Agnes Shanley is senior editor of Pharmaceutical Technology.

Agnes Shanley

Years ago, scaling up the production of a new therapy was a linear, iterative process that demanded patience and deep pockets. Best practices, and more than a little trial and error, moved processes from the bench to the manufacturing floor. Over the past decade, as manufacturing operations have become more complex, engineering advances in modeling, digitalization, single-use process equipment, and process intensification have made scale-up more systematic, for both small and large molecules.

	“A decade ago, most scale-up would have been a very linear process with lab-scale going to kilo lab-scale to pilot plant-scale and, for successful products, to an in-house existing batch production facility. Now there are many more nodes of partnerships and equipment choices to be made,” says Joe Hannon, CEO of Scale-Up Systems. “It’s no longer just about getting reaction, workup, and isolation steps to work as best they can in existing multi-purpose equipment, but also about reconfiguring modular components in a continuous process to fit the equipment to the requirements of the chemistry. This is allowing a lot more innovative thinking and is pushing the boundaries in developing chemistry in ways that were not considered feasible until recently,” he says.

	As a result, scale-up is becoming a more finely tuned process, with scale-out and scale-down playing more important roles. “The era of blockbuster production strategies has transitioned into scale-out and even scale-down for most of the bioproduction market,” says Michelle Stafford, global product marketing manager, Enterprise Solutions, GE Healthcare. “Multiproduct production models that utilize single-use technologies are now more common, and scale-out is becoming the norm, where increased flexibility and platform standardization enable faster changeover between batches and faster tech transfer,” she says.

	At the same time, contract development and manufacturing organizations (CDMOs) have assumed a dominant position in scale-up and must master a wide range of scales to meet clients’ needs. “Speed to clinical phases is faster now, and biopharmaceutical manufacturers are targeting smaller patient populations and specialized drugs in order to access the market faster, reach patients, and get financial returns sooner than ever before,” says Stéphane Varray, head of commercial development in Lonza Pharma and Biotech’s mammalian and microbial development and manufacturing business unit. “Based on that paradigm shift, manufacturing scale needs to be adaptable along the full lifecycle of the product. Moving from a small scale to a larger-or vice versa-must be done smoothly, in the context of a regulatory framework that is growing more complex,” he adds.

	“Overall, the scale-up process remains the same, regardless of the volume of product being produced, but strategies have changed,” says Sebastien Ribault, senior director of MilliporeSigma’s global Bioreliance end-to-end delivery and sales, and managing director of MilliporeSigma’s Biodevelopment Centers, the company’s CDMO division. “We are not using one scale in the lab and another one in the facility anymore, but multiple scales in the facility,” he says. In 2010 and 2011, he says, the company invested in developing models to help its scientists understand how biomolecules behave at various scales, he says. As a result, he says, it has scaled all biopharmaceutical processes up directly (i.e., in one pass) since 2012, where previously, efforts may have required going back to the drawing board.

	Within the past five years, pharmaceutical manufacturers have become more comfortable with advanced process control and are embracing other advanced manufacturing concepts, says Fernando Muzzio, professor of pharmaceutical science at Rutgers and head of the Center for Structured Organic Particulate System (C-SOPS), which has been focusing on tech transfer and commercialization of continuous manufacturing processes for solid dosage forms.

	Hannon, whose company specializes in scale-up simulation software for small-molecule APIs, sees a definite change in the way that manufacturers are approaching simulation. Where five years ago, they might simply require that a unit operation undergo simulation studies and specify the time required, they are now using simulation to couple batch kinetics with continuous equipment characteristics. “They want to operate flow chemistry optimally on the first run,” he says. “We are now seeing more of an appetite for deep process understanding around the reaction and impurity-formation mechanisms to support right-first-time operation in flow on scale,” says Hannon.

	Branded pharmaceutical and even generic-pharmaceutical manufacturers are increasingly interested in continuous manufacturing, says Muzzio, which allows development and manufacturing to take place at the same scale on the same equipment, eliminating the equipment scale issues that have dogged scale-up in the past. Most new chemical entities (NCEs) that are going from lab to larger scale still face scaling challenges, however, in that the next level equipment is significantly larger than the laboratory equipment, says Hannon. “Sound chemical engineering principles such as mixing and heat transfer must be engineered in for successful operation. And scale-down (i.e., being able to mimic production scale conditions closely in a laboratory setting) remains key to scale-up success, especially when troubleshooting a process,” he says. Beyond troubleshooting, Muzzio sees growing interest in scaling down to smaller lines for continuous solid-dosage form manufacturing. “Some companies are working with 5 kg/h, and a few, with 1 kg/h, which is ideal for products that use very small amounts of powder,” he says.

	He recalls one company that opted to start with a very small line and to duplicate that line four times to manufacture a large-volume product continuously. “This approach makes sense, because the integration of process, software, and sensors accounts for a significant fraction of overall costs. If you simply copy one line, the additional capacity will not add too much to overall costs.” However, he notes, scale down must be approached carefully to avoid having to redo scale-up later on. In addition, he says, scale-down can be particularly challenging with powders, which can be much more difficult to transfer and mix at small scale than at larger scales.

	Process intensification is becoming more important in biopharmaceutical manufacturing evidenced upstream by perfusion bioreactors and downstream by continuous chromatography. Tangential flow filtration (TFF) is becoming more prominent. In August 2019, Samsung Biologics scaled up alternating tangential flow (ATF) in a perfusion bioreactor at its manufacturing facility in South Korea. Samsung expects it to reduce processing time by 30%. Pressures to reduce cost of goods sold (COGS) and to meet smaller, specialized market demands are driving use of process intensification. “The goal is to increase productivity per some form of tangible measurement, such as in time, volume, or dollar,” says Andrew Yang, global product marketing leader for upstream bioprocessing at GE Healthcare. By integrating technologies, simplifying operations and offering insights via analytics, he says, it “offers a path to risk reduction and economic global manufacturing in an increasingly fragmented market.”

