In the early 2000s, the trade press was abuzz about an imminent “capacity crunch” in mammalian cell culture. Dire predictions of shortages were based on biopharmaceutical successes to that point, on bursting development pipelines, and on the lengthy timelines and high costs of assembling tens of thousands of liters of stainless-steel bioreactors and supporting infrastructure.
Those predictions failed to anticipate several positive developments that would render doom-and-gloom scenarios moot. Notably, yearly improvements in protein titers for MAb processes already were raising volumetric productivity from tens of milligrams per liter to as high as 10 g/L for monoclonal antibody (MAb) manufacture in Chinese hamster ovary (CHO) cells (1). Simultaneously, companies were beginning to apply single-use technology that previously had been limited to buffer preparation and hold tanks to more critical product-contact applications. Single-use bioreactors then surpassed the 200-L, 500-L, 1,000-L, and 2,000-L volume barriers. By the time US regulators issued guidance documents on process analytics, quality by design, and risk-based manufacturing — initiatives that provided a regulatory “comfort zone” for experimentation — concerns about capacity shortages had returned to normal levels.
Current biomanufacturing faces similar issues of preparedness, not for therapeutic proteins as much as for emerging therapies, particularly cell and gene therapies and vaccines. Added to those concerns are day-to-day capacity needs and supply chains strained because of trade restrictions during the pandemic. Because very little in the way of equipment and unit operations from the MAb/CHO world applies directly to emerging therapies, developers are scrambling for capacity and platform methodologies for expression systems, purification trains, formulation methods, and filling capacities. Such factors have led to a flurry of activity in the design and commissioning of new facilities.
One unexpected source of additional capacity may emerge when COVID-19 is behind us. We know that significant research and development resources have been devoted to vaccine discovery this past year. In addition to about a dozen coronavirus vaccines authorized (but not approved) globally as of February 2021, 60 more were in mid-stage clinical development, and another 150 or so were in preclinical testing (2) [Editor’s note: These numbers are updated twice weekly.] Most of the manufacturing capacity associated with those projects that have not already reached phase 3 is likely to be released back to the market.
Years in the Making
Because facility commissioning is such a long and expensive process, it is unlikely that COVID-19 has been much of a factor, aside from possibly diverting capacity from established programs toward making coronavirus vaccines and treatments. Rather, the past year served as a wakeup call (3) — perhaps suggesting a new paradigm for research, development, and approvals for future health emergencies.
“Demand for manufacturing capacity was growing before COVID-19, but the pandemic certainly is taking this need to another level,” says Dirk Lange (CEO of KBI Biopharma and Selexis). “We see many contract and development and manufacturing organizations (CDMOs) allocating significant capacity toward the manufacture of COVID-19 treatments and vaccines. This trend leaves less capacity available for clients to drive their first-in-human and late-stage programs forward. As a result, we are seeing increased demand, and we expect this to continue past the current acute phase of COVID-related manufacturing.”
Maik Jornitz (chief executive officer of G-CON, which manufactures flexible, podular facilities), agrees that the next capacity crunch has been years in the making. “First and foremost, the new therapeutic entities reaching clinical phases are in dire need of processing capacity for clinical material.” Such capacity needs previously were covered by CMOs and CDMOs. “But those also are operating at full capacity; ergo, the interest in new manufacturing facilities.”
Supply-chain vulnerabilities are another factor contributing to sporadic shortages, such as for small-molecule drugs used to treat a number of illnesses, including coronavirus infections. Jornitz continues, “We see other capacity needs coming from antibody boosters, the desire for continuous bioprocessing to allow more process-intensified facility layouts, and aseptic filling with smaller filling volumes and isolator-based designs for both biologics and small-molecule drugs to treat COVID-19. To summarize, new therapies and technologies require and allow, respectively, new processing capacity for both current and anticipated needs. The industry is changing in ways as fundamental as the continuing shift in focus from small molecules to biologics.”
That shift (and its implications for facilities) is reflected not just in peculiar requirements for making proteins rather than small molecules, but in changes occurring within biotechnology itself. “Most expansions are occurring for cell and gene therapies and small-volume MAbs,” Jornitz says. G-CON’s own market research into those three submarkets also suggests that demand is outstripping supply. “New cleanroom and facility designs and implementations are needed to meet those needs, because the old and slow approach no longer works. Cleanroom infrastructure suppliers understand that and have responded with standardization initiatives. This industry wants to move away from reinventing the wheel and to a leaner, more versatile approach that also will shorten timelines, reduce costs, and more reliably stay within delivery timelines and under budget.”
