To explore those issues, Contract Pharma gathered a panel of experts from the newly formed LNP Alliance, including:

• Syed Reza (MD-PhD), Licensing and External Innovation Alliance Management, NOF America Corporation

• Nicholas Boylan (PhD), Senior Director, Scientific Services, Phosphorex

• Jerry Williamson (MBA), CEO, Phosphorex

• Mandy Janssen (PhD), Scientific Director, Characterization, Nano Imaging Services (NIS)

• Sara Little (BS), Associate Director of Oncology, NeoSome Life Sciences

• Jake McDonald (PhD), CEO, Envol Biomedical

Together, the group discussed where LNP programs often stall, how the COVID-19 vaccine era changed the field, and why successful translation increasingly depends on tighter coordination across lipid supply, formulation development, characterization, and preclinical evaluation.

The Conversation

Contract Pharma: Where do LNP-based therapeutic programs most commonly falter on their way to the clinic?

Syed Reza: The fundamental problem with nucleic acid therapeutics compared to other modalities like antibodies and small molecules is that in vitro and in vivo behavior correlation is very challenging. With proteins or small molecules, you can do a lot of meaningful optimization in silico or in vitro, and that tends to translate well in vivo. With LNPs, the complexity of their interactions with human biology makes it very difficult to glean meaningful insights from in vitro work alone. The cost of conducting in vivo evaluation is quite high, so I have observed many innovators trying to delay extensive non-human primate (NHP) or advanced in vivo characterization, only to hit roadblocks as they try to transition to the clinic.

Jake McDonald: LNP characterization and performance testing are expensive, so you generally want to deploy a tiered approach. First, utilizing non-animal methods, then moving up the hierarchy. Over the years, we’ve developed imaging tools and other real-time methods to assess efficacy and safety more efficiently, but the tools available to answer these questions have historically been limited. The engineers can often create new LNP formulations more quickly than the biologists can screen them, so reducing that dimensionality is where we’ve seen the most progress in the last five to ten years.

Nicholas Boylan: From my vantage point in the LNP value chain, scalability is often the most significant challenge. It’s very easy to make small-scale LNP batches on a benchtop system. But when you try to scale up, numerous issues can present themselves. For instance, ionizable lipids with poor solubility can lead to aggregation, complicating downstream purification

Sara Little: The biggest concerns I consistently hear relate to off-target responses and safety. Many developers are creating LNPs without necessarily having deep expertise in targeting different tissues or understanding how delivery will affect the rest of the system. This knowledge gap is where programs tend to stumble.

Jerry Williamson: I’ll add a perspective that’s a bit broader. If you compare the level of extensive knowledge we’ve built over the decades for therapeutics such as small-molecule drugs and monoclonal antibodies, we simply don’t yet have the same depth of experience with LNP-based drug delivery. The success rate shouldn’t be expected to match that of other modalities because, to some degree, we’re still figuring this out.

CP: The COVID-19 vaccines were a watershed moment for LNP technology. Has that momentum resulted in a more robust LNP therapeutic pipeline?

Reza: The COVID experience absolutely demonstrated that ionizable lipids and mRNA can be delivered at scale and are reasonably well-tolerated in a vaccine context. But the industry quickly ran into a critical distinction: genetic therapeutics are administered to sick and frail patients, sometimes at doses 200 times higher than those used in vaccines, and frequently require repeated dosing. The immunogenicity required for a vaccine typically becomes a serious liability for a therapeutic. Although the LNP community is working on it, we don’t yet have a rational, systematic way to engineer immunogenicity out of an LNP system. There is still a lot of trial and error.

Boylan: The exploration of utilizing LNPs beyond vaccines and hepatic diseases is accelerating, but we’re still on the learning curve. When LNPs were first developed for liver targeting, the mechanism was fairly well understood. Specifically, serum proteins like Apolipoprotein E (ApoE) bind to the surface and facilitate hepatocyte uptake. Now that we’re trying to re-engineer these systems for extra-hepatic targets like the lung, immune cells, and CNS, we’re still learning how to facilitate that targeting, whether by re-engineering the particle surface or by conjugating targeting ligands. High-throughput screening approaches, including DNA barcoding methods that allow multiple formulations to be evaluated simultaneously in a single animal, are becoming increasingly important.

