In BPI’s June issue, we presented a supplement on geographical trends in biomanufacturing. We looked at the influence of a growing demand for biotherapeutics in emerging countries and the influence of new technologies that are driving interest in smaller, perhaps more geographically distributed production. We wanted to explore what a global bioeconomy would look like and where its primary capacity would be concentrated. Authors provided examples of how to balance cost with control issues. They talked about working in different legal and regulatory environments and addressed differences in intellectual property (IP) regulations and supply chain monitoring. Unavoidably, the issue of biosecurity was addressed: the need for pandemic preparation, mitigating supply chain contamination risks, guarding against bioterrorism — and so on.
A large part of our preparation for that supplement was a panel discussion at the 21 April 2010 Interphex conference in New York City. As we examined geographical opportunities and threats, an Icelandic volcano appeared eager to emphasize our topic by stranding our scheduled moderator in Paris. In Mike Ultee’s absence, Bob Broeze (Laureate Pharma’s president and chief operating officer) stepped in. This special report reproduces each speaker’s five-to ten-minute introductory comments (edited for space).
“Introduction and Outsourcing Trends” by Robert BroezeRobert Broeze brings 25 years of experience in the biopharmaceutical industry, with work spanning research, development, characterization, validation, and CGMP manufacture of biopharmaceutical products, from pilot to phase 3 clinical and commercial scale, with a strong emphasis on monoclonal antibody products for parenteral use. Before joining Laureate, Dr. Broeze held positions as vice president biologics research at Purdue Biopharma LP and was vice president of operations at Bard Biopharma LP. He’s held positions at Cytogen Corporation and was ultimately the vice president of operations there, with overall responsibility for development in CGMP manufacture of Cytogen’s products. He’s a graduate of Rensselaer Polytechnic Institute, where he earned both his BS and PhD in biology.
“Laureate Pharma (Princeton, NJ) works on the CMO [contract manufacturing organization] side of the supply chain within the biopharmaceutical industry. The company provides a full range of services from protein production through aseptic filling of products. Every biotech company, for every product in its pipeline, at some point needs to make a very strategic decision about how to make that product and how to fill the supply chain. It’s a make-versus-buy decision — or in other words, a decision whether to insource or outsource. Historically in the biopharmaceutical industry, insourcing was the norm. Companies like to have lots of control over their product, process, and inventory, so the trend in the past was to have fully integrated activities and companies.
“Presently, outsourcing is an increasing trend. It allows a company to focus on internal competencies, manage its internal resources for maximum efficiencies, manage expenses, and keep down capital costs. So the question every company has is, ‘What do we outsource, where are we going to do it, when and what’s our reason for doing it?’ This question applies to all aspects of the clinical and commercial supply train, starting with preclinical material, moving into the clinic, and then eventually into commercial supply.”
Capacity and Emerging Markets: “Around the world, capacity for manufacturing biopharmaceutical products is available in a number of areas. Historically, the United States and western Europe have been key areas: 70% of biopharmaceutical R&D takes place in the United States and Europe — the most popular biopharmaceutical manufacturing locations. But India, Singapore, and Korea are emerging in this market, with increasing acceptance. Additional capacity is coming on line, and it’s turning out to be a relatively large force in manufacturing. China’s portion of the industry is growing rapidly despite the country’s relatively late entry into the biopharmaceutical market. And I expect that over the course of time we’ll see more and more activity there. Beyond that, other areas in the world may play a more active role in the future of the industry. Eastern Europe and South America are two such emerging areas to watch.
“A number of factors go into choosing a location for biomanufacturing. For clinical trial material, proximity of appropriate expertise is critical. You want to have your research folks close to where you manufacture your early phase clinical product so they can get involved if problems occur. People in the plant need to oversee critical manufacturing areas and help troubleshoot issues when they arise. The track record of a biopharmaceutical manufacturing organization is critical. The track record of its quality operations is critical. And the track record — and depth — of its technical expertise is critical.