The Pandemic Effect — Not As Much As You Think: According to Jornitz, COVID-19 has shown, once and for all, that the “new normal” for cleanroom delivery is three to six months and not the prepandemic timeline of 12–24 months. “Rapid facility design and installation are not only possible, they are the new paradigm, the new normal that will require major changes to meet the capacity demands of both existing and emerging products. This new normal will drive possible standardization and turnkey facility templates and allow cloning of existing sites. The benefits are not just shorter construction and delivery timelines, but also cost reductions and the potential to accelerate qualification and staff training for a cloned site. Improving delivery speed and reliability has been discussed for many years; COVID-19 provided the impetus.”
Drivers for facility expansions include novel technologies and process strategies (e.g., single-use systems, continuous manufacturing, and “podification”) but no special pandemic-response measures. “We do not see new biopharmaceutical facility layouts incorporating the need for social distancing,” says Maik Jornitz, “because of the requirement for aseptic operating environments.” Most facilities with extensive single-use processing already take space into account. Cleanroom design is defined ultimately by the processes used within the infrastructure, and single-use processes already require mobility and flexibility, so space to fulfill distancing needs already are designed in.
“Pharmaceutical facilities are designed around processes and workflows,” Jornitz says. “The more process-intensified and flexible, the better. We see our autonomous HVAC design being used for multiple applications to ensure stringent containment, higher flexibility, and ease of scalability. Cleanroom pods can be scaled out without interrupting existing processes, which is important for cell therapy production and aseptic filling.”
Although applications and processes dictate facility design, opportunities exist for standardization. “Creating process and facility templates allows us to clone a facility or speed up the design, construction, and qualification, increasing the overall value,” Jornitz says. He makes the analogy of buying a car piece-by-piece or buying a standard model, with a firm price and delivery schedule, and either specifying or adding features. G-CON follows the latter model, delivering standard cleanrooms that are predesigned, prefabricated, and built off-site — and only then figuring out which features require customization.
A primary driver for bioprocessing expansion, says Paul Sullivan (senior vice president at Gilbane Building Co.), is the desire to bring manufacturing and research back to the United States. “COVID-19 shed light on the potential problems with looking overseas for manufacturing. That has been accompanied by rebalancing of manufacturing costs in the United States with costs overseas.” Government funding, the increased involvement of private equity, and expedited FDA review can position many organizations well for expansion. “We see expansion both for new facilities and in the repurposing of existing facilities for additional manufacturing and laboratory space.”
Sullivan also notes that such facilities already incorporate the need for social distancing in their industrial layouts and required clean operating environments. “From a construction standpoint, we focus our planning to overcome the effects of social distancing on project schedules and costs because the number of people who can work per shift may be lower. However, delays still occur in acquiring specialty process equipment or materials coming from overseas, which can affect commissioning and validation.” Some of those delays are, in fact, pandemic-related, particularly for filling lines and fill–finish facilities. “It’s important to identify such items early in project procurement, to evaluate alternative manufacturers where possible and typically well in advance of design, and to create flexibility in schedules.”
Innovation and Controls
Technologies, some new and some not-so-new, that are spurring on facility design encompass single-use and continuous bioprocessing, green initiatives, and alternative unit operations (e.g., precipitation instead of chromatography, membrane adsorbers instead of anion exchange for polishing, and so on). Designers and builders have incorporated those unit operations into facility design, where applicable. The current trend, according to Sherman Shwartz (business unit leader in Gilbane’s Life Science Center of Excellence), is to deal with what he calls “the common thread” of those initiatives: automation and, by extension, artificial intelligence (AI). Those affect installation, operational, and performance qualification (IQ, OQ, and PQ) activities.
“The initiatives are leading to even more sophisticated control systems, which will require reexamination of IQ/OQ/PQ protocols to address a unique paradox created by the incorporation of AI into facilities.” FDA guidance on validation is intended to ensure the repeatability of manufacturing processes, and by extension, drug product safety. “However, by its nature, AI facilitates continuous improvement — the opposite of repeatability. The question will be how to establish and measure IQ/OQ/PQ in the ever-changing environment brought about by AI.”
Flexibility and modularity for hardware and facilities are key aspects. We can see and touch bioreactors and the walls of purification suites. Behind that physical infrastructure, however, lies a separate world of computers, information, controllers, and supervision without which nothing happens. Software-as-a-service and cloud-based IT are now the rule rather than the exception, but until recently, initiating advanced supervisory software still involved a substantial on-site IT presence.
