September 15, 2017 (Vol. 37, No. 16)
Biosimilar Manufacturers, Take Note: This Stream Is for You
Although estimates vary slightly, many industry experts predict continuous manufacturing will, at the least, cut the cost of manufacturing biologics in half. Thus, it will be an attractive option for drug makers looking to trim manufacturing budgets. But not all biologics would be feasible candidates for a truly integrated processing stream, and a truly continuous line may require a larger initial capital investment.
To learn more about which companies and products will likely incorporate continuous manufacturing first—and to understand which “hot-button” questions regarding implementation still require attention and clarification—GEN spoke to pioneers in the field of continuous biomanufacturing, including Massimo Morbidelli, Ph.D., professor of chemical reaction engineering at the Institute for Chemical and Bioengineering at ETH Zürich; Andrew Zydney, Ph.D., distinguished professor of chemical engineering, The Pennsylvania State University; Michelle Najera, Ph.D., downstream development scientist, CMC Biologics; Gerard Gach, chief marketing officer, LEWA-Nikkiso America; Dana Pentia, Ph.D., senior application scientist, and John Bonham-Carter, director of upstream sales and business development at Repligen; Gerben Zijlstra, platform marketing manager, continuous biomanufacturing, Sartorius Stedim Biotech; and Karol Lacki, Ph.D., vice president of technology development at Avitide.
Common conclusions from the experts were that the use of surge tanks in continuous lines has both benefits and drawbacks (see the Bioprocessing Perspectives column on page 22 in the September 15 issue of GEN); enzyme-replacement products and monoclonal antibodies (mAbs) will likely be the first product candidate types selected for integrated continuous manufacture (more specifically, mAb-based biosimilars); and manufacturers are less likely to switch legacy products from existing batch processes to continuous operation. The high prices of affinity resins are not expected to decline significantly in the near future, and continuous production could result in resin cost savings—but the experts explain there are many benefits of continuous production of biologics besides those related to cost.
Most of the products that would likely be made in integrated continuous lines will be new products, and will also primarily be labile biologics or those which have uncertain demand. Continuous operation, says Bonham-Carter, will allow engineers to react to fluctuations in product demand and give companies the option to build late (or build out) significantly, if required. Dr. Morbidelli asserts that peptides, fusion proteins, scaffolds with mAbs, and other products “containing sensitive antennary glycostructures that are needed for biological activity” would also be types of therapies that could be made in continuous flows.
To tell which pharma manufacturers are likely to integrate end-to-end continuous manufacture into production lines first, investors and industry insiders should follow company patent filings and peer-review articles authored by company representatives. As Dr. Zydney points out, manufacturers that are already “clearly interested” in fully continuous biomanufacturing lines include Genzyme, Merck, Bayer, and Sanofi, among others. Also doing work in this space: Novo Nordisk, Novartis, Amgen, Shire, Pfizer, WuXi, and BiosanaPharma. In addition, some contract manufacturing organizations (CMOs) are in the continuous biomanufacturing field (such as CMC Biologics), and these organizations are poised to help biopharma clients lower the cost to clinic by way of fully continuous manufacturing strategies.
Regardless of who is or is not investigating continuous, “What is definitely a right development trend is that vendors do support the change and offer technologies that enable continuous operations,” says Dr. Lacki of Avitide, a company that makes affinity resins. “At the end of the day, it will be up to the end user to decide how a process needs to be operated, but the choice will be made on a thorough assessment of commercially available technologies.”
GEN: What types of biologic medications are the best/most feasible candidates for integrated continuous manufacture?
Dr. Najera: Any product with an expensive chromatography resin is a suitable candidate for continuous chromatography, especially for early-phase clinical products, where resin cost can be cost prohibitive. Continuous capture is also attractive for high-titer processes, since several small volumes can be used instead of a single large column. In fact, facility fit limitations are common for high-titer processes (>5 g/L), and continuous chromatography can help manufacturers avoid the capital costs associated with procurement of large stainless-steel columns (>80 cm diameter) and associated equipment. Finally, the concept of integrated continuous manufacture, (i.e., a fully continuous process), would be best suited for a well-established late-phase process with a stable market demand where cost-savings are realized through basic efficiencies related to batch versus continuous processing.
Dr. Zydney: This is a difficult question to answer—it very much depends upon one’s perspective. In some ways, the ‘best’ candidates for integrated continuous manufacturing are products for which there are particular challenges in using batch operations. For example, a highly labile product that degrades over time would benefit dramatically from the use of a continuous process—this is why perfusion bioreactors were originally used for production of unstable clotting factors. On the other hand, integrated continuous manufacturing is well suited for products that have very high-volume demand or significant cost constraints. The successful development of integrated manufacturing systems will likely require considerable product and process knowledge, which today is most readily available for mAb products due to the large number of these products already in commercial manufacture.
