Unlocking the Vast Potential of PAT in Solid-Dosage Manufacturing

June 2, 2015
Adeline Siew, PhD
Pharmaceutical Technology
Volume 39, Issue 6

PAT holds the key to real-time quality assurance and consistent product quality in pharmaceutical manufacturing.

 

Process analytical technology (PAT) has revolutionized pharmaceutical manufacturing by enabling the design and development of well-understood processes that consistently deliver quality products. The ultimate goal, under the quality-by-design (QbD) initiative, is to build quality into the product rather than traditional batch testing at the end of the manufacturing process. Emil Ciurczak from Doramaxx Consulting; Tim Freeman, managing director of Freeman Technology; and Alon Vaisman, product development manager, process systems at Malvern Instruments, spoke to Pharmaceutical Technology about the benefits of PAT in solid-dosage manufacturing as well as the strategies for PAT implementation.

Key drivers

PharmTech: What are the key drivers for the adoption of PAT in pharmaceutical manufacturing and why is uptake slow within the industry?

Emil Ciurczak (Doramaxx Consulting): The key drivers are and will be the economics of the industry. There is pressure from countries to reduce costs on both generics and proprietaries. New drug introductions in Germany, for example, are allowed to cost more than existing products only if a significant benefit is demonstrated.

The US Congress is bending to pressures from the largest lobby in the US: AARP. They are demanding lower drug prices and, with our new Republican-controlled Congress, cost-cutting will be the order of the day. PAT, well-executed, is the best way to immediately lower cost of goods sold.

Tim Freeman (Freeman Technology): PAT is a fundamentally important aspect of any modern pharmaceutical processing environment where the need for quality and efficient manufacturing is paramount. As the industry develops new products and processes, it is essential to scrutinize the characteristics of the in-process materials throughout the manufacturing cycle to ensure that quality is met and patient safety is guaranteed. This analysis requires suitable tools applied at each step of the manufacturing process. Significant progress has been made in the past 10 years since the introduction of the PAT initiative, but many processes remain poorly understood and un-optimized.

Unlike many other powder processing industries, the pharmaceutical sector is challenged by the need for rigorous validation of its processes and measurement techniques, which introduces additional hurdles to the adoption of in-process monitoring and control. This challenge is further exacerbated when implementing real-time release testing, where measurements are taken in real time and a decision about product quality is almost instantaneous.

Alon Vaisman (Malvern Instruments): One of the key drivers for the adoption of PAT is the growing demand for higher drug quality and efficacy, both from regulators and the public. Within the industry, there is increasing recognition that these demands cannot be met by the traditional approach of relying on final quality-control testing, but rather that a detailed understanding of all aspects of the product and of the associated manufacturing process is required. PAT can provide the information needed to exert and maintain effective process control, and consequently, the necessary assurance of quality and efficacy.

A second driver is the need to bring products through to market more quickly. Time to commercialization has a major and direct impact on the bottom line, both for generic manufacturers (being first to file) and innovators (time to manufacture under patent). More efficient commercialization calls for the use of enabling tools that produce data faster than can be achieved by applying traditional technologies in an analytical laboratory. Using appropriate PAT makes it quicker to implement the design of experiment (DoE) approach associated with QbD, and to fully scope the impact of critical process parameters (CPPs).

Several factors are responsible for the relatively slow uptake of PAT in the pharmaceutical industry despite these powerful drivers. Firstly, there are perceived regulatory hurdles, although it would seem that there is little evidence to support this view. Indeed, FDA is actively encouraging and promoting the use of PAT to demonstrate rigorous understanding and control of the manufacturing process and the parameters impacting product quality. In some instances, there is also a concern that the use of PAT will reveal previously unknown quality issues which will then need to be addressed-the potential downside of developing a greater understanding of process and product performance.

Finally, I would suggest that the scarcity of widely accepted standards for equipment and process design is also an issue, especially against a backdrop of historically low levels of equipment utilization. Almost every PAT implementation is essentially unique, or considered to be so by the end users who also have no clear guidance as to what analytical methods need to be applied in each case. Furthermore, PAT is typically being used to push manufacturing practice towards relatively unfamiliar performance standards. As a result, the introduction of PAT is often associated with demands for customization, increased validation burdens, and substantial engineering efforts on the part of instrument manufacturers and the end users, all of which are barriers to uptake.

Implementing PAT in pharmaceutical processes

PharmTech: What are the main challenges and key considerations when it comes to implementing PAT in pharmaceutical development and manufacturing?

