Standardizing Solutions to Enabling Platforming of Process Analytical Technology (PAT) Deployments

Gary McGeorge, Scientific Director at Bristol-Myers Squibb

The goal of process analytical technology (PAT), a mechanism to design, analyze, and control pharmaceutical manufacturing processes, is to understand these processes by defining their critical process parameters (CPPs) and monitoring them accordingly and in a timely manner. This allows for more efficiency in testing, while also reducing over-processing, enhancing consistency, and minimizing rejects. Gary McGeorge, Scientific Director at Bristol-Myers Squibb, spoke to us of the benefits and challenges associated with establishment of consistent resolutions while facilitating the steps associated with the implementation of this valuable practice.

Gary McGeorge is the Scientific Director at Bristol-Myers Squibb. Direct correspondence to: gary.mcgeorge@bms.com

Would you explain the meaning and benefits of standardizing process analytical technology (PAT) deployments?

Standardizing PAT technologies means that we are utilizing a smaller, parsimonious set of tools with common analyzer performance metrics, software workflows, and hardware integration. Due to the variety of PAT measurement technologies, it is complex to build consistent sets of solutions to systematically answer challenges in the pharmaceutical development organization. Each instrument vendor will build a technology solution that works for a set of applications in a certain way, but when trying to compare the data from those systems or the integration efforts across different vendors it becomes increasingly difficult to compare data across analyzer networks. As a result, analyzer standardization is very important, whereby measurement technology performance metrics are clearly established.

What are the main challenges you are facing?

There are several challenges that the PAT community is facing just now in the instrumentation space. There are a limited number of suppliers providing a range of solutions that vary in their maturity levels to the point that each installation becomes a bespoke solution. Since inline and online PAT solutions require interfacing with the processing equipment, there are integration hurdles beyond that of traditional analytical instruments. We can do a great deal of scouting and exploration of technology; however, building them out into a standardized toolbox needs significant organizational alignment and capital investment for both hardware and integration efforts.

Many pharmaceutical companies partner with various external contract research organizations (CROs) and contract management organizations (CMOs) to provide flexibility for development, during scale up and during clinical manufacturing, and hopefully into different sites for delivery of commercial supplies. This makes it an expensive process to ensure your network is fully PAT enabled. One good example as blend monitoring where a near-infrared blend analyzer is wirelessly connected to an intermediate bulk container (IBC) tumble blender. Here the system measures one spectrum every rotation and enables one to track the blending trajectory and gain insight into the rate at which the material attains a homogeneous state. Currently, vendor solutions that are on the market are all quite different from one other both in terms of the physical interface with the blenders, the software integration, and the software and experimental capabilities. A simple difference like the illumination spot size on these analyzers results in changes to the apparent spectral response and thus the results not transferable across vendors. The outcome is that one prefers to stick with a single vendor and build up standardized platform solutions with that one supplier. However, this single-vendor approach poses a risk to long term support, as we must maintain our ability to make the high-quality medicines using the control strategy that was filed with the Regulatory Authorities. The outcome is that many Pharma companies typically only include PAT applications in their regulatory filed control strategies for control points where it is essential and where there is no traditional or simpler analytical approach.

Is there significant resistance for standardization of PAT deployments?

I don't necessarily think it is resistance, but is more related to the challenge of having an unclear marketplace for vendors to understand where the opportunity space is for PAT. Each vendor is seeking to find a niche area where their unique solution or technology are best suited. Vendors may advertise about instrument stability, or instrument-to-instrument transferability of models, or portability, while others advertise sensitivity—all these features create a range of opportunities for customers to choose from. However, all these instrument choices have different interfacing overhead and software workflow integration needs. When you couple this complexity with the requirement that a commercial PAT deployment may need to be in place for 10 years or more—it then becomes essential to create a robust PAT solution, which has a direct impact on the choices that one can makes earlier in development.

One area where standardization has been seen is in the adoption from most instrument suppliers in embedding Object Linking and Embedding Unified Architecture (OPC UA) data transfer architecture on the analyzers. Leveraging OPC UA simplifies the integration of the analyzer into the industrial plant infrastructure for controlling the spectrometers and deploying models in real-time, while also allowing for integration with other dashboarding or advanced process control solutions that are becoming more commonplace. This reduces the software development overhead for smaller vendors, allowing products like Optimals SynTQ or Applied Materials PharmaMV to do the bulk of the data management, processing, and archiving. However, this does require PAT data management software solutions to be in-place, which is the topic for another day.

What PAT technologies would be the most deserving of standardization?

