Recent Advances in Hyphenated Chromatography and Mass Spectrometry Techniques and Their Impact on Late-Stage Pharmaceutical Development

This article reviews the changing role of mass spectrometry (MS) hyphenated to reversed-phase liquid chromatography (LC) and alternative separation techniques in late-stage pharmaceutical development. The impact of the changing portfolios within the pharmaceutical industry is discussed as the industry moves from a traditional small-molecule model to a more diverse portfolio. A new generation of high-resolution mass spectrometers and ion mobility mass spectrometers operating as orthogonal separation techniques has greatly increased the ability to resolve impurities and increase the level of knowledge gained from a single experiment. The continued impact and innovation of gas chromatography–mass spectrometry (GC–MS) in late-stage development is also discussed.

The introduction of small, compact mass spectrometers has widened the potential uses for this technique (1) These mass spectrometers may be considered as cheaper options for open access systems, and are used as supplementary and complementary detectors to UV for peak tracking and forced degradation studies, or as quantitative detectors for potentially mutagenic impurities, or for analytes without chromophores. The use of mass spectrometry (MS) to confirm the identity of an impurity during (accelerated) stability analysis and route development activities gives the analyst greater confidence in the data, and potentially highlights issues earlier than when using UV detection alone (for example, for the identification of coeluting peaks). The smaller size of these systems makes it much easier to take the mass spectrometer to the sample, for example, for on-line reaction monitoring (2); this has enabled self-optimizing routines to be used where the mass spectrometer is identifying when optimum conditions are reached (2,3).

Recent years have seen an increase in the use of different separation techniques, moving from traditional reversed-phase high performance liquid chromatography (HPLC) and gas chromatography (GC) to ultrahigh-pressure liquid chromatography (UHPLC) with shorter run times, hydrophilic interaction liquid chromatography (HILIC), supercritical fluid chromatography (SFC), and ion chromatography (IC). These can be a challenge to the mass spectrometer as a result of the need for faster scan speeds or issues with interfacing. In SFC–MS, the pressure reduces as the eluent leaves the column, the CO₂ can potentially boil off, and analytes can potentially precipitate. To overcome these challenges, the eluent flow can be split before the back pressure regulator, or the eluent can be mixed with a solvent miscible with CO₂. The use of a back pressure regulator alone can compromise the chromatographic integrity (4). SFC–MS has been shown to be applicable to a wide range of pharmaceutical compounds (5), including analysis from dosage forms (6), for chiral analysis (7), and preparative chromatography (8). SFC–MS has also been operated as an open access system in support of an academic MS facility (9). Capillary electrophoresis (CE)–MS has also been shown to have advantages in some instances (10).

The range and capability of mass analyzers available has continued to evolve. An increased number of these systems are capable of high mass resolution; as resolution increases, the mass accuracy and specificity increases such that it becomes easier to make structural assignments. The high resolution also offers an alternative to more traditional MS/MS experiments for quantitative analysis, where the specificity is gained by removing nominally isobaric impurities through mass resolution rather than the formation of different fragment ions (11). The robustness of modern analyzers and their ease of use has to some extent moved the operation of these instruments from MS specialists into the hands of analytical scientists.

The potential for application of ion mobility-mass spectrometry (IM-MS) within the pharmaceutical industry was first demonstrated by Eckers and co-workers in 2007 (12). The use of collisional cross-section (CCS) as an additional characteristic of an impurity, in addition to its retention and molecular weight, has significant potential as a tool within the pharmaceutical industry (Figure 1, reference [13]). The peer-reviewed literature contains abundant examples from academic research groups of the application of many different types of ion mobility techniques interfaced to MS for pharmaceutical analysis. The potential impact of the technology is illustrated by the 2018 review by Iain Campuzano and Jennifer Lippens (14), which discusses innovations in ion mobility technology and how they have been applied within research in the pharmaceutical industry. The review outlines the theory of different ion mobility technologies and describes applications to small molecules, metabolites, lipids, peptides, proteomics, proteins, and antibody–drug conjugates (ADCs). The authors note and reflect that ion mobility has seen broad acceptance and adoption within the academic community. However, within the pharmaceutical industry, it is still seen as a niche and specialist technique, which is reflected in its slower uptake and the resulting limited examples of applications originating from industrial research within the peer-reviewed literature.

