Trends in Pharmaceutical Analysis: A Technology Forum

LCGC spoke with experts in the pharmaceutical industry about current and emerging trends in pharmaceutical analysis, including the use of LC-MS instead of LC-UV for routine assays, best practices for impurity profiling, and areas where commonly used methods are likely to improve. Participants in this technology forum include Ann Van Schepdael, a professor at the KU Leuven in Leuven, Belgium, Tom van Wijk, a senior scientist at Abbott Healthcare BV in Weesp, the Netherlands, and Harm Niederlander, a project leader at Synthon Biopharmaceuticals in Nijmegen, the Netherlands.

Has there been any significant adoption of liquid chromatography coupled to mass spectrometry (LC–MS) for routine pharmaceutical analyses? Or is liquid chromatography–ultraviolet (LC–UV) still applied more often for routine assays and quantitative analysis?

Ann Van Schepdael: Many monographs still use UV as a detection technique and LC–UV for assays and related substances. LC–UV equipment is affordable and robust, and the available column chemistries allow the analyst to play with the selectivity of the system. LC–MS may also be in use in the industry on a routine basis, but it appears less in pharmacopoeial texts. It seems that LC–MS is very important for the preparation of regulatory files for a new chemical entity (NCE): It is significant for the structural characterization of unknown impurities on the one hand, and for quantitation of the drug and its metabolites in biological samples on the other. This is because of its better sensitivity and very good selectivity. The study of a drug’s pharmacokinetics (metabolite characterization, quantitation of excretion, kinetics of metabolism, and drug interactions) is quite well supported by LC–MS.

Tom van Wijk: LC–MS plays a crucial role in pharmaceutical analysis, but in contrast to biopharmaceutical analysis, LC–MS is hardly used for routine analysis. Although it is technically feasible to quantitate known impurities with variations within current requirements, the use of LC–MS for routine testing in pharmaceutical analysis is generally avoided. Aside from the fact that more technical details need to be in place, in general a number of non-technical issues, such as cost of analysis, transferability, and obtaining and maintaining sufficient knowledge levels, put off running this type of method.Quantitative LC–UV–MS methods are used in early development in cases where specific and sensitive detection is required; for example, for the analysis of low level genotoxic impurities or impurities without a chromophore. For the latter, often an alternative method will be developed for quality control (QC) purposes. Routine testing of genotoxic impurities in the final product can often be avoided by controlling impurities in the intermediate steps of the process.

Harm Niederlander: To answer this, we need to consider what is routine. LC–MS is used in analytical method development in pharmaceutical analysis more and more. As method development is increasingly becoming an automated task (though still requiring case-by-case expert evaluation), LC–MS can be considered an important tool in “routine analysis”. Furthermore, though less routine, the role of LC–MS in product characterization and structure elucidation of “unknowns” in both pharmaceutical and biopharmaceutical analysis is indispensable. If, however, you are talking about release and stability testing, LC–MS is largely avoided because its application is still not as straightforward as UV detection, for example.

What are the best practices for conducting impurity profiling of drugs? What is the best approach to finding unknown impurities?

AVS: In order to safeguard the quality, safety, and efficacy of medicines, impurity profiling of drugs is paramount. The chemical structure of these impurities is usually very much like that of the API and therefore the separation of API from impurities can be a challenge. As a result, drug producers use methods with the highest possible resolution to study the related substances present in a drug.

When aiming to find all the impurities in a drug, it is advisable to implement various kinds of separation techniques. For instance, LC can have a different separation selectivity from capillary electrophoresis (CE) so applying both techniques yields supplementary and complementary information about a particular sample. Moreover, within chromatography, it is often advisable to use a combination of columns with different selectivities, that is, to apply orthogonal methods.

Another way of finding all the impurities in a sample is to combine a separation technique with a variety of detectors. Each type of detector can highlight different types of compounds because the detector response depends on the chemical structure of the compound. Does it have a UV chromophore? Does it exhibit good ionization in an MS probe? Does it show good conductivity? The answers to these questions point to different detectors.

TvW: The procedure applied strongly depends on the level of knowledge of the drug substance, the phase of development, and the purpose of the impurity profiling study. For well defined processes, applying the defined method of analysis for different batches could be an acceptable profiling approach. However, state-of-the-art profiling would require a different approach, starting with performing a theoretical assessment based on product knowledge and the literature on the synthesis, degradation pathways, and interaction with excipients. Based on the long list of impurities from the assessment, selection of the analytical techniques and methods will be made, taking into account the physical and chemical characteristics of the components, such as presence of chromophores, (calculated) pKa values, and (predicted) ionization in MS. After tuning the methods based on a set of representative components, the data are processed, applying peak picking software to enable any impurities to be found or to compare (differences in) impurity levels. This approach, combining prior knowledge, good quality data, and suitable software, enhances the chance of detecting unknowns and increases the understanding of impurities found. When the information is added to a product knowledge document, it allows the information level to be monitored and estimations to be made at any point of time.

