Cocrystal Synthesis through Crystal Structure Prediction.

Area of Science:

• Solid-state chemistry
• Computational materials science
• Pharmaceutical development

Background:

• Crystal structure prediction (CSP) is crucial for identifying solid forms of active pharmaceutical ingredients.
• Understanding cocrystal formation is vital for optimizing drug properties.

Purpose of the Study:

• To employ CSP for predicting cocrystal coformers for MK-8876 and a triol intermediate.
• To explore the solid form landscape and identify novel cocrystals with enhanced properties.

Main Methods:

• Utilized a CSP-based cocrystal prediction method to rank potential coformers.
• Performed retrospective CSP for MK-8876 and prospective screening for the triol.
• Applied computational finite-temperature corrections for relative crystallization propensities and free-energy landscape analysis.

Main Results:

• CSP successfully predicted the known maleic acid cocrystal for MK-8876.
• CSP identified triol-1,4-diazabicyclo[2.2.2]octane (DABCO) cocrystal as rank one and triol-l-proline as rank two.
• The triol-l-proline cocrystal was experimentally obtained, showing improved melting point and deliquescence.

Conclusions:

• CSP is an effective tool for cocrystal screening and prediction in pharmaceutical development.
• The triol-l-proline cocrystal represents a promising alternative solid form for islatravir synthesis.
• Computational methods accelerate the discovery of solid forms with desirable physicochemical characteristics.