OR WAIT null SECS
New systems that combine Raman spectroscopy with automated imaging support the efficient gathering of such data, including information concerning size and shape distributions for individual components within a formulation.
The benefits of nasal and pulmonary drug delivery continue to prompt the development of all types of orally inhaled and nasal drug products (OINDPs), including dry powder inhalers, metered dose inhalers, nebulizers, and nasal sprays. The drug-delivery formulation can be a solution, but is more often a solid suspension or a powder blend that may incorporate multiple APIs and/or excipients. This mixture of ingredients can complicate the gathering of component-specific analytical data, such as the particle-size distribution of an individual API, which directly influences the success of drug delivery. Such information is also typically required by regulatory agencies to support new drug applications (NDAs), abbreviated new drug applications (ANDAs), and to prove bioequivalence.
For suspension-based nasal sprays, FDA requires the particle size of API to be measured before and after actuation when demonstrating bioequivalence (1). Similarly, dry-powder inhaler (DPI) development, whether for an innovator or generic product, necessitates probing the degree of agglomeration of a powder blend to assess dispersion and ensure an appropriate level of formulation understanding. Such requirements also extend into the manufacturing environment to ensure the effective control of parameters such as particle size, which define drug-delivery performance (2).
This article examines the regulatory requirements for API-specific particle-size information, with a focus on new technology that combines Raman spectroscopy with automated imaging. New, automated systems can rapidly generate data for particle size and shape, and enable chemical identification of different components in a formulation.
Regulatory requirements for OINDPs
The widely recognized correlation between the in vivo deposition behavior of OINDP drug particles and their size has resulted in extensive regulatory requirements for particle-size information and, more specifically, particle-size data for any present APIs. In addition, regulators emphasize the importance of detecting foreign particles in OINDP formulations, which is an application that similarly calls for reliable material differentiation. Examining guidance in these areas helps to identify analytical techniques that can help with regulatory submissions, and sets into context the benefits that can be achieved with newer technologies relative to traditional microscopy.
For nasal sprays, the chemistry, manufacturing, and controls (CMC) guidance pertinent to both NDAs and ANDAs, highlights the need to ensure that for suspension-based products particle size data are submitted to "...provide information and data on the presence of large particles, changes in morphology of the drug substance particles, extent of agglomerates, and crystal growth" (3).
Furthermore, for both suspension- and solution-nasal sprays there is a regulatory requirement to develop appropriate acceptance criteria for particulate matter that may come from the constituents of the formulation, or from the device–container and associated closure components. CMC guidance for metered-dose inhalers (MDIs) and DPIs also emphasizes the need to apply techniques (conventionally microscopy) to detect large particles and agglomerates that may define the morphology of either API or excipient, and to detect the presence of foreign particulate matter (2). In addition, the guidance highlights the need to identify and control any morphological changes that occur with time, for formulations where the crystalline form of API has an impact on the properties of the drug product, such as bioavailability, performance, and stability.
Particle size data are widely used to demonstrate bioequivalence in an OINDP, since this can be difficult and relatively costly to do using in vivo testing. Guidance relating to the demonstration of bioequivalence in a suspension-based nasal spray specifically suggests in vitro measurement of the particle size of the API before and after actuation (4). Such data confirm that the delivered particle size is as per the specification defined, ensuring an optimal deposition profile that is unaffected by the drug-delivery process.
Accessing component specific data
The preceding examples highlight the importance of differentiating particles within a formulation to gather information for discrete populations, such as for API, excipient, or foreign matter. This need extends right through from the earliest stages of formulation, where data support a knowledge-led approach to development, and then into manufacturing and quality control to ensure consistent production. Techniques routinely applied within this context include cascade impaction and microscopy.
A multistage cascade impactor divides a sample into a series of sized fractions on the basis of particle inertia. These fractions are then analyzed individually, usually to obtain a value for the mass median aerodynamic diameter (MMAD) of API alone (5). High-performance liquid chromatography (HPLC) is the most frequently applied chemical identification technique used to calculate API mass during this analysis; however, the sample-preparation approach used to perform this action means that specific information relating to particle size and shape is lost.
Issues that can be addressed via visual differentiation of the dose, as seen in many examples of foreign-particle detection, are still widely tackled using conventional microscopy methods. As with cascade impaction, microscopy techniques also share the same drawbacks as manual analysis; they are manually intensive and potentially inaccurate because of operator variability. In addition, microscopy techniques rely on manual interrogation of the sample and can be subjective, which compromises data quality. Finally, microscopy also cannot provide useful information where particles are morphologically indistinguishable.
Integrating automated imaging with spectroscopic methods
Automated imaging has evolved into a reliable and relatively quick method for size and shape analysis, prompting significant interest from the pharmaceutical industry for use in applications that have conventionally been approached using microscopy. With effective dispersion units and standard operating procedures, modern imaging systems can provide relevant size and shape data for both wet and dry samples. Images of thousands of individual particles can be rapidly captured to produce number-based distributions of both size and shape descriptors. By applying sophisticated classification, based on size and shape, these systems can:
As with microscopy techniques, however, automated imaging can only differentiate morphologically dissimilar particles. Developments made to combine imaging with spectroscopic analysis overcome this issue.
