GEN‘s contributing editor K. John Morrow, Jr., PhD, recently spoke with several industry leaders to discuss the issue of bottlenecks in the biopharmaceutical industry and how they can be addressed as the industry evolves. The panel included:
GEN: The issue of bottlenecks is long standing in the industry, but is it as serious as some analysts suggest, and if so, what are the possible remedies?
Dr. Çaylı: I don’t think that the bottleneck question is such a big problem anymore. Twenty years ago people were speaking about bottlenecks and new production plants that needed to be built. At that time this issue was more significant because the planning and building of a production facility required at least three years. In terms of capital demand, it called for a commitment of several hundred million dollars in the facilities’ construction costs. It was capital intensive, risky, and a long and demanding process. Today, if there is a demand, facilities can be built quickly, through the application of single-use technologies. Capacities can be expanded quickly because of modular building concepts, and big investments are no longer required.
Another transformation in the industry is that biopharmaceutical production expertise and capacities are now spread throughout the world. Previously, a few highly industrialized countries and companies were able to dominate the bioproduction sector of the world economy. The present ubiquity of the industry provides the flexibility to diminish (or even eliminate) the bottlenecks.
Mr. Varma: It is certainly not as serious as it used to be. This is in large part due to improvements in the quality of the mAbs being manufactured; with better performance the volumes do not need to be as great. And the industry can always build more bioprocessing facilities. The bottleneck problem becomes more significant if the profit margins on a particular product are too low. For mAbs the profit margins can be 90%, but if they are as low as 30% companies may not want to produce them.
Mr. Linz: No, because with the current unit operation, substantial improvements for downstream processing are feasible. Scheduling will have an impact on throughput and also more flow-through technologies will help to improve productivity in downstream.
Dr. Zijlstra: The industry is facing several manufacturing challenges due to overarching market trends, such as the advent of new product modalities which do not fit current platform processes, increasing cost pressures by biosimilars, and significant overall market growth on one hand—but with uncertain market demand per product and region.
These bottlenecks are especially serious for the existing large stainless-steel facilities operating in batch mode. With their fixed assets optimized for platform mAb production, i.e., multiple bioreactors feeding into one downstream train, they have limited flexibility in terms of process changes. Moreover, they may not be suitable for new products of a different nature, with different annual needs or titers. mAbs with higher titers may exceed the capacity of the installed DSP train, while mAbs or Fc-fusion proteins with limited annual needs may need only a single batch, thus requiring extensive downtime for changeover. Although these might still be dealt with creatively by using the N-1 bioreactor as a production vessel, this may result in increased COGs. Different modalities may even require substantially different downstream technologies or sequences and might not fit in those existing facility.
For these reasons, several different strategies are being employed, including outsourcing and risk sharing with CMOs, as well as investments in flexible intensified (hybrid) single-use facilities. By intensifying the upstream process, similar kg outputs (typically 20–40 kg) can now be obtained from 2000 L single-use bioreactors as previously obtained from 10-kL fed-batch bioreactors, while downstream can use similar scale (but mobile) downstream skids, achieving full flexibility in terms of product changeover and unit operation sequencing.
Dr. Bulpin: In my view, focusing on individual bottlenecks misses the context of what our industry is trying to achieve. The outlook for mAb market growth remains strong, yet significant uncertainties for product demand and competition from biosimilars and novel modalities such as cell and gene therapy require that supply chains and manufacturing capacity are flexible and reactive.
GEN: Bioprocessors have achieved great success over the years in ramping up upstream production by engineering cell lines. Is it conceivable that as upstream production continues to improve, the downstream end will never catch up, and we will be eternally faced with this issue?
Mr. Varma: No, I don’t believe this is a serious problem for us today; in fact, downstream technology has caught up with upstream technology. One of the issues that we have dealt with is the use of Protein A for mAb purification. We can reuse Protein A up to 100 cycles, but only for purifying the same protein. So if you are managing a number of different projects, you end up with a substantial inventory in your warehouse. Since the shelf life for Protein A is about two to three years, your number of actual reuse cycles is probably going to be much less, usually around 30 cycles.
