Process Analytical Technology (PAT): Measuring Return on Investment

Process Analytical Technology has the potential to dramatically change pharmaceutical manufacturing. But how can companies judge the Return On Investment (ROI) case for PAT? Despite its advantages, many companies remain hesitant about its implementation. The benefits are great but the investment cost is potentially high.

By offering companies the chance to understand and control their processes, both at the R&D and manufacturing stages, PAT enables continuous quality verification and with it, the chance to deliver:

• Consistent quality;
• Lower costs;
• Speed up product development and release;
• Improve market responsiveness;
• Reduce supply chain bottlenecks.

Companies will also be rewarded by a lighter regulatory touch if they can gain PAT-led control of the product design space.

Nonetheless, pharmaceutical manufacturers face a complex and, in some respects, contradictory set of demands. On the one hand they have the opportunity to make significant investments in automation and process technology but on the other hand, they face cost pressures, meaning that such investments must deliver maximum benefits.

Manufacturers face a future drug market that may be more personalized, posing key dilemmas on whether the manufacturing plant development should be on a large or small scale.



Short and Long Term Returns
An article by Johnson and co-authors investigated Multi-Variant Data Analysis (MVDA) applications in the biotech industry. There are four application areas that are typical for PAT in general where MVDA is combined with Design Of Experiment (DOE), and have added value to pharmaceutical production. They are:

• Optimization of large-scale production;
• Establishment of process comparability and trouble-shooting;
• Routine monitoring of manufacturing processes;
• Raw material characterization and screening.

These areas highlight the extent to which companies can start with a small application of PAT on one part of the production process, for example, on a single unit operation such as drying, before moving on to a more global view. A typical starting point, for example, could be the establishment of an end-point detection mechanism for a dryer or granulator. Some of the most common applications of PAT relate to the online monitoring of blending, drying and granulation steps.

In this way, companies are able to deploy PAT to resolve already identified problems in their current manufacturing processes. This identification of clear, current improvement goals is the first of two important yardsticks for the ROI evaluation of PAT.

The second yardstick needs to be a look at the bigger picture - where the company wants its manufacturing to be in the future. Therefore, besides identifying the current manufacturing problems that require improvement, companies should develop a clear manufacturing vision and involve all parts of the business in looking ahead on a 10-15 year time frame.

The approach outlined in the case illustration allows companies to prioritize specific problems within the context of long-term change. The range of specific concerns could include a need to fix or improve existing processes, speed up new product development, reduce site to site transfer risk and times, reduce validation costs or improve quality reliability. Most companies are likely to want to realize a blend of these benefits.

Their immediate priorities will be determined by the current state of play of their manufacturing and its fit with their regulatory, market and business goals. More importantly, they need to combine this review of current wider concerns with the type of longer-term scenario planning outlined in the case illustration.

Figure 1 outlines the steps that companies might take to put this process into practice.



Figure 2 provides an overview of the type of overall decisionmaking process that a company needs to undertake. The current manufacturing infrastructure has to be assessed in the light of the future manufacturing vision (in line with the global company's objectives). What are the current bottlenecks and the possibilities for improvement?



The resulting list of improvement proposals have to be evaluated, to judge what they bring to the company and whether they help to achieve the manufacturing vision and its objectives. Depending on which market the company is in, the regulatory constraints need to be superimposed to make sure that there are no unpleasant surprises. Even for those countries that are actively driving changes (such as the US FDA), it is important to involve the regulators early on in the process.

Four Change Scenarios
The outcome of this type of process will be a view about what type of manufacturing strategy and plant the company needs in the medium to long-term timeframe (5-10 years). The answer may be different from plant to plant and many companies are likely to need to plan for a mix of scenarios.

For example, a company may choose to implement relatively modest improvement investments in a plant to manufacture a product that is nearing the end of its patent period (scenario one in Figure 3).



Elsewhere, it may choose to plan for a rapid and full-scale move to PAT, enabling the full realization of the US FDA's vision of real-time product control and release, based on continuous manufacturing operations (scenario 2 in Figure 3).

Companies will also need to be mindful that a possible trend towards personalized medicines will increase manufacturing complexity and, in turn, pose challenges for Manufacturing Execution Systems (MES) and quality systems. A larger variety of products and a variation of the same products will require greater flexibility in production as well as closer integration along the whole pharmaceutical chain - R&D, manufacturing, sales and the end customer.

Scenarios three and four in Figure 3 highlight how companies will face a choice between big plants with flexible recipe production, versus small-scale development (pilot) plants which will also be production facilities with dedicatedlines.

For both models of production, industrial Information Technology (IT) systems will play a strategic role. This requires flexibility in the first model to support the flexibility of production that will be necessary and, in the second smaller scale model, to link production with continuous development and learning from clinical trials.

The regulatory stance will be a key factor in this mix and at present, regulators are investigating how to support this evolution with the appropriate regulations and guidelines.