Dr. Lacki: I would rather ask a different question: What type of expression systems or upstream technologies are more suited for continuous operation? And my answer would be perfusion operation or even [use of] six-pack fed-batch bioreactors would make an operation quasi-continuous. That said, I think that a successful continuous biomanufacturing process must be as simple, or rather, as robust, as possible. For instance, a downstream process that relies on as few chromatography steps as possible will be more suited for continuous operations. From that perspective, one could argue that an affinity step is an enabler for continuous downstream operation. Cost of an affinity resin will be lower if it is operated in a continuous manner, but even without that, the benefit of normalizing a product stream through a selective capture step would deliver so many advantages from the process reproducibility and controllability perspective that the cost of the resin need not even be considered.
Mr. Zijlstra: I would say [the most feasible type of therapies would be] primarily labile products that require perfusion and immediate capture from the cell broth to prevent product degradation. Some cost-model studies also suggest that beyond certain annual production scales, integrated continuous biomanufacturing leads to lower COGs for mAbs and are preferable in that case. Finally, evidence is mounting that even for mAbs, (critical) product quality attributes can be controlled much more tightly [in continuous than] in batch operational mode.
GEN: The manufacturers of Prezista (Janssen), a small-molecule drug, got FDA approval to change its processes to a continuous method. To your knowledge, are any biologics manufacturers looking into a similar manufacturing change? Do you think manufacturers are more likely to address the end-to-end manufacture of a totally new product, rather than for an existing product?
Dr. Najera: While it is easy to understand why people draw parallels between small molecules and large molecules, it is important to note that control of small vs. large molecules is significantly different. For small molecules, it is relatively easy to understand all impurities and variants through analytical testing. This is not always possible for large molecules. For this reason, continuous processing and process analytical testing (PAT) are often linked. Even with the challenges large molecules pose, there are many companies presenting compelling small-scale data on continuous purification strategies. However, these processes are inherently more complex and are perceived to pose higher risk. To the question, it is unlikely that an existing commercial manufacturing process would shift to a continuous process unless there was an extremely compelling cost benefit in doing so. The benefits would have to offset the cost of process development, establishing a new process at-scale, process validation, and potentially, new clinical studies. Additionally, significant facility modifications would likely be needed to convert a fed-batch facility to a continuous facility. For this reason, it is more likely that a fully continuous process would be developed for a new product.
Dr. Lacki: A change of a legacy process to a continuous version will only happen if a second- [or] third-generation of a process is being considered anyway. However, change of the infrastructure might prove financially prohibitive. A different story is if a company is building a new facility for the legacy product or for a brand-new API; in that case, continuous operation will definitely be evaluated as the capital investment associated with a continuous plant, at least from the equipment-side perspective [and] should be much lower compared with batch-based manufacturing.
Dr. Zydney: I am not aware of any biomanufacturer that is currently looking to replace an existing batch process with an end-to-end continuous process. I think it is more likely that a fully continuous process will be first developed for a new product that is particularly well-suited to continuous manufacturing, or potentially for a biosimilar where the cost reduction would be particularly attractive.
Dr. Morbidelli: It is now demonstrated that continuous processes provide higher quality and more homogeneous products, which are beneficial for the patient. This justifies the strong support that FDA always provided to the development of these new manufacturing technologies. In addition, these can also lower the production costs in the broad sense, which will obviously impact the growing market of biosimilars. This is why all major pharma companies are looking carefully at developments [in continuous manufacturing], although I do not know which approach they will select to implement this important transition.
Mr. Bonham-Carter: No biopharma would consider changing an existing commercial manufacturing process, in my view. Some might consider changes between Phase II and Phase III, for reasons of capital risk, uncertain market demand, tighter quality requirements, or manufacturing network management.
GEN: What are the financial implications of a truly continuous biomanufacturing line?
Mr. Bonham-Carter: Estimates range up to >80% cheaper in capital costs, but perhaps only 20–60% in cost of goods. Most models are relatively simple and do not take account of a company’s portfolio of drugs, risks of failure at different phases, quality requirements for different drugs, and existing manufacturing and engineering skill—each of which will impact true cost to the manufacturer.
Dr. Morbidelli: It depends which parts of the process are exchanged to become fully continuous, and those which would benefit from intrinsic step process improvements. A conventional fed-batch process can be nicely coupled to continuous downstream purification with one of several steps performing continuous batch (flip-flop) or countercurrent processes (capture and polish). If combined with membranes and single-use concepts, we expect a significant productivity increase having an implication on both CAPEX and OPEX. We estimate that besides the improvement in product quality, both CAPEX and OPEX could be lowered by 50%.
Mr. Zijlstra: Continuous and single production plants will be able to lower costs several fold and produce the same amount of product as current standard stainless-steel facilities. This facilitates greenfield investment decisions and reduces investment risks considerably.
Dr. Lacki: A standard argument is that equipment for a continuous line will be smaller. At the same time, truly continuous operations run 24/7, which means that even the downstream operation might need to be run in shifts, increasing labor cost. Undoubtedly, continuous operation will be more control-heavy. Cost of controllers, equipment maintenance, contingency plans—all of it needs to be considered when evaluating continuous processes. That said, advances in detection technologies and modularization of standard processing technologies (e.g., prepacked columns, flow paths, and even introduction of humanoid robots) will bring us closer toward continuous processing.
