Simon Webster

Therapeutic proteins are a powerful and versatile way to treat a wide range of diseases. Commercial products such as Remicade (infliximab), Avastin (bevacizumab), and Humira (adalimumab) treat a wide range of indications, and a significant proportion of drugs currently under development are proteins. When compared to conventional small-molecule drugs, however, the delicate nature of proteins makes them difficult to use as commercial pharmaceutical products. To be a successful pharmaceutical, efficacy alone is not sufficient; the medicine must also be robust enough to retain this efficacy and remain safe for patients during extended periods of storage over multiple years. 

The challenge to produce protein medicines that are stable over extended periods has further increased in recent years due to pressure to move away from lyophilized formulations, which need reconstituting by the patient, toward more convenient liquid formulations, which can be directly injected. In addition, an increasingly competitive commercial landscape means that there are considerable pressures to bring products to market faster and for these products to have longer shelf lives.  


  Proteins in solution can degrade by means of several mechanisms during extended storage, and a common degradation route is aggregation of the protein over time. Recently, this route has been a particular focus for biopharmaceutical developers and regulatory authorities due to increasing concerns that the presence of even relatively small amounts of aggregated protein may lead to unwanted immune responses in patients, thus potentially resulting in adverse reactions or loss of efficacy over time. 

Biopharmaceutical developers have two main strategies at their disposal to create liquid biopharmaceutical formulations with long shelf lives and a resistance to the formation of aggregates: engineering of molecules for aggregation resistance and the inclusion of additives that inhibit the formation of aggregates in the solution (i.e., formulation). There are, however, many possible ways to engineer a protein, as well as many different additives and combinations of additives that could potentially be used. How should a development scientist determine which of these many options will be the most effective at ensuring aggregation resistance during extended storage? One option is to try a range of protein engineering and formulation options, store them in the refrigerator, and come back in three years to measure the level of aggregates formed. Three years, however, is a long time to wait, especially if none of the protein engineering or formulations provide adequate aggregation resistance. What is needed is a rapid means of predicting at the start of the study, with a reasonable level of confidence, which molecules and formulations will resist the formation of aggregates. Such a method could be used by protein engineers and formulation scientists early in the development process to screen many molecules and formulations to identify the best options and ensure those taken forward will ultimately be suitable for use in a commercially viable product. Real-time storage-stability studies of the selected molecules and formulations will still need to be performed to satisfy the regulators, such as FDA and the European Medicines Agency, but early use of a predictive screening tool has the potential to dramatically reduce the risk of such studies failing.

Biopharmaceutical developers have historically performed a range of activities to try to predict and optimize the long-term aggregation stability of therapeutic proteins early in development. The scope and reliability of these efforts, however, have been restricted somewhat by the lack of suitable technologies and a limited understanding of aggregation processes. For example, methods such as differential scanning calorimetry (DSC)--traditionally used to experimentally screen the stability of proteins in a range of buffer compositions--use a relatively large amount of often scarce protein sample, which limits the number of conditions that can be tested. Additionally, DSC only probes one of a number of possible routes to aggregate formation and therefore may not be predictive if an alternative mechanism dominates.

Predicting aggregation during long-term storage


  In an ideal world, a computational tool would be available to predict which molecules and formulations will have optimal stability and how they will behave over extended periods of time. There are a range of computational tools available that aim to help predict the aggregation propensity of proteins. These are undoubtedly useful. Therapeutic proteins, however, are typically large and complex molecules, and industry’s knowledge and understanding of aggregation mechanisms, the effect of the solvent environment on these molecules, and the exact mechanism of action of many excipients remains incomplete. This incomplete knowledge currently limits the practical use of such computational tools. 

Empirical methods that can be applied early in the development process and that will provide the information necessary to make predictions about long-term storage stability must, therefore, be used. A fundamental question is: what experimental measurements of candidate molecules and formulations will, for example, predict the level of aggregation in solution after multiple years stored at 4 °C? Unfortunately for the biopharmaceutical development scientist, there is currently not a single, unambiguous answer to that question. There are, however, a number of attempts to tackle the problem reported in the scientific literature. 

. from the University of Delaware reviews the state of the art in 2008 and reveals that the science of predicting the long-term storage stability of proteins was something of a work in progress at the time (1). Despite some progress in the intervening years, it remains so today. A number of more recent publications, however, have tackled this difficult question and proposed some answers. This article presents a short review of some of the most notable proposed approaches to predicting long-term aggregation propensity of protein solutions.

Aggregate formation route determines predictive measurements


  There are a number of routes by which monomer proteins in solution can come together to form aggregates and, to make accurate predictions of long-term aggregation behavior, measurements must be found that probe in some way the crucial steps along this pathway to aggregation. Two pathways by which protein molecules can aggregate are illustrated schematically in 

 In the first pathway, some of the proteins in solution unfold either partially or fully so that aggregation-competent regions, such as hydrophobic residues, are exposed and cause the proteins to stick together--this is described as “non-native aggregation.” In the second pathway, the protein molecules retain their correctly folded, native conformation, but have aggregation-competent regions on their surface, such as localized charged regions or hydrophobic patches, that cause the proteins to stick together and aggregate. The aggregation rate-limiting steps for these two different pathways are quite different, and, therefore, different experimentally measurable parameters may be potentially useful, depending on which pathway is dominant for a particular molecule or formulation. 