“Logistical issues can come into play. Some specialized raw materials and supplies needed for a process may be available in some areas and unavailable in others. For commercial supply, proximity to the marketplace could be critical. Everyday operating activities must take into account differences in language, culture, and time zones to keep everybody on the same page. Obviously, costs are associated with manufacturing product no matter where in the world it is done, and this situation is also far from static. For example, five years ago labor in the Far East was relatively inexpensive, but it’s becoming more expensive as time progresses. Costs are associated with performing project management at a distance. Ultimately, with final product and drug product, there are distribution costs and shipping costs, particularly as they pertain to products that require cold-chain management.”
“Technology Trends Enabling Local Manufacturing” by Jim WilkinsSensorin is developing a calibration-free pH measurement technology that it hopes to implement in single-use technologies and bioreactors. In his opening comments, Jim Wilkins focused on disposables and the influence of high-titer processes, process analytical technology (PAT), and quality by design (QbD) and how these trends fit together. He commented that many products coming on line will be required and made in lower quantities than past technologies.
“Single-Use Technologies, as you all know, have ushered in a revolution in bioprocessing. They allow closed processing without clean-in-place (CIP) operations. This allows greater flexibility in facility selection, when you don’t have to think about building huge cleanrooms but can, instead, build basically in a warehouse. You could put in a closed, single-use vessel and benefit also from technologies that are increasing product titers.”
James A. Wilkins, PhD, has been the chief technology officer of Sensorin, Inc. since February 2009. He was director of technology assessment at Genentech, Inc. 2004–2009. Before that, he started the process development group and finally served as vice president of process sciences and manufacturing at Alexion Pharmaceuticals beginning in 1993. From 1989 to 1993, Dr. Wilkins was group leader of the protein chemistry department at Otsuka America Pharmaceuticals, Inc. From 1987 to 1989, he was a scientist in recovery process development at Genentech, Inc. Dr. Wilkins earned a BA in biology from the University of Texas and a PhD in biochemistry from the University of Tennessee.
Higher Titers and Downstream Innovations: “Some monoclonal antibody (MAb) titers have been reported at >10 g/L. This is another revolution that’s taken place in the bioprocessing area: high titers of protein in bioreactors. High-capacity, prepacked chromatography columns can couple with these upstream steps to provide a seamless process using single-use technologies. This provides small companies with an option: Given the idea that they might have the technology and technical know-how in house, they can actually build these operations on their own site.”
PAT and QbD: “The FDA’s PAT and QbD initiatives are encouraging manufacturers to obtain better processing understanding and control. The more technology we can implement and the more we understand our processes, the more flexible we can be at selecting a manufacturing site — whether we do it ourselves or contract out. Better analytical tools can include new sensors, spectroscopy, and perhaps rapid HPLC techniques — again providing us with greater flexibility.”
Purification Platform Technologies: “MAbs usually provide an opportunity for very targeted therapies and for very targeted and defined purification strategies. Affinity chromatography’s been used to a great advantage in this area. Most MAbs are amenable to protein A purification. The move to platform technologies is taking place mostly at large companies. Large companies are investigating ways to make MAb processes more uniform, trying to develop platform technologies. Defined technologies enable local manufacture through greater understanding of potential process issues. As we better understand the molecules we’re working with, we will be better able to define our strategies.”
The Role of Contract Manufacturers: “Over the years, companies in the contract manufacturing area have developed a wealth of experience in technology. We can look to them to help small companies manufacture their products. Of course, one key to selecting a contract manufacturer is matching your technology needs with the strengths of a potential manufacturing partner.
“All in all, I believe technology will play an increasingly important role in how we manufacture our products. The ability of a small company to manufacture its own products is becoming greater and greater. And I’m looking forward to seeing how this plays out over the next few years.”
“Vaccines and Global Health,” By Rahul SinghviNovavax is an emerging vaccine development company with technology based on recombinant virus-like particles and a number of product candidates for influenza. Singhvi focused on how manufacturing innovations can improve global health.
“In the past century, other than clean water, vaccines were the single biggest contributors to improved public health. These days, vaccines are often taken for granted, but they are incredible factors in increasing our longevity and reducing illness. The travesty is that many people in this world still die of diseases that can be prevented by currently available vaccines. How can we improve access to life-saving vaccines and prevent preventable disease? I don’t mean to do this through charity, because that’s not sustainable in our view. We have to do this in a way that is consistent with the principles of capitalism — that is, we have to make a profit. How do those two goals come together?
