From polymorphs to cocrystals and salts: successfully predicting axitinib's challenging crystal forms

Area of Science:

• Crystallography
• Computational Chemistry
• Pharmaceutical Sciences

Background:

• Experimental cancer drug axitinib development faced challenges due to unexpected discovery of new crystal polymorphs.
• Previous organic crystal structure prediction methods struggled to reliably model axitinib's complex conformational polymorph lattice energies.

Purpose of the Study:

• To demonstrate modern crystal structure prediction's capability in accurately modeling challenging axitinib crystal structures.
• To assess the prediction of distinguishing between salt and cocrystal forms in multi-component axitinib crystals, a known modeling challenge.

Main Methods:

• Utilized advanced crystal structure prediction methodologies.
• Addressed density-driven delocalization errors in generalized gradient approximation density functional theory (DFT) models.
• Applied intramolecular energy corrections and/or hybrid density functionals.

Main Results:

• Successfully predicted problematic axitinib crystal structures, including conformational polymorphs.
• Accurately distinguished between salt and cocrystal forms in three axitinib multi-component systems.
• Demonstrated that combining intramolecular energy corrections and hybrid DFT functionals yields superior lattice energy predictions.

Conclusions:

• Modern crystal structure prediction is now capable of reliably modeling complex organic crystal forms like axitinib.
• Improved computational methods can overcome long-standing challenges in predicting multi-component crystal behavior.
• These advancements enhance the de-risking of pharmaceutical development by preventing unexpected crystallization issues.