Alon Vaisman (Malvern Instruments): In many cases, the challenges of implementing PAT are linked directly with the reasons outlined for its slow uptake. For example, the lack of standards increases the workload associated with producing a fully fit-for-purpose solution for any given application. Engineered solutions are used to adapt available PAT tools to evolving needs, but in the absence of standards, the cost, time, and most importantly, the risk associated with each solution is higher than it should be.

Key considerations to address when it comes to using PAT include the need to rigorously assess what information is required and the ability of a proposed PAT tool to deliver it reliably. PAT brings value by enabling faster experimentation, more efficient scale-up, and/or by increasing confidence in the quality and stability of the manufacturing process, and ultimately, the end product. However, it will only deliver these benefits by measuring relevant data in a timely and robust way. Scalability is also an important factor to be considered within this context because ideally, the same PAT tool will deliver all of these benefits by transferring with the product, from R&D through scale-up into commercial manufacture.

Finally, method development is an important consideration when it comes to PAT because it affects how the information generated by the PAT device will relate to results measured using established quality control (QC) techniques. Correlations between QC and PAT data can be vital to the acceptance of a PAT tool.

Tim Freeman (Freeman Technology): There are now many examples of spectroscopic techniques being successfully applied to blending, granulation, and drying processes, as well as measurements such as in-line particle sizing and more traditional measurements of properties like temperature and moisture. However, a significant outstanding challenge, particularly within powder processing, is the rationalization of which material properties, for both raw materials and intermediates, are important in determining the quality of the finished product.

Attaining this information is the basis for improved product development and formulation, but also as a driver for the development of new or better process analytical technologies. Whilst there have been a great deal of solutions through the application of in-line spectroscopic techniques and particle sizing methods, there are other attributes of particles, such as density, flow, morphology and permeability, to name a few, that are currently not measured in-line and yet are broadly recognized as being influential in the critical quality attributes (CQAs) of the final product, whether this is weight variation or dissolution of a tablet, or delivery performance of an API from a dry powder inhaler.

Emil Ciurczak (Doramaxx Consulting): Quite simply, lack of experience in a very conservative industry gives managers pause before attempting anything new and not already being done. In other words, it is a lot simpler to convince a director to buy six new high-performance liquid chromatography (HPLC) systems than one new Raman or near infrared (NIR) unit. Adding more people and more equipment always worked in the past, so why not keep doing the same thing that made money before? The instrument manufacturers need to assure the producers of drugs that they will ‘hold hands’ throughout the early attempts at PAT and then follow through by having knowledgeable staff available.

PharmTech: What are the strategies for successful implementation of PAT in pharmaceutical development and manufacturing?

Tim Freeman (Freeman Technology): The implementation of a robust and appropriate PAT toolkit should be based on understanding the material properties and process parameters that are important in influencing the quality of the finished product. Some of these tools will be readily available, such as NIR, and others are still emerging or have yet to be commercialized, such as the ability to measure in-line particle morphology or granule density following a batch granulation process. A sensible approach would, therefore, be one whereby the identification of these variables is first achieved and the selection of available PATs is made.

Bear in mind though, that there may still be important material properties that cannot be measured in real time. Hence, the control strategy for the process needs to account for this absent information, either by retrospective testing or by showing through a QbD approach that such a material property will not vary outside of the acceptable criteria (defined perhaps by off-line measurements) as long as certain process parameters are controlled within a specific predefined range.

Alon Vaisman (Malvern Instruments): Regulatory documents relating to PAT and QbD, which is covered in detail in the International Conference on Harmonization (ICH) guidance, provide a framework for the successful implementation of PAT. The application of QbD involves the systematic identification of all CQAs and CPPs using a risk-based approach, thereby highlighting those variables that can be productively measured using PAT. The next step is to determine where in the process stream the measurement should be made to provide most value. Potential PAT solutions can then be usefully evaluated by considering the following questions:

  • How suitable is the analytical method? Does the PAT measure the parameter that you need it to?

  • Can you trust and validate the results that are generated? Does the method robustly report data that are representative of the process stream and allow you to securely differentiate between poor and acceptable process performance?

  • What calibration procedures are required to ensure data quality? How easy is it to store and use the generated data?

  • Is the technology suitable for its working environment? Are the materials of construction compatible with the process stream and are requirements for cleanliness/avoidance of product contamination met?