Rahul Singhvi, CEO of Novavax, Inc., is a recognized vaccine specialist, manufacturing expert, and business leader in the pharmaceutical industry. Since assuming his current position, Dr. Singhvi has restructured Novavax to focus on product innovation and development, in particular influenza vaccine development using the company’s novel virus-like particle and Novasome paucilamellar vesicle technologies. Before joining Novavax, he headed vaccine manufacturing operations at Merck & Co., where production increased by 25% at the two plants he oversaw. Dr. Singhvi received his MS and ScD degrees in chemical engineering from the Massachusetts Institute of Technology. He also holds an MBA from the Wharton School and serves on the Board of Directors for the TechCouncil of Maryland.
“In the value chain of vaccine development is the initial discovery process, development, manufacturing, regulatory approval and, ultimately, distribution. Most R&D is going on in the West. In developed countries it tends to be by large multinationals such as Sanofi, Novartis, Wyeth, Pfizer, Merck, and GSK. Some low-cost manufacturing is occurring in India. The idea of giving everyone access to vaccines has been partially successful in pediatric combination vaccines such as diphtheria, tetanus, pertussis, hepatitis B, and polio. They’re given very early in childhood. I think the World Health Organization and UNICEF have done a great job of procuring vaccines from low-cost manufacturers such as the Indian companies and distributing them through channels in African countries and other poor nations.
“So it’s not that people haven’t done enough to improve access, but that, unfortunately, new problems require new solutions. Some newer problems we’ve seen in public health, such as pandemic influenza, cannot be solved with approaches that have been used thus far. For example, we saw this past year at the start of the influenza pandemic that only the richest nations in the world had access to vaccine. They also happened to be countries with manufacturing capacities within their own borders. It became very obvious that there is no single arbitrator of supply and demand for vaccines. Ultimately, control is in the political hands of governments and those who wield economic power. If you’re going to provide equitable access to vaccines, you have to assure local control or manufacturing so that local governments control their own supplies.
“Some diseases erupting in different parts of the world are very local to specific geographies. For example, chikungunya came up in India, and EV71 is local to China, Hong Kong, and Singapore. You aren’t going to spend money develo
ping vaccines for these types of diseases unless you have capability within those geographies.
“Both of those problems are creating local capability for development and production of vaccines. I’m grateful for Jim’s comments because he pointed to success factors that enable local manufacturing. In my mind, you have to simplify the system. For good tech transfer, you cannot have a massive facility with a whole spaghetti of stainless steel piping. You have to reduce the capital costs and change the barriers. Commissioning time has to come down. People can’t wait four years to have a facility built and commissioned. You have to reduce the dependencies on utilities and ensure quality through technology. And you have to have a way to use capacity, not just in volume and economy of scale, but if you are working with lower volumes you need to either drive the cost structure to more variable costs or have capacity use through multiple products.
“One way to address these issues is through a combination of approaches. At Novavax, we use a platform vaccine technology that enables a multiproduct approach. We add single-use technology within a modular facility. Those three things together can create a solution.”
Case Study — Pandemic Flu: “The Novavax platform technology is virus-like particles. This recombinant technology expresses some genes of a virus. As those genes are coexpressed, they come together and self-assemble themselves into a particle that looks very much like the virus. They mimic exactly the influence of particles except that they don’t contain genetic materials required for replication. So these particles can be redecorated with antigens of other diseases, forming a platform. We can decorate them with the surface antigens of influenza or SARS or HIV or other type [of virus] — so with the same manufacturing process, just by manipulating the genes, we can produce different types of antigens. That gives us some economy of scope. Also, we use simple, ready-to-use equipment such as gamma-irradiated bags or liners and low–mechanical-energy based purification systems.
“We also use a modular facility we’ve built in Rockville, MD: a 10,000-ft2 facility with four cleanrooms. There are only electrical outlets in these rooms, no stainless steel. There’s no clean-or steam-in-place. This is a quintessentially simple facility that you can plug in anywhere. In fact, we used it last year. It was one of the things that enabled us to get in the game last year (even though we were a development-stage company) and create vaccines with VLPs for the H1N1 California strain against the 2009 pandemic strain. We produced this material and tested it in Mexico, where the pandemic originated. A large clinical study (4,500 people) demonstrated that our product works.
