PAT & QbD SUPPLEMENT


consistency against the quality specification or applicable monograph and if compliant is assumed to be identical. However, many aspects of material processing in manufacturing rely on the physical properties of the input material, e.g. particle size, bulk density, viscosity or particle morphology, and these details are not always specified in the quality specification. The absence of these critical parameters can result in potentially significant manufacturing events, as they dictate the success of important manufacturing properties such as granulation, powder flow, mixing etc. Resolving this type of issue is problematic as


it may require reworking the raw material at the manufacturer’s expense or a re-negotiation of quality requirement specifications with the supplier based on a new quality standard, including the critical material attributes. This contract may not bear any resemblance to the initial agreement, and may be considerably more expensive, assuming the supplier is able to produce material of the required specification. The alternative is less attractive as the process has to be re-designed to compensate for the new material and define a new set of operating conditions. The initial cost savings in raw material supply is soon outweighed by the cost of remediation.


QbD and design of experiments The ability to understand the product and process parameters through QbD, offers the ability to understand the relationship between critical input variables, (whether physical, material, chemical or engineering) and their effect on the quality specification. This is accomplished by defining the relationship between the input variables and output quality variables in a multi-dimensional design space. As quality is determined by a whole series of


parameters, insufficient understanding of the significance of material attributes may lead to the definition of all aspects as critical, resulting in an extremely complex multivariate model. It is therefore prudent to collate the existing sources of information and refine the list to exclude parameters known to be non-critical. Tools such as Ishikawa (fish-bone) or fault tree diagrams are particularly useful and aid production of a more effective model which readily extracts the significance of and inter-relationship between parameters. To establish a design space, the critical process parameters are related to the quality


14


European Pharmaceutical Review Volume 16 | Issue 3 | 2011


specification by a series of experiments. These are varied in a systematic way with the magnitude of the effect recorded. Once completed, this allows the user to interpolate within the model’s design space and explore conditions not included within the original experimentation and define the range of input parameters that will result in material of acceptable quality. A completed multivariate model provides


the opportunity to run simulations in the design space by feeding the model with a range of input variables from various experimental conditions. The effect on product performance can then be gauged, or alternatively a quality standard can be established retrospectively, by determining the manufacturing conditions required to attain it, e.g. dissolution rate. In many cases, sufficient historical information is present that can be used to validate the design space. Once the behaviour of the formulation is


understood, the method can be validated and appropriate controls put in place to ensure the critical attributes are controlled. The method can then be filed with the regulatory authorities.


“To fully utilise the benefits of PAT the feedback loop must be closed and the


instrumentation reliable”


Once accepted by the regulatory authorities, changes within the design space can be performed without significant regulatory oversight, enabling flexibility to the manufacturing operation.


Risk analysis Risk analysis is a fundamental part of QbD and is the formal process of evaluating and documenting the probability and severity of harm that any action or omission can cause. It is therefore valuable to assess the effect of changing the critical process parameters before embarking on experimentation, using the existing knowledge base or subject matter experts. Applied correctly, a risk analysis can provide


informed decisions and allow for measured risks to be taken through new experimentation, innovation and equipment, which ultimately may benefit the organisation as a whole. The ability to define risks through tools such


as Failure Mode Effect Analysis offers the opportunity to logically describe the assessment of the nature of the risk. As such it provides the ability to openly discuss the risk of a step, and functions most effectively as a collaborative tool, enabling different perspectives and expertise to be formally documented. Once identified and categorised, the risk should be mitigated. Ideally this will entail designing out the risk before implementation into the commercial environment so that it is eliminated completely. Alternatively, appropriate control strategies such as pressure gauges, real-time monitoring or PAT, will continually monitor the process and alert if a critical parameter is trending out of tolerance.


Process Analytical Technology (PAT) In continuous manufacturing or chemical processes, the benefits of QbD can be readily apparent to a manufacturer, but may be restricted by the absence of a rapid measurement system. Such a system can significantly contribute to the continued assurance of the quality of the active substance or final product. For systems that are well understood from engineering or clinical principles, these techniques may be precautionary or used to support product understanding, before being transferred to another opportunity. In this manner, PAT can be used as a risk mitigation technique. For example, the rate-determining step in dissolution may rely on the particle size of the active pharmaceutical ingredient as smaller particles dissolve more quickly because of their larger surface area. An on-line particle sizer integrated with feedback controls can ensure that the particle mill produces material of a consistent size. The output of the mill can be fed into a QbD design space relating the particle size produced from the mill to the resulting quality specification, allowing the ability to improve consistency, reduce variability and meet the ideal target profile. PAT facilitates real-time release and the


application of measurement, analysis and trending in real time ensures that the process is maintained within the range of the defined input parameter. To fully utilise the benefits of PAT the feedback loop must be closed and the instrumentation reliable. Therefore, PAT systems, as a system of control, need to be agile and robust across a design space. This can rule out certain techniques and methodologies e.g


Page 1  |  Page 2  |  Page 3  |  Page 4  |  Page 5  |  Page 6  |  Page 7  |  Page 8  |  Page 9  |  Page 10  |  Page 11  |  Page 12  |  Page 13  |  Page 14  |  Page 15  |  Page 16  |  Page 17  |  Page 18  |  Page 19  |  Page 20  |  Page 21  |  Page 22  |  Page 23  |  Page 24  |  Page 25