That changed in late 2020 when Honeywell announced that it had completed a control-system migration for Richter Gedeon, a Hungarian pharmaceutical/biotechnology company. Control systems sitting atop all operations require updates and migrations for continuity and improvement. “Customers who migrate benefit from new features and functions of newer releases,” says Owen Sillet (senior services manager at Honeywell Process Solutions). “New releases also have increased cybersecurity, which is vitally important in today’s world.”
The Hungarian project, which occurred entirely during a period of limited on-site access caused by the pandemic, involved updating a Honeywell Experion PKS R431.5 control system to the newest R511.3 release. But that was not just another run-of-the-mill software upgrade: It was the first such migration to occur remotely — “a game changer in industrial automation,” according to Honeywell.
Migrations normally are conducted on site and usually disrupt operations. Remote migration allows a process to be controlled off site by Honeywell experts from the company’s data center. That reduces potential risks, minimizes operational downtime, and makes migration less of a drag on operations. After an upgrade occurs in the cloud at a Honeywell site, customers can put the new system through its paces before implementing it. Uploading migrated images to the production system requires significantly less disruption to operations and reduced on-site time. “A concern with obsolete systems is that there always is risk of a fatal failure that could result in unplanned downtime and potential loss in production,” Sillet explains. “Older, obsolete systems may not be as secure as newer releases and may be vulnerable to cyberattacks. Newer releases may also have increased capacity in terms of system performance and/or increased connectivity that may not be present in older systems.”
According to Honeywell, remote migration improves execution cycle time by up to 50% and increases migration productivity by at least 60%, greatly reducing the resource footprint for installing or upgrading control systems for both existing and new production plants — including plants repurposed for public health emergencies such as pandemics. The service dovetails with current ideas on versatile manufacturing space and flexible, modular facilities.
Facility Expansion — Examples
KBI–Selexis Expansion in Europe: JSR Life Sciences announced plans to enlarge its European footprint through expansion of a state-of-the-art facility that will colocate primary European operations for JSR-affiliate companies KBI Biopharma SA and Selexis SA. Both companies will occupy a combined 94,000-ft2 space in Geneva’s Plan-les-Ouates Industrial Zone (ZIPLO) Stellar 32 campus. KBI Biopharma’s space will focus on clinical current good manufacturing practice (CGMP) biologics for Europe; Selexis will use the new facility for mammalian cell-line development.
At the Geneva facility, mammalian expression systems will be used to produce conventional and bispecific monoclonal antibodies (MAbs, BsAbs) and Fc-fusion protein constructs. The expanded capabilities include two 2,000-L single-use CGMP manufacturing trains with process development and analytical testing laboratories on site. KBI will perform CGMP quality control (QC) testing for biologic drug substance (BDS) release and in-process testing in the new Geneva facility, along with BDS CGMP and drug product stability testing.
KBI also is building a new commercial manufacturing facility in Research Triangle Park (RTP), NC, with an anticipated opening in early 2022. The US$150 million, 140,000-ft2 plant is located close to the company’s current RTP mammalian drug development laboratories. The facility will house six 2,000-L single-use bioreactors and associated downstream processing equipment.
“This setup gives us the ability to produce more than 100 commercial batches of MAbs annually, while retaining the ability to run more complex molecules such as multispecifics and Fc-fusion proteins,” says Dirk Lange. “To reduce risk during technology transfer, our new facilities will leverage the same process platform as our operations in Durham. This approach streamlines development, shortens manufacturing timelines, and provides flexibility to work across geographic boundaries.”
At the expanded Swiss facility, the new Selexis cell-line development laboratories will be operational by the third quarter of 2021. Operations at KBI Biopharma’s process and analytical development laboratories are scheduled to begin in the first quarter of 2022, with the 5,600-m2 bulk biologic substance manufacturing facility coming online by mid-2022. KBI expects the greenfield commercial manufacturing facility in North Carolina to be operational by the first quarter of 2022.
The turnkey facility was designed and constructed to accommodate KBI’s single-use production platform. KBI manages single-use supply risks through its vendors and with in-house assembly and sterilization capabilities. Both locations use ambient water for injection (WFI). Lange adds, “We designed the plants for an optimized client experience, with visibility at all key process steps. The facilities offer multiple upstream and downstream suites to ensure compliant and efficient operations in a multiproduct environment.”
The multiproduct Swiss facility is built around single-use technologies for both upstream and downstream processing. For greater efficiency and to mitigate supply risk related to consumables, however, operators will conduct buffer preparation in stainless steel and perform rapid microbial analysis for the WFI system. Lange adds that “our RTP expansion predominately leverages disposable technologies in upstream and downstream manufacturing while we use stainless-steel systems to improve economics and address supply risks in buffer preparation. Such operations can support larger campaigns for multiple products or run one product at a time.”