	As Ribault notes, manufacturing scales have come down overall. MilliporeSigma is no longer using 10,000–15,000-L volumes for most of its work. “Now, our facilities are running at 2000 L or below. When we need more volume, we operate with multiples of 2000 L,” he says. The philosophy behind scale-up has remained the same, however. “You still need to understand the platform and the differences not only for downstream but the impact of change of scale for upstream, which means gaining a clear understanding of the impact of agitation and power input, and also the quality of the molecule. Agitation and oxygenation give you a cell that will result in the right level of product quality. That is what we focus on when we scale up,” he says.Modeling and scale-down studies are playing a greater role in providing key process knowledge required for scale-up. AbbVie, for example, used a combination of mass-transfer studies (KLa), computational simulation (i.e., the use of computational fluid dynamics to model shear-rate distribution within the bioreactor), and scale-down (i.e., developing proof-of-concept on a small scale in the lab, then scaling up to 200 L) in one recent project, in which the time required to reach process performance qualification (PPQ) was reduced to 18 weeks (1).

	Ribault predicts that the use of digital twin technology will become more important. “In the future, we won’t need to run some steps at the bench and will be able to run process 

 before running at bench,” he says. Multivariate data analysis is also becoming more important in allowing developers to use production data to determine optimal characteristics for components and equipment parameters and use those characteristics in a model to optimize scale-up and manufacturing, says Stafford.

	In general, scale-up success demands coordination with tech transfer, and good project management and documentation are key, says Ribault, who notes that smaller research company clients may need to understand that research data alone cannot move a project to good manufacturing practice (GMP). “We don’t want to rush because we know the bare minimum that is required to scale up without intermediate scale and engineering runs. If we deviate from that bare minimum, we will not be managing the risk properly. In the attempt to save a few weeks, we may wind up wasting a few months, or even quarters,” he says.

	To address these challenges, CDMOs and customers together should take a careful and forward-looking approach to planning for commercial supply, says Varray. This should include a clear idea of the scale of development, planning for a biologics license application, and a sophisticated regulatory approach. “This plan may be drafted as soon as Phase I is cleared. External partners can work together with customers to adapt manufacturing set-up and regulatory strategy while the clinical phase is ongoing,” he says.

	Process intensification enables scaling out or building incremental capacity rather than scaling up to larger equipment. As this approach gains acceptance, the distinctions between scaling up and scaling out are blurring. “With the adoption of high titer processes, process intensification including continuous processing, and perfusion, the definition of scale-up and scale-out are synonymous such that requirement for product volumes are lower that can be achieved with single-use technology and replicated with increasing demand,” says Kenneth Dawson Green, manufacturing team leader with Samsung Biologics.

	At large scale (15,000 L), utilizing stainless-steel bioreactors, Samsung replicates the number of bioreactors required to achieve the optimal volumetric scale (up to 180,000 L), he explains. However, for single-use, the typical working volume is between 2000–5000 L and replicated to achieve desired scale, he says. Increasingly, single-use and stainless-steel technologies are being used in hybrid systems such that the bioreactor seed train may utilize single use bioreactors, and the production scale utilizes stainless steel. Samsung uses hybrid systems for both upstream and downstream processes, Green says, factoring applicability, cost, ease of operation, operational risk, and waste-minimization into equipment selection, he notes.

	It remains to be seen whether scale-out will become the preferred route for biopharmaceutical manufacturing in the future. At this point, both approaches are needed. “Adoption of continuous manufacturing, including perfusion, has led to the requirement for reduced product volumes with higher productivities for some products. The option to scale-out vs scale-up presents lower process risk and enables more efficient tech transfer when processes are transferred at the same scale,” says Green. However, he notes this approach is more practical with single-use technology and clinical products for which capital investment may be deferred. “For larger volume commercial products including biosimilars, large-scale stainless-steel equipment will remain the most cost-efficient platform,” Green says.

	Increased cell- and gene-therapy development also poses questions for scale-up. “In cell and gene therapy, we see both forms of scale-up and out,” says Phil Vanek, general manager of cell and gene therapy strategy at GE Healthcare, who says all three approaches are needed. While many of the therapies today are autologous or patient specific, Vanek says, for certain applications (e.g., acute injury or non-durable treatment duration) allogeneic versions of autologous therapies may be preferred for access or cost reasons. “We use scale up where we can make multiple doses to serve large markets; scale out to meet the needs of patient- specific therapies, and, perhaps most importantly, scale-down in the appropriate high dynamic range platforms to do process development,” Vanek says. Process development can typically be cost-prohibitive, so platforms that mimic large-scale (up or out) systems, but at smaller scale, will allow quicker process optimization, regardless of scale,” he says. At this point, Vanek says, “The jury’s still out on whether scale-out or scale-up will prevail.”

	1. Y. Wang, “Achieving Seamless Scale-Up and Tech Transfer: Case Study in Single-Use,” brighttalk.com, March 11, 2019.

Vol. 43, No. 9
	September 2019
	Pages: 16–20

	When referring to this article, please cite it as A. Shanley, “Scale-Up: Getting It Right the First Time," 

Articles

Front-end focus and new approaches are speeding scale-up and reducing costs, while scale-down and scale-out become increasingly important.

Scale-Up: Getting It Right the First Time

	Generic pharmaceuticals now account for more than 90% of all prescriptions issued in the United States.  However, the generic pharmaceuticals’ pipeline flow is anything but smooth. A report issued on Aug. 7, 2019 by the US Government Accountability Office (GAO) traces part of the problem to issues with FDA review. The 

 found that FDA approved only 12% of the 2,030 generic drug abbreviated new drug applications (ANDAs) submitted between 2015 and 2017 during the first review cycle.  Often, this could be traced to incomplete applications and quality problems with the drug or issues found during site inspections. In addition, reviews of complex formulations, notably ophthalmic therapies, were often held up.