  Lack of conformational stability in aggregation


  First consider the case of non-native aggregation processes, in which the conformational stability of the protein may play a role. One of the most widely used, experimentally determined parameters for screening molecules and formulations for stability is the temperature at which the protein is observed to unfold, which is the protein melting temperature (T

 may be determined experimentally by applying a temperature ramp to the protein in the solvent of interest and identifying the temperature at which the protein unfolds using either calorimetric methods, such as DSC, or spectroscopic methods, such as intrinsic protein fluorescence, as illustrated in 

 has been used as a crucial stability-indicating metric for candidate screening and formulation development by the majority of large biopharmaceutical developers for a number of years, and many reports of its application in high-throughput formulation screening have been presented (1-9). Conceptually, a higher T

 will mean that, in a refrigerator at 4 °C, there will be statistically fewer unfolded protein molecules in solution and, therefore, less chance of the formation of non-native aggregates over time. As a means of predicting long-term storage stability, this may be useful in cases for which the T

 of the protein is particularly low, but for comparing proteins or formulations for which the T

 is relatively high, the number of unfolded proteins predicted by this parameter will be very low indeed. An interesting illustration of both the potential--and the limitations--of the predictive powers of T

 was reported by Goldberg et al. from MedImmune, who observed a correlation between T

 and aggregation stability during storage at 40 °C for two monoclonal antibodies, but not for a third (2). The stability of the third antibody appeared to be predicted by its rate of aggregation when heated to 70 °C. T

is, therefore, a useful prescreening tool to identify particularly conformationally stable or unstable molecules or formulations, but may not, on its own, be predictive of long-term storage stability for all samples. It is worth noting, however, that T

 may be useful in understanding other aspects of protein stability relevant to its use as a biopharmaceutical, such as how the protein will behave at temperatures higher than 4 °C, due to unplanned temperature excursions during storage, the higher temperatures experienced during manufacture, and the elevated temperatures 

 Measurements with the potential to predict long-term storage stability. 

Probes of protein   conformational stability

Fluorescence, differential scanning calorimetry

Fluorescence, 
      circular dichroism (CD)

Static light scattering, self-interaction chromatography

Ammonium sulfate or polyethylene-glycol precipitation with static   light scattering, turbidity

Probes of   combined colloidal and conformational stability 

Aggregation onset temp., temperature at which aggregation rate   exceeds a certain value 

Static light scattering, size-exclusion chromatography­–high-performance   liquid chromatography (SEC–HPLC)

Time-dependant, thermally induced aggregation

Static light scattering, turbidity, SEC–HPLC

High-temperature aggregation kinetics as a predictive tool


An alternative or additional approach to the use of T

 as a predictor of storage stability (summarized in 

) is to observe the kinetics of aggregation when the sample is held at a temperature that accelerates the formation of aggregates. Such measurements essentially probe a combination of the time--dependent rates of unfolding and the aggregation of these proteins. Christopher Roberts and his group at the University of Delaware have demonstrated the value of measuring aggregation rates at elevated temperatures for rapid formulation and candidate screening (1, 10-12), and these authors emphasize that understanding the rate--limiting step in the non-native aggregation pathway is important to predicting behavior at lower temperatures and longer times. One approach to identifying the aggregation rate--limiting step is to try to fit various mechanistic aggregation models to accelerated (elevated temperature) aggregation data to identify the model that fits best. Once this best fit has been established, the selected model may be used to predict behavior at lower temperatures and longer times. Kayser 

 have also demonstrated the application of this approach (13). 

The Delaware group of Roberts et al. have also described a practical method which predicts the aggregation behavior for an immunoglobulin G (IgG) molecule of interest (11, 12). A thermal ramp was applied to the IgG in a range of solution conditions, and the amount of aggregation was monitored as a function of temperature. The authors determined that the temperature at which the rate of increase in aggregate content exceeded a certain set value (obtained in an experiment taking approximately an hour) was a useful predictor of aggregation behavior of the same sample when stored at 40 °C for 40 days. Experimentally, there are a number of possible routes to obtaining this kind of data in a high-throughput screening environment, with optical methods being particularly well suited to this type of application.

 shows typical data obtained using intrinsic protein fluorescence to monitor time--dependent unfolding and static light scattering to monitor the corresponding rate of aggregation. 

Protein-protein interactions as a predictive tool


The experimental approaches described so far largely probe non-native aggregation mechanisms arising due to attractive forces between fully or partially unfolded protein molecules in solution. As discussed earlier, attractive interactions between native proteins in solution can also potentially lead to the formation of aggregates, and it is therefore interesting to measure the strength and nature (attractive or repulsive) of these interactions for candidate proteins or formulations. The resistance to aggregation due to native protein-protein interactions in solution is often referred to as the “colloidal stability” of the protein. A number of experimental methods are available to determine this stability, including self-interaction chromatography and dynamic light scattering. Static light-scattering arguably provides the most accessible and developed method for measuring protein-protein interactions in solution and requires only the protein concentration-dependent light-scattering intensity from the protein of interest in the solution of interest. By combining these data with suitable physical and instrumental constants, one can generate a graph, known as a Debye plot, from which a value called the second virial coefficient (also known as A2 or B22) can be extracted (14). The sign of this value indicates whether the protein-protein interactions are attractive (a negative value) or repulsive (a positive value) while the magnitude of the value indicates the strength of the interaction. Generally, one might expect that molecules or formulations with net repulsive protein-protein interactions will be more resistant to aggregation. A number of studies have integrated use of the second virial coefficient, or its equivalent, into their formulation screening approaches (7, 14-16) and, like T

, it has proven to have a useful predictive value for some proteins, but not for all.

