Forum

Joint pharmaceutical analysis group

Particle characterisation in drug analysis and development

Understanding the physicochemical properties of particles in medicinal products was the subject of a meeting of the Joint Pharmaceutical Analysis Group held at the Royal Pharmaceutical Society’s London headquarters on December 7, 2000

Unanticipated changes in particle size, topography and morphology can have significant consequences for the performance of new medicinal products in vivo. This whole day meeting, organised in collaboration with the Royal Society of Chemistry's particle characterisation group, gave an opportunity to review the status of a number of established and emerging particle measurement techniques, as well as their application to the characterisation of drug substances and a range of drug products. The meeting attracted more than 100 participants, who heard presentations from academic, industrial and regulatory speakers.

Size matters

Mr JOHN SHERWOOD (Astra Zeneca) reviewed the choice of particle sizing procedures and their validation. "Different techniques can and do provide different results,” he said, before going on to emphasise the importance of microscopy as a confirmatory technique. Examples of drugs for which bioavailability, and hence safety and efficacy, were related to particle size included griseofulvin and digoxin. The presence of large particles greater than 50 microns could also affect product quality, such as the content uniformity of tablets and the texture of ointments ("grittiness"). Large particles might also lead to blockage of aerosol devices.

Methods selected might range from sieve measurements to laser diffraction procedures, but above all the final method should be simple, sensitive to any changes that might occur in the sample, reproducible, and robust to minor operational variation. Mr Sherwood also pointed out the importance of considering the time-scale for sampling and measurement compared to that of the manufacturing process. For instance, in-line testing techniques were generally more appropriate for monitoring micronisation of drug substances.

Reviewing a recent Pharmaceutical Analytical Sciences Group position paper, Mr Sherwood stressed the importance of carefully investigating and describing the critical parameters in the test procedure, because the sampling and dispersion technique itself could have a profound effect on the measurement of the particle size distribution.

For example, excessive feed pressures during dispersion of dry particles might lead to micronisation of the sample, while insufficient ultrasonication of suspended particles in liquids might not remove agglomerates.

Summarising, Mr Sherwood concluded that particle size measurement should be considered as an "organic" technique, which grew during product development as more batch-to-batch comparative data were established.

Real-time measurements

Two presentations featured developments in real-time measurement of particulate systems.

Dr MI WANG (University of Leeds) described the application of non-invasive electrical tomographic methods for the in-line measurement of particulate streams and the characterisation of complex multiphase systems. Varying the source direction of the electrical field applied to a system, multiple sensors were employed to detect changes in resistivity, which could then be transformed into three-dimensional images. The technique had been used to characterise the shape and dynamics of particle flow and mixing in systems such as bubble columns and fluidised beds, as well as powder conveyancing systems, leading to improved equipment selection and design, as well as optimum process control.

After briefly discussing the application of X-ray tomography as a high-resolution tool to characterise extruded pastes and granules, Dr Wang stated that tomographic methods offered real opportunities to confirm models of fluid dynamics, leading to superior understanding of particulate processes in the pharmaceutical industry.

Mr CARL SABIN (Malvern Instruments) described the integration of laser diffraction instrumentation into powder processes such as milling, bagging and filtration. In contrast to more traditional static measurement systems, a section of the powder stream was continuously sampled via a Venturi nozzle into the laser optic sample cell, before being returned, uncontaminated, to the manufacturing process. Since size measurements could be made in seconds instead of hours, the increased measurement frequency could have benefits in terms of optimising batch-to-batch repeatability, while improving process economics through reduction of operator, energy and product wastage costs. For example, smoothing of the hopper feed to a mill had been found to remove anomalous increases in power consumption and particle size, in turn extending the potential life of the mill.

Designer particles

The use of supercritical fluid technologies to control particle formation for drug delivery systems was reviewed by Professor PETER YORK (University of Bradford). As the sophistication of new types of drugs and formulations increased, traditional methods for preparing micron-sized particles or smaller became less suitable, often exhibiting particle damage and batch-to-batch variability, as well as being difficult to scale.