“These approaches using lower-cost manufacturing without compromising on quality have enabled our small company of <90 people to get in the game. We have also transferred this technology to our joint venture partners in India: Cadila Pharmaceuticals. We started tech transfer in May and facility design in June. The ground-breaking ceremony was in November, and we expect to finish completion of that facility in May 2010. By shortening the commissioning time (with tech transfer ongoing), we expect this facility to come on line with full validation by July 2010. This gives you the full picture of how we've taken advantage of a platform technology, modular facility, and single-use technology to move and shake things around. Hopefully we can, over time, take it to the next level. And what's that next level? What's the vision?”
The Vision: “Can we possibly extend this approach by turning it into a franchise model? Could we have a turn-key technology transfer package — like when buying a Dunkin’ Donuts franchise, you receive a package and make donuts exactly the same way. How do we do that with vaccines? How might we turn this into a place where all that technology, all the science, all the ‘art’ can be turned into documentation so everybody can make vaccines — and make vaccines of the same quality? Then to take it to the next level, can we have in these ‘franchises’ custom menus of products that address local needs? That’s the vision. And I believe this is a vision that will come to fruition with innovations from people such as you.”
“A Distributed Manufacturing Workforce” by Mani Krishnan“Presentations from Jim and Rahul have set me up very nicely to talk about disposable technologies. So I’ll focus on how we can capitalize on single-use technologies to enable distributed manufacturing. It’s not really outsourcing or insourcing; we are talking about making what you need to make where you need to make it. You can actually do technology transfer without relocating people. So how do we make these technologies easy to use, easy to implement, and easy to commission while minimizing the impact of moving people from one country to another — and perhaps to another? All these things have to do with minimizing start-up and commissioning time, making it easy for a distributed workforce to actually make manufacturing happen. The question is how to empower and train local employees?”
Single-Use Technologies: “Two key elements associated with single-use systems are flexibility and the ability to close a process. We talk about operational flexibility, and Rahul talked about the ability to manufacture different drug products in the same facility, even at different scales.
“The second element, particularly when you consider distributed manufacturing models, is that you may have a less experienced workforce. In such cases, how do you enable closed manufacturing — that is, manufacturing that minimizes open manipulations? I think single-use technologies significantly get you to that point.
“A flexible filtration solution is actually a simple concept similar to Lego blocks. You design some basic building blocks, then assemble them on demand depending on your operation needs. Our process development scientists situated around the world go to customer sites and do some sizing on the basic solutions a process uses. For different solutions and different volumes, they recommend a sizing. Based on that recommendation, we design some simple filtration assemblies. Then we developed a universal single-use bioprocessing container that can be used in multiple locations. This is highly unusual because in most places, upstream folks design a bag that works only upstream, and downstream folks design a bag that works only downstream. What we’re asking people to do is think about a bag that works everywhere.
Mani Krishnan is a program director for Mobius single-use processing systems in Millipore’s bioprocess division. He has been with Millipore for 15 years and has held positions of increasing responsibilities in process development, sales, and marketing. He has extensive experience in process optimization and scale-up of numerous separations-related applications with small molecules, proteins, and viral vectors. Mani has a BS in chemical engineering from IIT, Madras and an MS in chemical engineering from University of Calgary, Canada.
“Now when you start putting all these things together, they click together like Lego blocks to give you a significant amount of flexibility in your operations. One advantage of this solution is its quick implementation. You can go from nothing to getting a buffer kitchen or media kitchen in place in three to four months. Once you have the components designed, putting it together doesn’t take any time at all. Then you’re using modular components that have been tested for robustness. So you don’t have a plethora of assemblies; you test only a few. And what is more important for those of you in supply chain management is a significant, favorable impact on the bottom line. Think about the number of assemblies you have to manage in your warehouse. If you have 10 different buffers at 10 different scales, you need the capacity to manage these assemblies. But when they are assembled as modular components, you have to warehouse only those specific modules. So from a warehousing perspective, from a supply chain management perspective, single-use and modular assemblies provide a huge advantage.
