259445 Salt and Form Selection of a Chiral API Prone to Epimerization: The Importance of Determining the Nature of the Crystalline Racemate

Wednesday, October 31, 2012
Hall B (Convention Center )
Benjamin Cohen, John Traverse and William Leong, Department of Process Chemistry, Celgene Corporation, Summit, NJ

Selection of the final salt and form of an Active Pharmaceutical Ingredient (API) early in development avoids expensive and time consuming bridging studies if it is changed later in development.  The criteria for the selection are predominantly focused on having desirable physical attributes for the drug product (e.g. stability, flowability, solubility, low hygroscopicity).  In the case of chiral APIs, it is advisable to also understand the ability of the proposed salt and form to upgrade enantiopurity through crystallization by determining the nature of the compound’s crystalline racemate.  This  effort is particularly important for molecules that contain an epimerizable stereogenic center.   A case study of the salt and form selection process of an epimerizable API for which an understanding of the compound’s crystalline racemate was a significant contributing factor is presented.

To develop a thermodynamically controlled crystallization for enantioenrichment, it is important to determine the nature of the system’s thermodynamically most stable racemic crystalline form.  The most desired form for boosting enantiopurity is a conglomerate, however these comprise only ~10% of racemic crystals.   Racemic compounds make up the majority.  A third and relatively rare class, pseudoracemates, also exist.

The API of interest contains an epimerizable stereogenic center.  The route was designed to delay the incorporation of the functionality leading to epimerization to the final bond forming step.  Nonetheless, the route and process generates the API with moderate achiral and chiral purity, therefore a recrystallization is required to boost the assay, achiral, and chiral purity to an acceptable level.  A preliminary salt and form screen identified a sulfonate salt as a good lead for the final API salt.  However, the XRD, DSC, ternary phase diagram and crystallization data indicate that the RSO3H salt is a pseudoracemate (Type II).  As a consequence, the enantiopurity can not be increased in this system. It was decided that efforts should be directed towards finding an alternate salt, due to the risk that the manufacturing process would not deliver the API with acceptable chiral purity.

The free base (ee ~94%) was subjected to an extensive salt screen with achiral and chiral acids. The ee of the solids and filtrate were monitored in addition to the XRD and DSC of the solids.  One hit, the HX salt was found to have low hygroscopicity, high solubility, good stability and acceptable flowability. The XRD of the single enantiomer and racemic crystals of this salt are different, indicating that it is a racemic compound.   TGA, DSC and DVS analyses show that while the single enantiomer form is anhydrous, the racemic crystal preferentially forms a hydrate and methanol solvate. 

The eutectic ee of a racemic compound is typically independent of the solvent system except when one or both of the components is a solvate.  Because the racemic HX crystals form solvates in the presence of water or methanol, the eutectic ee has a strong dependence on solvent composition and temperature.  With the understanding of the eutectic’s dependence on solvent and temperature, a proof of concept experiment was successfully run to demonstrate the ability of the methanol system to elevate chiral purity.


Extended Abstract: File Not Uploaded