Last year, I attended a presentation from Joseph Cancelarich, director, Pfizer Global Engineering, about the company’s Intellicentic program, a new offering of consulting services and advanced platform technologies that Pfizer is rolling out with selected partners. Intrigued by the presentation and the “open source” business model that Mr. Cancelarich described, I made an appointment to meet with him and Mojgan Moshgbar, who serves as director/team leader, Advanced Manufacturing Technology, Global Technology Services at Pfizer Global Supply. We discussed the company’s history in developing process technology, its willingness to sell these systems to competitors, and the regulatory Catch-22 of trying to advance standards.

Mr. Cancelarich, Dr. Moshgbar, and Dr. Jun Huang, senior manager at Pfizer, will give a presentation on Intellicentic during Interphex 2014, on March 20, 11:45 a.m. to 12:45 p.m., “Intellicentic — A unique business model to leverage Pfizer experience to deliver advanced technologies and solutions for pharmaceutical manufacturing.” You can find out more about Intellicentic at www.intellicentic.com

—Gil Y. Roth

---

Contract Pharma: Tell me about Intellicentic: what is it, how will it work, and who can benefit from it?
Joseph Cancelarich: The pharma industry’s unique in the sense that you create a product and it stays in use for 20 or 30 years. In other businesses, a product fades and you replace it with a new one. The expectation in pharma is that those products are made the same way, after decades. However, raw material suppliers change, equipment changes, other factors change. So the question becomes how we maintain products attributes throughout this. Our Process Analytical Technology (PAT) systems, is one way to understand these changes.  We think using technologies like PAT, are better than the standard procedure.

The PAT systems developed and implemented in Pfizer add real analytical value to traditional procedural uniformity exercised in the industry.

Mojgan Moshgbar: The development of the capabilities that are now called Intellicentic , started about a decade ago, when the Pfizer’s PAT group (PASG) started to look at the possibility of using PATs not only to monitor a process but to also use the data to increase process reliability. We started to look at PAT based Advanced Process Control (APC) as a way to make process more predictable and repeatable.

APC is a mature technology and used widely across many industries like petrochemical, food, and semiconductor sectors. Process Analytical Technology (PAT) by itself is very important for monitoring processes, but when you combine it with APC, its impact become transformational. Over the years, we have identified significant opportunities for better monitoring and control in different units of production.

Early on, we realized that we cannot bridge the technology gap between pharma and other industries by gradual technology introduction. It would take too long. We realized that to really make a difference, we must make a leap, rather than an incremental improvement. Pharma processes are complex and we needed to develop technologies that would solve the challenges in our industry. This meant we had to use the state of the art in other industries as a starting point and customize the available technologies to meet our specific needs. 

CP: What sorts of problems are those?
MM: When you look at processes in any industry, the best way to use data is to convert it to knowledge and capture and preserve that knowledge. In the pharma industry we produce a lot of data, captured either on paper or electronically. However, Pharma is not as good as other industries in utilizing the data it captures and making it into cause-and-effect-related knowledge that be  subsequently used to reduce manufacturing or development time and effort.

CP: In what areas?
MM: Let’s take tableting as an example. Development of formulations in most cases are still based on trial and error. It’s a science, but it involves so much trial and error that you could say it’s partly an art.

Typically, transition occurs at large companies — scientists and formulators come and go, facilities get shut down, products get transferred — There’s a lot of effort around knowledge management within the industry, however, it is easy to lose the knowledge that was present, say, 20 years ago when a particular scientist or team that is now gone.

The majority of processes can be defined scientifically. Just because we’re making pharmaceuticals doesn’t mean it’s a dark art. I believe the best way to capture process knowledge is to have a systematic approach to extract cause and effect relationships from the data we gather and put it in the process itself. The idea of intelligent systems for pharmaceuticals is about turning this vision into a reality.

In our day-to-day life, we use very advanced technologies like smartphones for everything. We expect our new smart gadgets to get smarter, delivered custom-made a few days after ordering them, and we expect them to be of the best quality. But when it comes to processes on the shop floor, the expectations are not the same. Fire fighting and dealing with the aftermath of deviation in our industry uses a lot of resources. If a proportion of those resources are invested in the technology that capture the right data and transforms that data into knowledge that is engineered back into the process to prevent deviations and variability, and to control the system for predictability, then we’ll be entering a new frontier.

For us, this technological journey  has also been a cultural one. To get where we are now, has taken a lot of effort and resources. We’ve assembled some of the best scientists to work on development of these intelligent platforms and control systems over the years. One thing we’ve learned is that even if we change the culture and our technology benchmark internally, the regulatory environment for pharma means the change is not as easy as introducing new technology in, say, semiconductors. In a regulatory environment, what is practiced most commonly is considered lower risk than embarking on a new path.

CP: That sounds like a Catch-22 situation, where regulators won’t be comfortable with a new technology until lots of companies are using it, but lots of companies won’t roll it out until regulators are comfortable with it.
MM: And that’s why we’ve established the Intellicentic to offer the generic platforms we have developed over the years in Pfizer We want to help develop a new technology standard for the pharma industry. Intellicentic is an attempt to move the existing standards forward.

