(241b) Cocrystals: Latest Advances in 'in Silico' Methods for Predicting Cocrystal Coformers and Their Properties

The formulation, design, and implementation of active pharmaceutical ingredients (API) is an area of great interest. Changing the structure and composition of an API by cocrystallizing it with a coformer can have a significant influence on the properties and bioavailability of the drug, and can even offer the potential for new patents and associated revenues.  For example, changing the solubility can influence the formation of crystals, cocrystals or solvates, as well as the pharmacokinetics through dissolution during drug release.

The traditional experimental approach to screening for possible new coformers and measuring their properties is time-consuming and expensive.  However, in silico methods are potentially orders of magnitude faster and are becoming more accurate, such that it is now possible to prescreen huge libraries of potential coformers in minutes or hours on a laptop computer, to determine the most likely candidates.  This enables the chemical engineer to take into account a much wider range of possible candidates and then focus the experimental work on the most promising ones, so greatly improving the chances of success in the lab.

This presentation will review the capabilities and limitations of some of the available in silico tools for predicting cocrystal coformers and their properties, and compare them with experiment, including group contribution methods, quantum chemistry, and statistical thermodynamics.  In addition, the latest algorithmic and method advances for improving the prediction of compounds that can form cocrystals and their properties in a range of solvents, will be discussed.