McDonald: There has definitely been a momentum shift toward LNP-based delivery. Historically, many developers relied on viral vectors, but we’re seeing far less of that now, and the pandemic almost certainly accelerated the shift. 

Little: In terms of the work we’re doing at NeoSome, a greater percentage of programs now involve gene editing that relies on LNP delivery. We’re definitely seeing a lot of innovation and advancement in the space.

Mandy Janssen: From a characterization standpoint, the mRNA vaccine era did something important: it brought an enormous volume and diversity of LNP formulations into analytical workflows, and what that revealed is how structurally variable LNPs can be, even within a single “platform.” The assumption that formulation success in one context readily transfers to another can obscure the fact that small changes in lipid composition, cargo, or process conditions yield meaningfully different structures.

CP: When should the biodistribution strategy enter the development process, and what can getting it wrong cost a program?

Boylan: Biodistribution strategy work should begin at the outset of a program, ideally before you lock in a lead formulation candidate. These considerations will directly inform your choice of ionizable lipid, the overall LNP composition, and whether an active targeting approach is necessary. For liver-targeted applications, LNPs will naturally accumulate there; it’s relatively low-hanging fruit. But for extra-hepatic targets, there’s always a balance between how much of the formulation reaches the target tissue versus how much ends up in the liver. Confirming biodistribution early in the program ensures that the most promising candidates are advanced and minimizes the risk of getting the wrong biodistribution, which can result in significant delays, added cost, and potential safety concerns due to off-target effects.   

Little: Early biodistribution studies can make or break a program. You can have positive in vitro results but still get unexpected results after administering a formulation to an animal. If the therapeutic doesn’t achieve the desired targeting in a mouse, it is unlikely to work in a human system, and proceeding to more expensive primate models is, at this point, premature.

Reza: We encourage clients to screen a variety of ionizable lipids and LNP compositions early, checking for potency and biodistribution as a fundamental screening step. The goal is to have candidates that are well-validated in both rodent models and in vitro human cell-based immunogenicity assays before moving into NHP studies, where you will generate a much higher level of confidence in the therapeutic’s performance, but testing costs are much higher.

McDonald: The tools available to answer biodistribution questions have advanced considerably. Over the last five to ten years, we’ve seen meaningful progress in the use of imaging, both ex vivo and in vivo, to evaluate whether a formulation is hitting its intended target or producing off-target effects. We’ve also seen innovations on the analytical side, including DNA barcoding approaches that allow you to tag multiple test articles and evaluate them simultaneously in a single animal. The reality is that engineers are often faster at generating new LNP permutations than biologists are at screening them, so the name of the game is reducing the dimensionality of the screening problem. Smarter integration of test articles and more disciplined down-selection are where the field has made its biggest strides.

CP: Cryo-EM characterization can reveal structural details about LNPs that conventional analytics simply cannot. At what stage of development should structural characterization be integrated, and how often does it change the direction of a program?

Janssen: Structural characterization should be integrated from early formulation development, not held until late preclinical stages when the window to act on findings has largely closed. In practice, cryo-TEM changes program direction more often than teams expect. Bleb formation, lamellar disorganization, elevated empty particle fractions, and overall heterogeneity are not detectable with standard tools such as dynamic light scattering (DLS), nanoparticle tracking analysis (NTA), and bulk encapsulation assays. Yet, each has direct implications for potency and stability. Finding these issues early, during excipient screening or process optimization, is what makes structural data formative rather than merely confirmatory.

Cryo-TEM plays important roles beyond early-phase development. For IND-enabling work, it provides the morphological identity data that regulatory submissions increasingly expect. For troubleshooting, it serves as a source of truth when bulk assays disagree or when a batch fails to meet specifications. For targeted LNPs specifically, cryo-electron tomography (cryo-ET) is particularly valuable, as it enables direct visualization of surface-bound antibodies or ligands in three dimensions, making it uniquely suited to confirming conjugation and assessing ligand distribution, which no bulk assay can provide.