GEN: To date, why do you think so few biologic manufacturers have explored the use of an end-to-end continuous line?
Dr. Morbidelli: We think there are actually quite a few [therapies] in the exploration phase for either perfusion culture or continuous downstream or both. We know that manufacturers are currently exploring continuous systems at the pilot scale [that] would move into clinical trial manufacturing in 2018. FDA has consistently supported and guided continuous manufacturing and we believe that the regulatory risk for a new product development is low. As with any new technology, continuous manufacturing for biologics will need some evaluation time to be taken into the commercial setting, as the technology has to show that it is robust and provides the expected benefits. There might be a level of confusion as to the quality of available technology platforms with different claims on performance/productivity. We believe that a strict scientific approach in evaluating the benefits of each technology may drive the choice to platforms that are simple and robust, reducing the validation effort and likeliness of hardware failure in continuous operation.
Dr. Lacki: The short answer could be, a lack of real need. The industry is still young, and fairly profitable. The focus from the beginning was more on drug safety than on reducing the cost of goods. But as in the case of all industries, with technology maturation comes challenges related to implementation of technological solutions that will be able to address the market pressure, without sacrificing the safety profile of a product. The pressure is both related to the pure amount that needs to be produced (number of medicines keeps increasing) and to the expectation from society to make these medicines affordable for all. Personally, I don’t think the [biologics] industry will ever approach the commodity industry where every tenth of a cent counts—but it will definitely aspire to lower the manufacturing costs, to deliver affordable medicines, and to stay competitive against its peers. The patent cliff is here to stay.
Dr. Zijlstra: Many large pharma companies are actively testing intensified and continuous approaches in their advanced research and development labs. Intensified technologies are being implemented as we speak. However, before the fully continuous technology is accepted by manufacturing as a replacement for the existing, proven, and robust platforms, there is still some technological and regulatory strategy development required.
Dr. Najera: I think it is related to a struggle within the industry to be able to define a single batch or a lot of drug product from a regulatory perspective. A drug product is not complete without the associated paperwork that describes the processing (e.g., batch records, analytical results). Continuous manufacturing really challenges us to rethink how to present this paperwork to sufficiently demonstrate how the purification process to make that drug is under control. For example, a continuous process may rely on time-based process analytical data showing that the product quality is well controlled, rather than analytical results for intermediates at specified points in the process. In sum, although the regulatory agencies have encouraged the introduction of more continuous processes, a strategy for how to define a ‘continuous batch’ it not straightforward.
Mr. Gach: The ‘continuous’ market is in a stage like when single-use was first introduced—it was adopted in the areas where [it would have] the highest benefit or were the most developed areas, and only recently has grown into a full-on production scenario. Presently, leading companies are actively applying continuous technology in unit operations where they are feeling the most pain and the technology is straightforward. With these experiences, they will gain confidence and (regulatory) buy-in to expand further up- or downstream. Some will eventually grow into full continuous, while others will gain significant benefit using ‘continuous’ technologies to debottleneck batch operations.
Dr. Zydney: Biomanufacturing tends be ‘conservative’ in its approach—the primary goal is to provide a manufacturing environment that is capable of delivering the desired biotherapeutic in a fully robust fashion, while satisfying all safety and regulatory requirements. Fully continuous biomanufacturing is currently ‘untested’; there is inherently greater uncertainty regarding the manufacturing process, including the approval of that process by the appropriate regulatory agencies. Many biopharmaceutical manufacturers would be happy to be ‘second’ in the development of a continuous process, but they are very reluctant to be first. Some of this is cultural/organizational—the individuals involved in the development of new technology solutions that will be needed in continuous bioprocessing may not be the people making the final decision on the design of the manufacturing facility. However, considerable progress is being made, both by technology providers and biomanufacturers—I think it is only a matter of time before we see a continuous process implemented at pilot scale (e.g., for production of a biotherapeutic for use in clinical trials), and then ultimately, at manufacturing scale.
Dr. Pentia: It is still early [for many companies to consider] continuous biomanufacturing. The idea only started to be taken seriously approximately five years ago. The industry is still trying to understand where it is appropriate and how to fully control continuous manufacturing at large scale (tens of products are already produced in perfusion at commercial scale). I would not say there are too few manufacturers that are exploring this—it might be more accurate to say that every manufacturer is considering this approach, but some are further advanced than others, while others are just in the conceptual phase.
Read the rest of GEN’s supplement on Continuous Bioprocessing.
Continuous Bioprocessing Is Coming
Continuous Bioprocessing: Is It for Everyone?
Moving toward Continuous Bioprocessing
Bioburden Control in Continuous Capture of Monoclonal Antibodies
Single-Use Sensors in Continuous Bioprocessing
MilliporeSigma Sees Opportunity in Continuous Processing
Case Study: Insulin Increases Influenza Virus Yield