Mr. Linz: No, because with the current unit operation, substantial improvements for downstream processing are feasible. Scheduling will have an impact for throughput and also more flow-through technologies will help to improve productivity in downstream.
Dr. Zijlstra: The main advantage of increased upstream titers is that now smaller single-use bioreactors can be used, allowing a more flexible approach, which is needed to cope with current product portfolio and market trends. So far, the kg outputs of these 2000-L bioreactors can still be processed in a batch mode with existing equipment. When increasing titer beyond the 2000-L scale, and up to the 10,000-L scale (>5 g/L), downstream productivity (g/L/hr) will need to increase. In the current batch downstream processing approach, productivity can still be further increased by resin capacity increases (e.g., MabSelect Prisma) or by column size increases (e.g., by radial flow principles).
However. a more (cost-) efficient method to increase downstream productivity may be to switch to more “continuous” and connected approaches, such as Multi-Column Chromatography (MCC) and flow-through chromatography, with resins or even with membrane absorbers (MA). These technologies can increase productivity dramatically (e.g., 4–6-fold for MCC and more than 10-fold for MA), thereby dramatically reducing the downstream footprint, installed capital, and consumables while increasing flexibility.
Dr. Bulpin: It is true that bioreactor productivity has increased significantly in the past decade; however, there are downstream technologies that offer orders of magnitude improvements in throughput. One area that is particularly exciting is membrane-based capture chromatography. Protein A resins are limited by the relatively high residence times, on the order of 4–8 minutes, required to achieve high capacities. Membrane-based devices can operate at residence times of 1–10 seconds, making these processes 10- to 100-fold more productive. These devices can be cycled greater than 100 times while processing a batch, enabling a truly single-use capture process.
GEN: One bottleneck is the lack of adequate staff to fill industry positions. How could this be addressed, and is the challenge made more difficult by current immigration laws?
Dr. Çaylı: It is indeed true that to find the appropriately skilled personnel in the bioprocessing sector is difficult. We finally gave up the search to find someone with the precise experience required for a specific position and to hire that person. So now we search for individuals with a solid basic education, flexibility in their thought processes, and the ability to easily integrate into our company culture. Then we train our new employees in our goals, culture, and technology in an intensive, year-long program.
My feeling is that current immigration laws throughout the world and their possible future adjustments will not change the recruiting picture. The industry is constantly searching for highly specialized professionals. Those professionals are characterized by their flexibility in relocation choices until now as well. In other words, these people could move from one country to another and find employment easily. Modifications of present immigration laws will not change this dynamic.
Mr. Varma: This is an increasing issue, depending on location. At our New Jersey facility, most of the local talent was engaged in making small molecule API and less in biologics, so we had to hire outside the local area, requiring more from the immigrant pool. This means that it is important to have a fairly liberal immigration policy. In biotechnology hubs like San Diego the job pool is large, but there is a lot of competition from the many technology companies in the region. But overall, we are not anticipating a big problem in filling our positions.
Dr. Bulpin: We recognize the need to develop the next generation of talent to be able to act on the innovative methods that are being developed today. These skills are not necessarily currently taught in universities. As an active Tier 1 member of the National Institute for Innovation in Manufacturing Biopharmaceuticals (NIIMBL), [I can say] there is a key focus on the workforce development activities to build these skills for the future. Our company pursues collaborations across borders, cultures, and specialties; we have found that immigration laws are not a limitation. This involves working with our global drug manufacturers to help bring drug therapies to market, in an expedited timeline.
GEN: Provision of adequate buffer volumes for large scale manufacturing is a possible source of bottlenecks. It has been suggested that outsourcing production of buffers could constitute a possible solution. Is that a valid approach, or are there other, more feasible alternatives?
Mr. Linz: Although other industries have proven that smart outsourcing will lead to efficiency increases, facility designs addressing this approach still need to be developed. But undoubtedly, it would be a clear benefit for manufacturers.