KBI has integrated its parent company’s JSR ONEDigital platform at the RTP plant, which incorporates manufacturing 4.0 principles. The digital plant will feature data collection systems, including electronic batch records and logbooks, paperless materials management, and laboratory information management systems (LIMS) for QC data management. “These tools will enable real-time process and product- quality monitoring and allow for future AI technology adoption,” Lange says. “We are building our Geneva plant to be digitalization ready. Following successful implementation, KBI plans to implement these tools in Europe as well.”
Thermo Fisher Scientific’s Multiuse Facility: In October 2020, Thermo Fisher announced that it would build two sterile filling lines in Singapore to extend its ability to develop and manufacture therapies and vaccines for pharmaceutical and biopharmaceutical customers in the Asia–Pacific region. Established with the support of the Singapore Economic Development Board, the $130 million facility will make product to meet regional demand and to respond to future health emergencies.
“Responses to the COVID-19 pandemic have shown the importance of industry–government collaboration in ensuring the availability of critical diagnostics, therapies, and vaccines,” says Lukas Utiger (Thermo Fisher’s vice president of Pharma Services APAC). “This partnership expands our sterile fill–finish network in the region, while strengthening the Singapore government’s ability to respond quickly to future health crises.”
The new facility initially will include a high-speed sterile line approved for live-virus filling — the first such large-scale capability in Singapore — followed by an additional line for standard fill–finish. The facility also will house cleanroom capacity, laboratories, warehousing, and offices. Once operational in 2022, it could manufacture up to 30 million sterile doses per month and employ more than 300 people.
The filling lines will have state-of-the-art compounding capability based on single-use technology. “So far, standard sterile-filtration technology is planned, but additional technologies can be retrofitted,” Utiger says. With the same general layout, the two lines will be segregated fully in terms of material and personnel flow; both can fill biosafety class II compounds (e.g., live viruses as well as typical sterile products) and will incorporate “the latest isolator design,” according to Utiger. One line will have two 30-m2 lyophilizers. “In the end, cleaning validation and customer quality agreements will determine which line will be dedicated to which product category. Based on current cleaning methods and analytical detection levels, the filling lines can be used for live- virus and normal sterile products,” depending on regulatory approvals, cleaning validation, and customer acceptance.
The Samsung Biologics “Super Plant”: In November 2020, Samsung Biologics began construction on Plant 4, a multistory, 238,000-ft2 “Super Plant” taglined as “The Future of Biopharma.” When completed, it will boast 256,000 L of manufacturing capacity in what Samsung reports will be world’s largest mammalian cell culture contract biopharmaceutical manufacturing facility. Located in Incheon City, South Korea, Plant 4 features a highly flexible design with multiple bioreactor sizes and operation modes, with an n – 1 perfusion cell culture system, QC capabilities, and full digitalization.
Recent “scientific and technological innovation,” says Samsung’s senior vice president, James Choi, provides the biopharmaceutical industry with “an excellent opportunity to adopt modern digital solutions,” which Samsung has applied to the very earliest planning and construction of Plant 4. “Plant 4 will leverage our recently launched Enterprise Quality Unified Information System (EQUIS) platform, which provides clients with secure access to request, review, and approve documents and quality records on demand,” he says. “Mobile tablet devices, which replaced paper binders on our manufacturing floors and in QC labs years ago, allow us to streamline day-to-day tasks. These devices are EQUIS-connected to increase efficiency, reduce compliance risk, and improve visibility and data integrity.”
Samsung also will launch a manufacturing execution and control system and tools to enable real-time process simulation. In addition to n – 1 perfusion, Plant 4 will be equipped with stainless-steel production systems, which Choi says deliver more robust services in terms of supply, increased scale-up or -down capability, less waste generation, decreased operational costs, and lower risk of supply chain disruption. “Samsung Biologics also has extensive experience in operating single-use systems to provide flexibility and to meet the demands of processes suited to single-use technology, including scaling up to larger stainless-steel bioreactors.”
Upon completion, Plant 4 will house fully one third of total global CMO biomanufacturing capacity, offering a combined sum of 620,000 L at a single site. Plant 4 also features a modular design to enable limited-volume manufacturing at the site by the end of 2022, with full operations beginning in 2023.