It is also worth noting that the aggregation rates of unfolded monomer proteins will be determined, in part, by the nature of interactions between these species. For example, attractive hydrophobic forces caused by exposure of the hydrophobic residues could potentially be overcome by repulsive electrostatic forces if all the molecules carry a significant net charge.


An alternative approach to investigating the effect of native protein-protein interactions is to determine the solubility of the candidate protein in the solution of interest. An interesting study was recently presented by Banks et al. from Amgen, which experimented with two strategies to formulate an IgG (17). One approach sought to stabilize the conformation of the protein, and the other sought to improve its colloidal stability. The effects of the formulations on conformational stability were evaluated with DSC, and the effect on protein solubility was investigated using ammonium sulfate precipitation. Importantly, the formulations were also stored at 4 °C for nearly a year, and the rate of aggregate formation was measured to directly assess the effectiveness of the formulation in preventing aggregation during long-term storage. The key result of this study was that, for the molecules and formulations studied, the effect of the formulation on the solubility of the protein, rather than the effect on the conformational stability, was key to improving the long-term aggregation resistance of the protein at 4 °C. 

Complications at high protein concentrations


  To be delivered subcutaneously, many therapeutic proteins need to be supplied at very high concentrations, often 100 mg/mL or more. At these high concentrations, the protein molecules are very close to one another, and the nature of the dominant forces acting between them can change from long-range forces to shorter-range forces. This change has the potential to further complicate the process of predicting shelf life because, in some instances, there may be different processes limiting the rate of aggregation for dilute and concentrated solutions. 

An interesting example of this phenomenon was reported by Kumar and colleagues from Abbott, who studied two molecules, a monoclonal antibody (IgG1) and a dual variable domain antibody (DVD-Ig), and obtained a range of analytical data at low and high (up to 150 mg/mL) concentrations (18). The data were correlated with the rate of aggregate formation when the samples were stored at 5 °C for an extended period. The results suggested that, for the monoclonal antibody studied, the second virial coefficient (A2) was a good predictor of aggregation behavior for both the low and high concentration samples. However, for the DVD-Ig molecule, although the second virial coefficient was a good predictor at low protein concentrations, it failed at high concentrations, for which the thermal conformational stability was found to be a better predictor. The authors rationalized this as being due to the interactions leading to aggregation of the IgG at both high and low concentrations and of the DVD-Ig at low concentration being electrostatic and, therefore, relatively long-range in nature. In contrast, at high concentrations of DVD-Ig, short-range, attractive hydrophobic forces became the dominant interactions, leading to aggregation during quiescent storage. The authors hypothesized that, in conditions where the DVD-Ig molecule had low thermal conformational stability, there may be a greater chance of the normally buried hydrophobic residues becoming more exposed. This would enhance attractive hydrophobic interactions and explain the apparent correlation between T


As can be seen from the previous discussions, a number of rapid analytical measurements have been used to successfully predict protein-aggregation behavior during long-term storage at refrigerator temperatures (2-8 °C). Currently, however, no single measurement can be considered predictive of long-term storage stability for all proteins in all formulations and at all protein concentrations. To improve the predictive capabilities of early-stage protein candidate and formulation screening, multiple measurement types are needed to probe different potential pathways to aggregation. These measurements can be used to optimize protein candidates and formulations. In addition, these measurements can be used to investigate the dominant mechanisms that lead to aggregation for a particular molecule, so that strategies to alleviate these may be rationally designed. Measurements probing conformational stability, such as T

 and time-dependent rates of thermal unfolding, as well as measurements probing colloidal stability (e.g., second virial coefficients from static light scattering or solubility from ammonium sulfate precipitation) should all be performed to give a full picture of the mechanisms which may have an impact on long-term storage stability. Some measurements, such as aggregation rates at elevated temperatures approaching the T

 of the protein, might be expected to contain a convolution of information about conformational and colloidal stability (e.g., rates of unfolding and rates of aggregation of unfolded species). 

Conveniently, recent advances in analytical instrumentation mean that many of the measurements described above and in the literature referenced in this article can be rapidly performed in an automated, high-throughput manner using modest protein- sample volumes. Some of these instruments can obtain multiple conformational and colloidal stability parameters from a single sample, further streamlining the process of obtaining the empirical information needed to make accurate predictions. These technologies make it feasible to obtain a wide range of potential predictive parameters routinely and cost-effectively as part of the protein candidate and formulation screening and optimization workflow. Although this still may not provide guaranteed predictions of what will happen to every product after three years in a refrigerator, it should greatly reduce the risk of any unpleasant, and expensive, surprises in the future.

W.F. Weiss IV, T.M. Young, and C.J. Roberts,

R.K. Brummitt, D.P. Nesta, and C.J. Roberts, 


Simon Webster is cofounder and former chief scientific officer of Avacta Analytical, Wetherby, UK, optim@avacta.com.

Articles

Predicting long-term storage stability of a given protein and formulation is desirable for effective screening and optimization early in the development process. Multiple routes to aggregation during storage to suggest that multiple measurement types should be made to probe different aspects of protein behavior.

Predicting Long-Term Storage Stability of Therapeutic Proteins

FDA's process analytical technology (PAT) initiative that asserts "quality cannot be tested into products, it should be built in or should be by design" has generated a large amount of interest in new technologies for pharmaceutical analysis.