“A media filtration train is sized appropriately for a given volume. You put the manifold together, put in three 250-L bags, and click them together. You can do this in gray space because each module is independently gamma radiated. You bring it in house and connect them, and voila, you have a system for filtering 750 L of medium. Now for the next product you have to filter — say, a buffer at almost 2,000-L scale — you do the same thing but just plug in more bags. And you have a system that can provide the two or three times more volume. So the concept is very simple. But what it requires is working in close partnership with a single-use provider to design universal assemblies that work across multiple processes.”
Fill–Finish: “I also want to talk about single-use fill–finish. Here the key element is reducing the risk of contamination while still providing for rapid product change-out. So after you run your filling process, you can remove the single-use assembly and put a new filling line in, then run your next process. This was presented by Nigel Bell from GlaxoSmithKline, who reported significant reduction of intrusion time along with significantly reduced operator time. And there’s no worry about cleaning validation or legacy from contaminated vessels. All these things play very well into the concept of distributed manufacturing.
“Training is a critical component of taking your process to geographies with a less-experienced workforce. One thing we’ve done is expanded our training locations in Asia. We have training centers in India and Singapore, >70 engineers and scientists distributed worldwide that work with our customers. They assist in process development, in operator training, and scaling up operations. The idea is to provide both the technology that’s necessary to operate in distributed manufacturing environments and the requisite training to keep your operators up to speed. Single-use technologies can aid distributed manufacturing. And global training consistency is absolutely necessary to ensure manufacturing consistency.”
“The Global Biomanufacturing Environment” by Tom Ransohoff“As biomanufacturing consultants, we work with a wide range of companies to develop manufacturing strategies. These are obviously different for every different company because a manufacturing strategy needs to be aligned with your business strategy. Every company has a different pipeline, different risk profile, and different cost of capital, and these all need to be factored into the strategy. But one area that is relatively consistent is the industry environment. Over the past eight years, we’ve been building a database to help us look at supply and demand of biomanufacturing capacity. Our focus is on recombinant proteins in MAbs made in microbial fermentation and mammalian cell culture. We published a report at the end of 2008 to describe this type of analysis and data set in detail. Here I summarize an update we recently completed on the industry environment for global biopharmaceutical manufacturing capacity.
“Key drivers for capacity demand include the sale of commercial biopharmaceutical products. Sales of mammalian-cell-culture–derived MAbs and Fc fusion proteins are the fastest growing segment of the market over the past decade. About eight tons of material were required in 2009 to support ~$40 billion in sales. Other mammalian-cell-culture–derived recombinant proteins include factor VIII, erythropoietin, and other mostly historical products, although a number are in the pipeline too. To support ~$25 billion of sales for these products, only 100 kg (0.1 ton) were required. So antibodies and Fc fusion proteins introduced the need to produce significantly more mass of material per unit sales volume increased. This led to more and larger facilities and caused us to tell process development groups to improve volumetric productivity. As an industry, we were very successful at both strategies, leading to an embarrassment of riches: We now have a surplus of mammalian cell culture capacity. We also look at recombinant proteins, antibody fragments, and Fc fusion proteins made by microbial fermentation with bacterial or yeast systems.
“The other factor in projecting demand is to know what the pipeline looks like. About 60–65% of the products we’re tracking are derived from mammalian cell culture. This is actually decreasing as microbial-fermentation–derived products seem to be increasing gradually. Antibody products represent >80% of the mammalian pipeline. Demand for products in the foreseeable future (three to five years) is more of the same in a pipeline still dominated by antibody products.
Thomas C. Ransohoff, MS, vice president and senior consultant at Bioprocess Technology Consultants, has >20 years of experience in the biopharmaceutical industry. His expertise covers development and scale-up of biopharmaceutical processes, separation and purification technologies, CGMP manufacturing, and management of technology-based start-up ventures. He is a cofounder of Tarpon Biosystems, Inc. and of BioFlash Partners LLC, where he was President and CEO before selling the business in 2010. Before joining BioProcess Technology Consultants, Mr. Ransohoff was vice president of operations at TranXenoGen, Inc., responsible for purification process development and facilities, and before that vice president of bioseparations at Dyax Corp., where he was instrumental in establishing a business unit to develop novel affinity products using phage display. At Repligen, Mr. Ransohoff was senior director of manufacturing, responsible for CGMP pilot plant operations producing material for clinical trials and reagent products such as protein A for commercial sale. He has a BS from MIT and an MS from UC-Berkeley, both in chemical engineering.