Figure 1: (a) Conformation of cefpodoxime proxetil, obtained through molecular modeling, which had a theoretical CCS value 0.65%, different to that of lower experimental CCS value; (b) bimodal arrival time distribution of cefpodoxime proxetil annotated with the experimental CCS values; (c) conformation of cefpodoxime proxetil, obtained through molecular modeling, which had a theoretical CCS value 0.97%, different to that of the higher experiment CCS value. Adapted with permission from Hines et al., Anal. Chem. 89, 9023 (2017), copyright 2017 American Chemical Society (13).

An area of particular interest in the pharmaceutical industry is enantiomeric analysis of small molecules and this has been explored by IMS-MS. A recent example is the publication by Donald and co-workers, where differential ion mobility spectrometry (DMS)–MS was explored for the rapid and quantitative chiral recognition of small molecules (tryptophan and phenylalanine) using a chiral selector (N-tert-butoxycarbonyl-O-benzyl-L-serine [BBS]) that formed proton bound diastereomeric complex ions (15). The formation of gas-phase charge isomers (protomers) has been shown by Sobott and co-workers to be an additional complication during ion mobility analysis because multiple peaks are observed for the same molecule (16); this has also been observed by Hines and associates (13).

The biggest challenge to the analytical chemist or MS specialist working in late-stage pharmaceutical development is the now immense diversity of molecular entities that are being developed as drug molecules, with a notable shift towards larger molecules (17); these may be peptides, oligonucleotides, or drug delivery systems such as ADCs. This shift can require adoption of new techniques or a retraining in old techniques that have to some degree fallen out of favor (CE and size-exclusion chromatography [SEC], for example). These molecules provide challenges, especially around the identification and quantification of impurities. For example, CE–MS has shown some complementarity with LC–MS for the analysis of peptides through orthogonal separation (18).

Oligonucleotides present a particular challenge as a result of the large number of chiral isomers. The complex structure and multistep synthesis and purification lead to a broad range of impurities such as N - 1 and N + 1 shortmers and longmers where the impurities have either one less or one more nucleotide (and the similarity between the main component and the impurities). The separation of these molecules are typically based around ion-pair chromatography (19,20), but the presence of coeluting impurities means that MS is used to quantify the purity of the main peak. The importance of therapeutic oligonucleotides is clearly reflected in their increasing prevalence within the peer-reviewed literature. The potential impact of oligonucleotides was illustrated in the 2011 review paper by Niessen and van Dongen, which discussed bioanalytical LC–MS of therapeutic oligonucleotides (21). This review recognized the increasing importance of LC–MS to characterize the parent oligonucleotide and its metabolites in biological fluids. The extensive review covers many of the key aspects of LC–MS of oligonucleotides, including chromatographic retention, ionization efficiency, ion-pair chromatography, pH, organic modifiers, the distribution of multiple charges, and fragmentation efficiency. Bartlett and co-workers have been notably active and this is reflected in two recent publications. A review published in 2018 focuses on the application of chromatographic techniques (including ion-pair reversed phase–HPLC–MS) for the determination of a broad range of oligonucleotide impurities and degradation products (22). The review also describes in detail the vast range of impurities and their synthetic origin. The importance of the characterization of the impurities and understanding their origin in the context of both process optimization and design of commercial synthetic processes is highlighted. In addition to this thorough review, Bartlett and associates have also recently described the application of IP–reversed-phase LC–MS/MS for the in-depth characterization of the degradation products formed from four different antisense oligonucleotides under stressed conditions (different pH values and temperatures) (23). There have been a number of recent examples of research in the area of oligonucleotide characterization originating directly for the pharmaceutical industry. Smith and Beck at GlaxoSmithKline described the application of LC–MS and 31P NMR to quantify a low-level coeluting impurity in a modified oligonucleotide (24), and Breda and co-workers at Aptuit have published a validated (10–10000 ng/mL) bioanalytical ion pair LC–MS/MS assay for the quantification of a 13-mer oligonucleotide in rat plasma to support a four-week toxicology study (25). Though less prevalent within drug project portfolios, therapeutic peptides are of increasing interest within analytical science. This has been reflected in the growing market for counterfeit biopharmaceuticals and the impact on analytical science has been investigated by Vanhee and associates (26). Their 2015 paper discusses the analysis of illegal peptide biopharmaceuticals frequently encountered by controlling agencies. It describes the development of a general screening method employing LC–MS/MS for both the identification and quantitation of illegal injectable peptide preparations that covers a range of therapies including oncology. The method was selective for the characterization of 25 different peptides (based on MS/MS fragmentation), and also validated for quantitation according to ISO-17025.