 A key aspect of FDA's PAT philosophy is that every stage of the manufacturing process should be monitored and understood so that the end product is right first time, every time. In practice, the implementation of this philosophy will require a range of analytical tools capable of operating throughout the manufacturing process: at-line, on-line and in-line. 

These technologies must be capable of generating detailed, meaningful results in real time, which may be fed back into the control mechanisms of the process to ensure the final product quality is consistently maintained. Near infrared (NIR) spectroscopy has been widely discussed and applied in this role.

 Raman spectroscopy also has great potential in this application and has indeed formed part of FDA's PAT initiative education programme

 and reports are now beginning to appear in the literature discussing Raman's application as a PAT-type tool.

The Raman effect was first observed in 1928 by C.V. Raman

 and has been extensively studied and applied, generating an enormous body of scientific literature. This superior scientific foundation enables a deeper, molecular level understanding of the manufacturing process and gives confidence to both the manufacturer and the regulatory authorities in the value of this technique. 			

 The sharp features of the Raman spectra enable individual constituents in complex mixtures to be identified. Characteristic features of the chemical species of interest can often be identified in isolation from interfering features from other materials allowing simultaneous monitoring of a large number of chemical species and the use of simple, robust univariate prediction models. This is illustrated in Figure 1(a) where a simple model was generated to determine the relative concentrations of two solvents in a mixture. The Raman bands at 840 cm

 are characteristic of solvent A and solvent B respectively, and are sufficiently well resolved to be used without the need for multivariate analysis. The intensity of the Raman bands is expected to be linearly proportional to the concentration of the scattering species and this was found to be the case (Figure 1(b)). 

Additionally, the large amount of chemical information present in a typical Raman spectrum can contribute to a fuller understanding of the process at a molecular level. 

 Figure 1: An example of the use of Raman spectroscopy to determine the component concentrations in a binary solvent mixture. 			

Aqueous samples, slurries, solids, liquids and gases.

 Water has a very strong absorption in NIR and mid-infrared spectroscopy, which often totally obscures the signal from the materials of interest in aqueous environments. The Raman spectrum of water is, however, very weak and generally does not interfere with the spectra of the materials in solution — this is one of Raman spectroscopy's key advantages when compared with infrared (IR) absorption methods. This is illustrated in Figure 2 where an IR absorption spectrum and a Raman spectrum from an aqueous suspension/solution of an active pharmaceutical ingredient (API) are compared. 

Figure 2: Comparison of (a) IR absorption and (b) Raman spectra for an aqueous suspension/solution of an API.  

Although for the IR measurement the optical path length was only ~10–20 microns, an extremely large feature because of water is observed at ~1640 cm

 as well as a number of sharper features because of atmospheric water vapour. By contrast, the Raman spectrum was obtained by focussing through the wall of a sealed cuvette ~0.5 mm into the sample; the Raman band of water, which also lies at ~1640 cm

, is hidden under the signal from the API. 			

 In principle, all that is required to obtain a Raman spectrum is to shine monochromatic laser light on a sample, and collect and spectrally analyse the resulting scattered light. There are many possible configurations to achieve this. Fibre optic delivery of the laser light and collection of the Raman light is particularly suitable for PAT applications. Figure 3 shows a schematic cut-away view of such a probe. Fibres several tens of metres long may be used to connect a central Raman instrument to remote parts of a production line. The sampling (distal) end of the fibre optic probe can be very compact and robust, facilitating easy integration with the production line. A number of commercial manufacturers supply a range of probes for assorted sampling arrangements. 

Figure 3: A cut-away schematic of a fibre optic couple Raman probe system suitable for pharmaceutical PAT applications. 

Commercially available Raman spectrometers designed for PAT-type applications are available with up to 60 fibre optic inputs allowing potentially 60 individual points on the production line to be monitored simultaneously. Use of a suitable optical 'relay' can potentially increase this degree of multiplexing further. 			

It is also worth noting that Raman spectra may readily be obtained through glass windows, or from samples inside sealed-glass containers. Optics are available with a working distance of typically a few centimetres (although the acquisition of Raman spectra at much greater stand off distances has also been demonstrated). The flexibility with which Raman spectroscopy can be implemented also increases the potential for integration with other analytical techniques to perform multiparametric measurements and, therefore, increases the amount of information obtained from the sample.  

Challenges In Applying Raman Spectroscopy

The practical implementation of Raman spectroscopy as a PAT tool also presents a number of technical challenges, many of which stem from the weakness of the Raman effect.  			

Low signal level and long acquisition times.

 For dilute samples, simply obtaining a good signal-to-noise spectrum in a reasonable time may present a challenge; long signal accumulation times may be required and even then, in some circumstances, it may not be possible to acquire a satisfactory spectrum. This limitation can be overcome to some extent by employing a more powerful excitation source and a more sensitive detection technology, but both of these options can be costly. It should be noted, however, that in favourable circumstances excellent quality spectra may be readily acquired in fractions of a second.			

 The cost of these will depend to a large extent on their performance (e.g., the narrowness and stability of the wavelength emitted and the power output). The required performance will very much depend on the specific application. For example, many PAT-type applications may only require a relatively low specification laser. The cost and performance of the laser source is also highly dependant on the exciting wavelength required. The optimum wavelength will again be application specific. The Raman scattering efficiency is inversely proportional to the fourth power of the laser wavelength. A combination of this fact and the spectral response of silicon detectors generally used for Raman spectroscopy means that blue, green or red laser excitation is generally preferred. For many samples, however, problems with sample fluorescence dictates the use of near infrared excitation wavelengths (see later). Obviously, wherever lasers are employed eye safety is of primary concern and this will affect how the technique is integrated within the production line.  