This is a difficult question as there are so many potential technical opportunities for PAT to make an impact that the need will be different for each company. Traditional tablet and capsule manufacturers will see a different opportunity space compared to biopharmaceutical manufacturers, while drug substance and product will be similarly differentiated.

In my opinion, we are probably in need of a white paper that puts forward some simplified agreements that pharma companies are aligned on, which would help the suppliers see that there is commonality across many companies. Several examples of historical PAT applications have been based on single company or supplier conversations that had good success but were limited in acceptance by the broader community. The result is the supplier is either going out of business, selling off the technology, or simply shifting gears to another industry or product. This is not a special issue for PAT, but I have seen it with many spectrometer companies over the years.

The example I gave earlier of the blend BU analyzer is one potential area that would be interesting for me personally as we have seen several vendors enter and leave this space, likely due to the low sales volume. However, by leveraging such a standardized whitepaper approach, we could get alignment across various pharma companies on the design features and requirements then it would simplify adoption deeper into organizations. There is an Installation Qualification (IQ) consortium effort that is looking into modernizing blend uniformity testing in Pharma and one output could possibly be a standardized PAT approach.

Are there any new technologies that would assist you in the ability to standardize deployments?

New technologies would imply that there is a need but no current solution. Looking at this from the needs side of the equation, sensitivity is still the biggest challenge for spectroscopic based PAT instrumentation. So, when looking at low dose drug products, let’s say <0.5 %wt/wt, it is currently difficult to find solutions that can be quantitative for the analyte of interest. We have been partnering with the Enabling Technologies Consortium (ETC) to explore novel fluorescence instrumentation that can do this routinely for monitoring and control of low concentration ingredients within the continuous manufacturing space. This is easier to adopt as there is no competing technology to displace. There are also advances in spatially offset spectroscopy and how these tools highlight the promise of fast (sub-second) measurements, potentially enabling 100% inspection solutions to be integrated into manufacturing equipment. It will be interesting to see how these two technologies play out over the years.

What sort of support are you receiving from the FDA or other organizations for these efforts?

There is quite a bit of support across a variety of consortia and partners. One example is the ETC which is a forum for pharmaceutical and biotechnology companies to discuss ideas, share information, and collaborate on the development of new enabling technologies. Member companies can vote on problem statements that need innovation and through a bidding process companies can apply to provide prospective solutions to the challenge. Projects like, spatially offset spectroscopy to support drying monitoring or automation solutions to rapidly determine solubility of active ingredients have been put forward by member companies.

The International Consortium for Innovation and Quality in Pharmaceutical Development (IQ Consortium or IQ) is also a valuable resource where a range of initiatives are worked on by member companies to drive innovation. There is an actual PAT chapter within IQ.

Health Authorities like the FDA, MHRA and EMEA, along with USP are all innovating to provide industry guidance that enables innovative technologies and approaches to be more readily adopted by the industry. Within the FDA’s Office of Pharmaceutical Quality (OPQ) they have created the Emerging Technology Program (ETP) to familiarize the FDA assessors on new technology that being presented by Pharma companies. This will hopefully ease the pressure on all involved to get PAT more readily embedded in Regulatory filings.

There has been a perception that Regulators are concerned about innovating in the PAT space, limiting the number of marketing applications that include PAT. By modernizing the ICH guidelines to reflect innovation by including concepts like analytical target profiles (ATP) it has enhanced the flexibility that Pharma companies have in embedding novel analytical solutions, including PAT in their filings.

What are some of the best sources of information for the latest details on PAT developments?

Personally, this is a challenge. Much of the knowledge exchange is happening within either conferences, like the International Forum for Process Analysis & Control (IFPAC), SciX, or The American Institute of Chemical Engineers (AiCHe), or in vendor sponsored events, or consortia driven meetings. Little of this work has been published as of late, resulting in difficulties staying abreast of PAT advances in various areas if you are not part of these meetings.

LinkedIn, however, provides me the best and easiest means to see what is going on overall across the industry by seeing what my connections have been sharing on the platform. Building and maintaining a network outside my own company was critical during Covid and appears to still be the case. People are eager to share, in an easily digestible manner, what meetings they are targeting, which vendors they are excited about or what research they noticed that caught their eye. Similarly, vendors are leveraging the platform to highlight their innovation, what meetings they are attending and where they are taking their solutions. I highly expect such platforms to become increasingly important as the pace of innovation continues to accelerate.

About the Author

John Chasse is the Managing Editor of Spectroscopy and LCGC. Direct correspondence to: jchasse@mjhlifesciences.com