A Perspective on the Application of Preparative Supercritical Fluid Chromatography Using Achiral Stationary Phases in Pharmaceutical Drug Discovery and Development

• Merck & Co

Abstract

Supercritical fluid chromatography (SFC) has been widely used for chiral separations and purifications in the pharmaceutical industry as a powerful and environmentally “greener” separation technology. However, traditional high-performance liquid chromatography (HPLC) and flash chromatography techniques still dominate the purification of various achiral intermediates and active pharmaceutical ingredients (APIs) in pharmaceutical drug discovery and development areas. In this article, successful examples of purifications by SFC using achiral stationary phases help to illustrate the capability, uniqueness, and efficiency of SFC for achiral preparative purifications. Further investigation of the use of preparative SFC for the purification of achiral mixtures will broaden the scope of this important technology in pharmaceutical discovery and development.

Introduction

Preparative supercritical fluid chromatography (SFC) has been adopted as the technique of choice for the small scale preparative purification of chiral compounds and it is now widely used in the pharmaceutical industry. The benefits of preparative chiral SFC purifications, including savings in cycle time and cost have now been fully realized. The rapid purification of several grams of enantiopure material is now a relatively routine task. Compared to preparative high-performance liquid chromatography (HPLC), SFC also provides a unique selectivity and a “greener” alternative, where carbon dioxide mobile phase which is non-toxic, non-flammable and recyclable, replaces hexane or heptane as a solvent^1-6.

However, the use of preparative SFC for achiral purifications has not been as widely used, and traditional preparative HPLC and flash chromatography still dominate the achiral purifications of different intermediates and active pharmaceutical ingredients (APIs) in the pharmaceutical drug discovery and development areas^7-9. Some of the reasons for the more limited use of preparative SFC for achiral purifications include the more challenging method transfer from HPLC to SFC for achiral compounds, the lack of a universal stationary phase for achiral separations by SFC, such as the C18 stationary phase in reversed-phase HPLC (RP-HPLC), and the overall lack of expertise and instrument availability for more routine purifications of compounds by achiral preparative SFC as compared to preparative HPLC and flash chromatography. However, the more recent introduction of new achiral stationary phases for use in preparative SFC¹⁰, as well as the introduction of new preparative SFC instruments by different suppliers provides the potential for increased use of preparative SFC for achiral purifications.

Drug discovery and development is an intensive, time consuming, and expensive process that can take more than twenty years to discover, develop and market a new drug¹¹. The pharmaceutical industry is faced with an increased demand to produce medicines that are safer and more effective. In order to meet these challenges, the potential drug candidates not only need to be active, but they need to have “drug-like” properties to advance to clinical development. Druglike properties such as solubility (i.e. logarithm of partition coefficient logP), potency (i.e. IC50 or high potency at the biological target), and permeability (i.e. permeability across biological membranes for drug absorption and distribution) are all important factors for the successful drug development. Additionally, other factors such as substructures with known toxic effects, including mutagenic/genotoxic, carcinogenic, and teratogenic properties need to be considered when discovering and developing new and safer drugs¹².

Recent advances in chemistry at Merck and elsewhere have enabled new synthetic methods that can generate thousands of different compounds in a very short amount of time¹³. Despite these significant advances made for the high-throughput reactions on a small scale, the purification and isolation of these compounds for their further evaluation is a bottleneck, thus making the need for faster and more efficient purification tools a critical need.

In this study, the application of preparative SFC using achiral station-ary phases for the successful purification of various compounds is illustrated through different case study examples. These examples illustrate the advantages of preparative SFC vs. preparative HPLC or flash chromatography, especially for the purification of poorly soluble and polar compounds.

Preparative SFC vs. Preparative HPLC or Preparative Flash Chromatography

The speed and efficiency advantages of preparative SFC often leads to significantly more productive purifications relative to HPLC. Figure 1 provides an illustration of this point in which purification of a development compound on 5 g scale containing a 30% major impurity is compared with both HPLC and SFC. The analyte mixture was first screened by analytical SFC and HPLC conditions to find the best purification method. The RP-HPLC purification required 40 L acetonitrile mobile phase and 46 hours of run time (with no stacked injection capability). The inability to do stacked injections is a software limitation of the particular HPLC instrument used in this study, and not a general disadvantage of HPLC.

