Companies that manufacture IVDs have a lot to think about these days: contamination, outsourcing, and domestic and international regulation and labeling, to name a few.
Jorgine Ellerbrock is senior vice president, operations, Gen-Probe Inc. (San Diego). She previously served as vice president, operations, for Invitrogen Corp., most recently for its molecular biology business. Before joining Invitrogen, Ellerbrock held positions with GE Healthcare Biosciences and Roche Molecular Systems. She can be reached at firstname.lastname@example.org
While IVD manufacturing and processing technologies pose some unique challenges, IVD makers also deal day to day with the same issues that all manufacturers in regulated industries do. These include constantly working to improve processes, be more efficient, cut costs, and attain compliance.
To learn more about current manufacturing and processing technologies and the issues surrounding them, IVD Technology editor Richard Park spoke with Jorgine Ellerbrock, senior vice president of operations at Gen-Probe Inc. (San Diego). In this interview, Ellerbrock talks about Gen-Probe's Process Integration and Enhancement team, her positive experiences with FDA, the importance of preventing contamination, and the role of the pilot plant in transitioning to production scale-up. She also discusses the current outsourcing trend and whether Gen-Probe is planning any overseas outsourcing in the near future.
IVD Technology: What have been the most significant advances and the latest trends in the area of IVD manufacturing and processing technologies during the past few years?
Jorgine Ellerbrock: IVD manufacturers, like all manufacturers these days, constantly have to look for ways to improve quality, reduce cycle time, and lower the cost of goods. The advances that come to mind are those that would deliver that kind of benefit. One example that we've been able to put in place at Gen-Probe is computational fluid dynamics software. We've been using it to minimize our costs and time-to-scale. It allows us to simulate mixing for liquids or solids. That gives us assurance that we have the homogenate we need and shortens our process development times. It speeds up the process when we're trying to do scale-up and lets us do that without wasting reagents. That's been pretty important for us.
Another trend is the increased sophistication that comes with automation. As you look at different pieces of equipment now providing real-time quality metrics to the operators, you can download the data right into a manufacturing execution system. As a result, many companies now have paperless manufacturing floors.
Please expand and elaborate on Gen-Probe's use of computational fluid dynamics software.
It is still relatively new. We purchased it before the end of last year, so we've had it for about six months. The software was recommended by a group within Operations called Process Integration and Enhancement. It's a small group of people that really help us look at how we can quickly scale-up processes and make them more robust. The leader of that group had a discussion with me about how the software can take the guesswork out of a lot of the scale-up efforts that we have going on, because typically we have to produce multiple batches to fully develop our mixing processes, as do most IVD manufacturers. When you scale up from a small tank to a big tank, the changes in mixing parameters are not necessarily linear. We looked at that and at the amount of reagents that we have to make at the higher scale in order to get the validations through and thought, “You know, if we can do this kind of scale-up ‘virtually' first, we can get it right the first time,” and that's pretty much what's happened. So we've been able to use fewer reagents, get the scale-up done faster, and really save time and money.
Does your Process Integration and Enhancement group constantly look for ways to improve Gen-Probe's manufacturing and processing technologies?
Yes. It's actually a small department within a larger team called Operations Technical Support, and that's one of their key responsibilities.
Provide some information about how exactly the group works and its responsibilities, and how it arrives at such ideas as the implementation of this computational fluid dynamics software.
The group consists of scientists and process engineers, each with a particular technical specialty. When the need for their expertise arises, such as process-scale-up or other process improvement, a Product Support Project is initiated. The request or idea can come from anywhere in the organization. For example, Planning often looks 18 months out at demand, and if they see that we're going to have to make a product very frequently, they go to this group for counsel. The members help us determine how to scale-up in a manner that permits us to make the product a little less frequently but without risk to the process. We have a pretty sophisticated program by which their projects are prioritized, and obviously it's those that are business-critical that would happen first. The group's basic job is to continue to improve our manufacturing and QC processes.
Challenges to Manufacturing
In terms of sheer technical complexity, what are the most troublesome areas in manufacturing IVDs?
As a company continues to produce tests that are more and more sensitive and detect fewer and fewer numbers of organisms primarily through nucleic-acid amplification, it is very easy to contaminate your lab with the product of that amplification, which is called amplicon. Whether it's PCR or TMA, you can get out-of-specification results because you've got amplicon present in your laboratory that's contaminating everything that you do, and it's very difficult to determine where the source is in order to eradicate it. It's not like radioactivity, where you can check the soles of your shoes before you leave the lab and make sure that you're not tracking it around. We have gone to pretty great lengths to segregate those activities that are postamplification from those that are preamplification. Even so, we're constantly on the lookout for this kind of thing. While I'm sure overall it makes us less efficient, it keeps us effective. So it's something that we strictly enforce.