JC: Pfizer has invested a lot of time and resources in PAT based technologies over the decades. They’ve led to a lot of successes within our plants, and we’ve gotten good regulatory feedback. With that said, there are very few companies that would undertake this, because of resource constraints.

As far as the business model goes, moving the industry forward will create tremendous value. Intellicentic is taking applications that are working in our own facilities and allowing other players to see if they can utilize them to improve their own processes.

CP: Have you had conversations with FDA about rolling out these technologies?
MM: We have had some opportunities for dialogue with the agencies. Some of the regulators are very excited about the idea of having more predictable pharma processes. One thing to remember is that majority of the Intellicentic platforms can provide improved process performance without impacting regulatory aspects of the processes.

CP: Do you think these platforms you’re making available would be more useful for new product development, or for retroactively working them into existing, late-in-lifecycle products?
MM: The technology is extremely useful in development. Take crystallization processes, for example. In those, you must develop a design of experiments (DoE), use different reactors in different conditions, and then harvest the crystals, measure the particle size, etc.

The majority of those tests are just to weed out what is not going to work. One or two reactions give you close to optimal conditions in terms of crystals purity, yield and particle size. We’ve developed a system that can look at the supersaturation of crystallization slurry and see what is happening and adjust it during the test. In our studies we have shown that you can significantly reduce the number of experiments needed to optimize cooling curves and fix problematic crystallization processes . The reduction of the number of trial-and-error DoEs, help reduce cost and time to market.

JC: The Intellicentic technologies really help when there are changes. If you change a supplier, you have to run your batches, send them to the drug product group so they can make tablets, then put them on stability shelves. Now there’s a capability that let us say, “This is how the process looked before we changed raw materials, and it’s looking the same after.” You have real data.
As we get more experience in larger deployments and reach similarity, I think there’ll be a change in expectations. Having data makes the argument much stronger. But if we don’t introduce technologies like this, then the old system will never change.

MM: And when you look at smaller CMOs, the dynamics are such that they can’t really afford to constantly battle issues like problematic crystallization, or other variations. Being able to make a single investment that enables them to reduce the number of DoE runs, scale up reliably, and control the process — that would be a fantastic business opportunity.

One thing we’ve found with that crystallization platform is that we can monitor and control particle size distribution, which is a major issue. It can also consistently provide the same type of yield from the crystallizer and reduce cycle times. In trials, we’ve seen as much as a 40% reduction in crystallization cycle time when you control the process like this.

CP: Sounds like quite a savings! Who had the initial idea to sell these platforms?
JC: In today’s pharma environment, it’s not so easy to see where one company’s walls stop and where those of its suppliers
and partners begin. There’s a lot of blend-ing in the industry. For example, you couldn’t imagine, 15 years ago, that Pfizer and Merck would partner on diabetes treatments.

MM: I’m a major advocate of open innovation. I believe the business model of the industry is changing such that it’s not possible to continue unilateral investment in transformational technology. Open innovation makes sense to us and Pfizer leadership.  This is a new business model for technology development in diverse industry where the overall performance is not driven internally alone. Our performance is impacted by our partners like external suppliers. This an attempt to find a better way to share knowledge and best in class technology and know-how within that complex network, because at the end it will help us and our patients. We believe it is a true win-win opportunity.

CP: What’s Pfizer’s ultimate goal with Intellicentic?
JC: Our long-term objective is to make this program self-sustaining. However, considering Pfizer’s scale , we know that revenues from a program like this won’t move the needle. We’re not looking at a return in term of immediate dollars.

CP: So it’s not about trying to recoup Pfizer’s investment in developing Intellicentic platforms?
JC: That’s correct. Pfizer Global Supply has a brand. It’s respected for quality, for being a technology leader, for its position in knowledge management. We believe that open innovation will only make that position stronger. In the end, it’ll impact the big line; that is, manufacturing and moving our products more effectively. These systems will maintain that quality we’re renowned for at a lower cost, with lower wastes, and make our processes more environmentally sustainable.
The reality is that someday, members of Mojgan’s team will be retired and Pfizer will still be selling products they were working on. So how do you put in systems that assure the Right-first-time quality will remain after the operators have moved on?

CP: Do you have an example of a process where you’ve faced that problem?
JC: Take sugar-coating. It’s a bit of a black art, in some respects. There was a small product where we had an operation that had perhaps 10 people working on it. It reached a steady state where one person could do the process, but when that one person retired, we were left with a gap in knowledge. There are issues out there that these systems solve.

CP: And you’ll be able to reduce other players’ initial investment?
JC: By orders of magnitude. Intellicentic isn’t just about the hard assets and pieces of equipment that you buy. In a situation like this, you’re not paying for the actual stainless steel and electrical setup. You’re paying multiples for the engineering that went in before. We’re offering all that engineering for (essentially) nothing, along with the asset base and the handling to set it all up.

CP: Why are you working with an outside partner, rather than marketing it directly?
MM: The majority of these systems are integrated solutions, not just one individual sensor or control system. Where we use PAT, it’s integrated with othersensors, pieces of automation and software to create a complete platform that has a communication strategy with existing automation.