Solution enhanced dispersion by supercritical fluids (SEDS) technology had successfully addressed many of these issues. Supercritical carbon dioxide was ideal for use as an antisolvent in this process where, by changing the temperature, pressure and feed rates of drug solution and the supercritical fluid to a coaxial nozzle which exited to a particle formation vessel, drug particle characteristics, such as size and polymorphic form, could be exquisitely controlled.

Professor York then described several important applications. SEDS could consistently produce small, uniform steroid particles, with tight particle size distributions suitable for inhalation. Unlike micronised materials, these particles possessed "clean", undamaged surfaces that confer attractive aerodynamic properties leading to efficient delivery to the lung, as well as having low amorphous content. Professor York also demonstrated how the controlled formation of sulfathiazole and salmeterol polymorphs could be achieved by manipulating simple parameters in their phase diagrams. As a result of these advances, a "spin-off" company has been formed with a small SEDS manufacturing facility capable of manufacturing at scales of up to 20kg per day to current good manufacturing practice standards.

Surface imaging

Dr NIKIN PATEL (Molecular Profiles Ltd) highlighted advances and applications of surface science technology for the in situ chemical and physical characterisation of pharmaceutical materials. Several powerful methods could be employed to produce chemical maps of both organic and inorganic species residing on surfaces, including time-of-flight secondary ion mass spectrometry (ToF-SIMS), X-ray photoelectron spectroscopy (XPS), and vibrational spectroscopy (infrared and Raman).

Dr Patel presented the application of ToF-SIMS to the high resolution imaging of multilayered controlled release pellets, where the cross-sectional distribution of drug could be differentiated clearly from the silicaceous pellet core and the ethylcellulose overcoat. XPS could be used to probe more deeply into surfaces (up to 10nm), and could provide elemental composition data as well as information about the oxidation state and bonding of atoms present. Infrared microscopy, including more sensitive attenuated total reflectance (ATR) microscopy, and Raman microscopy, could also provide valuable complementary chemical composition information. Additional applications included the characterisation of polymer films in biomedical coatings, and the mapping of other drug delivery systems including polymeric or sugar-based microparticulates.

A family of related topographical techniques was then presented, focusing mainly on atomic force microscopy (AFM). This technique was akin to a stylus responding to the track of a gramophone record, except that the highly sensitive AFM probe and cantilever system responded to the variations in topography (at the nanometre scale) or material properties of the sample under investigation. AFM could be used to define crystal morphology and surface roughness, discriminate polymorphs (complemented with microthermal conductivity measurements), and to study in situ crystal dissolution/growth (eg, of aspirin) with chemically functionalised probes.

In conclusion, Dr Patel pointed to the utility of these techniques for problem solving as well as support for research and development, formulation, quality control and patent litigation activities.

Particulates in parenterals

A "steady hand, a clear conscience and an open mind" were the qualities necessary for a microscopist investigating particulate contamination in parenteral products, asserted Mr GARY NICHOLS (Pfizer Global R&D). Microscopical methods were ideally suited to the characterisation and identification of particulates in injectable products, and were often superior to more high-tech approaches. While pharmacopoeial monographs limited the number of subvisible particles, establishing their nature and origin was not a requirement. However, even if particulate levels in a product were low, their presence could forewarn of potential production or storage problems, so their identification could allow the problem to be eliminated at source. The presence of particulates compromised product appearance, indicated a failure of process controls and represented a potential patient safety issue.