Many peptide separations can require buffers, salts, or additives that render them incompatible with MS. Hao Luo and colleagues at Merck have sought to overcome this challenge by developing two-dimensional (2D)-LC as an online desalting tool to allow peptide identification directly from these MS-unfriendly HPLC methods (27). Their method employs a heart-cutting 2D-LC system coupled to a quadrupole time-of-flight (QTOF)-MS. Fractions separated in the first dimension using an MS-incompatible mobile phase are transferred to the second dimension, where fast desalting with an MS-compatible phase allows subsequent MS characterization of impurities. In a novel method, Gammelgaard and associates have investigated the use of selenium as an elemental label for the quantification of the cell-penetrating 16 amino acid peptide penetratin (28). Using the labeling method in combination with flow injection combined with inductively coupled plasma–mass spectrometry (ICP-MS) (for total Se), LC–ICP-MS (for quantitative peptide uptake), and liquid chromatography–electrospray ionization-mass spectrometry (LC–ESI-MS) (for the characterization of degradation products) provided detailed information of the peptide cellular uptake.

Another class of compound that is becoming increasingly prevalent is the ADC. The challenges involved in the mass spectrometric analysis of these compounds have been investigated by Friese and co-workers (29). For characterization of ADCs, Cianferani and colleagues have described a proof of concept study on the application of an on-line four-dimensional hydrophobic interaction chromatography (HIC)×SEC×ion mobility-mass spectrometry (IM-MS) methodology (Figure 2). The approach allows several critical quality attributes required for process and formulation development, lot characterization, and stability testing to be monitored in a single analysis (30).

Figure 2: Flow chart of the analysis from brentuximab vedotin. Adapted with permission from Ehkirch and associates, Anal. Chem. 90, 1578 (2018), copyright 2018 American Chemical Society (30).

Polymeric materials have long played an important role in the pharmaceutical industry, for example as excipients in oral solid-dose drug product formulations. Fiebig and colleagues from Boehringer Ingelheim have taken a novel approach to characterizing the regularly used formulation constituents, polyethylene glycol 400 and polysorbate 80. Their publication describes the application of traveling wave ion mobility spectrometry (TW-IMS) quadrupole time-of-flight high resolution mass spectrometer (QTOF-HRMS) and the use of both the collision cross-section and accurate mass for this characterization challenge (31). The methodology was applied to in vivo metabolite studies allowing rapid identification of the formulation constituents.

More recently polymeric materials are being developed as nanocarriers for targeted drug delivery in biomedicine. Examples include nanoparticles that encapsulate an active pharmaceutical ingredient (API) and dendrimer drug conjugates, where a number of API molecules are attached to the surface of a hyperbranched polymer (32). As a result of their relatively recent emergence and novelty, reports on the characterization of dendrimers is limited, however poly(amidoamine) (PAMAM) dendrimers have found some focus, notably by Fernandez-Alba and colleagues in 2013 (33,34). The group have described the application LC–ESI-MS and LC–ESI-MS/MS (using both QTOF and hybrid quadrupole–linear ion trap) to the characterization (accurate mass MS/MS) and quantitation (SRM) of PAMAM dendrimers (generations G0 to G3) in simple aqueous media and more biorelevant urine. The quantitative method was validated and shown to have sensitivity in the micromolar range.

Finally, we should not lose sight that GC–MS remains an essential tool within the pharmaceutical industry for many qualitative and quantitative applications. Continued innovation in GC–MS technology has been demonstrated by the introduction of a number of high-resolution GC–MS systems (35). The authors of this article have themselves demonstrated the capability of GC coupled to an orbital mass spectrometer for structural characterization to deliver process development and understanding (36). Accurate mass GC–electron ionization (EI)-MS and GC–chemical ionization (CI)-MS data were used to characterize key impurities of a synthetic building block for an important drug substance that was under development. Such characterization and impurity tracking of small synthetic building blocks is an essential aspect of process development and design for long-term product quality and patient safety. The quantitative potential of GC with orbital trap MS was also evaluated.