 Sensitive, low-noise detection of the Raman scattered light is required with potential cost implications. As with the lasers discussed earlier, the specification of the detector should match that required by the application. Where plenty of signal is available and only modest spectral resolution is required, relatively inexpensive detectors may be employed. In more challenging applications a high specification cooled charged couple device (CCD) may be required. Such a device can represent a significant proportion of the overall cost of an instrument, although its suitability for simultaneous multiplex measurements (e.g., from 60 or more fibre optics from various parts of the plant) and its potential to improve sensitivity, resolution and speed of measurement may justify this cost for many applications. 

 A particular problem often encountered with 'real world' applications of Raman spectroscopy is that even relatively small amounts of auto-fluorescence from the sample can totally swamp the Raman signal. This problem can be reduced somewhat by the use of NIR excitation. Traditionally, this has been done by employing 1064 nm Nd:YAG lasers together with Fourier Transform Raman spectrometers. However, the relatively low sensitivity of such systems means consequent increases in required laser power and data accumulation times. In recent years, instruments employing ~780 nm excitation have become popular since at this wavelength fluorescence problems are reduced for many applications, yet sensitive multiplex detectors may still be used.  

Alternatively the use of far ultraviolet (UV) excitation also reduces problems because of sample fluorescence. This is demonstrated in Figure 4 where four different excitation wavelengths ranging from the far UV to the NIR have been used to obtain spectra from the same API.  

Figure 4: An example of the effect of different exciting wavelengths on the fluorescence background generated when recording Raman spectra from an API. 

It can be seen that when 488 nm excitation is used a fluorescent background is generated which obscures much of the detail in the Raman spectrum. However, when 782 nm NIR excitation is used, the fluorescent background is reduced revealing a Raman spectrum rich in detail. When 244 nm far UV excitation is used the fluorescence background is once again absent and resonance effects enhance the intensity of certain Raman bands simplifying the spectra and aiding interpretation. 

The optical design of in-line implementations also needs to ensure ambient light from the plant does not enter the Raman spectrometer since even low levels of stray light can obscure the Raman signal. 

This article has outlined some of the general strengths and weaknesses of Raman spectroscopy as a PAT tool. When considering this technique in a particular application these strengths and weaknesses must be taken into account to ensure that Raman spectroscopy is a viable technique. If this is the case then the detailed requirements of the application must be determined so that appropriate instrumentation is used.  			

Factors such as the nature of the material to be studied, its physical state, how the instrumentation will be integrated in the plant and, importantly, what information the spectroscopy is expected to provide play a crucial part in determining the required performance specification of the instrument. This performance specification will determine the complexity and price of the instrument. 

Applying Raman spectroscopy in a manufacturing environment or a regulatory environment places additional demands on the instrumentation over and above those required for typical laboratory-based instrumentation. Other considerations include: 					

Environmental sealing to protect the delicate optical components from dust and moisture in the plant and general robustness/resistance to shock and vibration. 

 Ease of use by production line staff or automated operation. 

 Long-term stability (e.g., optical alignment, spectral calibration and intensity calibration). 

Built-in calibration and validation routines (with reference to American Society of Testing Materials [ASTM] and the National Institute of Standards and Technology [NIST] standards, for example).

 Fibre optic coupling to multiple remote probes.

 Overall performance must be sufficient to satisfy regulatory requirements. 

 Data handled and stored in suitable fashion (e.g., 21 CFR Part 11 compliance). 

 Raman spectroscopy is not yet widely used as an at-line, on-line or in-line analytical technique in pharmaceutical manufacturing. It has, however, been used for some time in a manufacturing support role in other non-regulated industries such as semi-conductors and synthetic polymers, and a range of proprietary instrumentation tailored towards particular PAT-type applications is available.  

Increasing interest in applying the technique in an industrial environment has lead to the instrument manufacturers and standards organizations working towards developing suitable performance verification standards. Indeed the ASTM standards organization in the USA has a committee dedicated to Raman spectroscopy and has a standard for the spectral calibration of Raman instruments. NIST, also in the USA, currently supplies Raman intensity standards and are developing frequency standards.

The practical features of Raman spectroscopy offer great potential for pharmaceutical PAT applications. At the time of writing, however, its actual use is confined to a few niche applications and, generally, it has yet to make a significant transition from the laboratory to the production environment. This has probably been because of, at least in part, the industry's reluctance to adopt the new technology because of a lack of familiarity, concerns about the willingness of the regulatory authorities to accept new technologies, and the traditionally relatively high cost and limited performance of Raman instrumentation. 			

This situation, however, is now changing. FDA's PAT initiative seeks to generate an environment where the implementation of 'new' analytical technologies is positively encouraged (with the obvious proviso that their implementation is scientifically sound). There is also the accompanying prospect of future changes in the regulatory environment that will require far greater application of analytical techniques throughout the manufacturing process. The implications for the pharmaceutical industry are potentially enormous. 