  • Does the proposed solution address user requirements? How easy is the hardware and software to use? What routine maintenance is needed?

  • Are any requirements for versatility met? Can the technology be used for more than one product if necessary?

Alongside this technical assessment, it is also essential to conduct a business-value review to determine the economic value that an investment in PAT will deliver to offset its cost. An important part of this review includes rigorous assessment of the cost of implementation-the investment associated with the equipment and any associated installation, qualification, and validation work.

Applying PAT in solid-dosage manufacturing

PharmTech: PAT has been used across many types of operations in the solid-dosage manufacturing process. Can you provide some examples of how PAT has been applied?

Emil Ciurczak (Doramaxx Consulting): In raw materials testing (i.e., every container of every lot), for example, there have been a number of cases where improper particle sizes were delivered. In one specific case, the micronized drug was delivered instead of 100-mesh, granular material; and it would have been a disaster if used for a suppository batch had the problem not been detected by NIR. The United States Pharmacopeia (USP) testing had already approved the lot-sieving only required that there be ‘no less than 1% on a 100 mesh screen.’ There wasn’t an allowance for smaller sizes being sent.

In another example, a granulating/drying process was run to ‘less than 1% moisture’ by Karl Fisher titration. It was discovered that the drug could exist as either a monohydrate or hemihydrate after drying. Both forms were physiologically equivalent, but the hemihydrate had a six-month-shorter stability profile, leading to an earlier recall. This situation is easily controlled by using NIR to control the airflow and temperature.

Extending to clinical studies, errors in packaging (leading to erroneous correlation) can easily be avoided by scanning 100% of the blister packs with NIR or Raman to assure compliance with the protocol non-destructively. Also, HPLC results may take days, while spectroscopic results are immediate.

Alon Vaisman (Malvern Instruments): Examples of PAT from Malvern Instruments that has been used to improve solid- dosage manufacturing include: the Parsum in-line particle sizing probe for the optimization and scale-up of granulation processes and the Insitec online laser-diffraction particle-size analyzer that finds application in milling and spray drying control.

High-shear wet granulation is used routinely to improve the properties of tablet blends ahead of tabletting. Such processes, however, are notoriously difficult to optimize and scale-up because of their sensitivity to small changes in the formulation or in the mixing regime. A QbD approach can be used to develop the knowledge required to minimize and control the risk associated with moving such processes through to commercial manufacture, and real-time PAT is used to facilitate this process. One of the tools deployed by several global pharma companies is a Parsum in-line particle-sizing probe that measures the size of the growing granules in situ within the granulator.

Using real-time particle-sizing data, scientists can confirm that a granulation is proceeding identically at different scales to deliver granules with well-controlled properties. Experimental studies have shown that batches of granules identified as being comparable, on the basis of particle-size data, go on to produce tablets of comparable quality. Real-time particle sizing is thus helping to speed up process development and to enable successful scale-up, while at the same time reducing the number of trial runs needed, thereby saving highly valuable API.

Spiral jet mills are used extensively in the pharmaceutical industry, for example, to control the particle size of active ingredients. Although these mills are seemingly simple in design and easy to operate, micronizing a batch to the right (and often very tight) specification can be challenging due to gradual changes in the performance of the mill and variability of the feed material. Common process optimization by iterative sampling of small sub-batches to find the right parameters and then ‘blind’ processing of the main batch results in a higher risk of failing final QC and of wasting valuable API.

Using an Insitec laser diffraction analyzer to measure the particle size of the material leaving a jet mill, in real-time, makes it possible to quickly assess the impact of milling parameters such as material feed rate and injector pressure. This information can be used to fully scope a milling process or to identify conditions that will produce a specified particle-size distribution. The continuous analysis of mill output enables a timely response to any deviation from the set specifications during the run. As a result, this PAT can save time and money, by reducing the length of trials and the material required to identify an optimal processing setup, as well as by lowering the overall risk of out-of-specification production.

Tim Freeman (Freeman Technology): The enthusiasm for, and the recent adoption of, continuous manufacturing within the pharmaceutical industry has necessitated the successful implementation of a number of PATs across a range of process steps incorporated within the continuous manufacturing of tablets.

Work published by Vertex Pharmaceuticals (1) in recent years shows the use of a GEA Pharma Systems ConsiGma continuous manufacturing suite incorporating spectroscopic characterization methods for measuring water content of product in a dryer and blend uniformity of an intermediate prior to compression. It also features an in-line particle sizing technology in the form of a Malvern Insitec for measurement of particles following the milling process.