CP: Ionizable lipid selection is often described as one of the most consequential decisions in LNP development. How should developers approach this selection process?

Reza: Some people think of this as searching for El Dorado; one golden lipid that does everything. But that doesn’t exist. There are currently about half a dozen ionizable lipids that have been used in the clinic. When we speak with partners, the preference is almost always to work with a lipid that already has human exposure data. But beyond science, the decision is also heavily driven by business considerations like intellectual property, licensing access, and whether you’re acquiring just the lipid or an entire validated LNP formulation package. If you’re licensing a lipid with a validated formulation, you can streamline your efforts. If you’re working with a novel ionizable lipid, you’re starting with more freedom but also more unknowns.

CP: How is the LNP Alliance collaboratively working to address patient safety concerns? 

Reza: Immunogenicity is a huge topic, but we are developing a better understanding of which components of an LNP drive inflammatory responses. The choice of ionizable lipid is a primary consideration, as some are more immunogenic than others. Our in vitro macrophage assay is very useful for predicting immunogenicity, and we’re closely monitoring the induction of anti-PEG antibodies. For instance, we recently concluded a study with Envol in NHPs comparing two PEG lipid types, and there are early signals that the choice of PEG lipid meaningfully affects this response. A third factor is impurities, particularly from RNA or lipid degradation products. That’s where Phosphorex’s manufacturing controls and sophisticated analytics are critically important.

Little: While there are patient safety concerns related to repeat dosing, LNP therapeutics tend to fare a bit better than some other strategies on this front. Developers are, in many cases, moving away from viral vectors due to patient safety concerns about redosing because of off-target effects, and their larger particle size can limit biodistribution, with certain viruses being rapidly cleared by the liver and kidneys.

CP: Can you describe how the Alliance has functioned in practice and some of the benefits of the collaborative effort?

Williamson: The fundamental premise of the LNP Alliance is that a biotech or pharma company trying to develop an LNP-based drug has to bring together all of the expertise that we collectively represent. Assembling all of the experience we have within the LNP Alliance would take a lot of time and would incur high costs. We each have our areas of specialization: NOF provides expertise and supply in ionizable lipids; Phosphorex handles formulation development and manufacturing; NeoSome provides in vitro and in vivo evaluation in rodent models; Envol provides non-human primate studies and preclinical toxicology; NIS offers advanced characterization expertise. Combined, we cover the full translational arc. As I said, the collective we have built would take a lot of time and money for a biotech to build, and large CDMOs are often serving multiple modalities, so they tend not to have this level of specialized depth.

Reza: The first major collaborative project we completed was a translation study in rodent and non-human primate models evaluating our ionizable lipids, and was planned jointly by Phosphorex, NOF, Envol, and NeoSome. One very practical issue we encountered was that the animals gained weight between selection and dosing, resulting in unexpectedly higher required material volumes. We had to redesign batch production at Phosphorex to keep up with rising material requirements. The lesson from the effort was to plan for at least a 25 percent material overage and build in close communication between partners throughout the effort. It’s that kind of real-world, cross-organizational problem-solving that is so promising for future work.

McDonald: Companies come to us to evaluate a wide range of LNP permutations, systematically controlling variables to understand their impact on efficacy or toxicity. In many cases, though, they’re using whatever screening tools are most readily available rather than the tools best suited to answer their specific questions. That’s one of the real advantages of the LNP Alliance: we’re not just deploying existing tools, we’re developing new models and new capabilities that give developers a genuinely better chance of identifying what’s going to work.

Boylan: Each time we work together, we become more efficient. The planning, the lead time, the anticipation of challenges, all of that improves with shared experience.

Where Translation Happens

The LNP field stands at an inflection point. The science has matured enough to move well beyond hepatic targets and vaccine applications, yet the gap between what developers need to achieve, available tools, and what empirical knowledge can reliably support remains significant. Closing that gap will not happen through any single organization working in isolation.

Translating LNP-based therapeutics to the clinic is, at its core, a multidisciplinary problem that demands expertise in lipid chemistry, formulation engineering, structural characterization, and preclinical biology working in close, continuous coordination. The LNP Alliance was built on this exact premise and is working to provide specialized expertise to dramatically increase the clinical and commercial success of LNP therapeutics.