 Figure 1. Preparative SFC purification of a 5 g development compound containing 30% major impurity. SFC conditions: Chiralcel OD-H column (250 x 21.2 mm i.d.), isocratic 45% methanol in CO₂, 60 mL/min flow rate, 35 °C, 100 bar outlet pressure.

Overall, the workup and drying of isolated fractions required 8 hours to provide desired compound with overall 57% recovery from the crude material (or 80% recovery from the HPLC purification step). The SFC purification, on the other hand, was completed in only 3 hours (using stacked injection capability) and using less than 5 L of methanol (Table 1). Also, due to the concentrated fractions in the methanol modifier, the workup and drying of isolated fractions was completed in less than an hour to provide desired compound with about 70% recovery from the crude material (or >95% recovery from the SFC purification).

Table 1. Comparison of preparative SFC vs. preparative HPLC

It is also important to note that the preparative SFC purification used a chiral column for the achiral purification. Due to the already well established chiral SFC screening capabilities and the availability of different chiral columns for preparative SFC purifications, the chiral columns can also be used successfully for the purification of achiral separations, especially when dealing with various isomers or closely related impurities^14-16.

Next we explore the power of preparative SFC vs. flash chromatography for the purification of achiral mixtures. Flash liquid chromatography is the purification workhorse in many discovery areas due to its fast, simple, and cost effective attributes. This technique uses a simple setup with relatively inexpensive and even disposable columns or cartridges, operating at low pressures, and conditions that are easily scaled from thin layer chromatography (TLC). However, there are also many disadvantages of flash chromatography compared to preparative SFC⁹. Flash liquid chromatography provides low efficiency separations due to the use of large particles, often making it difficult to effectively purify compounds with especially closely related impurities. Flash chromatography also uses large amounts of organic solvents, which are expensive to purchase, but also expensive to dispose.

Figure 2 illustrates the advantages of preparative SFC vs. flash liquid chroma-tography. A development intermediate compound at 4 g scale needed to be purified before next performing a subsequent synthetic step. First, flash chromatography was attempted on 1 g scale using 1.6 L of organic solvents in about 40 min run time. This flash purification provided material with only about 70% purity, which was very difficult to further purify. The remaining 3 g of crude material was submitted for preparative SFC purification. SFC column screening showed excellent resolution of different impurities using an achiral column, and the 3 g material was successfully purified using 2.3 L of methanol in less than 2 hours, providing material with >99% purity. Table 2 highlights the comparison of the two methods, with flash chromatography affording incomplete separation and prep SFC affording fast, green, and efficient purification.

 Figure 2. Preparative SFC purification of a crude 3 g sample. SFC conditions: 2-Ethyl pyridine (2-EP) column (250 x 30 mm i.d.), isocratic 30% methanol (with 0.2% NH₄OH) in CO₂, 70 mL/min flow rate, 35 °C, 100 bar outlet pressure.

Table 2. Comparison of preparative SFC vs. preparative flash chromatography

Preparative SFC Purifications for Compounds with Solubility and Stability Issues

Good solubility is critical for the successful and efficient chroma-tographic purification of compounds on preparative scale. Moreover, intermediates and APIs prepared from a chemical synthesis in any organic solvent can be either directly injected into SFC or diluted with an organic solvent for purification. For example, about 3 g of a development intermediate compound was difficult to purify by preparative HPLC due to its poor solubility. The compound was dissolved in methanol/dichloromethane mixture and purified by preparative SFC in only about 2 hours and using 3.6 L of methanol (Figure 3). Separation of many closely eluting impurities was difficult, but ‘heartcutting’ of the main peak provided material with acceptable purity for next reaction step.

 Figure 3. Preparative SFC purification of a 3 g crude mixture dissolved in methanol/dichloromethane mixture. SFC conditions: Diol column (250 x 30 mm i.d.), isocratic 40% methanol (with 0.2% NH₄OH) in CO₂, 70 mL/min flow rate, 35 °C, 100 bar outlet pressure.

Purification of about 2 g of another development compound prepared from a biocatalytic reaction using aqueous/dimethyl sulfoxide (DMSO) mixture as a reaction solvent was attempted by RP-HPLC conditions using acidic water and acetonitrile as mobile phase, with about 35 mg being purified in a 75 min total run time. However, the compound was found to quickly decompose upon drying and concentration. The rest of the crude mixture (1.8 g) was then submitted for preparative SFC purification and it was successfully purified using about 1.1 L of methanol in only about 50 min run time. The desired compound was isolated with high purity and recovery with no issues (Figure 4).