	However, GAO also found that FDA reviewer comments varied in clarity and substance. Applicants that had filed incomplete ANDAs in the past were also found to be at higher risk of having their applications rejected during the first cycle, GAO’s analysis found. Another problem was that labeling changes required for the equivalent name brand drug were often made late in the ANDA review cycle, delaying approval of the generic. 

	FDA recently introduced changes designed to improve communication between reviewers and applicants and developed templates designed to help make the review process more consistent, the report noted. However, GAO recommends that the agency take additional steps to improve the consistency of reviewer comments and that it also looks into the timing required for brand-name drug labeling changes.

	The GAO report focused on FDA’s contribution to the problem, but also noted a tendency for applicants to rush to file applications despite gaps in process knowledge that translate into incomplete ANDAs. Focusing on this same tendency, but from the developer’s side, was an 

 of FDA data (1) by the National Institute for Pharmaceutical Technology and Education (NIPTE), which noted that this tendency to rush to file is having a serious impact on the generics pipeline. Over half of the generic drugs approved by FDA either never reach the market or are only commercialized after a delay. Of the 1,600 generic pharmaceuticals that FDA approved in 2017 and 2018, about 43% were not on the market by the start of 2019. In addition, many potential generics either never even make it to the review stage. 

	The NIPTE analysis concludes thatscale-up issues, process validation, and repeatability can pose challenges, worsened by pressures to file first without incorporating adequate risk analysis and understanding potential sources of variability, which NIPTE describes as gaps in prior knowledge. 

	To counteract these problems, NIPTE has proposed a “New Prior Knowledge” concept that would make crucial process and product information available to improve the quality of the first application and increase the likelihood of more ANDAs being approved during the first review cycle.

	Ajaz Hussain, director of NIPTE, discussed the concept with 

“The generics system works only when more than five generic companies can compete for the market,” he said. Although simple dosage form small-molecule generics may not pose a problem, generic pharmaceutical formulations are now more complex, he said, and mechanisms such as the technical citizen’s petition to block approval or assessment of therapeutic equivalence require additional work,” he says.

PharmTech: What is the New Prior Knowledge concept and what is its goal?

	Hussain:  The ‘time-cost-quality’ constraint and the Hatch-Waxman incentive to ‘file first’ in the development of generic drugs can often be a cause for a ‘file-first figure it out later’ approach, in which a manufacturer receives multiple complete response letters, necessitating multiple review cycles. Furthermore, the available information and prior knowledge about API and patient, product and process related failure modes are often not published in peer-reviewed scientific literature. What is published can be highly variable, which increases the likelihood that such failure modes will remain unaccounted for, despite the multiple review cycle process.  

	As a result, the efficiency of the nation’s generic drug system remains suboptimal and prone to frequent negative media coverage in the form of warning letters and import alerts, which erodes the trust and assurance that patients need in order to experience equivalent therapeutic outcomes in the real world. ‘New Prior Knowledge’ (NPK) would fill this gap by having NIPTE universities conduct research to make available critical information and knowledge needed for quality by design of generic drug products so that multiple generic companies would be able to use the information and compete with high-quality ANDA submissions.  

	NPK would provide a mechanism for defining important generic products and identify what are the technical challenges that have to be addressed, design research programs around those challenges, and then fund it so that information and knowledge can be curated and published so that it eventually becomes available to all generics companies to inform their own development programs, and to FDA to improve its own programs. 

	Hussain: The notion of NPK is based on my experience at Sandoz and collaboration with Momenta; particularly lessons that were learned in the development of generic enoxaparin, generic glatiramer acetate injection, and biosimilars. Sandoz identified sources of variability and reference products and studied reference product so exquisitely, in order to know the lot to lot variability and failure mode of every component. It took a lot of effort and we were analyzing and understanding failure modes and sources of variability with the most modern set of orthogonal analytical tools.  We wanted a deep understanding of the sources of variation and the failure modes so that we could design equivalent products with more confidence, and without failure rates in manufacturing. Only a few companies have the resources to do that.

	Where this information was private at Sandoz, NIPTE would create a public program to generate NPK by characterizing multiple lots of reference products via a set of advanced orthogonal analytical methods (going way beyond basic USP testing requirements). 

PharmTech: Knowing the inevitability of patents ending and generics being developed, could branded drug companies be expected to share any of this information near the end of a drug’s life cycle?

	Hussain: I do not believe they would.  NIPTE and its 18 member schools seeks to organize a disciplined mechanism to select drugs and drug products that are difficult to develop and pose a risk of therapeutic inequivalence in the real world.  NPK research would be multidisciplinary and among other things, identify failure modes and methodology for analytical characterization.  This NPK would be peer-reviewed, curated, and disseminated publicly, in time for generics companies to inform their development programs and improve the quality of their ANDA submissions.  FDA would also be able to utilize NPK for their guidance development and to organize their review process.  Associated with NPK would also be considerations for education, training, and certification of a community of pharmaceutical knowledge.

PharmTech: Is NIPTE doing anything to help train FDA reviewers? 

	Hussain: Several years ago, NIPTE had developed and conducted an elaborate pharmaceutical technology education program for FDA chemistry, manufacturing, and controls (CMC) review staff. Over the years NIPTE has continued to conduct other specialized training programs for FDA reviewers. At FDA’s request, NIPTE is currently in the process of developing a comprehensive education program for FDA review staff in the Office of Pharmaceutical Quality.