Dr. Zijlstra: Outsourcing buffer manufacturing could definitively reduce the on-site medium and buffer prep effort dramatically. To make it cost efficient, ideally a limited number of concentrated buffers, from which the final process buffers can be reconstituted in-line, should be defined.
Dr. Bulpin: We see a key need for reduction in the footprint and resources required for buffer prep at biomanufacturing facilities. Working with our customers directly and through the BioPhorum Operations Group, we are developing an optimized set of buffer concentrations with a precision buffer dilution system that allows for effective outsourcing of buffer manufacturing. Buffer concentrates dramatically reduce the cost of shipping, and the automated buffer dilution system allows for preparation of numerous buffers for the downstream process without the need for powder weighing, dispensing, and hydration.
GEN: As modern biopharmaceuticals become more potent, this should decrease the bottleneck problem, but is this really happening?
Dr. Çaylı: Yes, in some aspects it is indeed occurring. As innovative biopharmaceutical products become more potent, the amount of protein per dose decreases dramatically. A well-known example from the early days of biopharmaceutical production is mAbs. These were required in extremely high doses, and especially for the worldwide market, volumes in the range of many kilograms were required to meet demand. Since the potency of drug molecules has increased, the required amount per dose decreases, so that the pressure on companies to meet these requirements decreases as well. In the future, change may not be as dramatic or as rapid as we would hope, but it will surely continue.
Mr. Linz: Different molecular types will require different process approaches. The current pipeline consists of highly potent molecules where the annual demand is low. Flexibility of facilities for high potency drugs and larger scale manufacture of drugs represents an important change in the facility design.
Dr. Zijlstra: Indeed, more and more potent molecules are being developed, which leads to lower annual needs; however, at the same time, the global market for these molecules is also gradually increasing. Furthermore, having more and more small products will increase the product changeover challenge.
Dr. Bulpin: Novel modalities certainly create uncertainty for the future of biopharmaceuticals. Technologies such as antibody-drug conjugates (ADCs), cell and gene therapies, and viral vector therapies bring their own set of process challenges. While with mAbs the challenge is high-volume production, cell therapies are often patient-specific, produced at low volumes. This puts our production model on its head with high diversity at low process volume.
Through our cell and gene therapy program, we are developing efficient bioreactor methods that allow for sterile, closed processing of these treatments in collaboration with many of the emerging leaders in this field. In addition, our M Lab™ Collaboration Centers are designed to offer a lab setting where viral vector manufacturers can experiment without disrupting their production line.
GEN: Advances in bioprocessing are, on the whole, incremental. What disruptive technologies might result in quantum leaps for the industry? That is, are there concepts on the drawing board that could partially or completely revolutionize downstream processing?
Dr. Zijlstra: All downstream processing begins upstream. With the increasing ability to genetically alter and optimize biopharmaceutical production strains, one can imagine that difficult-to-remove host cell proteins will be partially knocked-out or otherwise reduced. Subsequently, starting with much purer product, a full flow-through downstream, and only removing contaminants rather than binding and eluting the product may revolutionize bioprocessing.
Dr. Bulpin: Delivery of revolutionary changes will require a journey across many disciplines. Through our BioContinuum™ platform, we are bringing a holistic, cogent effort to advance process intensification, single-use technologies, process analytics, software and controls, and digital integration. Adoption of this suite of technologies will enable future improvements with significantly reduced footprint and capital and operation expenses. The platform features next-generation technologies to obtain incremental process benefits now, with a mind to the continuous process of the future.
We believe this platform will deliver greater flexibility in the quantity and number of molecules that can be processed while improving quality and significantly reducing cost. Our platform will revolutionize drug manufacturing by setting the standard for improvements in process efficiency, simplified plant operations, and consistency in manufacturing.
John Morrow, Jr., PhD ([email protected]), is president of Newport Biotechnology Consultants. He is also the co-editor of “Biosimilars of Monoclonal Antibodies: A Practical Guide to Manufacturing, Preclinical, and Clinical Development, published by John Wiley and Sons.