 Figure 4. Preparative SFC purification of a 1.8 g sample dissolved in DMSO. SFC conditions: 2-EP column (250x30 mm i.d.), isocratic 30% methanol in CO₂, 70 mL/min flow rate, 35 °C, 100 bar outlet pressure.

Finally, SFC is also a very useful technique for the purification of compounds that are unstable in the presence of water or alcohols. In such cases, RP-HPLC or flash chromatography is not suitable, but SFC with anhydrous acetonitrile as modifier provides a unique alternative choice. For example, a few grams of an intermediate compound sensitive to water and alcohols was attempted to be purified using flash chromatography with a silica column, but the compound quickly decomposed during the purification process. However, 3.5 g of this compound was successfully purified using preparative SFC with anhydrous acetonitrile as modifier in less than 3 hours and using only 1.8 L of acetonitrile (Figure 5).

 Figure 5. Preparative SFC purification of a 3.5 g compound that is labile to water or any alcohol. SFC conditions: 2-EP column (250 x 30 mm i.d.), isocratic 15% acetonitrile in CO₂, 70 mL/min flow rate, 35 °C, 100 bar outlet pressure.

Preparative SFC Purifications of Very Polar Compounds Using Water as Additive

Despite the fact that SFC is very similar to the normal phase HPLC separation mode, preparative SFC using achiral stationary phases can be efficiently used to purify very polar compounds by simply adding water to the modifier. Recent studies have shown the advantageous chromatographic benefits of water as an additive in carbon dioxide-based mobile phase in SFC analysis of achiral compounds^17-18. Water alone has very low solubility in supercritical carbon dioxide, but water as an additive in the presence of methanol modifier is miscible with supercritical carbon dioxide. Therefore, water is typically added up to 5 (v/v) % in the methanol modifier. The addition of water to the mobile phase has been shown to improve peak shape and extend the polarity range of SFC to allow the separation of very polar analytes, including nucleobases, peptides, and proteins¹⁹. This provides an excellent alternative to purification by traditional RP-HPLC, where drying down aqueous chromatographic fractions can be very slow. For example, about 5 g of a highly polar development compound in dimethylformamide (DMF) reaction media was purified by SFC using water as an additive. The preparative SFC purification used only 2.1 L of methanol (modified with 5% water), the separation was completed in 50 minutes and the desired product was isolated with high purity and recovery (Figure 6).

 Figure 6. Preparative SFC purification of a 5 g very polar compound using water as additive. SFC conditions: Basic column (250 x 30 mm i.d.), isocratic 60% methanol (with 5 v/v% water) in CO₂, 70 mL/min flow rate, 35 °C, 100 bar outlet pressure.

Conclusions

Preparative SFC is a powerful and environmentally “greener” separation technology compared to traditional preparative HPLC and flash chromatography. The technique can be employed in achiral separation and purification as well as in chiral resolution in support of pharmaceutical drug discovery and development. The different examples highlighted here illustrate the successful application of SFC for purification of many different compounds, including compounds with poor solubility or stability. As the technique of preparative SFC continues to be deployed for the separation and purification of intermediates and APIs in drug discovery and development, additional improvements in performance and scope can be expected.

Acknowledgements

We thank Ingrid Mergelsberg, Ian Davies, and Chris Welch for insightful discussion and suggestions.