Another challenge in manufacturing IVDs today is the drive to use ever-smaller volumes of reagents. As nanotechnology becomes more the norm in diagnostics, I can really see this concept of lab-on-a-chip and the use of very small reagent volumes being used gaining ground. And from a manufacturing standpoint, I can see the challenges that we're going to face when dispensing that very small volume accurately, then keeping it from evaporating until we can get it packaged.
Please elaborate on the efforts Gen-Probe makes to prevent amplicon contamination.
We restrict people from working in a preamplified area if they've already worked in an area where amplification has taken place. Some of that comes down to process and people knowing how they have to schedule their work because they'll be in an amplification area and don't want to contaminate anything later. That sort of scheduling is one of the primary things we do. We don't take papers out of that area. We don't take SOPs from desks into that area. And most importantly, training of people is really critical so they understand how easy it is to move that contamination around. But also very, very important is the rigorous cleaning regimen we go through in the labs where we have amplified materials. So it's a combination of different things, some of which are pretty well within our control. Of course, you'll have the well-meaning person who thinks they're doing something right by coming in to help, and maybe they haven't realized or remembered that they've been in a post-amp area first; they're not automatically locked out of the labs. It's on the QC side where most of the amplification takes place. But even in manufacturing, we have teams that are separated by what we call template-positive and template-free areas to further ensure that there's just no opportunity for templates to get introduced to an area where they can be inadvertently amplified.
In today's market there is tremendous pressure to get products out as quickly as possible but without sacrificing quality. How do you balance these conflicting demands?
It comes down to planning and scheduling. We work very closely with the planning team and the supply-chain team here. The quality-control group schedules people to work in the amplified areas of the lab in the later parts of the day so that they can have more flexibility in the morning working in the nonamplified areas. What you don't want is to do all your amplified assays in the morning and then not be able to do pre-amp work in the afternoon and have the whole QC group sitting at their desks. That's what I meant by the process being possibly less efficient because you're not necessarily maximizing the amount of hours that you can get out of a technician's day. Sometimes there's a schedule change that requires us to reprioritize, but it's so worth it not to have the amplicon problem. Amplicon contamination can shut down a lab and necessitate making alternate arrangements for testing. So, avoiding contamination is a big deal.
One of the other things we're doing within Gen-Probe to reduce manufacturing cycle times and move products through the process quickly is teaching people some tools for lean manufacturing. We want people to think about how to eliminate waste, whether it's waste of motion of their steps, waste of their time, waste of inventory, etc. We hold specific events called “kaizen” events to make rapid changes that have good impact, and to try to get individuals to think about how they can change the way they do things to become more efficient. And that's something that a lot of manufacturers in our space are really moving toward right now.
What special difficulties are presented by manufacturing tests for molecular diagnostics?
Speaking narrowly and with hats off to our development group, there really isn't much difference between the processes that we use to manufacture our human papilloma virus assay and our chlamydia or gonorrhea assays; there are even some common reagents in there. From a manufacturing standpoint that's been a real plus when you're looking at training of operators, minimizing errors upon transfer, and use of existing production automation. There really hasn't been much difficulty there, I'm happy to say.
Regulations and Standards
What was your experience and reaction when you first confronted and had to deal with meeting FDA and other regulatory requirements?
I actually do remember my very first GMP training. I was working for a device manufacturer at the time, and I remember thinking that the GMP made very logical sense, and I didn't understand why people were making such a big deal about it. The instructions to write down everything that you do, make everything legible, and follow the rules just seemed to make a lot of common sense to me.
I guess the first area where it got to me a little bit and felt like a bit of a breach of trust was when I was in the lab and I had to have a checker make sure that I was adding the right amount of chemical to a solution. I think with age comes wisdom, and you realize that it makes a lot of sense to have someone else double-check that you're doing something the right way the first time, not only for the patient so that you can be sure that you're not making a simple human error, but also from the manufacturing-cost standpoint. You realize that if you can catch an error before it happens, you have a chance to correct it. So it's really never been much of an issue for me. It makes sense. It's called “good manufacturing practices” for a reason.
In general, would you say that the various regulatory requirements and standards that are out there—the GMPs, the QSRs, etc.—make sense? Would you say they are fair and are good policies that have been established and executed by FDA?
From my perspective, yes. It helps us do things right the first time, it goes beyond manufacturing back to the development of the products, and it keeps everybody on the same page in terms of what the expectations are, making sure we're designing something right the first time, making sure that we're ensuring reproducibility of results of manufacturing. Ultimately, as a consumer, it comforts me to know that we've got something like that in place.