	1. US Government Accountability Office (GAO), Report to Congressional Committees, Generic Drug Applications, August 2019.
	2. A Hussain, V. Gurvich, and R. Morris, 

A GAO report analyzes some of the reasons why so few generics are approved the first time around. NIPTE has proposed a program that would address these issues

Generics: Getting Beyond “File First, Figure it Out Later”

Manufacturing for clinical trials requires addressing complex packaging, labeling, and logistics issues and maintaining blinds, so that the identity of the drug being administered remains unknown. New patient-centered trial models and the growth of personalized gene and cell therapies have added additional degrees of difficulty to the process. Creative management and use of interactive response technologies (IRTs) solutions are helping sponsors and contract partners improve overall efficiency and reduce waste. In this article, Matt Blume, general manager of global business operations for Catalent Pharma Solutions, shares insights and best practices. In a separate interview, Stefan Dürr, senior director, client delivery, with Cenduit, LLC discusses the use of IRT in optimizaton efforts. 

 How are you addressing the typical disconnect between chemistry, manufacturing, and control (CMC) and good manufacturing practice (GMP) specialists in your work with clients? 

 We include clinical project managers in early phase manufacturing discussions with the sponsor. For example, in our San Diego facility, we have co-located clinical project managers in the CMC plant so that they can be included in the earliest possible project discussions. Even though these discussions may not focus on clinical supply issues, the earliest stage of the product is the perfect time to highlight some of the areas that sponsors should consider (e.g., general project timelines as well as  packaging configuration options). Early development batch production is a good time to place product down on stability in multiple material configurations. Doing this very early in the process allows meaningful data to be gathered as soon as possible, ensuring that the most patient-friendly package is designed, and allowing for potential savings in Phases II and III if data can be established that allow for higher shipping temperatures. 

	There is no downside to ensuring that all team members are engaged throughout the process. The interaction of the different specialties not only helps the specific project being discussed, it also builds cross-functional knowledge, strengthening the CMC team’s understanding of clinical concerns and the clinical team members’ knowledge of CMC challenges. 

 What are the most challenging packaging and logistics issues involved in clinical trial manufacturing, and how (and when) should they be addressed?

 One challenging task is the labeling of biologics and gene therapy products prior to freezing.  At the point of fill/finish, minimum label content must be determined so that it can be used for many different countries and destinations. This needs to be addressed very early on because many of the facilities that focus on biologics and gene therapy are not clinical supply experts. Co-locating clinical supply team members to focus on label design and clinical challenges within the manufacturing team is extremely helpful, especially since these teams have little clinical trial experience, unlike their peers at traditional solid dosage form manufacturing plants.

	With smaller vials and syringes that will be subjected to cold temperatures, it is very important to ensure that the adhesive used for the label is appropriate and that as much of the label text as possible will fit within a small area. For example, many gene therapy treatments are packaged in 3mL vials, minimizing the real estate available for vial labels. Early involvement with clinical specialists is key to thinking through minimum label requirements and design.

	Another challenge is the blinding of increasingly complicated therapies including biologics and pre-filled syringes. Many factors must be considered when blinding an injectable or intravenous therapy. With injectable products, the solution’s appearance must be verified before administration, so the matching placebo must match all of the characteristics of the active product.

 Where do you see direct-to-patient models having the greatest change on clinical trial manufacturing and logistics? 

 By delivering directly to the patient from a clinical packaging site, the manufacturing distribution site now becomes a pharmacy dispensing site. In the United States, the regulatory and licensing requirements for establishing a pharmacy and licensure in all 50 states must be met before patients can be supplied. This requires licensure and expertise in areas that have not traditionally been required at clinical supply facilities. In addition, shipping direct-to-patient requires that data privacy procedures be in place to ensure compliance with all data privacy laws. With traditional supplies, materials are shipped by kit number to a clinical site. No patient names or addresses are ever part of the clinical supplier chain. In direct to patient, however, the established pharmacy will have this information and must control it appropriately. 

	The new models also require flexibility (e.g., the initial patient visit of a trial may be with the investigator, and future shipments will be made directly to the patient’s home). Having the materials co-located in one site with the ability to ship to site or directly to patient is important when the total amount of available supply is low. Direct-to-patient orders will need to be flagged differently through IRT and processed differently through the site. When establishing the trial setup, careful thought must be given as to whether an order should be triggered as a site shipment or as a direct-to-patient shipment. 

	In addition, at this point, care must be taken not to include patient names or addresses in the IRT request. The shipping site must have processes in place that will allow orders to be split based on their intended destinations, and the site must also maintain patient record privacy standards.

	Optimizing clinical trial performance with IRT

	Interactive response technologies (IRTs) and e-clinical tools have become indispensable in ensuring that protocols and followed and in optimizing clinical trial manufacturing, logistics, and other operations. Stefan Dürr, senior director of client delivery and head of Cenduit LLC’s Drug Supply Center of Excellence shares insights into how IRTs are being used in the clinical trial materials supply chain.

 What do you see as the most challenging aspects of clinical trial manufacturing and supply today, from each participant’s perspective?

 The most challenging aspects in the clinical trial supply chain are to reduce costs through reduction of wastage, and at the same time to ensure that clinical trial supplies are never on the critical path for completion of the clinical trial by always having the needed clinical supplies available across the depots and sites in the study. From a sponsor perspective, the challenge is to decide what part of the clinical supply chain is managed in-house and what part is outsourced. The sponsor must ensure that the different internal and external stakeholders work together efficiently to achieve a lean clinical supply chain that achieves the above goals. From a contract development and manufacturing organization’s (CDMO’s) perspective, the challenge is to reduce timelines and increase flexibility to react quickly to changing demands in the clinical supply chain. The faster the needed comparator can be sourced, the next packaging run planned and executed, and the quicker we can ship between the different depots, the less uncertainty you will need to accommodate in your clinical supply plans. From a contract research organization’s (CRO’s) viewpoint, the biggest impact on the clinical supply chain is the ability to predict the recruitment and execution as accurately and as close to the plan as possible. From an IRT perspective, the challenge is to have a resupply algorithm that can adapt to the study needs and minimize wastage through using real time data to determine the lowest buffer stock level at the site that will allow for the treatment of all the patients.   