Both techniques applied in this appropriate manner provide real-time assessment and the ability to feedback relevant information to permit control of the process should changes need to be made to ensure the product attributes remain within the target specification.

Recent advances in PAT

PharmTech: What recent advances in PAT tools have you seen over the past five years? Or what would you identify as significant advances in PAT for pharmaceutical solid-dosage manufacturing?

Emil Ciurczak (Doramaxx Consulting): I would say without hesitation that continuous manufacturing is the biggest boon to PAT/QbD in the past 20 years. A number of companies are going full bore into this mode of production. Several immediate benefits have been seen:

  • The footprint of a continuous production facility is far smaller than a traditional plant set-up, which means lower land costs, lower heating, ventilation, and air conditioning (HVAC) costs, lower warehouse costs (intermediates need not be stored), and less cleaning costs and time.

  • There is no scale-up. The experimental batches are the production batch size. This fact alone could add 12–18 months to the patent lifetime.

  • Design of experiments, geared to give the design space for QbD and run in a conventional manner, would cost up to 20 times as much as continuous manufacturing and take weeks, whereas in continuous manufacturing this DoE takes days. This approach saves API, cleaning time, allows more experiments, and is run at the level that the final batches will be produced.

  • Finally, with constant monitoring, any out-of-specification (OOS) excursion can be immediately spotted and production halted. The amount of product lost is minimized and costs contained.

Tim Freeman (Freeman Technology): There are a number of emerging technologies within the PAT toolkit that include numerical modelling tools as well as techniques for measuring attributes of the in-process material, such as ribbon density of a product leaving a roll compactor. Certain attributes are amenable to in-line measurements, yet others remain challenging to quantify within the process, such as particle morphology and powder flow properties. Nevertheless, both are important powder attributes that have the potential to influence the characteristics of the finished product.

For example, a blend may be defined as having acceptable content uniformity as it exits a continuous blender, as measured by an NIR probe, and it may have a suitable particle size distribution following a milling process. However, neither of these measurements ensures that the powder has the same particle-shape characteristics as a previous functional ‘batch,’ nor that it has suitable flow properties. Given that both morphology and flow can lead to capping, weight variation, and hardness problems, the absence of this information explains why variation in product CQAs is still observed, even for processes employing a number of PATs. In this circumstance, the ability to predict and to control all of the CQAs has not been achieved.

Whilst there have been many advances in PAT, there are still a significant number of material attributes that need to be understood in regard to their influence on CQAs. Once these relationships are established, the next challenge is to develop on-line measurement techniques that permit the measurement of these additional properties to further interrogate and control the complex processes employed.

Alon Vaisman (Malvern Instruments): Over the past five years, there has been an increasing push for ‘information rich,’ multi-parametric analysis, especially in batch unit operations prone to variability, such as blending and granulation. The term PAT is often synonymous with on- or in-line measurement, because of the value of such instrumentation for process monitoring and control. PAT, however, is actually defined more broadly as ‘a mechanism to design, analyze, and control manufacturing processes…’ (2). So, for example, we are finding that there is appetite to use our Morphologi G3-ID, a laboratory-based instrument, as a PAT, because of the relevant data and insight it is able to provide. The Morphologi G3-ID enables particle size and shape measurement and chemical identification, by combining automated imaging with Raman spectroscopy. It can, therefore, be used to assess, for example, the homogeneity with which an active is distributed within a blend or to assess changes in the size and shape of an active caused by specific processing steps.

A second trend is the drive towards continuous manufacture now that the successful, cost-effective implementation of this approach has been amply demonstrated. Because continuous manufacture calls for analyzers to work together and for automated process control, it has helped to stimulate the development of new software solutions that simplify analyzer integration, such as Malvern Link II. In addition, new instrumentation such as the Parsum IPP80 probe has been developed for easy data transfer and use to enable the more complex control strategies required for continuous manufacture.

References

1. P. Hunter et al., AAPS News Magazine 16 (8) 14–19 (2013).

2. FDA, Guidance for Industry PAT-A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance (Rockville, MD, Sept. 2004).

Article DetailsPharmaceutical Technology
Vol. 39, No. 6
Citation: When referring to this article, please cite it as A Siew, “Unlocking the Vast Potential of PAT in Solid-Dosage Manufacturing,” Pharmaceutical Technology 39 (6) 2015.