HN: It is not really possible to point out a single procedure that is routinely applicable for impurity profiling. The very diverse nature of (potential) impurities demands significant expertise in the selection of separation and detection techniques to be included in impurity profiling studies. A theoretical expert assessment of impurities that might be expected should therefore always precede any effort of practical profiling. Based on the estimated properties of these potential impurities, a choice can be made for an array of separation and detection techniques to include in such studies. These separation techniques should preferably be selected so as to be orthogonal [that is, relying on significantly differing separation mechanisms; for example, reversed-phase LC (various modes), hydrophilic interaction chromatography (HILIC), CE] to minimize the chances of missing out on impurities that may potentially co-elute, elute without retention, or not elute at all. In addition, for detection, it is desirable to include more than just a single technique (note that no single detection technique is really generic). Finally, after profiling, investigation of mass balance proves to be a versatile tool to estimate if important impurities may have been missed.

What are important topics for research in pharmaceutical analysis of small molecules?

AVS: For small molecules, analysts always aim for an improvement in the sensitivity of the analysis and an improvement in selectivity of the separation technique, as well as an improvement in efficiency and speed. This is why we will most probably witness in the near future more and more methods using ultrahigh-performance liquid chromatography (UHPLC), entering into monographs. Companies can save a lot of time and expense by adopting the newer miniaturized separation techniques, and they have an advantage in doing so from the start, that is, when submitting the regulatory file. They can avoid the time-consuming method transfer and adjustment process needed to transfer a standard LC method to a miniaturized one.

TvW: In general, activities that support impurity profiling offer opportunities for improvement; for example, the improvement of method development strategies, orthogonality of methods and techniques, column selection, prediction of degradation pathways, and interaction with excipients. Analysis of polar components is of special interest as these show little retention in the classic LC–UV methods on C18 columns; investigation into the use of HILIC separation methods have grown in the last few years as a result. Control of potential genotoxic impurities is also an important area for research. In contrast to impurity profiling of regular impurities, which focuses on the detection of any unknowns above a specific threshold, the control of potential genotoxic impurities today fully relies on assessments. Although the impurity threshold for genotoxic impurities is much lower, technical capabilities allow screening for toxic impurities based on their intrinsic reactivity, in addition to the assessment. Although not required, these screening methods have already been developed for alkylation agents and a similar approach would allow other classes of toxic compounds to be screened for. With new EMA (CHMP/SWP/4446/2000; 2013), USP (232/233; pending), and ICH (Q3D; pending) guidance on the horizon for heavy metals, there is a strong increase in work performed in this field, mainly using inductively coupled plasma–mass spectrometry (ICP–MS).

HN: Important areas include: genotoxic impurities; residues of (heavy) metal (catalysts); and process analytical technology.

How are methods used for quality control and characterization of biopharmaceuticals different from those used with small-molecule pharmaceuticals?

AVS: Biopharmaceuticals tend to be more complex in primary and secondary structure. In the past decades we have seen the arrival of various forms of biopharmaceuticals, all with their own specificity. Following the first set of compounds made through genetic engineering, we have seen the coming of monoclonal antibodies and the conjugated forms of biopharmaceuticals, made in order to enhance their pharmacokinetic performance. All of these biopharmaceuticals require proper characterization such as a study of glycosylation patterns and checking for the presence of deamidated products.Thinking about nucleic acid-based materials as oligonucleotides, the determination of their sequence can be done using MS coupled to a separation technique. Purity testing can gain from the combination of different orthogonal techniques, such as ion exchange liquid chromatography and sieving capillary electrophoresis.

For proteins, sequencing techniques and tryptic maps can also perform structure confirmation. But the biopharmaceutical field is in need of techniques that allow quality assessment of intact proteins. The latter are indeed the compounds that will be administered to the patient, and their activity and quality are determined by the structure of the intact protein.

TvW: Biopharmaceuticals cover a wide range of compound classes and, when compared to small molecules, the classification of purity and impurity is not that well defined. Historically, many people working in biopharmaceuticals have a background in small molecules and the ICH Q3A/B guidance may be followed as a way of ensuring quality, however, this may not always be feasible or required. Often “fingerprints” are used for characterization purposes. For biopharmaceuticals, higher reporting and identification levels of impurities are acceptable because of the larger process variation anticipated.

For both small molecule pharmaceuticals and biopharmaceuticals, high-end technology is available and is more often applied as supportive data for product characterization in regulatory filings. For routine quality control analysis, however, the classic methods, such as ELISA and SDS-page for antibodies, are still in place.