Mr Nichols revealed the diversity of particulate sources, including in-going ingredients and the manufacturing environment (eg, fibres, metal flakes, microbiological growth), as well as the product container and closure system itself (eg, from glass delamination in high pH formulations, rubber particles from stoppers). Often formulations were incompatible with the container, as in the case of a lyophile product containing a sulphate drug, which produced an unacceptable haze on reconstitution. Microscopic examination revealed the presence of miniature barite "roses", which were confirmed as barium sulphate using energy dispersive X-ray analysis. Substitution of a low barium glass container removed the haze effectively. Electron microscopy, X-ray diffraction and most microspectroscopic techniques (eg, Fourier transform infrared spectroscopy, mass spectrometry) also provided valuable complementary information. However, "no laboratory should be without a light microscope", concluded Mr Nichols.

— Contributed by Stephen J. Robinson.

Regulatory aspects of particle characterisation

Giving a personal overview of the regulatory requirements for particle characterisation, Dr LESLEY ANDERSON (Medicines Control Agency) emphasised the importance attached to the understanding and control of physicochemical characteristics liable to affect bioavailability by reference to European directives, regulations and guidelines.

Dr Anderson said that particle characterisation could be relevant to all sections of a marketing authorisation dossier, and was related to the physicochemical characteristics of the input ingredients, the presentation of the dosage form and its intended use.

In terms of particle size and polymorphism, particle characterisation could be particularly important in product development. For solid oral dosage forms such as tablets and capsules, particle size might determine the product’s rate of dissolution and could significantly affect the bioavailability of low solubility drugs. In the case of dry powder and metered dose inhaler products, careful consideration had to be given to the particle size of both the active and the drug transporting excipients to ensure the most efficient delivery of small drug particles (typically less than 5 microns) to the lungs. In contrast, intranasal products might require larger drug particles (greater than 10 microns) to ensure suitable nasal deposition.

In addition, common manufacturing operations such as milling, granulation, drying and compression could induce spontaneous polymorphic transformation of crystalline materials, Dr Anderson warned, adding, "Polymorphs may behave like different compounds."

Dr Anderson went on to discuss the selection of test procedures and acceptance criteria for particle size distribution and polymorphism, making extensive reference to the decision trees developed by the International Conference on Harmonisation of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH). Marketing authorisation applicants should not only provide data to demonstrate reproducible batch manufacture to the requisite quality but should also monitor these parameters, where deemed critical to the product's performance, throughout the proposed shelf-life of the product.

A new impactor for aerodynamic particle sizing

Reviewing the development of the "next generation impactor" (NGI) for the characterisation of inhalation drug products, Dr STEVE NICHOLS (Aventis Pharma) said that the site of deposition of drug particles in the respiratory tract depended on their size, shape and density. While larger particles might be deposited in the throat, and very small ones exhaled, particles 2-5 microns in size were generally deposited in the target organ of the lung. In vitro measurement of this "respirable fraction" could be correlated with in vivo measurement of lung deposition using gamma scintigraphy. This made aerodynamic particle size an important parameter for product development.

Because the current workhorse for aerodynamic particle sizing, the Andersen cascade impactor, had not been designed specifically for that purpose, an industry consortium had been formed in 1996 to develop a new cascade impactor with more suitable operational and technical characteristics. In both the Andersen and NGI apparatus, the dose emitted by the device was entrained in a stream of air, which then flowed through a series of stages with decreasing size "cut-offs". Drug collected on each stage could then be assayed, usually by high performance liquid chromatography, and the aerodynamic particle size distribution reconstructed. Among the NGI's novel design features was that drug deposition occurred in cups that were easily removed, singly or as a set. This feature enabled the use of a flat bed design (rather than the traditional "stack"), which better facilitated manual use and automation of the impactor. The aerodynamic properties of the NGI had been optimised to enable operation over a wide range of flow rates and sample loading, with minimal inter-stage losses. Importantly, full manufacturing specifications and calibration data would be available and published.

After extensive inter-company testing, the NGI would be commercially available in early 2001. Summarising, Dr Nichols stated that the consortium was extremely confident that the NGI would not only be practically more efficient and deliver better quality data, but would also be acceptable to regulatory agencies and pharmacopoeias, several of which had been consulted during its development.