	CMC requirements, distribution, and randomization

 How can IRT solutions help address the following challenges: chemistry, manufacturing, and control (CMC) requirements (i.e., stability requirements, risk of contamination, labeling, overall quality); distribution and logistics; randomization; and comparator sourcing?

 IRT can help manage some CMC-related actions during the trial (e.g., blocking kits or managing recalls in case of quality issues or management of expiry date following new stability data). Additionally, it can be used to manage temperature excursions during shipments and storage at sites, and to work with temperature stability data from CMC to determine whether an excursion exceeded the allowed temperature stability range. In distribution and logistics, IRT solutions can help users evaluate the distribution service level agreements (SLAs) to best manage the depot and site shipments. Based on how long shipments take, the right moment to schedule a shipment can be determined in IRT to ensure that the shipment arrives in time at the required destination. IRT can also be used to reduce the number of shipments in clinical trials to save costs and reduce the burden on sites to manage many incoming shipments. Of course, IRT is central to managing randomization and can support any requirements. From a drug supply perspective, certain randomization models would allow the reduction of site buffer stocks. Therefore, a close collaboration between clinical supplies and biostatistics can have a positive impact on clinical supply budgets. For comparator drug sourcing, the challenge is getting the right amount of comparator at the right time with minimum waste. IRT can help manage studies with high comparator costs with minimum buffer stocks to keep the wastage as low as possible. IRT can also track and manage all drug-related interactions, allowing users to manage returns or on-site destruction of the drug after conclusion of a trial. In addition, the IRT ensures that all the relevant data are available for inspection readiness.

 Are pharmaceutical clients sharing more information upfront to help improve the overall process?

 We see more companies use clinical supply forecasting solutions and services to plan their clinical supply chains. It is very important that such planning and execution efforts be closely aligned in IRT, and that real-time data are used regularly to re-forecast different future scenarios to ensure that there are no surprises.


	Vol. 43, No. 8
	August 2019
	Pages: 30–31

	When referring to this article, please cite it as A. Shanley, “Orchestrating Success in Clinical Trial Manufacturing," 

Personalized medicine and direct-to-patient trial models have made the difficult even more challenging.

Orchestrating Success in Clinical Trial Manufacturing

Manufacturers of over-the-counter (OTC) drugs have always been subject to the US 

) Parts 210 and 211, but recently, FDA has begun to step up enforcement of cGMPs in this sector, says Nick Capman, CEO of The FDA Group. So far, 2019 has seen an increase in both FDA 483 notices (483s) and warning letters to OTC drug manufacturers.

	For an area that had traditionally been less scrutinized than generic or name brand drugs, the new regulatory emphasis has signaled the need for a culture change, particularly for business owners and CEOs, who, in the past, may have been disconnected from day-to-day operations, says consultant Gary Ritchie. “This is an industry that has operated without much oversight in the past. Now they are being scrutinized, and change is proving difficult,” he says. 

	OTC company managers must first develop compliance strategies, says Ritchie, and generally, the best way is first to apply the predicate rule to operations and then run everything under current good manufacturing practices (cGMPs), taking special care to ensure data integrity. Ritchie cites ABC Labs, a contract testing lab for OTCs, as an example. The company received a warning letter (1) from FDA based on inadequate response to a prior 483 in 2012, for failing to ensure that lab record data were complete, for inadequate access controls to computer systems as well as master production and control records, and for failing to validate its test methods for key assays. Based on previous inspection findings, FDA had asked the firm to suspend testing until it could ensure that its laboratory testing met cGMP requirements. The company had not provided that assurance, yet resumed testing services in 2016.

	“Just about the worst thing any company can do is have a standard operating procedure (SOP) and fail to follow it, but even worse is doing something that FDA has expressly asked you not to do,” says Ritchie. “With heightened scrutiny, the fundamentals need to be practiced, and it is management’s responsibility to see that they are,” he says.  Recent FDA citations for OTC facilities have followed similar patterns, including everything from failing to have a quality department in place, to failing to act on problems mentioned in previous inspections and 483 notices (2,3). In one case, the Canadian firm, Petra Hygienic Systems International, showed violations that had originally been noted by FDA in 2012 and 2014. In addition, the firm had inadequate and unvalidated equipment cleaning methods and failed to follow requirements for finished product testing. Imports from the company were banned (4).

	In a webinar (5), J. Lawrence Stevens, senior regulatory and quality consultant, and Alan Greathouse, senior director of quality and service assurance, both with The FDA Group, discussed trends that have resulted in OTC drug manufacturers receiving 483s and warning letters from FDA for cGMP violations.  The most frequent problems can be sorted into six categories, Stevens said during the program:

		Inadequate understanding of CFR 210 and 211, and lack of understanding of “unwritten regulations” (e.g., the need for corrective and preventive action [CAPA] programs), which is never actually called out in CFR 210 or 211

		Inadequate analytical and microbial testing

		Inadequate nonconformance and CAPA management

		Common quality management systems (QMS) problems.

	Even a cursory look at recent OTC warning letters and 483s reveals repetitions of all these issues. One problem that Stevens noted in the webcast is document control. “Beware of employees using cheat sheets to remind them of entire procedures. FDA will view these as showing that there is insufficient control over manufacturing documentation,” he said.

	CAPA is another area often cited during FDA inspections because inspectors are applying a quality systems approach, and for 

 211, in nonconformance, which is where CAPA fits in, Stevens told the audience.

	Companies are also expected to perform internal audits, using their own people or external consultants, but all testing must be validated and documented, Stevens said. In addition, he noted, many companies are performing nonconformance testing, but once they find problems they simply reject product, but don’t investigate the root cause of the problems. FDA will expect such an investigation, he said.

	Regulators often cite water systems as a problem area, Greathouse told webcast listeners. It is important to ensure that FDA and US Pharmacopeial Convention (USP) guidance are being followed for these systems, and that sufficient testing is validated and fully documented for total organic carbon and microbial testing. “If you use deionized water in product, you need to know requirements for the water in order to test it, and you procedures must be in place in case there are any deviations from requirements,” he said.
	 