A key influence will be on the demand side and it is likely to see a mix of large scale, high throughput facilities, handling generic production and micro-process centers, concentrating on higher end personalized medicines. Pharmaceutical companies therefore need to investigate the investment in planning for a potentially different manufacturing future as well as responding to pressures on their current manufacturing set-up.

Manufacturing Infrastructure
Once they have chosen between the different possible manufacturing visions and have completed some scenario planning, companies will need to decide on the manufacturing and IT infrastructure that will be required for the chosen scenario. One inhibiting factor has been the absence of software that enables the full integration of PAT tools and all information flows during the processing and online comparison of process data with previous or historical data.

A PAT project at the Process Development Laboratory of the Netherlands Vaccine Institute (NVI) was used to develop software that would fill this gap. The software, known as SIPAT, has helped to develop a real-time process verification tool for all critical process attributes of the cultivation process step of the Bordetella pertussis bacteria, used for whooping cough vaccine.

The goal of the project was to develop an alternative process development methodology and advanced process monitoring and control techniques that could lead to the realtime release of the end product without final quality testing.

The software enables the collection of data and the full integration of all information flows during processing and online prediction calculations as well as the comparison of the actual batch trajectory with the "golden batch" trajectory. With SIPAT, the institute can integrate lab models with sophisticated near-infrared and mass spectroscopy measurements and data gathered by sampling, to build a full process model.



Wider Implementation
This technology is being used to develop the process on a two-liter research-scale bioreactor. The knowledge gained will facilitate an up scaling of the process to commercial manufacturing. There are plans to implement PAT both in other cultivation processes and in other unit operations, such as freeze-drying or process chromatography.

The software has now been developed for wider pharmaceutical R&D and manufacturing applications. It gives pharmaceutical companies a common system architecture. It serves as a common user-friendly interface for all PAT tools (process analyzers, multivariate statistical tools, LIMS, MES, process control, historian) and can be fully integrated into the manufacturing and development architecture.

The incorporation of PAT is expected to enable the rapid migration of development of pipeline processes into large scale manufacturing.

Companies can frame a ROI case for PAT implementation on a range of scenarios, ranging from future development and manufacturing facilities, to the fixing of specific snags in current manufacturing processes, such as end-point detection for a dryer or granulator. An effective ROI framework needs to combine a review of both immediate, short-term improvement issues with longer-term scenario planning that is built around a future manufacturing vision. Companies that succeed in bringing together a holistic view of the short- and long-term are more likely to make PAT implementation decisions that deliver a more effective ROI.

----------------------------------------------------------------

----------------------------------------------------------------

Case Illustration: Manufacturing Vision Development

A pharmaceutical company has a product that will soon run out of patent and generic manufacturers are becoming strong competitors. Reducing manufacturing costs has been defined by this pharmaceutical company as a key business objective.

Typical Response
The company decides to appoint a team of experts whose task is to review manufacturing and propose optimization proposals. After a couple of months this team presents the cost reduction initiatives to their management. A list of suggestions have been made, such as better planning to remove Work-In-Progress (WIP) and to lower inventory; optimization of manufacturing yields and costs by enlarging the batch size (higher filling levels in manufacturing equipment); inline inspection instead of manual inspection; installation of process analyzers to detect batch end-points, for example for drying and blending, The team shows that these measures will deliver a reduction in manufacturing costs.

A 'Manufacturing Vision' Response
Another company takes a different approach. Instead of appointing a team to look for optimizations and improvements, it first organizes a high level meeting with representatives from various departments - R&D, manufacturing, sales and marketing, regulatory affairs. The aim of the meeting is to investigate what the needs will be in 5-10 years, taking account of business challenges, technological options and regulatory opportunities.

The group has already looked at their current product portfolio and future portfolio, based on their pipeline. It has investigated the consequences of this new portfolio on the current manufacturing infrastructure. It has considered what the future manufacturing landscape will look like in order to be able to cope, not just with the new product portfolio, but also with the future market and environmental requirements, business model requirements, regulatory changes etc.

A scenario planning exercise has supported the exploration of possibilities and future scenarios. This study results in the identification of a manufacturing vision, which describes the required future manufacturing landscape that will best fit with the most likely scenarios.

This vision makes it easier to identify the gaps between the current "as is" manufacturing situation and the future "to be" one. It also helps to indicate the improvements and changes that the company can already start to implement. A roadmap linking the "as is" and the future "to be" situation enables the company to focus on the improvement and optimization projects that help it move to the future situation.

The company can avoid investments which, taken in isolation, might have a sufficient ROI to implement, but when looked at in a fuller context, would not achieve a more sustainable advancement for the company. This broader perspective enables the company to move forward in the knowledge that it is not just investing in little islands of optimizations but is linking them to a bigger quantum leap forward.

----------------------------------------------------------------