(411b) An Optimization-Based Approach to Identify Thermodynamically Stable Blends for Spray Drying Dispersions

Many new compounds in the drug discovery pipeline are classified as poorly soluble chemicals which result in low bioavailability and dissolution rates. Spray drying dispersion (SDD) is a commonly used method in drug manufacturing to improve solubility, bioavailability and drug delivery of poorly soluble compounds [1]. SDD formulations consist of an active pharmaceutical ingredient (API), preferably in its amorphous state, dispersed and stabilized in a polymeric matrix to increase solubility. Despite the benefits of this technique, solid dispersions are often thermodynamically metastable. Therefore, it is important to select suitable API-polymer formulations that meet desired properties and ensure high solubility and long-term stability of the drug [2-4]. In current practice, heuristic approaches and experimental investigations are extensively employed for pre-screening small sets of commonly used polymers and solvents for the spray drying of APIs. Such protocols lead to high material consumption and thus they result in increased cost and environmental footprint. Clearly, there is an important need for developing more sophisticated and streamlined methods to reduce the number of time-consuming and costly experiments.

In this work, we present a systematic approach for identifying optimal (i) API-polymer and (ii) API-polymer-solvent formulations that meet desired physicochemical properties and can lead to drug products with improved bioavailability [5]. In particular, we employ property-prediction models to estimate solubility, miscibility and the glass transition temperature of a wide range of binary and ternary blends. In addition, we use optimisation models [6] to identify improved formulations that yield high solubility and stability of formulations with high drug loading. Multi-objective optimization is also utilized to model and identify optimal solutions of competing objectives. The design methodology is demonstrated with the help of a model drug, naproxen, where optimal polymers and solvents are selected to maximize drug loading while ensuring phase stability and a sufficiently high glass transition temperature of the final blend. A ranked list of optimal blends with different chemicals and compositions is obtained by introducing integer cut inequalities into the model, and proper phase diagrams of the best binary and ternary mixtures are constructed. Finally, the proposed model is used to generate a library of optimal formulations for various SDD systems that is used to identify better-performing designs and guide experimental work.

[1] Duarte, I., Santos, J.L., Pinto, J.F., Temtem, M., 2015, Screening methodologies for the development of spray-dried amorphous solid dispersions, Pharmac. Research 32, 222-237.

[2] Newman, A., Zografi, G., 2022, What Are the Important Factors That Influence API Crystallization in Miscible Amorphous API−Excipient Mixtures during Long-Term Storage in the Glassy State?, Molecular Pharmaceutics 19, 378-391.

[3] Lehmkemper, K., Kyeremateng, S.O., Heinzerling, O., Degenhardt, M., Sadowski, G., 2017, Impact of polymer type and relative humidity on the long-term physical stability of amorphous solid dispersions, Molecular Pharmaceutics 14, 4374-4386.

[4] Dohrn, S., Reimer, P., Luebbert, C., Lehmkemper, K., Kyeremateng, S.O., Degenhardt, M., Sadowski, G., 2020, Thermodynamic modeling of solvent-impact on phase separation in amorphous solid dispersions during drying, Molecular Pharmaceutics 17, 2721-2733.

[5] Jonuzaj, S., Burcham, C.L., Galindo, A., Jackson, G., Adjiman, C.S., 2022, Optimizing the selection of drug-polymer-solvent formulations for spray-dried solid dispersions in pharmaceutical manufacturing, Computer Aided Chemical Engineering, Accepted for publication.

[6] Jonuzaj, S., Cui, J., Adjiman, C.S., 2019, Computer-aided design of optimal environmentally benign solvent-based adhesive products, Computers & Chemical Engineering 130, 106518.