To learn more about the LNP Alliance, visit www.lnpalliance.org.

Meet the LNP Alliance Panel

Nick Challoner, CEO of technology-driven CRDMO Ingenza, believes the next phase of growth in biotechnology will depend not only on scientific innovation but on the industry’s ability to translate discovery into scalable, commercially viable solutions. In this Q&A, he shares with Contract Pharma his perspective on what customers now expect from their partners, and where companies like Ingenza must focus to remain competitive.

Contract Pharma: Briefly describe your background and how it has shaped your approach to leadership.

Nick Challoner: I think my career path was a little bit different from others at Ingenza, because the majority of my experience is within the commercial areas of high-performance ingredients, but I did spend a good part of my early career in R&D. One of my drivers throughout my career has been to really understand the R&D process. I have spent a lot of time commercializing science and innovations, and I love working with really talented people within the research sphere. A key skill I have developed over the years is decoding complex science into something non-scientists can understand, and aligning it with current or future market needs.

CP: What attracted you to Ingenza?

Nick: I was looking for an organization where I could make a difference very quickly, to help them grow or achieve their objectives in a short period of time, but mostly, I was excited about the technology and the space that they are in. I know the pharmaceutical, consumer care, and sustainability sectors well from my career experiences, and I was excited about helping Ingenza prepare for growth and execute its strategy.

CP: What capabilities will be most important for CRDMOs to remain competitive in this environment?

Nick: The pharmaceutical sector is massive, with many players. CRDMOs need to understand exactly what their customers want to ensure that the service they offer and the way they do business align with the market’s needs. For customers, speed in this industry is everything; the longer you take to develop something, the higher the cost becomes, which creates a disadvantage for the company trying to bring that innovation forward. Organizations are therefore typically looking for CRDMOs that can accelerate development and reduce development risk. For example, at Ingenza, our InGenius chemistry manufacturing and controls (CMC) platform enables rapid, risk-managed project execution to develop and optimize bioprocesses and protein development for a range of complex biologics and enzymes.

“For customers, speed in this industry is everything; the longer you take to develop something, the higher the cost becomes, which creates a disadvantage for the company trying to bring that innovation forward.” — Nick Challoner, CEO, Ingenza

CP: Technology is transforming the way development is approached. How do you see AI influencing biotechnology?

Nick: AI is a tool that can help us significantly in moving towards customers’ end goals of improving speed and reducing risk. But unless you integrate AI closely with expert knowledge and real-world scenarios, you will never be effective in meeting customers’ expectations. Customers’ views are starting to mirror ours, and we’ve heard from the industry that they also see the need for AI approaches in areas such as gene design to be supplemented by skilled biotechnological capabilities to deliver effectively in practice. This understanding has been driven by frustration with AI-only front-end models, whose results have often been unachievable in the real world.

The key to Ingenza’s success in the past and in the future lies in accelerating the uptake of our dual-engineering biotechnology program. Dual engineering is essentially the concept of applying AI or machine learning front ends—including our codABLE gene design algorithm and UNVAIL enzyme discovery platform—then filtering them through the combined intellectual capacity and experience of our team, and into our wet laboratory to deliver what the client needs. Because we offer both the AI front end and the wet lab, these parts of the business can keep talking to each other, which is continuously improving our hit rate.

CP: Ingenza recently conducted a strategic review. What was the main objective of that process?

Nick: There is one fundamental thing that’s important for the future of this organization: understanding exactly what its customers and the market need. That is central to how we operate. Part of the strategy review was to understand exactly what our customers want and, therefore, how we can ensure our service offering is 100 percent aligned with their needs.

CP: What are your priorities for Ingenza’s next phase of growth?