	Another recurring theme in FDA enforcement actions has been OTC companies failing to respond adequately to a 483 and having the same problems discovered by FDA inspectors during a subsequent inspection. Repeat problems will trigger a warning letter.“Respond to 483s as if they were warning letters,” Stevens suggested.

	Another potential vulnerability is in finished product testing and validation, said Greathouse.  He suggested determining whether to perform microbial and other testing and validation studies inhouse or outsourcing it. “If you use an outside lab, you must ensure that the lab has validated its test methods and that they have demonstrated that to you,” he said. This requires formally auditing the contract lab to determine when and how they perform the tests, and what equipment and procedures are in place for those tests.

	Another area where OTC manufacturers may stumble is in nonconformance management and CAPA, an “unwritten regulation” for OTC drug companies, said Stevens. “CAPA is a formal way to identify problems, investigate the root cause of problems, and implement corrective action.  You may already be doing this informally but will need to formalize procedures to comply with cGMPs,” he said.

	Beware of inadequate, incomplete, or undocumented investigations, which will show up in warning letters. “Firms may fix problems but may not investigate the reasons for the problems in the first place to ensure that they don’t repeat. FDA expects you to use CAPA to remediate problems,” he said.

	Root cause analysis techniques will be crucial, and your CAPA procedure must identify the tools that you will be using. Another point to remember is to define the product conditions that will trigger a product recall. Stevens also suggested that companies not dismiss out-of-specification (OOS) and other problems as “laboratory errors” in documentation without examining their cause.

	As noted previously, another problem that FDA often finds, especially with small companies, is that there is no official quality department at the company. “A company’s quality unit has responsibility and authority from senior management. One way to tell how seriously a company’s management considers regulatory compliance and quality is how they budget for it, the kind of people they hire for it and what their authority is, who listens to them, as well as what kinds of quality decisions can and cannot be overturned,” said Stevens.

	The company’s most senior executive is legally responsible for the quality unit, and problems may result in the CEO being charged with violation of the US Food Drug and Cosmetic Act, he said.  It also behooves OTC company staffers to be on top of all previous FDA inspection files. After all, said Greathouse, the FDA investigators visiting the facility will have pulled these documents out a week ago, and they will determine how today’s inspection will be run. “Repeated observations are a red flag,” he said.

Warning Letter to Advanced Botanical Consulting Testing, Inc.

, June 11, 2019.
	2. B. Unger, “Week of March 10, 2019: FDA Sent These Warning Letters,” 

, March 19, 2019.
	3. B. Unger, “Week of June 23, 2019: FDA Sent These Warning Letters,” 

, July 3, 2019.
	4. A. Mulero, “Repeat Violations Spur FDA to Ban Canadian OTC Drugmaker Imports,” 

, May 14, 2019.
	5. FDA Group Webcast, “cGMP and Quality Management for OTC Manufacturers: Six Critical Questions,” 

Vol. 43, No. 8
	August 2019
	Pages: 40–41

	When referring to this article, please cite it as A. Shanley, “FDA Steps Up Enforcement of cGMPs for OTC Drugs," 

Senior managers of OTC drug companies are on a learning curve, as FDA warning letters cite insufficient understanding of cGMPs and inadequate responses to prior 483s.

FDA Steps Up Enforcement of cGMPs for OTC Drugs

The past few years have seen the commercialization of personalized therapies (e.g., chimeric antigen receptor [CAR]-T and other genetically modified cell therapies and gene therapies) and a shift from traditional to direct-to-patient clinical trials. Both developments are having a major impact, not only on pharmaceutical manufacturing and quality, but on supply-chain management, as third-party logistics companies and shipping specialists emphasize solutions designed for sensitive biologics and for new clinical trial models.

	While the cold-chain packaging designs targeting this market continue to improve, companies within the pharma supply chain are moving to digitization. More are embracing technologies that allow for continuous real-time product monitoring during transport, using advanced data analytics and leverage the Internet of Things (IoT) and artificial intelligence (AI). 

	A 2018 survey of 100 biopharmaceutical and pharmaceutical company managers (1) found that most of them were embracing new technology to improve traceability and efficiency, to intercept shipments to prevent loss, and to increase forecasting accuracy.  Considering the type of technology, 63% said they were using supply chain analytics, 37% were using advanced data analytics, 29% were actively using artificial intelligence, and 19% were using IoT, although 35% said they were implementing AI-based solutions. They also saw new technology as a way to optimize cold-chain spending. The survey found that 54% of respondents said they’d overspent on temperature control in over 25% of shipments, while 24% said they had underspent (1).

	Working directly with technology vendors such as Controlant, Elpro, OnAsset Intelligence, TSS, and Cloverleaf, or collaborating with shipping specialists that use vendor-supplied or in-house platforms, pharmaceutical companies find that the technology increases visibility and offers the opportunity to respond quickly to problems during material transit. Allergan Pharmaceuticals has found that Controlant’s real-time monitoring platform has improved data retrieval and removed supply-chain blind spots, according to Tommy Jennings, associate director of Allergan in Ireland (1). 

	Having access to data analytics can allow the root cause of logistics problems to be determined quickly, and, over time, permits companies to take a “manage by exception” approach, said Ada Palmadottir, Controlant’s business development director in a webcast (2). In one case, during truck shipment, the real-time monitoring system detected a temperature excursion, although the data logger on the truck did not. Ultimately, the problem was traced to the fact that the back window of the truck had not been shut completely after a customs check and let cold air into the vehicle, explained Palmadottir during the webcast.

	As patient-centered clinical trials transform biopharmaceutical logistics, personalized therapies demand transport at much lower temperatures. Packaging providers have responded by improving existing active and passive temperature-control methods and both reusable and single-use designs.

