(366g) Efficient Production of Multi-Component Amorphous Solid Dispersions By Spray Drying

One of the major challenges in drug development nowadays is low water solubility (BCS classes II and IV). A widely used approach that had been proven successful in improving drug solubility is the formulation of amorphous solid dispersions (ASDs). Typically an ASD is composed of the poorly soluble drug and the hydrophilic polymer. If the combination is correct, the polymer propels the concentration of the poorly soluble drug into a supersaturated state during dissolution (spring effect). However, often the drug precipitates back into the solution, causing the concentration and thus bioavailability to drop. This issue can be circumvented by using a second polymer, which acts as a precipitation inhibitor (parachute effect). Yet, finding out the correct ratios in this tri-component system (drug: polymer 1: polymer 2) can often prove to be difficult.

In this work, we produced multi-component solid dispersions by spray drying in an efficient manner. As a widely recognized model system, we chose cyan, magenta, and yellow (CMY) printer dyes and lactose as a carrier. We can conclude that the ratio between the components in certain collected batch (experimental values) can be different from the ratio between the components in the inlet feed (theoretical values) as the residence time of particles in the system can be greater than the collection period. We evaluated the cross-contamination of samples by UV-Vis spectroscopy. Across all 15 samples, which formed a ternary diagram, the average deviation from the theoretical compositions was approximately 1.7 %. The first part of the figure below shows all collected samples arranged in the ternary diagram, forming visually confirmable color gradients. In the second part, the results from UV-Vis spectroscopy are plotted into the ternary diagram. These results show that we can produce multicomponent solid dispersions efficiently without significant cross-contamination between samples.

Acknowledgements:

Authors acknowledge financial support by the Technology Agency of the Czech Republic, project number TJ02000095 (Zeta Program).