References

1. Welch CJ, Leonard WR, DaSilva JO, Biba M, Albaneze-Walker J, Henderson DW, Laing B, Mathre DJ. Preparative chiral SFC as a green technology for rapid access to enantiopurity in pharmaceutical process research. LCGC North America 2004;23(1):16,18,22,24,26-29.
2. Welch CJ, Wu N, Biba M, Hartman R, Brkovic T, Gong X, Helmy R, Schafer W, CuffJ, Pirzada Z. Greening analytical chromatography. Trends in Analytical Chemistry 2010;29(7):667- 680.
3. eonard WR, Henderson DW, Miller RA, Spencer GA, Sudah OS, Biba M, Welch CJ. Strategic use of preparative chiral chromatography for the synthesis of a preclinical pharmaceutical candidate. Chirality 2007;19(9):693-700.
4. Biba M, Welch CJ, DaSilva JO, Mathre DJ. Impurity isolation in support of pharmaceutical process research. American Pharmaceutical Review 2005;8(2):68-72,84.
5. Miller L, Potter M. Preparative supercritical fluid chromatography (SFC) in drug discovery. American Pharmaceutical Review 2008;11(4):112-117.
6. Taylor LT. Past, current, and future directions in: supercritical fluid chromatography. LCGC Europe 2013;28-31.
7. Lazarescu V, Mulvihill MJ, Ma L. Achiral preparative supercritical fluid chromatography. Supercritical fluid chromatography 2014;97-143.
8. White C, Burnett J. Integration of Supercritical Fluid Chromatography into Drug Discovery as a Routine Support Tool: II. Investigation and Evaluation of Supercritical Fluid Chromatography for achiral Batch Purification. Journal of Chromatography A 2005;1074(1- 2):175-185.
9. Ashraf-Khorassani M, Yan Q, Akin A, Riley F, Aurigemma C, Taylor LT. Feasibility of correlating separation of ternary mixtures of neutral analytes via thin layer chromatography with supercritical fluid chromatography in support of green flash separations. Journal of Chromatography A 2015;1418:210-217.
10. McClain R, Huyn MH, Welch CJ. Advances in achiral stationary phases for SFC. American Pharmaceutical Review 2014;17(3):32,34,36-41.
11. DiMasi JA, Hansen RW, Grabowski HG. The price of innovation: new estimates of drug development costs. Journal of Health Economics 2003;22(2):151-185.
12. Lipinski CA. Drug-like properties and the causes of poor solubility and poor permeability. Journal of Pharmacological and Toxicological Methods 2000;44:235-249.
13. Kutchukian PS, Dropinski JF, Dykstra KD, Li B, DiRocco DA, Streckfuss EC, Campeau LC, Cernak T, Vachal P, Davies IW, Krska SW, Dreher SD. Chemistry informer libraries: a chemoinformatics enabled approach to evaluate and advance synthetic methods. Chemical Science 2016;(Ahead of print).
14. Pirzada Z, Personick M, Biba M, Gong X, Zhou L, Schafer W, Roussel C, Welch CJ. Systematic evaluation of new chiral stationary phases for supercritical fluid chromatography using a standard racemate library. Journal of Chromatography A 2010;1217(7):1134-1138.
15. Schafer W, Chandrasekaran T, Pirzada Z, Zhang C, Gong X, Biba M, Regalado EL, Welch CJ. Improved chiral SFC screening for analytical method development. Chirality 2013;25(11):799-804.
16. Regalado EL, Welch CJ. Separation of achiral analytes using supercritical fluid chromatography with chiral stationary phases. Trends in Analytical Chemistry 2015;67:74- 81.
17. Taylor LT. Packed column supercritical fluid chromatography of hydrophilic analytes via water-rich modifiers. Journal of Chromatography A 2012;1250:196-204.
18. Liu J, Regalado EL, Mergelsberg I, Welch CJ. Extending the range of supercritical fluid chromatography by use of water-rich modifiers. Organic & Biomolecular Chemistry 2013;11(30):4925-4929.
19. Ashraf-Khorassani M, Taylor LT. Subcritical fluid chromatography of water soluble nucleobases on various polar stationary phases facilitated with alcohol-modified CO₂ and water as the polar additive. Journal of Separation Science 2010;33:1682-1691.

Author Biography

Mirlinda Biba is currently the Science Lead for the Purification Group in the Process and Analytical Chemistry Department at Merck in Rahway, NJ. She received her PhD in Analytical Chemistry at Drexel University under Prof. Joe Foley and Dr. Chris Welch from Merck. Her research focused on analysis and separation of short RNA oligonucleotides by different liquid chromatography methods. She also has over 15 years experience with SFC, including small to large scale purifications.

Jinchu Liu is a separation scientist in the Process and Analytical Chemistry Department at Merck in Rahway, NJ. He received his BS at Xiamen University, China, and his MS at the University of Alabama. After working as a process chemist for 13 years, he joined the separations group in 2008. His work focuses on chiral resolution, purification, and impurity isolation by SFC technology to serve the research of process and medicinal chemistry.