Nick: In the recent strategy review, we redefined our sectors into Human and Animal Health, which is where our traditional pharmaceutical business sits, and Sustainable Technologies, markets where sustainably delivered performance and broader trends are important drivers of demand. And when we talk about ourselves being a biotech company, what we really mean is a protein company. We see ourselves as ‘the’ protein company because we believe we are better positioned to help customers with their protein problems or development than anyone else out there. The reason that’s important for our pharmaceutical customers is that, if you look at modalities outside of the small molecule space, virtually everything is protein or protein-related, i.e., biologics, monoclonal antibodies, and mRNA are all based on protein systems. The market for protein therapeutics is about $300 billion and growing at a 10 percent CAGR, making it a significant opportunity for us.

CP: What are your ambitions for the future?

Nick: Our 2030 vision is to strengthen our performance to become the leading protein CRDMO solution provider across the Human and Animal Health and Sustainable Technologies markets. We’re going to build on what is a strongly performing company with a stronger customer focus, ensuring a greater understanding of the market and the sector’s performance drivers.

To learn more about Ingenza, click HERE.

Currently, 49 cell and gene therapies (CGTs) have been approved within the U.S and pipelines are proliferating and maturing. As more therapies move from clinical development to market, demand is increasing for the capabilities needed to deliver these therapies at scale, namely manufacturing and supply chain solutions.

With the increasing complexity of CGT programs, sponsors are focusing on strategic partnerships to expand their offerings. Sponsors often engage contract development and manufacturing organizations (CDMOs) earlier in development, whether it’s for specialized platform technologies or for smoother tech transfer and scale up. This shift signifies the growing technical demands of CGTs as analytical complexity and scale-up constraints create challenges following the discovery stage.

The following perspectives from SMEs in the CGT CDMO space highlight the key reasons sponsors outsource, what they’re looking for in a CDMO, and what these partnerships entail. 

Factors Driving Outsourcing in the CGT Space

Developing and manufacturing CGTs involves a myriad of complexities, requiring specialized capabilities, infrastructure and GMP-compliant facilities. Building these in-house is challenging and costly. With available capacity, experience navigating evolving regulatory requirements and supply chain complexities, CDMOs have the necessary expertise to support the development manufacture of CGTs.

Sponsors outsource CGT development and manufacture for a range of reasons, and the starting point is not always the same, according to eXmoor Pharma’s Harvey Branton, Senior Translation Consultant, and Drew Hope, Senior Bio Manufacturing & Compliance Consultant & UK QP. “The most common reason is a lack of in-house GMP manufacturing infrastructure. But many organizations have some internal capability and still choose to outsource because their existing resource does not match the capacity, specialist experience, or timelines a specific program requires,” Branton and Hope contend. 

CDMOs provide established and readily available capacity, experienced technical staff and validated quality systems to help navigate production and support progression to clinical trials and into the market, according to Katherine Schewe, Director, Strategic Accounts, Advanced Therapies, Fujifilm Biotechnologies. “CDMOs also bring vast experience in supporting INDs and BLAs, institutional knowledge in CMC documentation and inspection readiness, and guidance in navigating evolving regulatory requirements,” Schewe said. 

Regulatory expectations continue to rise, with increasing emphasis on manufacturing robustness, comparability, and commercial readiness, according to Thomas Fellner, Vice President and Head of Commercial Development, Lonza Specialized Modalities. “As a result, sponsors look to CDMOs not just for capacity, but for risk reduction, speed, and execution certainty. Outsourcing is ultimately about creating a competitive advantage and enabling a smoother transition from development to reliable commercial supply,” he said.

Product complexity is another significant driver of outsourcing. CGTs can present unique manufacturing challenges, particularly where multiple drug substances or processing approaches need to be integrated within a single drug product, according to Branton and Hope of eXmoor Pharma. “RNA/LNP-based non-viral gene therapies are a good example: oligonucleotide synthesis, plasmid production via microbial fermentation, and lipid nanoparticle formulation each require distinct manufacturing approaches, making in-house delivery difficult without substantial investment and breadth of expertise,” Branton and Hope said.

As such, sponsors often seek specific, high-value capabilities from a CDMO. Luca Alberici, Executive Vice President of the Cell and Gene Technologies Division, AGC Biologics, said, “The market is facing some interesting structural imbalances, with an overcapacity in AAV vector manufacturing on one hand, and a persistent bottleneck in complex, labor-intensive autologous cell therapies on the other. This dynamic means sponsors are looking for partners with specialized expertise in their exact modality, making a CDMO’s scientific and technical skill the real differentiator.”