Raman technology has advanced greatly in the last few years giving vastly improved performance, ease of use and general applicability to 'real' world problems. Encouraged by the opportunities that will arise as a result of the PAT initiative, the Raman instrument manufacturers and scientific community are now putting considerable effort to develop the technique as a PAT tool. Indeed, a range of proprietary Raman PAT instrumentation is already available. This increased interest has accelerated the development of Raman technology, increasing performance and reducing costs. 

While not necessarily the best choice for every application in the pharmaceutical industry, a critical appraisal reveals that in certain circumstances Raman spectroscopy possesses a number of compelling practical advantages. This has been demonstrated to good effect, for example in the analysis of pharmaceutical polymorphs. But perhaps Raman spectroscopy's greatest single advantage is its flexibility of implementation and it is this that will surely lead to it becoming an integral tool in the pharmaceutical industry's analytical armoury. 

The first part of this article introduced the basic features of Raman spectroscopy and presented some examples of its application in the pharmaceutical industry. This second part focusses on the technique's application as a PAT tool within the pharmaceutical manufacturing environment. FDA's PAT initiative has provided motivation to explore the application of 'new' analytical technologies to the pharmaceutical manufacturing process and Raman spectroscopy shows great promise. The strengths and weaknesses of the technique as a potential PAT tool are discussed together with some examples of how this works in practice in a pharmaceutical manufacturing environment.

Raman as a PAT Tool

Until recently, Raman spectroscopy was not widely applied in the pharmaceutical industry. However, in the last few years, developments within the pharmaceutical industry, coupled with improvements in Raman instrumentation, have generated renewed interest in the technique. It is already proving to be a valuable tool in the analysis of pharmaceutical polymorphs and now looks destined to find increasing application throughout the pharmaceutical industry. 

After 80 years of maturing in academic laboratories around the world, Raman spectroscopy has finally begun to make a significant impact in the industrial environment, finding particular application in the semiconductor, chemical and synthetic polymer industries. While still not as widely employed in the pharmaceutical industry as its cousins, near- and mid-infrared (IR) absorption spectroscopies, it has recently found a particular niche in the analysis of pharmaceutical polymorphs. Raman spectra have proved highly characteristic of different polymorphic forms of pharmaceutical compounds and the ability to obtain spectra without the need for potentially polymorph changing sample preparation has proved ideal for applications from polymorph screening to patent litigation. This high profile application of Raman spectroscopy has lead to greater acceptance of, and interest in, the technique within the pharmaceutical industry. With the advent of the US Food Drug Administration's (FDA) process analytical technology (PAT) initiative, Raman spectroscopy is increasingly being considered for applications in the pharmaceutical manufacturing environment. 

So that a reader unfamiliar with Raman spectroscopy can begin to determine its suitability to their own particular application, the basic principles and features of the technique are outlined with particular reference to practical application in a pharmaceutical environment. Comparisons are made with IR spectroscopy with which the reader may be more familiar. Some example applications are presented including the detection and identification of pharmaceutical polymorphs. The potential of the technique as a PAT tool will be discussed separately in Part II. 

A large body of excellent literature exists documenting the basic theory and practical application of Raman spectroscopy, some of which are recommended.

 The essential features of the technique are, however, outlined here for the benefit of those readers unfamiliar with the technique. 			

Raman spectroscopy is a form of vibrational spectroscopy and as such is related to near- and mid-IR absorption spectroscopy.

 Similar to the IR absorption techniques, Raman spectroscopy measures the vibrational frequencies of various parts of a molecule. These frequencies depend on both the bond strength and mass of the bound atoms as well as other factors such as intermolecular interactions. The 'pattern' of vibrational frequencies from a molecule is, therefore, highly characteristic of a given molecular species and, for solid samples, of the crystalline arrangement of those molecules. Raman spectra may be readily recorded from gasses, liquids and solids.

 Figure 1 Schematic overview of the basic experimental arrangement to perform Raman spectroscopy.			

While IR spectroscopy is essentially based on illuminating the sample with a broad range of wavelengths of IR light and measuring which are absorbed, a Raman spectrum is obtained by illuminating the sample with a single wavelength of light from a laser and collecting and analysing the resulting scattered light (Figure 1). Briefly, the basic experimental arrangement to perform Raman spectroscopy includes the following steps: 					

To obtain a Raman spectrum the sample is illuminated with light from a laser.

The molecules in the sample 'scatter' the incident laser light. Some of this scattered light contains detailed information about the molecular properties of the sample.

The scattered light is collected and analysed using a Raman spectrometer to produce a Raman spectrum.

Raman spectra are generally clear, well resolved and rich in features facilitating detailed and unambiguous interpretation.

The spectra contain large amounts of information about chemical composition, intra- and intermolecular phenomena, and the longer range structure of a sample.

The wavelength of the exciting light can lie anywhere from the ultraviolet (UV) to the near-IR ([NIR] with the choice of wavelength used being determined by the details of the application). When the scattered light is collected and spectrally analysed it is observed that a relatively large amount of light is scattered at the wavelength of the incident light; this is known as Rayleigh scattered light and in practice is filtered out and discarded. In addition to this, much smaller amounts of scattered light are observed at various wavelengths shifted with respect to the exciting wavelength; this is the Raman scattered light. Closer inspection reveals that the wavelength difference between the exciting light and each of the various components of the scattered light, when converted to units of frequency, corresponds to the frequency of the molecular vibration giving rise to the scattering. 