	Pelican Biothermal Co., for instance, offers reusable passive shippers, such as Credo thermal containers, the Credo Cube and Sherpa single-use container system (3,4). Sonoco ThermoSafe offers a variety of shippers for use in cell and gene therapy clinical trials (5). Marken recently introduced SmartBox (6), while Aerosafe Global‘s Aerocore packaging can operate from -65° C– 25° C for 24 to 120 h.  In 2018, the company introduced A-20 single-use packaging (7).

	Diversified transportation companies offer web-based monitoring systems that allow shipments to be tracked, some of them in real time. Cryoport, which has become increasingly active in cell and gene therapy over the past four years (8), offers recyclable Cryoport Express shippers which are used in the company’s Cryoport SmartPak II condition-monitoring system, connected via the company’s CryoPortal cloud-based network (9).

	Sunoco ThermoSafe’s PharmaPort platform offers shipping containers with sensors that allow for real-time condition monitoring via the company’s web-based system (10). Envirotainer offers active temperature control technologies with products such as its RAPT2 container and introduced a customer portal with expanded functionality in June 2019 (11). Marken provides the SENTRY condition-monitoring system, which provides uses with hourly updates (12).

	The company offers specialized capabilities for shipping both single-patient (allogeneic) and multi-patient (autologous) cell therapies (13,14). AmerisourceBergen’s World Courier subsidiary recently introduced Cocoon, a pallet-loading device that is designed for applications where very strict temperature controls are required (15). In May 2019, World Courier teamed up with GE Healthcare to integrate parts of GE’s Flex Factory with World Courier’s controlled temperature solutions for cell and gene therapy (16).

	Manufacturers can also turn to a wider number of solutions that allow for continuous monitoring of cargo-not only for temperature but other vital conditions such as humidity and vibration. Cloudleaf, for example, offers Sensor Fabric, a network of sensors using IoT servers and gateways that collect information that is monitored via a control center. The technology has been used to optimize transport of plasma samples (17). Sendum Wireless Corp., meanwhile, offers the Findum real-time monitoring platform and recently introduced a tracking system and sensor that can handle cryogenic level temperatures (18).

	1. Logipharma, Increasing Visibility with Real-Time Data, 

, August 2018.
	2. Controlant, How to Leverage Digital Transformation and Real-Time Monitoring, Webcast, May 2019.
	3. Pelican Biothermal, Pelican Credo Cube Product Data Web Page, 

.
	4. Pelican Biothermal, Sherpa Single-Use Packaging System Product Data Web Page, 

.
	5. Sunoco Thermosafe Product Data Web Page, Pharmaport, 

.
	6.  Marken, Smart Box Product Data Web Page, 

.
	7. N. Basta, “Aerosafe Expands Its Product Line,” 

, Nov. 16, 2018.
	8. V. Barba, “Cryoport’s CEO on Cell Therapies Market Robust Demand,” 

, April 22, 2019.
	9. Cryoport, SmartPak Condition Monitoring System, 

, 2018.
	10. N. Basta, “The New PharmaPort360: A Solution for Challenging Journeys in Global Logistics,” 

, May 3, 2019.
	11. Envirotainer, “New Online Services on Envirotainer Customer Portal,” Press Release, June 2019.
	12. Marken, Sentry Demonstration Video, 

, 2017.
	13. Marken, Services Capabilities Web Page, 

, 2018.
	14. Marken, “Speed and Flexibility to Deliver Temperature-Sensitive Immunotherapies Anywhere,” 

, July 19, 2018.
	15. World Courier, Cocoon User Guide Video, 

, May 2019.
	16. World Courier, “GE and World Courier Collaborate to Drive Commercialization of Cell and Gene Therapies,” Press Release, June 2018.
	17. Cloudleaf, Case Study, 

, June 2018.
	18. Sendum Wireless, “New Temperature Probe,” Press Release, May 2019.


	Vol. 43, No. 8
	August 2019
	Pages: 46–48

	When referring to this article, please cite it as A. Shanley, “Real-Time Monitoring Drives New Logistics Solutions,"

As patient-centered therapies and clinical trials become more important, innovation is providing more control and transparency in the way that pharmaceuticals are delivered. 

Real-Time Monitoring Drives New Logistics Solutions

Biopharmaceutical manufacturers have often described operations as data rich but information poor. While advanced analytics and sensors collect more data than ever before, much of it may not be used or shared with the operations that need it most to prevent lost batches and quality or compliance problems. The rise in outsourcing has only intensified the challenge.

	IDC Health Insights surveyed 126 biopharmaceutical and pharmaceutical executives in the United States and the United Kingdom and found a significant gap between their need and their strategies for harnessing data (1). More than 98% of respondents said that cross-functional data access was important or very important to their business strategies, and 94% described the ability to apply advanced analytics and/or artificial intelligence the same way. Respondents believed that data access would be crucial to improving overall quality and productivity as well as the return on investment of their R&D investments.

	However, 51% of those surveyed said that they did not have a clear strategy in place to help them reach either of those goals, citing regulatory uncertainty, budget prioritization, and the need for more action from functional operational groups. As Kevin Julian, senior managing director in Accenture’s Life Sciences practice, the survey’s sponsor, commented, “Important insights that could lead to the discovery, development, and delivery of promising new treatments are too often trapped within the functional silos of ... biotechnology companies” (2).

	While some pharmaceutical manufacturers are still using paper-based record systems, a growing number are digitizing processes and making more data accessible in the most accessible context. On a fundamental level, open control systems and a common data structure have helped allow this to happen, according to a recent report (3) from Rockwell Automation, resulting in distributed control systems (DCS) that enable integration using unmodified Ethernet, and allowing two-way communication between enterprise resource planning (ERP) and manufacturing execution systems (MES). Work is underway to improve collaboration in preclinical and quality labs, where good laboratory practices (GLPs), rather than good manufacturing practices (GMPs), drive operations (Sidebar, p. s14) Much progress is being made in the area of batch records, and Â­expanding connections between electronic lab notebooks, laboratory information management systems, MES, and ERP to increase access to information. Augmented reality offers one way to do this at the basic data recording and recovery level.