At the same time, therapeutic pipelines have sponsors seeking specialized partners. “There’s growing excitement around next-generation modalities such as in-vivo CAR-T and gene editing, which avoid complex ex-vivo cell handling,” said Alberici. “Sponsors are choosing to outsource to tap into a CDMO’s established excellence in these advanced platforms, which helps them accelerate their own entry into these cutting-edge fields.” 

Outsourcing has also become a key strategy for managing risk in an unpredictable market. According to Alberici, a CDMO with a global, multi-site network offers built-in supply chain resilience and operational continuity. He said, “We’ve seen how valuable this is when a program changes ownership during the clinical phase; a consistent CDMO partner preserves crucial process knowledge and ensures a smooth transition.” 

Meanwhile, Wade Macedone, CEO of Andelyn Biosciences, explained, “Outsourcing is not just about increasing capacity, it’s about readiness. Sponsors are looking for partners who can help them make informed decisions, avoid unnecessary risks, and maintain momentum as programs advance. 

“There’s also a practical reality: many CGT programs are still in early development. Outsourcing offers flexibility, enabling sponsors to scale thoughtfully without overbuilding too soon, while still maintaining a clear path toward clinical and commercial supply.” 

Selecting a CDMO

Beyond availability, key factors for sponsors when selecting a CDMO to support CGTs include advanced technology platforms, modality expertise, analytical capabilities, regulatory knowledge and importantly, a commercial track record. 

Choosing a CDMO for a cell or gene therapy should be viewed as selecting a strategic long-term partner that meets both technical needs and operational reliability, according to Schewe of Fujifilm Biotechnologies. “Established modality expertise, such as AAV, lentivirus, adenovirus, autologous or allogenic cells, robust GMP quality systems, and a strong regulatory track record are critical factors to consider when evaluating a CDMO partner,” she said. “Other factors to consider include proven tech transfer and analytical capabilities, and supply chain robustness.”

Meanwhile, manufacturability and cost of goods are also critical considerations according to Thomas Fellner of Lonza. “Many advanced therapies face challenges achieving commercial viability, so sponsors value CDMOs that can apply platform-based approaches and design efficient, scalable processes early in development. Strong regulatory expertise is also essential, given evolving CMC expectations and the need to advance programs without unnecessary delays,” said Fellner. Sponsors are also assessing a CDMO’s continued investment in next-generation technologies and emerging modalities beyond today’s programs.

AGC Biologics’ Alberici added, “While many CDMOs can produce GMP batches, only a handful have successfully guided multiple CGT products through the rigors of FDA and EMA approval. Sponsors are now scrutinizing a CDMO’s history, prioritizing partners with a portfolio of approved commercial products. This is the ultimate proof of their regulatory skill, quality systems, and ability to execute at scale.” The ability to prepare a single, robust CMC package that meets the distinct requirements of both the FDA and EMA is an invaluable capability.

Additionally, a CDMO’s analytical capabilities are key. “With the understanding that ‘you cannot control what you cannot measure,’ sponsors are choosing partners with a deep arsenal of in-house analytical tests and the expertise to develop robust, phase-appropriate methods to fully characterize a product,” Alberici added.

Sponsors want to know how CDMOs operate dependably in a regulated setting, according to Wade Macedone of Andelyn Biosciences. They evaluate a CDMO’s quality systems, how processes are documented and deviations are managed, as well as how consistently the organization performs over time. “Beyond that, experience is important. Not just technical skills, but also in different stages of development and program phases from preclinical to commercial within CGT. Sponsors want to see how a partner has handled complexity before and what they have learned from those experiences,” said Macedone.

Availability is also critical. Branton and Hope of eXmoor Pharma added, “CGT programs frequently run to tight timelines, and a CDMO that cannot commit capacity at the right point in a development plan offers limited value regardless of its technical capability. Cost matters, but needs to be weighed against quality, communication, regulatory readiness, and the downstream consequences of delays or rework.”