HN: Typically, the “purity” of biopharmaceuticals extends beyond the level of identifying or quantifying components that are not the intended active ingredient. Biopharmaceuticals may consist of mixtures of iso-forms and slightly (differently) modified proteins that can all represent (some) activity. Therefore, profiling the composition of these mixtures is an important part of biopharmaceutical analysis in characterization and quality control. Parameters evaluated often include: Folding and association using spectroscopic techniques (circular dichroism, fluorescence); oxidation, deamidation, and N- and C- terminal heterogeneity using typtic peptide mapping; charge heterogeneity using cation exchange chromatography (CEX) or capillary isoelectric focusing (CIEF); and glycosylation using digestion or deglycosylation with reversed-phase LC, anion exchange chromatography (AEC), or matrix-assisted laser desorption–ionization time-of-flight (MALDI-TOF), and receptor assays.

Following on from the fact that drug activity results from the combined effect of many individual contributions, at least one (overall) activity assay (often cell-based) is always included.

Furthermore, the diversity of product- and process-related impurities is generally much wider for biopharmaceuticals than for small-molecule pharmaceuticals. As a result, the number of methods needed to cover all of these is generally much wider too. These include:Product-related impurities: Soluble aggregation is tested using size-exclusion chromatography (SEC); and cleavage, decomposition or proteolysis is tested using SEC, SDS-page, or CE.Process-related impurities: Host cell proteins are tested using immunological techniques; DNA impurities are analysed using real-time polymerase chain reaction (qPCR); and individual generally xenobiotic process additives are analysed using immunological, chromatographic, or spectroscopic techniques.Other: Bioburden or virus-related testing is carried out using compendial techniques; and general parameters are also tested using compendial techniques.Please note that my focus here has primarily been on antibody biopharmaceuticals.

Are spectroscopic techniques (without chromatography) still important for pharmaceutical analysis?

AVS: The answer to this question depends on the purpose of the analysis. What type of information does the analyst aim for? A separation step might not be needed when the compound is present as the only active pharmaceutical ingredient (API) in a pharmaceutical mixture, and is simply dissolved in a simple matrix. In such cases, it may be possible to apply a stand-alone spectroscopic technique.

One example is identity testing of a bulk pharmaceutical ingredient. This is most conveniently performed by infrared (IR) spectroscopy. Indeed, the typical fingerprint region in the IR spectrum allows confirmation of the identity of a compound after comparison with the spectrum of a standard.

When the purpose of an analysis is assaying the medicine, UV testing can be a good choice. Formulations containing a single API can be conveniently analysed with UV spectrophotometry if the excipients do not interfere in the UV absorbance. It may or may not involve a simple sample preparation procedure, and the analysis could be applied routinely because of its simplicity. It is also possible to carry out UV analysis in the form of a flow injection analysis (FIA). In FIA the samples are injected into a flowing stream of liquid that continuously passes through a detector cell. UV measurement in the cell allows very fast and automated analysis of all the samples.

When people are testing for the presence of impurities in medicines, in the majority of cases a separation technique will be implemented. In this case sample preparation may be needed for the analysis of drug products containing the formulated drug in the presence of excipients.

There are also areas in which the use of particular spectroscopic techniques (without chromatography) is emerging, such as quick and initial detection of a counterfeit drug in suspicious medicines. Raman and near infrared (NIR) spectroscopy have shown their strength in this field. These spectroscopic techniques could become as important as UV in the future.

TvW: Separation prior to detection generally results in higher specificity. However, since this is not always required, direct application of spectroscopic techniques is therefore desired in cases where timely and cost-effective analysis is paramount. Direct UV measurement is the preferred detection for dissolution testing. Vibrational spectroscopic techniques such as near infrared (NIR) and Raman are used in process control and for anti-counterfeit analysis where fast and nondestructive analyses are required. Implementation of quality-by-design (QBD) has resulted in a strong increase in the use of these techniques. In addition, spectroscopic techniques can replace visual evaluation or comparison of colour against European Pharmacopoeia reference standards, to make colour assessments more objective. These examples show that new opportunities can still be found for applying direct spectroscopic techniques.

HN: Maybe less so in small molecule pharmaceutical analysis, but in biopharmaceutical analysis, spectroscopic techniques (without chromatography) still have a very important position in product release, stability testing, or characterization. A few examples include: UV absorbance in the content analysis of protein products; absorbance or fluorescence in immune- and cell-based assays (and even in some chemical assays like those that test for free SH groups); and fluorescence and circular dichroism in the characterization of secondary and tertiary structure of proteins.

More on LC–MS and SFC in pharmaceutical analysis

The Power of Liquid Chromatography–Mass Spectrometry in the Characterization of Protein Biopharmaceuticals