Overall, have you found FDA to be fair and open with its regulations and standards? And have you found the agency to be cooperative, and clear in conveying its expectations?
Yes. My career started with devices for a short period of time and then switched to IVD for about seven years, and then it was a combination of working with research-use-only products and in vitro diagnostic products at another company. And then I moved to two companies following that; they were entirely in the life sciences realm, so there was no regulation for those products. For the most part, however, we followed GMP because it made sense. I haven't been in a regulated industry for my entire career. But from a manufacturing perspective, I think FDA's expectations are pretty clear, and I wouldn't say that they're unfair. I've been in situations—not here at Gen-Probe, but at previous companies—where, when the inspectors came through and found something, the problem was easily and quickly correctable. I've been fortunate that they haven't found any huge deviations in any of my workplaces. FDA has been pretty open and honest during the inspections.
Gen-Probe is a global company. What are some of the unique regulatory challenges that global companies face, from manufacturing and processing standpoints, when selling to such overseas markets as Europe and Japan?
I'm not aware of anything unique that overseas markets such as Europe and Japan require of us, other than the labeling. Our quality systems comply with FDA, and they comply with the European standards. To my knowledge, we have never had anyone from Japan come to inspect or audit us.
How can IVD manufacturers balance their compliance activities for FDA and for international standards (for example in the EU) with their IVD directive to eliminate redundancies? Or are there still areas where independent efforts are required?
Whether you are talking about FDA or European standards, the bar is constantly being raised for IVD manufacturers. We are lucky to have a quality assurance group and a regulatory group that are pretty plugged into regulatory trends. They help us understand what the FDA is thinking or which new IVD directives are headed our way that we might need to comply with. What concerns me when we talk about redundancies is that you don't want to have two quality systems operating within the same plant. I believe it is entirely possible to have a single quality system that complies with both the U.S. and EU regulations.
I can tell you that we just had an ISO 13485 recertification audit. There is a fair amount of effort and energy that goes into hosting the audits and making sure that the auditors have the information and access they need, and if we could have one inspection instead of two, that would probably be a good thing.
Regulatory requirements and international standards are intended to keep companies from designing or marketing bad products. In what ways do those requirements help to make technically complex processes simpler for IVD manufacturers?
When you've got a rigorous product development process with design reviews and the right people involved operationally as well as in development, and you ensure that you've got your production processes, your specifications, and your scale-up all delivering reproducible product at the right quality level, those kinds of processes really help ensure correctness the first time. The regulations give more muscle to R&D and manufacturing working together to deliver that kind of result and keep us aligned with the expectations.
Are the various regulatory requirements and international standards equally useful when dealing with the development of such cutting-edge products as new cancer tests?
Yes, they are equally effective and useful. That's the beauty of them. They're general enough that you can apply them whether you're doing something cutting-edge or very simple.
I assume that when you're developing something completely brand new, changes need to be made to the manufacturing process and technologies. So, accordingly, what sort of extra attention or efforts need to be made to ensure that such new manufacturing and processing are compliant with the various regulatory requirements and standards?
At Gen-Probe, we would do that kind of thing in our pilot plant and make sure that we had all of our i's dotted and t's crossed when it came to the processes before we turned the project over for prime-time manufacturing. And the pilot plant actually works under GMP, so even though they may make some changes, the changes are fully documented, and they go through change control. There are no changes made “on the fly,” so to speak. When you're working on a new technology, you still want to make sure you've got reproducibility, traceability, and a process that's in a validated state. I think the regulation continues to make sense in this case. I don't think it takes away from the ability to do something innovative. It just helps you do it reproducibly once you've got it in manufacturing.
Organizational Structure and Culture
What role does an IVD company's organizational structure play in making the manufacture of a product line easier or more complicated?
I don't think organizational structure in and of itself comes into play at all. What makes our lives easier or more complex is more about the processes that we implement than about the reporting structure that we've got in place. It has been my experience that it's easier and more common to create a complicated process, with excess bureaucracy and multiple approvals, than it is to create an elegantly simple process that enables compliance and process velocity. Once complex processes have been put into place, they are relatively hard to change. But that doesn't have anything to do with the organizational structure, it's more about organizational culture.
I believe the organizational culture is far more important than the organizational structure. If you have clear accountability, especially when you're talking about the product development process, and you have clear expectations so that there's shared accountability of the team to do the right thing for the company and the customer, then whom they report to doesn't really matter. It's about the total outcome. And I think that's the culture we have at Gen-Probe. I don't think that anybody in our company would ever say, “I don't report to this department, I report to that department, and that's not my job.” It's just not the way we work.