While some sponsors want a collaborative relationship, with the CDMO functioning as an extension of their own team, others prefer a more defined, transactional arrangement, according to Branton and Hope. “Neither approach is inherently better, but a mismatch in expectations can create friction that affects program delivery,” they said. “Location is also worth considering; proximity to clinical trial centers or target markets can reduce logistical complexity and regulatory burden in ways that are easy to underestimate at the outset.”

Sponsor, CDMO Partnerships 

In the CGT space, partnerships are often less transactional and more about a shared-risk-reward model where the CDMO’s investment in innovation helps to advance the sponsor’s program. Advancing CGTs requires partners have a sense of shared responsibility and goals, technical collaboration, and close coordination.

A true partnership in CGT is active and continuous with a shared responsibility across development, manufacturing, and quality, according to Wade Macedone of Andelyn Biosciences. “This involves early alignment on process strategy, analytical methods, and regulatory expectations. It requires ongoing data sharing and joint decision-making as programs develop,” he said. “It also means being prepared to pause, review, and modify when necessary to safeguard quality and patient safety.”

Katherine Schewe at Fujifilm Biotechnologies added, “Ongoing technical collaboration improves process development and design, helps avoid program delays, and ensures a strong analytical strategy. Ultimately, enduring partnerships are built on a foundation of trust between a sponsor and their CDMO and thrive as an alliance with mutual investment and accountability.”

Strong partnerships typically begin early, with CDMOs working alongside sponsors to shape development strategies, identify high-risk steps, and design processes that are scalable from the outset, according to Thomas Fellner of Lonza.

Early alignment is critical as decisions made at this stage directly influence cost, timelines, and the likelihood of commercialization, Fellner noted.  “As programs advance, the partnership extends across tech transfer, process verification, comparability planning and regulatory alignment across regions,” he said. “This requires close coordination, transparency, and shared accountability, particularly as programs increase in complexity.”

In today’s market, a CGT partnership is a deeply integrated strategic alliance, according to Alberici at AGC Biologics. “The CDMO acts as a stabilizing force, making proactive investments in next-generation technologies and capacity before the market demands it,” he said. “This de-risks the sponsor’s development path by reducing their upfront capital costs and giving them access to state-of-the-art infrastructure.”

Meanwhile, Branton and Hope of eXmoor Pharma pointed out that the nature of a sponsor, CDMO partnership in CGT varies considerably depending on the size and maturity of the sponsor. “Smaller companies tend to look for a deeply collaborative relationship, where the CDMO takes on a meaningful share of the CMC thinking and operates as an extension of their technical team, often because internal resource is limited,” said Branton and Hope.

On the other hand, larger organizations typically have more defined expectations and require a well-scoped manufacturing service with clear boundaries rather than broad collaborative input, according to Branton and Hope. “That said, a larger sponsor may still choose to work with a smaller, specialist CDMO for capability in a newer or more complex area and may accept that some mutual support is part of that arrangement as the CDMO develops towards commercial-scale supply.”

Importantly, a successful sponsor, CDMO partnership goes beyond timelines, costs, and manufacturing capacity, according to Susan D’Costa, Chief Technical and Commercial Officer, Genezen. “Sponsors value partners that demonstrate flexibility, creativity, and a willingness to problem-solve when unexpected challenges emerge,” she said. “Effective CDMOs take a long-term, lifecycle-focused approach, considering future scale-up and commercialization needs from the outset. Proactive risk management across technical, quality, and regulatory areas is also key to maintaining program momentum.”

Sponsors bring extensive knowledge of their therapy and program goals while CDMOs provide manufacturing expertise, operational discipline, and regulatory experience. When these perspectives come together effectively, the result is stronger, more resilient programs, according to Wade Macedone of Andelyn Biosciences. “At its best, the partnership provides a consistent way forward, able to adapt to complexities without compromising the integrity of the work,” he concluded.

The difference between the two is not subtle. It shapes the quality of communication, the reliability of timelines, and ultimately the probability that a therapy reaches patients. The CDMO environment is defined by tight budgets, specialized modalities, lean biotech teams and rising regulatory expectations. Therefore, the approach to risk taken by a CDMO is a key differentiator, as well as an operational necessity.