Another key difference between Raman spectroscopy and its IR cousins is that although all the techniques measure the vibrational frequencies of a molecule, the fundamental physical process giving rise to the effects is different. In practice this means that while some molecular vibrations may be observed in both the IR and the Raman spectra, often vibrations observed using one technique will be extremely weak or totally absent in  the other. This complimentary nature can have practical consequences. For example, the IR absorption spectrum of water has very strong features corresponding to the O-H bending vibration (frequency ~1655 cm

) and the O-H stretch vibration (frequency 3500 cm

), which tend to obscure the features from other compounds that may be present in solution or in a slurry. By contrast, the Raman signal from water is very weak making the technique particularly suitable for measurements from aqueous samples. 

An important feature of the Raman effect is its low efficiency and the consequent very low intensity of scattered light; for every 10

 photons incident on the sample, typically only one Raman scattered photon is produced. This has a range of practical implications ranging from problems as a result of fluorescent interference from the sample to the requirement for sensitive, low noise detectors. 

The features in the Raman spectrum corresponding to the various vibrational modes of a molecule are generally sharp, well resolved and, for many active pharmaceutical compounds, numerous. This is well demonstrated in Figure 2, which compares the spectra of a number of common pharmaceutical compounds. In this example, it is worth noting the large number of sharp, well-resolved peaks and the many differences between the spectra. Spectra were obtained using a Renishaw Raman microscope with 785 nm excitation and 10 s exposure time. A zoomed in view of the 400–1000 cm

 region of the spectrum is shown. Example characteristic features for each of the spectra are highlighted.

Figure 2 Example Raman spectra from some common pharmaceutical compounds.			

. The instrumentation required to perform Raman spectroscopy is, in principal, relatively straightforward. The instrumental details will depend entirely on the functional specification of the instrument, but typical components are described. 

 The sample is illuminated with a laser of a suitable power, wavelength and stability. Examples commonly employed in modern instrumentation include diode pumped solid state lasers (most often 1064 nm and 532 nm, but many other wavelengths are available) and stabilized NIR diode lasers (typically ~780–820 nm) operating at powers of a few tens of mW to ~1 W. These lasers are generally compact, reasonably robust and reliable. 

. The exciting laser light may be delivered to the sample either via a conventional optical system of mirrors and lenses, or via fibre optic cables, which may extend many tens of meters (note that this is not possible with mid-IR absorption spectroscopy because of absorption in the optical fibre). A lens system focusses the exciting light onto or into the sample and collects the resulting Raman scattered light. It is common to couple lab-based Raman spectrometers to optical microscopes so that delivery and collection of light are performed by a high numerical aperture objective. In this arrangement a micrometer sized sampling volume may be achieved, which is particularly suitable for obtaining spectra from small samples, samples embedded in a matrix or to produce high spatial resolution 'Raman maps' such as that shown in Figure 3. Here, the tablet was sectioned using a microtome to reveal the core. An array of Raman spectra was recorded at 13 μm  intervals from a 1.5 mm

area on the sectioned surface of the tablet. The sampling volume for each spectrum was ~5 mm

and the exposure time for each spectrum was 5 s. Some examples of the recorded Raman spectra are shown. The instrument software was then used to look for characteristic features in the recorded spectra and thus, determine the identity of the material at each point. This information was then used to generate the colour coded map illustrating the distribution of each of the tablet ingredients. It is worth noting that much smaller sampling volumes are possible (by an order of magnitude) with Raman spectroscopy than with IR spectroscopy as a result of the shorter excitation wavelengths used (giving a smaller diffraction limited spot) and because of the fact that all the collimated light emitted by a laser radiation can be readily focussed to an intense diffraction limited spot. This is not the case with traditional, nonsynchrotron IR sources.

Figure 3 Example Raman map of tablet core showing the distribution of APIs and the principal excipient.			

Collecting and filtering the scattered light.

 Efficient collection of the scattered light is essential in Raman spectroscopy because of the weakness of the effect. The collection optics must, therefore, be positioned as close to the sample as possible and have as large an aperture as possible to give a large collection angle (i.e., large numerical aperture collection optics should be used). The collected scattered light must then be spectrally filtered to remove the Rayleigh scattered component otherwise this would swamp the Raman signal. Various types of optical narrow band rejection or edge filters are available to perform this function. 

 The collected Raman light must then be spectrally analysed. This may be done using a low stray light, dispersive spectrometer, with a bandpass filter arrangement or a Fourier transform (FT) type spectrometer. A dispersive spectrometer is generally preferred for excitation wavelengths from the UV to ~820 nm while an FT system is generally required for excitation wavelengths at longer than ~820 nm (e.g., 1064 nm Nd:YAG).

 A sensitive, low noise detector is required to detect the Raman scattered light. For UV-NIR excitation slow scan, cooled charged couple device (CCD) cameras are the detector of choice. Such a detector will, however, typically represent the single most costly component of a Raman system. For some applications it may be possible to employ a point detector or a simple array detector which, while providing only very limited spectral information, will be much less expensive. Array detectors suitable for Raman spectroscopy with IR excitation wavelengths have until recently been prohibitively expensive and so a single element detector coupled to an FT spectrometer is generally used.

These, however, are less sensitive than a system comprising a good dispersive spectrometer and a cooled CCD detector. Recent developments in indium phosphide/indium gallium arsenide phosphide array detectors, however, show promise for IR excited Raman spectroscopy.

. The way in which the various components are combined to build a Raman spectroscopic system is critical to the ultimate performance of the instrument. Careful design is required to ensure system reliability, stability and robustness — parameters of vital importance for operation in an industrial environment. 

