dissipation and the significant role heat generation plays in determining the final product attributes. Using a 30%w/w dipyridamole-in-copovidone formulation, 11 mm-, 16 mm- and 24 mm-diameter extruders with L/D 40, solid-state characterization tools, a geometric scaling equation, and Ludovic® twin-screw extrusion software, we compared the total imparted material energy, the conducted energy and the difference between barrel and melt temperature at die exit for various feed rates and screw speeds. Numerical simulation identified desirable adiabatic conditions at multiple extruder scales in agreement with the chosen scaling factor. With the use of computational tools, the energetics in an extrusion process can be evaluated and processing conditions can be selected to identify the most efficient scaling of a HME process.

In vivo predictive mini-scale dissolution for weak bases: Advantages of pH-shift in combination with an absorptive compartment.

The purpose was the evaluation of a new miniscale biphasic dissolution model with pH-shift (miBIdi-pH). Its capability to predict supersaturation and precipitation of weak bases (e.g. dipyridamole) and the in vivo performance of various formulations of the model compound BIXX (weak base, poor solubility, good permeability) was investigated with respect to dissolution, precipitation and re-dissolution. Single phase dissolution with and without pH-shift [small scale dissolution (V = 20 ml) and USPII] and miBIdi-pH (50 ml aqueous phase covered by 15 ml octanol) were used for analyzing crystalline dipyridamole and the four BIXX-containing formulations. Precipitate was analyzed via X-ray diffraction. Bioavailability of the formulations was tested in dogs. Phoenix WinNonlin(®) was used for IVIVC. For dipyridamole, precipitation upon pH shift was less pronounced in the miBIdi-pH in comparison to the single phase dissolution (35% vs. 90%). In case of four BIXX-containing formulations, USPII revealed significant differences in their dissolution, whereas the final amounts of BIXX in the octanol phase in the miBIdi-pH were alike. Different partitioning rates into octanol were observed. The miBIdi-pH was superior to single phasic dissolution in predicting in vivo precipitation of dipyridamole. In case of the BIXX-containing formulations, it was superior in ranking the formulations and it was capable to capture the kinetics of different absorption processes in vivo.

Rational development of solid dispersions via hot-melt extrusion using screening, material characterization, and numeric simulation tools.

Effective and predictive small-scale selection tools are inevitable during the development of a solubility enhanced drug product. For hot-melt extrusion, this selection process can start with a microscale performance evaluation on a hot-stage microscope (HSM). A batch size of 400 mg can provide sufficient materials to assess the drug product attributes such as solid-state properties, solubility enhancement, and physical stability as well as process related attributes such as processing temperature in a twin-screw extruder (TSE). Prototype formulations will then be fed into a 5 mm TSE (~1-2 g) to confirm performance from the HSM under additional shear stress. Small stress stability testing might be performed with these samples or a larger batch (20-40 g) made by 9 or 12 mm TSE. Simultaneously, numeric process simulations are performed using process data as well as rheological and thermal properties of the formulations. Further scale up work to 16 and 18 mm TSE confirmed and refined the simulation model. Thus, at the end of the laboratory-scale development, not only the clinical trial supply could be manufactured, but also one can form a sound risk assessment to support further scale up even without decades of process experience.