Integrating PAT with MES

With all the challenges that the manufacturing industry has had to deal with over the last ten years — growing compliance demands, increased competition and price pressures — it is perhaps not surprising that pharmaceutical firms are increasingly trying to streamline their manufacturing processes to maintain profit margins, speed up the time-to-market, as well as comply with market regulations that are becoming increasingly stringent.

With the advent of process analytical technology (PAT), an initiative promulgated by FDA, pharmaceutical companies are now studying a framework that enables them to improve efficiency while introducing controlled process quality throughout all manufacturing processes.

By emphasizing process understanding, PAT helps pinpoint the sources of process variability with control mechanisms and measurements. Although it represents a groundbreaking shift in focus for the pharmaceutical industry, PAT implementations can only be successful if they are seamlessly integrated with existing manufacturing execution systems (MES). This is the only logical way to guarantee coherent data management and analysis before, during and after the manufacturing process.

Quality, efficiency, compliance

PAT is an innovative measure in that it is centred on simultaneously introducing higher quality into both the product and the manufacturing processes that produces it. It also represents a framework that helps harmonize compliance with regulatory requirements.

Implementing PAT implies a deep scientific understanding of all production processes, acquired by experience of course, but also by using powerful data collection and analysis tools integrated throughout the enterprise. These are among today's most effective solutions for process integration in manufacturing companies and therefore have direct relevance to PAT.

They help close the information gap between enterprise resource planning (ERP) systems and the factory floor, and connect previously separated systems and company departments. MES are indispensable when planning a PAT project because they can contribute to a better understanding of the existing processes needing optimization.

Questions have arisen about FDA's reaction when pharmaceutical companies implement PAT initiatives on existing processes or products, and discover issues or errors in the production environment as a result.

FDA is conciliatory on this potential problem. In its guidance document PAT — A Framework for Innovative Pharmaceutical Development, Manufacturing, and Quality Assurance, it indicates that "The FDA does not intend to inspect research data collected on an existing product for the purpose of evaluating the suitability of an experimental process analyser or other PAT tool. FDA's routine inspection of a firm's manufacturing process that incorporates a PAT tool for research purposes will be based on current regulatory standards, for example, test results from currently approved or acceptable regulatory methods."

An important premise of the PAT initiative is that the quality factor should be present at every stage of the manufacturing process, and even designed into the finished product from the start of its life cycle. This focus on quality, efficiency and compliance has been hailed as a major step forward for pharmaceutical companies and a response to ever-present goals such as reducing high levels of rejects, maximizing profits and enhancing competitive advantage. Historically, managing quality in pharmaceutical manufacturing implied laboratory analyses to check quality after manufacturing the finished product. Many of these inefficiencies are based on cost considerations and a reluctance to change. The disadvantages of this approach are continual process optimization, high levels of rejected products and limited adoption of new technologies.

Potential benefits of PAT

The mass of information available on PAT make it clear that the initiative can generate dramatic benefits, both short- and long-term. Some of these will likely include:

• Shorter production times and faster time-to-market through the use of in-process measurements and controls.

• Reduction in the amount of rejects because of the prevalence of consistent quality checks throughout the manufacturing process.

• Reduction in the instances of human error following increased automation of processes.

With careful planning and budgeting, the acquisition of tools, systems and solutions need not be too onerous over the longer terms and can generate measurable results. They can also show an impressive return on investment (ROI), but it is important to note here that the key to achieving a satisfactory ROI will be watertight data management throughout the enterprise as well as tight integration of all the systems involved.

Interconnecting MES and PAT solutions

Why is there a need for integration with MES? PAT techniques are focused on the online monitoring and control of manufacturing processes within a particular work cell or piece of equipment. Specifically, they include the ability to perform multivariate analyses on critical operating data and to feed back results by adjusting the production process.

This is unquestionably a step in the right direction, but a more holistic PAT approach could produce further gains in efficiency.

For example, electronic batch recording (EBR) systems, which are at the heart of an MES, could steer the overall production processes on a higher level. An EBR system can decide on the process flow within the boundaries of the predefined drug recipe based on the PAT information gathered within a work cell. During the tablet manufacturing process, the need for an additional milling step can be evaluated by an EBR system based on particle size information collected online and delivered by certain items of PAT equipment.

The PAT layer will provide an insight into the process by supplying additional process-critical data, most of which are relevant for batch documentation. The EBR system must be able to filter and present these data in a user friendly form.