GC–MS plays an important role in the characterization and quantitation of extractables and leachables that may result from devices used within the pharmaceutical industry. GC coupled with HRMS has proved particularly effective in extractable and leachable analysis (37,38).

A recent example of this is the report by Lacorte and associates who have assessed the migration of plasticizers from poly(vinyl chloride) and infusion bags both qualitatively and quantitatively using selective extraction and GC–MS (39). PVC is widely used in the pharmaceutical industry for the manufacture of a wide range of medical devices, including tubes, probes, bags, and primary packaging. Therefore, the characterization of the migration potential of plastic additives (for example, phthalates, various phenols, and benzophenone) is of great importance in the context of patient safety and adherence to international regulations.

Summary

The use of mass spectrometry in all areas of the pharmaceutical industry has increased markedly over the last ten years as instruments become smaller and cheaper, or smaller with increased resolution. The changes in the project portfolios across the pharmaceutical industry with novel (larger) molecules and complex drug delivery devices means that there are many challenges where mass spectrometry will be the analytical technology of choice. However, there is also a requirement to shift to differing separation techniques in front of the mass spectrometer or for ion mobility mass spectrometry, after the ionization has occurred. It is clear that mass spectrometry coupled to a wide range of separation technologies continues to play an essential role throughout the pharmaceutical industry, from discovery to development, to supporting a long-term supply of essential medicines to patients. The continuing evolution of MS technologies will only further strengthen the future impact and importance of MS in the pharmaceutical industry. LC–MS is still a predominant technique and its impact will not only continue, but will be enhanced over the coming years.

References

(1) D.T. Snyder, C.J. Pulliam, Z. Ouyang, and R.G. Cooks, Anal. Chem. 88, 2–29 (2016).

(2) D.E. Fitzpatrick, C. Battilocchio, and S.V. Ley, Org. Process Res. Dev. 20, 386–394 (2016).

(3) N. Holmes, R. Akien, R.J.D. Savage, C. Stanetty, I.R. Baxendale, A.J. Blacker, B.A. Taylor, R.L. Woodward, R.E. Meadows, and R.A. Bourne, React. Chem. Eng. 1, 96–100 (2016).

(4) D. Guillarme, V. Desfontaine, S. Heinisch, and J.L. Veuthey, J. Chromatogr. A 1083, 1578–1586 (2018).

(5) J.D. Pinkston, D. Wen, K.L. Morand, D.A. Tirey, and D.T. Stanton, Anal. Chem. 78, 7467–7472 (2006).

(6) P. Prajapati and Y.K. Agrawal, Anal. Methods 8, 4895–4902 (2016).

(7) K. De Klerck, D. Mangelings, and Y. Vander Heyden, J. Pharm. Biomed. Anal. 69, 77–92 (2012).

(8) D. Spekbrouck and E. Lipka, J. Chromatogr. A 1467, 33–55 (2016).

(9) J. Herniman and J. Langley, Rapid Commun. Mass Spectrom. 30, 1811–1817 (2016).

(10) M. Pioch, S.C. Bunz, and C. Neusüss, Electrophoresis 33, 1517–1530 (2012).

(11) http://www.chromatographyonline.com/meeting-challenges-bioanalysis-time-flight-high-resolution-mass-spectrometry-tof-hrms-0

(12) C. Eckers, A.M. Laures, K. Giles, H. Major, and S. Pringle, Rapid Commun. Mass Spectrom. 21, 1255–63 (2007).

(13) K.M. Hines, D.H. Ross, K.L. Davidson, M.F. Bush, and L. Xu, Anal. Chem. 89, 9023–9030 (2017).

(14) I.D. Campuzano and J.L. Lippens, Current Opinion Chem. Biol. 42, 147–159 (2018).

(15) J. Zhang, J. Diana, K.M. Mohibul Kabir, H.E. Lee, and W.A. Donald, Int. J. Mass Spectrom. 428, 1–7 (2018).

(16) J. Boschmans, F. Lemière, and F. Sobott, J. Chromatogr. A 1490, 80–88 (2017).