As you know, there's a big push now to become more environmentally aware and eco-friendly. What sort of efforts are being made throughout the IVD industry, at IVD companies and at Gen-Probe, to look at manufacturing and processing technologies to make them more green?
We are just at the beginning. We've just recently started a sustainability council at Gen-Probe to drive some of those activities across the entire corporation—not just Operations. But the place we're starting within Operations is on the packaging engineering side of things. We are looking at our packaging and asking, “Do we have too much packaging for some of the different products?” “Are there ways that we could make it more streamlined, more recyclable, use more environmentally friendly materials?” We are just at the beginning of that awareness.
Does Gen-Probe outsource any of its product development or manufacturing activities, and what is the business rationale for a company's decision in this regard?
I can't think of a company, at least that I've worked for, that definitely outsources product development—although one might certainly exist. But manufacturing is pretty commonly outsourced, and that runs the gamut from a small part of the process to the entire process, the entire product. When you talk about business rationale, I think that really depends on the organization and ultimately, capability and costs. I've seen start-up organizations that have really good product development, but they don't have the infrastructure or the equipment or maybe the know-how to do large-scale manufacturing, and they may outsource the whole thing.
There are lots of companies that do contract manufacturing. That's their business model. Other companies may outsource a piece of it, sometimes because there is another organization that has a specific competency. Or, more often now, I would say cost is a driver. Looking at offshore manufacturing is something that's pretty common, even in this business.
For Gen-Probe specifically, we make a lot of our own raw materials—or what a lot of other companies might call raw materials—but we do buy some in. But the transformation process typically happens here. There are components that we're considering outsourcing right now and are actively conducting a make-buy analysis for, and questioning the rationale of why we have always done something in-house if we can do it cheaper out of house, freeing up that manufacturing space for further expansion.
Does Gen-Probe currently outsource any of its manufacturing to OEMs and suppliers?
I've been speaking solely of the reagent side of things, the assay manufacturing, but we do have sole-source instrument manufacturers that we work with. A lot of the development is done here, but our outsource partner companies do the final box manufacturing. We don't do any instrument manufacturing in-house.
Gen-Probe takes its manufacturing and the various quality and regulatory standards very seriously. What sort of challenges does that present to you from a quality standpoint in terms of making sure that your outsourcing partners are up to the same level of compliance that you demand for your internal manufacturing?
They are put through a rigorous audit at various stages in the process, and that audit is typically performed by engineers or subject matter experts together with the members of the quality assurance team. There are contracts in place, quality agreements, supply agreements. We just try to be as transparent as we can with our partners regarding expectations right up front, so that there are no surprises later in the game. We also do routine business reviews with the outsourcing partners after product launch. Our general method is that we let contracters know what our expectations are, and then we audit them to those expectations. So far, it's worked pretty well.
Small-Scale Manufacturing: The Pilot Plant
Please provide some information about what your pilot plant is and how it works.
The pilot plant, like our Process Integration and Enhancement group, is under our Operations Technical Support team. That's the group that does process development. They work very closely with the team that develops the product, and they serve as an intermediary between the development group and full-scale manufacturing. They are highly technical. They work under GMP but at a lesser scale. As the need arises, they are the ones who scale-up a process, pulling in the Process Integration and Enhancement group as needed, although that group is usually more involved after a product has been in manufacturing. So the pilot plant is a transition from development to full-scale manufacturing. It works out the bugs.
When Gen-Probe is developing, producing, and marketing a new product, at what stage in the product development life cycle does the pilot plant become involved?
The pilot plant gets involved shortly after the business planning for the new product development process is completed. Often at this stage, the Development team needs variations of the product's active ingredient to optimize formulations. The pilot plant can rapidly provide these at small scale. As such, the pilot plant scientists are either on the core team or the extended core team. Once the final formulation is “locked,” the pilot plant will lead scale-up and process development activities, and transfer these into production and QC for validation.
Is the pilot plant unique to Gen-Probe, or is it a common part of the manufacturing process across the industry?
My experience is that it is pretty common. In some companies it is done in development. In some companies it belongs to or reports to Operations. But for most high-scale IVD manufacturers, there is usually a separate group that does the optimization and the process development. This is not even specific to IVD; even in the life sciences industry there is typically a technical group that's not considered direct labor that leads efforts for scale-up and process optimization.
The pilot plant is separate from the plants that Gen-Probe uses for its actual products?
It is. It's like a plant within a plant. There is a separate entrance, and it's like a miniature version of the larger operation. Just about anything we can make in manufacturing we can make in the pilot plant at a smaller scale.
Copyright ©2009 IVD Technology