Spectroscopy techniques have broad relevance for organic compounds routinely handled in the pharmaceutical industry. Raman spectroscopy, for example, provides comprehensive pharmaceutical entity identification and is a familiar tool for compositional analysis. More sophisticated imaging systems combine Raman spectroscopy capabilities with imaging technology to exploit the synergistic potential of the two techniques, enabling in-depth characterization of pharmaceutical products based on integrated size, shape, and chemical entity measurement. Such systems complement and extend analytical options for OINDPs because they allow for more detailed investigation of the size-fractionated samples than is afforded by HPLC. By probing the collected samples particle by particle, it is possible to determine the particle size distribution of API on each impactor plate, or the state of dispersion of API. Such information is lost within any analysis that requires dissolution. Importantly for the detailed scoping of OINDP performance in line with regulatory requirements, these new systems can boost the speed, accuracy, and robustness of key analyses, as the following study demonstrates.
Case study: Using integrated size, shape, and chemical identity analysis to characterize DPIs
In an experimental study, the size and shape of particles in a commercially available DPI were measured using an automated imaging system with integrated Raman spectroscopy (Morphologi G3–ID, Malvern Instruments) probe to obtain information about the two APIs present in the formulation. During measurement, samples were automatically dispersed on to a metal-coated microscope slide, and size and shape data were then collected by standard operating procedures. From these data alone, it was impossible to identify the two different APIs because of their morphological similarity, so Raman spectral analysis was applied to chemically differentiate the components and to determine the particle size distribution of each API individually.
For OINDPs, it is the sub-ten micron fraction that is typically of most interest because it is particles in this size range (more narrowly the sub-five micron fraction) that will tend to deposit in the lung. Raman spectral analysis targeted only particles in the 1–10 µm size range, which was defined through the application of an appropriate size classification.
Figure 1: A scattergram plot of the correlations scores (lower diagram), developed by referencing against the library spectra (upper diagram), enables the identification of particles as either API 1 or API 2.
A spectral reference library was created by using Raman point spectra of the pure components to provide a basis for identifying the particles of each API in the sample. The spectrum of each individual particle was correlated with that of a library component, and scored according to the closeness of this correlation; a correlation score close to one indicates a high degree of similarity. Figure 1 shows two discrete particle populations differentiated on the basis of chemical composition, and shows how the system allows the exploration of relationships between particle morphology and chemical parameters. This plot graphically illustrates the relative proportions of each type of particle within the sample, showing that there are far more particles of API 2 than of API 1. The associated data can be manipulated to generate a separate particle size distribution for each of the two different APIs by applying appropriate classification criteria defined in terms of the measured chemical parameters (Figure 2).
Figure 2: Overlaid CED distributions and individual particle images confirm the morphological similarity of the two APIs present in the DPI formulation.
The overlaid circular equivalent diameter (CED) distributions of the two APIs shown in Figure 2 suggest they are similar in terms of particle size, while images confirm their comparable shape. However, when the relative proportion of APIs in the 1–10 µm size range is analyzed, the qualitative observation that there is a greater quantity of API 2 in the sample is confirmed (see Figure 3). The ability to chemically differentiate the samples yields information that the proportion of API 2 in the dose is around ten times higher than that of API 1, which is important information when investigating product performance and the precise composition of the dose delivered to the patient.
Figure 3: Chemically differentiating the particles makes it possible to accurately measure the relative proportions of each API within the DPI formulation. The proportion of the particles in API 2 is much greater than those in API 1.
The development, and indeed replication, of OINDP performance is an exacting challenge that relies upon effective control of several parameters, including particle size, which directly influences in vivo deposition and the success or otherwise of drug delivery. In some instances, it is sufficient to measure the particle size of the formulation in its entirety, but in other cases information is required specifically for individual components, usually for any present APIs, but also for the excipient, or to detect contaminants. Regulatory guidance in the area of CMC and demonstrating bioequivalence points specifically to the need for efficient particle differentiation to provide sufficient information for development and to ensure consistent production.
Combining a spectroscopic technique, such as Raman, with automated imaging technology creates a powerful analytical technique for OINDP development that supports the efficient gathering of component-specific information. By integrating size, shape and chemical identity measurement, such systems enable the accurate differentiation of particle populations within a sample. The resulting data readily quantify the relative proportion of different species in a sample, such as the presence of foreign contaminants, or the relative proportions of multiple actives within a size range likely to deposit in the lung, even for morphologically identical species. In addition, these systems permit the generation of size and shape distributions for individual components within a formulation. These capabilities are highly, but not uniquely, valuable to those developing OINDPs as they streamline and accelerate the information gathering required for regulatory submission and subsequent commercialization.
Carl Levoguer is product marketing manager, laser particle sizing and imaging, at Malvern Instruments, Enigma Business Park, Grovewood Road, Malvern, Worcestershire, WR14 1XZ, UK, tel. +44 1684 892 456.
1. FDA, Bioequivalance (BE) and Bioavailability (BA) Studies for Nasal Sprays and Nasal Aerosols for Local Action (Rockville, MD, April 2003).
2. FDA, CMC (Draft—Not for Implementation (CDER, October 1998).
3. FDA, Nasal spray and Inhalation Solution, Suspension and Spray Drug Products—Chemistry, Manufacturing and Controls document (Rockville, MD, July 2002).
4. FDA, Bioavailability and Bioequivalence Studies for Nasal Aerosols and Nasal Sprays for Local Action (Rockville, MD, April 2003).
5. USP 29–NF 24 general Chapter <601>, "Aerosols, Nasal Sprays, Metered Dose Inhalers and Dry Powder Inhalers,"