. The details of the instrument control and data analysis software will be application dependent. Various proprietary software packages are available that provide a basic interface and often quite sophisticated data analysis facilities coupled with facilities to control instrumentation and tailor the software to an individual application. Increasingly, Raman instrument manufacturers are developing software tailored to the needs of the pharmaceutical industry. 

. Various other features may be required of a Raman system when used in specific applications. For example, when operating in a regulatory environment, automated system performance validation to suitable standards and handling of the data (e.g., compliance with 21 CFR Part 11) are desirable. Instrument robustness, stability and good technical support provided by the manufacturer are important for 'mission critical' applications. 

. There are a number of established manufacturers of Raman instrumentation.

 These offer a range of different instruments targeted at different applications including Raman microscopes and instruments designed for industrial and manufacturing applications. Alternatively, a number of companies exist that will design and build custom instrumentation optimized for a particular application. 

 A given solid state pharmaceutical compound can often have different polymorphic forms. This can affect the ease of manufacture, efficacy and stability of the final formulation. Detection and identification of different polymorphic forms is, therefore, important throughout the drug development and manufacturing process, for example: 					

During the drug discovery/development process so that the optimum form for eventual manufacture can be selected. Analytical identification of this form will be an integral part in its patent protection. 

During scale up and manufacture it is important to monitor the material being produced to ensure the drug is of the correct polymorphic form for inclusion in the final product.

Once at market, it may be desirable to determine the form of a competitor's material to help establish whether any patent infringement has occurred. 

The molecules in different polymorphs experience different intermolecular forces resulting in modifications of both the frequency and intensity of some of the modes of vibration observed in the Raman spectrum. In practice, Raman spectra have proven to be very sensitive to the polymorphic form of a material with often substantial spectral differences evident between polymorphs. The sharp, well resolved features typical of pharmaceutical compound spectra also facilitate the identification of more subtle spectral differences which may, for example, indicate the presence of low levels of one polymorph within another. 

In the Ranitidine spectra shown in Figure 4 there are more than 40 clearly resolved peaks in the limited spectral range shown. A zoomed in view of the 900–1400 cm

 region is shown and example characteristic features highlighted for each of the polymorphs.

Figure 4 Example Raman spectra from two different polymorphs of Ranitidine HCl obtained from tableted formulations from different manufacturers.			

Note the large number of sharp, well-resolved peaks and the large number of differences between the two forms demonstrating the ability of Raman spectroscopy to unambiguously discriminate between and identify different polymorphs of the same material. The highlighted characteristic features for each polymorph do not overlap and would thus be potentially well suited to use in a univariate calibration model for the quantitative analysis of mixed samples. 

The spectra were recorded with a Renishaw Raman microscope with 785 nm excitation and 15 s exposure.  The sensitivity of Raman spectroscopy to polymorphic form has been demonstrated for a number of pharmaceutical compounds.

The absence of the need for sample preparation and short spectral acquisition times makes Raman spectroscopy particularly suitable for high throughput identity verification of raw materials. The facility to record spectra through optically transparent containers/windows gives the potential to test materials without opening their containers or in-line as they are fed into the production line. 

. The high spatial resolution (small sampling volume) available when a Raman spectrometer is coupled to an optical microscope enables chemically specific spectral maps of tablet cross sections to be generated.

 These can typically have a spatial resolution of ~0.5–2 μm and the highly characteristic, well resolved nature of Raman spectra enables the simultaneous mapping of several ingredients. An example Raman map of a commercially available analgesic formulation is shown in Figure 3. The spatial distribution of the three active pharmaceutical ingredients and the principal excipient is clearly revealed. These measurements are straightforward to perform and can give information about the homogeneity of mixing on a micron length scale. 

Contaminant identification and cleaning verification.

 The high spatial resolution of a Raman microscope makes it particularly useful for identifying small particulates such as contamination within formulated products or on swabs as part of cleaning verification. 

Raman spectra are rich in information about the molecular nature of a sample. In particular the Raman spectra of crystalline pharmaceutical compounds typically contain many well-resolved features and are highly characteristic of a given material. These spectra have also been shown to be highly sensitive to the different polymorphic forms of pharmaceutical compounds. Little, if any, sample preparation is required to record a Raman spectrum making it particularly suitable for polymorph identification or high throughput raw material testing. Raman spectra may be recorded from very small sample volumes enabling micron-sized particles to be identified or micron spatial resolution chemically specific maps to be generated. 			

Part II of this article will focus on the potential for Raman spectroscopy as a PAT tool. The technique's strengths and weaknesses are discussed regarding its implementation within a pharmaceutical manufacturing environment. 

 is a senior scientist specializing in spectroscopic characterization of pharmaceuticals and biologics and 

 is a senior scientist specializing in spectroscopic and optical instrumentation both at Avacta Ltd, UK.

Modern Raman Spectroscopy: A Practical Approach

 (John Wiley and Sons Ltd, Chichester, UK, 2004). 			

Analytical Applications of Raman Spectroscopy

 (Blackwell Science Inc., Malden, MA, USA, 1999). 

3. I.R. Lewis and H.G.M. Edwards (Editors), 

Handbook of Raman Spectroscopy: From the Research Laboratory to the Process Line

 (Marcel Dekker Inc., New York, NY, USA, 2001). 

Detection and identification of different polymorphic forms is, therefore, important throughout the drug development and manufacturing process.