(M1330-08-45) Quality by Design Approach of Chiral Method Development of Compound E

Purpose: To develop a chiral separation method for Compound E to evaluate stereoselective catalysis in support of process optimization. Using the Quality-by-Design (QbD) approach provides a deliberate process that takes into consideration project and technical challenges before starting laboratory work, including the following: Aggressive timeline and budget to prepare for GMP manufacturing, compatibility of method with CMO equipment, training, and expertise, analytical method robust and validate-able, 2 chiral centers (4 stereoisomers), weak UV chromophore, and primary and tertiary amines and limited physico-chemical characterization (e.g. volatility) Because of the urgent timeline to finalize the route of synthesis and confirm adequate stereoselective control, it was hypothesized that the QbD approach would save significant time and money. We performed this work, while the CDMO worked in parallel executing the work per its unfettered strategy.

Methods: In this work, the physico-chemical properties of the intermediate were assessed to make predictions about chromatographic behavior. Based on those properties, we hypothesized that both GC and HPLC could provide viable options. For GC, it was hypothesized that capillary GC with FID detection would be ideal for providing high resolving power and for overcoming the limitations of UV detection. The low-polarity low-molecular-weight free base could be volatile enough to transverse the inlet, followed by separation of the diastereomers on a chiral stationary phase (CSP). It was also hypothesized that the neutral free base would also be able to dock with a cyclodextrin stationary phase, given the suitably-sized, chiral, and neutrally-charged cavity. For HPLC, the analyte’s pKa’s, the mobile phase pH and buffering capacity, interaction of the molecule with the CSP, the molecule’s charged state, and chiral column manufacturer’s recommendations based on molecule class were all considered. Based on those, it was decided to investigate the separation of the free base diastereomers on 3 of the 7 recommended columns, coincidentally containing CSPs modified with aromatic groups. As mentioned above, the lack of a strong UV chromophore dictating low wavelength detection limited mobile phase options. The combined challenges of wavelength and CSP selection prompted a structurally-based hypothesis, which was that the molecule’s chromophore and retention behavior could be mitigated simultaneously. Thus, we proposed that derivatizing the primary amine using benzoyl chloride would provide the following advantages:  Add a strong aromatic UV chromophore, allowing detection at longer wavelength for a more stable baseline with less interferences, and allowing more options for mobile phase composition, provide a structural moiety that could π-stack with the stationary phase, thereby enhancing the residence time and chiral recognition, convert the primary amine to an amide to improve peak shape.

Results: Results GC: After a preliminary volatility screen by increasing inlet temperature, the separation on the cyclodextrin CSP was realized almost immediately, and due to the resolving power of capillary GC, only needed some minor thermal gradient adjustments. The chromatogram of the GC chiral separation is shown in Figure 1. No GC solution was attempted by the CDMO. The method employs industry standard ALS-GC-FID, a commercially available column, and no special training is required. Synthesis of the individual diastereomers is underway to identify each of the 4 isomers. Results HPLC:  With the detector set at 254 nm, a basic mobile phase pH to render the intermediate free base, and with a typical normal phase methodology, a nearly baseline-resolved chromatogram of the 4 diastereomers was achieved almost immediately, and only required some modest gradient adjustment. The chromatogram of the normal phase chiral separation is shown in Figure 2 . The derivatized diastereomers are detected at 230nm, taking advantage of the aromatic chromophore. The method employs industry standard HPLC-UV, a commercially available column, and modest training. Synthesis of the individual diastereomers is underway to identify each of the 4 isomers.

Conclusion: Chromatographic methods (GC and HPLC) were developed for the chiral separation of Compound E in a timely manner by using a Quality-by-Design (QbD) approach that drove thoughtful consideration of the molecule’s properties, including two chiral centers, weak UV chromophore, amine-containing, and unknown volatility.  Both methods are compatible with the equipment and experience of most CMOs, therefore easily transferable and Quality Control friendly. The QbD approach significantly reduced the time and cost versus a trial-and-error approach.

ALS-GC-FID of Intermediate Free Base Analyzed on β-Cyclodextrin



HPLC-UV of Compound E Base Derivatized on Diacel IF-3