Over the past decade, Samsung has constructed three state-of-the-art contract manufacturing facilities, with each of them surpassing the previous one in terms of capacity and advanced technologies. “Based on our experience, we decided to design Plant 4 with two separate bioreactor halls to allow for speedier validation. You could say there will be two mini-plants within Plant 4, and one will be constructed and validated first to meet increasing client demand,” Choi says.
He agrees with our other experts that demand for biopharmaceutical CDMO services had been rising before the pandemic. “However, the sudden, exponential surge in demand during COVID-19 has accelerated implementation of plans for Plant 4. Many other CDMOs are planning similar expansions.” He cites demand as both direct — for pandemic-related therapeutics and vaccines — and indirect, from the biopharmaceutical industry’s strategy shift from in-house manufacturing to multisourcing, with greater focus on lean operations and risk mitigation.
Thermo Fisher Scientific’s Plasmid Facility: In December 2020, Thermo Fisher Scientific unveiled plans to build a CGMP plasmid DNA manufacturing facility in Carlsbad, CA. It will expand the company’s ability to produce plasmid DNA in microbial expression systems as a raw material and for drug substances. The 67,000-ft2 facility, expected to begin operations in mid-2021, will feature advanced technologies, including single-use processing ranging from 30-L to 1,000-L scale, digital connectivity, and data visibility to support operational efficiencies and operator training.
Volumetric requirements for plasmid production are generally lower for mRNA and cell/gene therapy applications. “Producing plasmids in E. coli is simpler and much faster than in mammalian cells,” says Darren Leva (commercial operations director of microbial manufacturing services and cell therapy at Thermo Fisher Scientific). “A fermentation takes two or three days at most, and the entire process can run in about a week or less. The upstream process alone on mammalian cells might take two to four weeks.”
Thermo Fisher has been in the research-use–only plasmids business for many years. However, the demand for cell and gene therapies and DNA vaccines has increased, especially in the wake of the pandemic, and the global supply of plasmids is not meeting current and projected demand. This is where facilities capable of high-quality GMP plasmid production come in. “We have been expanding capacity strategically by partnering with Arranta Bio and improving our facilities by building new production lines, a mission-critical initiative driven by the cell and gene therapy market and our adjacent viral vector business,” Leva adds. “This has been ongoing since before the pandemic.”
Leva notes that although interest in emerging therapies heightened because of COVID-19, capacity shortages have been a chronic issue that predated the pandemic. The Carlsbad facility “achieves a commercial plasmid scale that has not been achieved in the past. If plasmids prove to be an effective delivery method for a vaccine or therapy, this facility would be capable of rapid technology transfer and scale up as one of the largest producers of plasmid drug substance globally.”
Leva cites improvements in single-use fermentors, which were previously volume- and performance-limited, as critical drivers for the design of this facility. Volumetrically speaking, large-scale, single-use microbial fermentors are several years behind their single-use mammalian counterparts. “The ability to support the gas and heat transfers required by high-density microbial cultures is relatively new. The single-use systems we will be using support very aggressive and reduced startup times.”
But their incorporation into the design of this facility required consideration of equipment sizes and areas needed to stage, store, unpack, and implement single-use fermentors. “The use of proven single-use technologies also adds more pressure on supply-chain functions for procurement and stock management, but this is an advantage for us as a provider of single-use technologies.”
Another interesting feature of the Carlsbad facility not commonly found in typical MAb production lines is a continuous extraction (alkaline lysis) step currently in development for processing the large volumes of clarified and unclarified lysates from 1,000-L single-use fermentors. Continuous extraction offers impressive scalability, control (e.g., over homogeneity, mixing, and reaction time), volume management, and performance (extraction yield and quality), Leva says.
1 Hub IE. CHO Media Development for Therapeutic Protein Production. Am. Pharm. Rev. 23 December 2019; https://www.americanpharmaceuticalreview.com/Featured-Articles/559489-CHO-Media-Development-for-Therapeutic-Protein-Production.
2 Draft Landscape and Tracker of COVID-19 Candidate Vaccines. World Health Organization: Geneva, Switzerland, 2021; https://www.who.int/publications/m/item/draft-landscape-of-covid-19-candidate-vaccines.
3 Macdonald G. CHO Media Development for Therapeutic Protein Production. BioProcess Insider. 25 December 2020; https://bioprocessintl.com/bioprocess-insider/global-markets/covid-19-has-spurred-long-lasting-innovation-says-expert.
With a PhD in organic chemistry from the State University of New York at Stony Brook, freelance writer Angelo DePalma (email@example.com) was a chemist first at Brookhaven National Laboratory and then at Schering-Plough.