The 3rd Cambridge Area Analytical Chemistry Group Meeting
Nick Taylor commenced proceedings with an update on Mass Directed Auto Purification (MDAP) at GSK and how it has been applied to chiral purification including in an Open Access environment. He detailed the solvent changes that were implemented on a Waters FractionLynx prep system including the MS make-up solvent and how the installation of an active splitter has allowed reliable collection whatever the mobile phase. Then Nick went on to explain how the use of SIR (Single Ion Recording) gives a 10-100 fold sensitivity improvement which has enabled low level purification and much more accurate setting of the collection threshold both for achiral and chiral purification. He showed several examples of how SIR has been implemented on Open Access MDAP systems to overcome various scenarios that lead to collection failure on a standard system, e.g. very low levels and poorly ionising compounds that would otherwise have required a specialist purification service.
The figure below shows a low level chiral MDAP of flavanone (0.01mg each enantiomer)
Continuing the purification and chiral themes, Mark Bayliss (CSO, Virscidian Inc.) spoke on the selection of purification methods using a data-driven approach in a highly automated workflow-driven purification setting. He walked us through a thought-provoking example of the optimisation of an SFC separation of 2 diastereomeric pairs of enantiomers in a sample also containing two isobaric impurities. The screening method matrix (8 columns x 3 modifiers x 2 temperatures) and optimisation goals were explained. Mark then introduced Virscidian Analytical Studio, how it scores the results obtained from the different methods and how the data is visualised and reviewed. He showed how the variables used for scoring affect the ranking of the methods. In other words, simply optimising on resolution produces one result but as other variables are factored in such as asymmetry, tailing, height and peak width alternative methods may emerge as superior. In the case in question, it wasn’t until a new variable, namely the difference in Retention Time between peaks that the best separation was highlighted, showing that resolution is not the only driver for a successful purification method. Although the initial choice based on resolution only did generate a usable separation, the multi-variate approach led to a 30% reduction in purification time.
Mark then turned to achiral purification and throughput optimisation. He discussed a number of purification workflows and how the tools available within Analytical Studio enable productivity gains by automating review of pre-QC LCMS data for method selection and prep sample list creation, intelligent fraction selection and review of post-QC LCMS data. By automating the majority of decision-making and only flagging a minimal amount of data for user-review where necessary, great productivity gains are realisable without compromising quality.
Alex Brien of Reach Separations explained the chiral column QC approach in use at Reach Separations including the reasons for its implementation, the system used and the QC procedure and column regeneration procedures in some detail as well as the data that is captured. He gave an excellent example of how an apparently defunct column could be rescued following their procedure.
Overall, they have been able to reduce spend on prep columns by 40% and have much greater control over column performance and can identify scale-up issues and trends in samples.
Then we had a change in direction as Ioanna Stamati, Bioconjugation Team Leader at Antikor Therapeutics based at the Stevenage Bioscience Catalyst, gave a presentation on Fragment Antibody Drug Conjugates (FDCs). Ioanna gave a quick introduction to the hot topic of ADCs of which there are nearly 100 in Clinical Development, the downside of whole Mab IgG ADCs and the advantage of using fragments that are 1/5 the size of an IgG. Antikor’s linker technology produces high DAR (Drug:Antibody Ratio) products (12:1) with good spatial separation to reduce aggregation and improve solubility.
Turning to analytical techniques for ADCs, Ioanna stressed the importance of DAR analysis alongside characterisation of purity, aggregation and stability. A very wide range of separation techniques is required including HIC (Hydrophobic Interaction Chromatography), SEC (Size Exclusion Chromatography), SDS PAGE (Polyacrylamide Gel Electrophoresis), Ion Exchange Chromatography and cIEF (Capillary Isoelectric Focussing) along with amino acid analysis, peptide mapping and, of course, Mass Spectrometry. In addition, functional methods to assess binding such as SPR (Surface Plasmon Resonance) are used.
Ioanna explained how both the drug (payload) and linker contribute to hydrophobicity and charge (pI). A high DAR on a small protein leads to bigger effects on charge and hydrophobicity which, in turn, determines the ease of analytics. She showed a range of UV/Vis, SEC, SDS PAGE and LCMS data for the measurement of DAR, purity and aggregation, including showing how the linker can have an effect on aggregation.
Christophe Auberger of Crawford Scientific gave us a good round-up of column selection and how to navigate your way through the >700 Reverse Phase columns available including >190 C18’s. He gave a detailed description of the Hydrophobic Subtraction Model (HSM) of reversed-phase column selectivity :
log α = log k/kEB = η’H – σ’S* + b’A + α’B + κ’C
by which columns can be compared quantitatively for differences in selectivity and explained the various terms in the model:
- Hydrophobic selectivity (H)
- Steric Resistance (S*)
- Column Hydrogen Bond Acidity (A)
- Column Hydrogen Bond Basicity (B)
- Ionic Interaction pH7.0 & 2.8 (C)
He then explained how the F(s) metric can be used to define columns as equivalent, similar or orthogonal.
A useful resource is to be found at www.hplccolumns.org which is a database of most commercially available columns in a user-friendly web page interface. It contains column parameters for over 500 reversed-phase HPLC columns characterized using the HSM.
Finally, Christophe introduced the Crawford Scientific Model and Database and how this can be used to find similar and orthogonal columns by comparison to a given reference column, depending on the acid/base nature of the analytes and the pH of the mobile phase.
Our final speaker was Terry Soteriou from Malvern Panalytical who spoke on applications of X-ray Diffraction and X-ray Scattering, a topic unfamiliar to the majority in the audience. He explained how X-ray powder diffraction (XRPD) is THE technique of choice when working with crystalline substances to determine which polymorph you have (from API to formulation) and to detect low levels of polymorphic impurities with little sample prep and with recovery of the sample. Interestingly, it is possible to study in situ study crystallization. He also spoke about the use of XRD for polymorph and salt screening in a high throughput setting.
XRD can also be used to monitor stability at different temperatures and relative humidities and on freeze-drying, in case of possible phase transitions during production processes, storage or transport which could have an effect on drug efficacy. Crystallization can be monitored on-line by use of a flow cell.
X-ray scattering is used for particle size analysis and comes in two flavours: Small and Wide-angle X-ray scattering (SAXS and WAXS). Terry showed an example of the use of SAXS for biomolecules, namely for the particle size distribution of proteinase K samples.
Following conclusion of the presentations, lively discussion continued during the buffet lunch kindly provided by GSK for a good hour and a half!
If you wish to receive information about future meetings, please contact me on: firstname.lastname@example.org