The examples above demonstrate the need for tight integration between PAT systems and instrumentation and the MES layer.

With the rapid adoption of open-system architecture in today's business environment, access to data is becoming less of a challenge. Many pharmaceutical companies are using single, integrated platforms that span the automation controller level, batch engines, EBR and recipe management systems, as well as the integration with ERP systems.

With its broad focus on effficiency, PAT encompasses the whole range of process steps from the process unit up to the enterprise level.

However, PAT also needs to be addressed throughout the drug life cycle. PAT data gathered in earlier phases of the formulation process or during clinical manufacturing can help to accelerate the commercial scale-up process.

The reverse procedure can also deliver important efficiencies, especially if knowledge gathered for a certain product throughout the drug life cycle is used as a baseline for the development of new drugs.

Integrated MES systems that support the formulation process, clinical manufacturing, clinical supply processes and commercial production provide an optimal knowledge base for cross-functional integration.

Phased PAT implementation: appraisal and analysis

Understandably, the thought of implementing a closed-loop PAT strategy throughout the global enterprise and for all products can keep pharmaceutical executives awake at night. PAT is far more likely to succeed if a phased approach is adopted — perhaps starting with one product — and if the whole management team has expressed its support for the plan and sees it as a strategic initiative.

The project starts with a detailed appraisal of internal capabilities, assets and resources — human as well as IT and engineering. Many pharmaceutical companies realize during this appraisal that their data management systems will need to be upgraded or even replaced, because it is possible that the volume of data being generated will exceed the capacity of existing systems. In addition, advanced automation tools will be required together with powerful visualization and analysis tools.

Here, too, it is clearly important to tightly integrate PAT and MES functions and systems because of the requirement for data feedback (and, indeed, feed-forward) between chemometry, process control, MES and business systems. The appraisal of internal capabilities and assets will also need to take into account other potential areas for optimization or improvement. These might include:

• the identification of best practices

• identifying critical operating data components that can be reused to cut costs and prevent wastage

• other areas of cost reduction

• key performance indicators

• reliability of equipment.

An overview of potential costs and benefits can be drawn up and aligned to the project plan, especially if the project is initially limited to a single product or production line. Some pharmaceutical companies that have reported on their experience with PAT implementations have noted that the first project could be for a 'robust' product that would benefit from enhanced efficiency rather than dramatically different quality processes.

Subsequent projects can then focus on current products where the improvement of quality processes is a higher priority than manufacturing efficiencies. The various teams involved will then need to analyse current work practices and process chemistry, as well as their current manufacturing techniques and validation methods.

The appraisal phase is followed by a detailed analysis of product information, including historical data, regulatory filings, manufacturing data and information on quality at each stage of the production process.

Real-time parameters can be applied to these data to ascertain whether, for example, it is necessary to combine ingredients for a set length of time, or whether the blending process can be optimized using sensors that measure the fineness of the powder and stop the blending process when the required consistency is achieved.

The results of these analyses will make it clearer to the teams involved — and, of course, to management — what replacements of, or additions to, their existing systems and tools will be necessary within the PAT solution. A powerful collaborative knowledge management system is likely to be a key part of that solution, as well as analytical tools specific to chemical processes, and process and endpoint tools for the real-time monitoring and control of all attributes. It is vital to consistently relate technology investment decisions and requirements back to the goals of the PAT project and how they relate to enterprise goals as a whole. This approach speeds approval procedures and provides a certain discipline for the elaboration of the 'shopping list'.

Long-term outlook

Following the successful completion of a single-product PAT effort, it is much easier to assess the impact of process changes and the initiative can be gradually broadened to include multiple products and processes. Driven from the factory floor and integrated with MES, companies will start to see measurable benefits and might want to envisage the implementation of a sort of 'enterprise-level' PAT initiative.

At this level, the project can start to embrace non-manufacturing processes such as those inherent in clinical trials or R&D. The specialized, dedicated PAT teams from the existing manufacturing and quality areas, perhaps in collaboration with appropriately experienced external consultants, can apply what they have learned to underpin enterprise PAT on a global scale.

Even when applied solely to manufacturing processes, pharmaceutical companies will reap the benefits of the hard work they have put in to the development and implementation of their PAT strategy: improved profits and an enhanced competitive edge, higher and more predictable quality, as well as efficient manufacturing processes and optimal compliance with FDA requirements. Continuous quality improvement becomes a way of life and the business benefits are long-lived — and sustainable.

Martin Dittmer is a life science solution architect at Rockwell Automation, Germany.