(17) A. Gautam and X. Pan, Drug Discov. Today 21, 379–384 (2016).

(18) J. Klein, T. Papadopoulos, H. Miscak, and W. Mullen, Electrophoresis 35, 1060–1064 (2014).

(19) L. Gong and J.S.O. McCullagh, Rapid Commun. Mass Spectrom. 28, 339–350 (2014).

(20) L. Gong, Rapid Commun. Mass Spectrom. 29, 2402–2410 (2015).

(21) W.D. van Dongen and W.M.A. Niessen, Bioanalysis 3, 541–564 (2011).

(22) N.M. El Zahar, N. Magdy, A.M. El-Kosasy, and M.G. Bartlett, Biomed. Chromatogr. 32, e4088 (2018).

(23) N.M. El Zahar, N. Magdy, A.M. El-Kosasy, and M.G. Bartlett, Anal. Bioanal. Chem. 410, 3375–3384 (2018).

(24) M. Smith and T. Beck, J. Pharm. Biomed. Anal. 118, 34–40 (2016).

(25) S. Franzoni, A. Vezzelli, A. Turtoro, L. Solazzo, A. Greco, P. Tassone, M.T. Di Martino, and M. Breda, J. Pharm. Biomed. Anal. 150, 300–307 (2018).

(26) C. Vanhee, S. Janvier, B. Desmedt, G. Moens, E. Deconinck, J.O. De Beer, and P. Courselle, Talanta 142, 1–10 (2015).

(27) H. Luo, W. Zhong, J. Yang, P. Zhuang, F. Meng, J. Caldwell, B. Mao, and C.J. Welch, J. Pharm. Biomed. Anal. 137, 139–145 (2017).

(28) L.H. Moeller, C. Gabel-Jensen, H. Franzyk, J.S. Bahnsen, S. Sturup, and B. Gammelgaard, Metallomics 6, 1639–1647 (2014).

(29) O.V. Friese, J.N. Smith, P.W. Brown, and J.C. Rouse, MAbs 10, 335–345 (2018).

(30) A. Ehkirch, V. D'Atri, F. Rouviere, O. Hernandez-Alba, A. Goyon, O. Colas, M. Sarrut, A. Beck, D. Guillarme, S. Heinisch, and S. Cianferani, Anal. Chem. 90, 1578–1586 (2018).

(31) L. Fiebig and R. Laux, Int. J. Ion Mobil. Spectrom. 19, 131–137 (2016).

(32) N. Larson and H. Ghandehari, Chem. Mater. 24, 840–853 (2012).

(33) M.M. Ulaszewska, M.D. Hernando, A. Ucles Moreno, A. Valverde Garcia, E. Garcia Calvo, and A.R. Fernandez-Alba, Rapid Comm. Mass Spectrom. 27, 747–762 (2013).

(34) A. Ucles, B. Martinez, J. Maria, M.M. Ulaszewska, M.D. Hernando, C. Ferrer, and A.R. Fernandez-Alba, Rapid Comm. Mass Spectrom. 27, 2519–2529 (2013).

(35) I. Špánik and A. Machyňáková, J. Sep, Sci. 41(1), 163–179 (2018).

(36) S. Baldwin, T. Bristow, A. Ray, K. Rome, N. Sanderson, M. Sims, C. Cojocariu, and P. Silcock, Rapid Comm. Mass Spectrom. 30, 873 (2016).

(37) https://www.agilent.com/cs/library/applications/5991-8198EN.pdf

(38) https://assets.thermofisher.com/TFS-Assets/CMD/posters/PN-64741-Leachable-Impurities-ASMS2016-PN64741-EN.pdf

(39) Z. Haned, S. Moulay, and S.J. Lacorte, J. Pharm. Biomed. Anal. 156, 80–87 (2018).

Tony Bristow is the Principal Scientist for Measurement Science (Chemical Development) within Pharmaceutical Technology and Development at AstraZeneca, in Stockport, United Kingdom. Andrew Ray is the Associate Principal Scientist in Mass Spectrometry within Pharmaceutical Technology and Development at AstraZeneca, in Hurdsfild, United Kingdom. Direct correspondence to: Anthony.Bristow@astrazeneca.com