“Switching to the supply side of the equation, what capacity is out there to make all these products? We have a separate database to track all the facilities that we can find public information on in terms of their bioreactor capacity and the type of technology they have. This came from work done at MIT by Liz Reynolds and Charlie Cooney’s group. We collaborated by providing some of the data, but Liz Reynolds evaluated why biomanufacturing companies choose to locate
in different areas and how that geographical distribution capacity is shifting. It’s really great work, available through the MIT industrial performance center Web site (http://web.mit.edu/ipc/publications/papers.html). One thing it shows is that as capacity construction levels off in the United States and Europe (where the majority of our industry’s capacity is held), capacity construction in Asia is picking up. That is certainly the fastest growing segment on the supply side. We can see distribution of capacity shifting worldwide. In terms of total liters, in 2010 we have just under 3,000,000 L of equivalent fed-batch bioreactor capacity for mammalian cell culture on line, and we forecast that to grow to about 4,000,000 L by the middle of the decade. Capacity growth is forecast to be primarily in Asia.
“Where does that leave us when we bring the two sides together? How well are we using our capacity? We need to make a number of assumptions, but we do it consistently every year, so it gives us a good picture of where we are and where we’re heading. We estimate the current use at about 50% worldwide, industry-wide. That capacity is fairly closely held, and it’s not exactly a liquid commodity. So we don’t think the industry could really operate at 100% capacity. Probably when you get to 75–80%, you’ll start to see individual product shortages because again, not everybody has perfect access. We do expect capacity use rates to increase gradually over the next five years. But we forecast a period in the coming decade with sufficient industry-wide capacity for mammalian cell culture.
Trends Affecting Global Biomanufacturing Capacity: “Jim [Wilkins] mentioned continued improvement in the throughput and use of existing facilities, particularly in process yields and process productivities. We have been successful at developing very high-titer processes for production, particularly of MAbs. Additionally we see better ability to operate existing facilities as the industry matures. So operational excellence initiatives have also had an impact on our ability to improve the use of capacity without any change in process. These improvements are moving the bottleneck downstream in these facilities, and that’s led to a need to develop new downstream technologies because the downstream side hasn’t kept pace in technology development. We and others are seeing fewer blockbuster products. This tells us that products will need more and more multiproduct facilities. That’s already happening, but we will see more in the future. Companies also need to match their supply rapidly with changes in demand, which as the other speakers have noted, increases industry-wide use of disposable technologies. I think we’re seeing that coming at us very rapidly for the reasons they mentioned: reduced capital investment, increased speed, and the ability to transfer a process that’s embodied in the disposable technologies you’re transferring.
“In summary, we think that industry-wide there’s likely to be capacity on the mammalian cell culture side through the middle of the decade. But this capacity’s very closely held. If you don’t have capacity, your manufacturing strategy may be to build it. Geographical distribution is shifting. Improvements in overall process productivity have been significant over the past decade, and we expect that to continue. It’s pushing the bottleneck downstream and enabling disposable technologies to be considered for making commercial supplies. This accelerates the drive to disposable technologies across the flow path — which is, I think, driving innovation in areas such as continuous chromatography and membrane absorbers. As a result, developing a manufacturing strategy is becoming more complex. It’s no longer just make-versus-buy of a stainless steel facility. Now acquisition’s increasingly an option. Whether you use traditional operations or modern disposables is also a question. And risk assessment is more important than ever in making that decision.”
Audience Questions and AnswersQuestions here are paraphrased. The first asked for the speakers’ sense of where companies are in designing fully disposable facilities and at what point the industry is likely to see such facilities become the norm.
Ransohoff: “The industry invested heavily in stainless steel, large-scale facilities, and once you’ve made that investment, there’s an economy of scale that is difficult to beat. But the picture looks very different depending on where you sit. If you’re Genentech or Roche, invested in very large-scale capacity, you have those on line already, so the impetus to implement disposable-based processes is not there and may not even make sense. But if you’re making smaller-scale products and you don’t have capacity on line, then the time and cost to build a stainless steel facility is still quite large. Now you have the option of bringing a low-cost, low-capital facility on line.