	In addition, according to Emerson Process Management’s consultant Johan Zebib, quality review management and review by exception are being used with MES (4). Data historians can also be developed as a point of access, a concept that Eli Lilly has leveraged with its contract manufacturers in medical devices (5). Vendors are offering tools that make this task easier, with applications that use machine learning and other facets of artificial intelligence, allowing users to make Â­connections between data points that might otherwise have seemed unrelated.

	Apprentice.io has developed augmented reality and database applications that allow users to get feedback as they perform their jobs and to compare equipment performance and different batch runs (6). Contract development and manufacturing organizations (CDMOs) are becoming a more important market for the technology, says CEO Angelo Stracquatanio, using it not only to share data, but for training and troubleshooting in real time. “There are so many silos within manufacturing, and data are not being leveraged at different levels,” he says.

	In 2019, the company has made improvements to its augmented reality products with augmented batch recordkeeping products that extend batch connectivity to laboratory information systems (LIMS) and electronic lab notebooks (ELNs), allowing inputs to be captured for every batch. Users can analyze process data to compare batch runs and implement continuous improvement programs and analyze specific runs to isolate deviations, Stracquatanio says, leaving an audit train of data that can be mined to decrease variability.

	A growing number of CDMOs are using Apprentice.io’s Tandem remote telepresence tool to collaborate with their clients. The Apprentice System also collects voice, picture, and other types of data to create a rich audit trail. “For example, for a single-use filter, one can scan its bar code. Users now know what filter that is and can create an audit trail for it … and put data together in real time, using a hierarchy of importance to present data in a way that doesn’t overwhelm the user,” says Stracquatanio.

	Use of advanced analytics and artificial intelligence (AI) is growing in the pharmaceutical industry (Sidebar, p. s15). However, many companies are still at the earliest stages of developing strategies for using it. So far, the technology is farther along in clinical trials and in discovery. Accenture launched INTIENT in May 2019 to focus on discovery, clinical, and pharmacovigilance applications. Amgen has been working with Tata Consultancy Services on a Holistic Lab digital platform using Dassault Systemes’ BIOVIA, for process development (7).

	However, some vendors are focusing on pharmaceutical manufacturing applications. Quartic.ai, for example, has launched an AI-driven platform to provide feedback to operators and to monitor and improve processes (8). The platform, which includes a data engine, designed to extract data from DCS, quality management systems (QMS), and data historians, as well as a connector that allows disparate software systems to communicate with each other, was designed to be integrated into existing plants and equipment, but Quartic is also working with a pharma company to embed the platform into a new facility.

	Quartic.ai cofounder and CEO Rajiv Anand has an extensive automation and reliability background and previously worked at Emerson. The company’s management team members all come from pharmaceutical and automation backgrounds. “We didn’t want pharma users to feel that they needed to be coders or data scientists,” says vice-president of life sciences Larry Taber.

The platform is geared to the fact that every potential user will have a different level of digital maturity, says Anand. Once legacy data sources have been connected, the artificial intelligence engine can be used to solve a specific problem (e.g., monitoring an asset’s performance for deviations), he adds. Some clients are using it for complex predictive work.

	The company has used its platform in a number of situations, including an effort to monitor and improve fermentation yield in a highly variable process where all critical quality attributes were under control. Quartic extracted data and identified a few key batches, Anand explains, and then built an algorithm to study relationships between the batches, clarifying eight years’ worth of data and fingerprinting each phase of the process. Ultimately previously unknown sources of variation were discovered. Work will now focus on learning more about them. The company has also done work with pharmaceutical companies in the area of predictive maintenance. Anand recalls one project designed to baseline the performance of an autoclave.

	The equipment was modeled and industrial Internet of Things (IIoT) sensors used to get additional vibrational and ultrasound information. Once deployed, machine learning models could then predict potential failures with the ability to trace the source of the failure down to an individual component (i.e., a damaged valve).

	There are many other applications where artificial intelligence could be applied to pharmaceutical manufacturing operations, particularly in visualizing end-to-end processes. As data analyst Jonathan Lowe (9) found at one company, more than 100 quality events were being investigated at any one time, stretching staff capacity. A machine learning model was designed to predict which events would take the longest to resolve. The model predicted more than 85% of the severe delays weeks before they happened, allowing quality managers to prioritize tasks more equitably.

	1. Accenture, “The Case for Connectivity in the Life Sciences,” Infographic, 

, May 22, 2019.
	2. Accenture, “Accenture Introduces Intient,” Press Release, May 16, 2019.
	3. B. Sisk, “Analytics Informs Decisions in the Facility of the Future,” Whitepaper, Rockwell, July 2019.
	4. J. Zebib, “Review by Exception: Connecting the Dots,” Whitepaper, Emerson Process Management, July 2019.
	5. D. Berg, et al., “Data Sharing in a Contract Manufacturing Environment,” Presentation at the OSI Soft Users Conference, 2017.
	6. A. Shanley, 

Pharm Tech Bio/Pharma Laboratory Best Practices eBook

 2018 (2) 8–9, November 2018.
	7. A. Louie, “Empowering 21st Century Science in Life Sciences,” an IDC Perspectives Paper, 

Quartic Introduces AI Platform for Process Manufacturing Operations at INTERPHEX 2019

 April 4, 2019.
	9. J W. Lowe, “Why Biopharma Manufacturing Needs to Leverage Network Optimization Techniques via Data Science,” medium.com, June 19, 2018.


	Supplement: Outsourcing Resources
	August 2019
	Pages: s12–s15

	When referring to this article, please cite it as A. Shanley, “From Data to Information,"

Pharmaceutical Technology Outsourcing Resources 

Making siloed data accessible across functions and to contract partners is the first step to facilitating continuous improvement and enabling use of artificial intelligence in manufacturing.