This integrated solution requires components coming from instrument manufacturers, software vendors, advanced control experts and a lot of knowledge and resources in terms of understanding a particular pharmaceutical process. So it makes sense to partner with the companies that provide the individual pieces of the puzzle. It makes more sense in terms of market accessibility. Our partners do the marketing and we continue to concentrate on our day job, developing technologies for Pfizer.

CP: How did you select the vendors you’re working with?
MM: We are partnering with these particular vendors because we have worked with internally. They’re not necessarily unique in their capabilities or product portfolio, but they’re currently our preferred suppliers.

Intellicentic is about integration of more than one technology component, to create customized platforms geared towards biopharmaceutical manufacturing. The original customized platform has been designed and commissioned internally at Pfizer, so we hold the customized solution. Intellicentic partnership allows our partners (that include instrument and software providers as well as system integrators) to market these propriety solutions to other customers. This means that we don’t necessarily get involved in contracts. Our partners will work with the customers and carry the standard system and engineering warranties.

JC: They’re the ones we prefer to use internally. They are willing to stand behind their systems. The partnership means they work closely together on a project to make it successful. We know we’re not going to be carrying the risk when a company uses one of these platforms on its own process. The responsible party has to take ownership of that. What we bring to the table is  the knowledge base we have on the configuration and usage.

CP: What other benefits do you see from helping this sort of model proliferate?
JC: It can help us apply these platforms in our own network. The heavy lifting was in developing them, but there is still resources spent on deployment and maintenance. We’d like to see a better value there. So as more people develop the skill sets to run these platforms and processes, the better we – as Pfizer and as an industry – can deploy our resources.

MM: What’s unique about our technical approach is that we’ve intentionally have developed generic platforms that can monitor and control a class of processes, rather than a specific product. They can monitor and control a step in biopharmaceutical production, like wet granulation, drying, API reactions, crystallization etc. The platforms are not specific to a given product or formulation. This increases their implacability across different plants and companies.

The other thing is that we have a lot of ongoing research. We certainly haven’t reached the end. We want to expand the use of these types of systems to gain even more insight and innovation. More experience in diverse situations will trigger new ideas that helps expand innovation. Pfizer can benefit from that, rather than just experimenting with these systems internally.

CP: So feedback from other customers is a part of the value-equation?
JC: Absolutely! We want our partners, like Federal Equipment, to be a center for the flow of information.

CP: Is Federal Equipment going to be the main partner?
JC: Yes. They know a large peer group of manufacturers, CMOs and other companies, and they understand project management and execution. Our technology partners include Perceptive Engineering, Kaiser Optical Systems, Stottler Henke Associates, Middleton Research, and Metrohm USA, who will support the relevant technical aspects.

CP: How’s the reception been for Intellicentic?
MM: We showcased a number of the platforms at IFPAC recently. The reception and enthusiasm was very diverse and very encouraging. It has definitely created a buzz, and we expect an increase in awareness and interest, as our partners increase their marketing efforts.

CP: Does the interest come more from smaller companies or from Pfizer’s peers?
JC: Some peers have shown interest. It comes down to execution. The challenge is how to put the funding together to execute. That’s part of our efforts: developing business models that allow people to get to those points without getting hung up on putting together the dollars for implementation.

There’s been a lot of consolidation in the industry, but has the total number of products on the market decreased? More drugs and prescriptions are being sold worldwide now than there were 10 years ago, but they’re being sold differently. The unit operations are the same, so the need for execution has increased.

In the past, people have heard “PAT” and decided they could commit one person in a plant to handle it. One guy is not going to develop these solutions. But if that one guy has a standardized package, he can convince people higher up that this is the right value proposition. More to the point, if you have a system available, people will find new applications for it.

CP: How many platforms will you make available through this program?
MM: There are close to 20 of them. It’s a comprehensive suite of solutions. They platforms are not limited to shop-floor production. We’re also offering very effective solutions that can drastically reduce the efforts and costs associated with planning, production scheduling and changeovers. The platforms also have significant impact on the environmental sustainability. That’s both direct, by saving energy, and indirect, by preventing failures. The full list of the platforms can be found at www.intellicentic.com.

CP: What opportunity does this represent for contract manufacturers?
JC: I think it could be very useful to them. Not only is there a direct RoI for them, but there’s also the idea that, if you’re looking at two otherwise equal CMOs and one of them is using technologies like Intellicentic and the other isn’t. Customer look for manufacturing reliability, so having a good technology base probably factor into the decision-making process.

CP: Is there going to be a Pfizer branding connected to Intellicentic? And is there any concern that peers might be averse to using a platform provided by a competitor?
JC: These are the systems we use in our sites. So, while a customer would be buying from a third party, the Intellicentic trademark is only delivered if the customer uses our knowledge base and our implementation. The implementation of the platforms doesn’t  reveal any trade secrets about the buyer’s product or Pfizer manufacturing operations. You won’t learn how to manufacture Lipitor, for example, by employing these systems. What they’ll provide, if implemented well, is production at higher consistency, with less rejects, lower cost, and more confidence in process outcomes.