This is very attractive, and it’s potentially enabling for smaller companies. So I think we are seeing disposable facilities come on line. They tend to be smaller in scale now. People are talking about 2,000-L and 5,000-L bioreactors, but there are very few of those in operation. The technology right now is more at the 500-L to 2,000-L scale. But I have no doubt that we’ll see commercial facilities that are heavily based on disposables in the coming decade.”
Are those new facilities coming on line indeed designed to be flexible for manufacturing multiple products?
Ransohoff: “Absolutely. We’re seeing such technologies being implemented more by CMOs who need to, by definition, handle multiple products. And Acceleron up in the Boston area is a good example of a small company that’s decided to do some manufacturing in house. Once you develop a disposable manufacturing capability, you can much more easily implement multiproduct operations because you really do have fewer cleaning and changeover validation activities.”
Broeze: “We’ve used single-use technologies at Laureate for many years, going all the way back to hollow-fiber, migrating into Wave, and now the larger single-use stir-tank technologies. They’re very amenable to multiuse and multiproduct manufacturing. And some efficiencies are definitely gained in using this technology, specifically the elimination or decrease of cleaning requirements and validation and cross-product contamination. So it’s great for our industry, and I see it being a moving force in the future.”
Singhvi: “We’re seeing customers migrating to single use for clinical trial manufacturing. A big biotech may have large-scale facilities, maybe 10,000-L or 20,000-L bioreactors for commercial production. But for preclinical phases 1 and 2, we see some larger companies migrating to single-use technologies because of multiproduct requirements.”
What is the most important area to be improved first to really make disposables into the next generation technology? Would it be process-specific, as a platform?
Singhvi: “From my perspective, you have to have high-yielding processes for disposables to work. You have to have a high-data process. If the yields are not there, it doesn’t make sense. When you move to disposables, you move to a more variable cost structure. The cost of goods makes investment work only if you have a high number of doses coming out of the process. You have to have scalability. At this point, 1,000–2,000 L is probably the ‘workhorse.’ But you need to go higher. Also, interconnectivity of unit operations has to be much more refined so you can truly have a high-quality closed system that allows you to transcend the facilities.”
How will the technology be modified for personalized approaches?
Ransohoff: “The ability to manufacture personalized medicines and/or smaller-scale products is a natural f
it for disposables. I think the verdict is still out as to whether we’ll see truly personalized medicines. Some aspects of drug development costs are not going to scale down as easily as some think. Quality and development costs for developing drugs are still going to be significant. So just for me, as much as I might like the idea, it’s probably not going be practical in the near term. But certainly smaller-volume, niche products appear to be a big feature of the pipeline. I think those will benefit significantly from disposables, both in lower capital investment and the process portability and lower tech transfer and cleaning costs.”
Singhvi: “There have been a couple of very interesting publications on this topic. Lindsay Leveen from Genentech has published extensively on this subject. What I’d like you to think about is a concept of total waste. When you talk about single use, what you get is a lot of visible waste. Although that is visibly unacceptable, in traditional facilities there is also a significant amount of total waste, meaning the liquid that’s going under your feet in the drain line. It may not be your problem, but it’s somebody else’s problem. If you look at the total cost of treating that waste and the amount of energy used in generating hot water, single use comes up as almost the equivalent cost or in some cases actually more — but with a smaller carbon footprint than traditional technologies. Just imagine the amount of heat required to convert water into steam.”
For products used in CGMP biomanufacturing, are there any regulatory hurdles in terms of materials used in their construction?
Wilkins: “We are seeing increasing focus on testing for leachables and extractables as well as interest in ensuring that all materials are animal-component–free. You wouldn’t think that would be so difficult for plastics, but a number of plasticizers are actually derived from beef tallow.”
Broeze: “Another regulatory challenge is assuring maintenance of a sterile system. With stainless steel, you can steam sterilize. With single-use technologies, you don’t. So a whole technology has evolved around making sterile connections, making sterile welds, sealing tubes, and similar types of operations.
“The process analytical technology (PAT) initiative made it very clear that the FDA wants manufacturers to have a better understanding of their process and implement those technologies into their processes so they can control and understand them better.”
More Information On Our Website
Find the full presentation slides from each speaker online at
www.bioprocessintl.com/bpiextra!