When Does A CDMO Create Risk?

At its core, risk in CDMO partnerships emerges when complexity, uncertainty and siloed working converge. Drug development is unpredictable by nature, and manufacturing pathways are often conditional, non-linear and sensitive to every change in input. When a CDMO overlays this inherent uncertainty with created fragmentation: multi-site operations, unclear ownership, multiple handoffs, multiple project management teams, or overly rigid processes, the result is not simply slower progress. It is the systemic amplification of risk.

Many sponsors recognize this pattern all too well. They enter a partnership hopeful, yet quickly find themselves navigating communication gaps, inconsistent data flow, or shifting points of contact. Each handoff becomes a point of vulnerability; each delay a missed milestone. Risk compounds quietly in the background, long before it emerges in the form of a late batch, an unexpected deviation or an unanticipated cost.

This is the scenario biotech companies work hardest to avoid, and it is why the most forward thinking CDMOs have taken a fundamentally different approach, designing their operating models not only to control risk, but to actively reduce it.

Mitigating Risks in CDMOs

A key consideration for CDMOs is how to mitigate risk in an environment where risk is abundant. Many CDMOs rely on generic project management structures or traditional operational silos, creating miscommunication and blind spots in key projects. Meanwhile, forward thinking CDMOs structure their team structure and ways of working around early anticipation, cross-functional integration and continuity. Risk mitigation is not a separate workstream; it is embedded into every interaction, every lab process and every stage of a program’s lifecycle.

Knowledge Continuity as a Risk Mitigator 

Many CDMOs treat project management as an administrative function, a scheduling and documentation exercise that exists adjacent to scientific execution. The result is an immediate gap: a project manager may be involved, but not truly embedded, and therefore cannot foresee scientific risk early, nor influence technical decisions with a full understanding of program context. CDMOs that actively mitigate risk take a different route. Technical and project delivery staff operate as a single, integrated unit that moves through development and manufacturing as one continuous team. This eliminates the vacuum in which risks usually grow. A shift in how risk is mitigated means that issues are identified proactively, as opposed to reactively. This allows risks to be identified and dealt with before they have material impact because the people managing timelines are the same people who have been working on the project from the first day it entered the research and development labs. With consistent people on your project, knowledge is maintained at every phase of the project, reducing the risks associated with knowledge loss.

CDMO Operational Structure as a Risk Consideration

One-site CDMOs significantly mitigate risk through reduced handovers, reduced complexity in moving projects through each phase, and by improving the communication by working with people who work together every day. In some CDMOs, each phase: preformulation, formulation development, analytical development, process optimization, tech transfer, and GMP manufacturing is handled by separate, loosely connected teams. Each transition requires a handover, and introduces the possibility of misalignment, misinterpretation or delays. By contrast, one-site operations like Upperton’s Trent Gateway Facility, are built around continuity. The same program management teams remain with the program throughout its journey, maintaining technical memory and ensuring that every decision made early on is carried forward with clarity. This reduces uncertainty, compresses timelines and creates a risk mitigating through line from day one. 

The physical design of one-site operational facilities plays a role as well. Many CDMOs operate legacy sites that have been expanded organically. They have multiple buildings, disconnected labs, and process flows that force teams to work around the architecture rather than with it. One-site operations such as Upperton’s purpose-built Trent Gateway facility take the opposite approach. They streamline the movement of materials and people, reduce contamination and deviation risk, and support efficient, linear progression from development to clinical or commercial manufacturing. By removing complexity from the environment, it removes risk from the process.

Cross Team Collaboration as a Key Risk Mitigator

Cross team collaboration may be the most significant differentiator. Risk management is not solely about tools or infrastructure, it is also about communication workflows, behavior, and people processes. CDMOs that inadvertently increase risk often do so because teams operate under pressure, communication is reactive, or internal priorities overshadow client outcomes. Through a one-site operational model, and consistency through projects teams and ownership, teams can be encouraged to flag uncertainties early, challenge assumptions, and make decisions collaboratively. This openness is essential: when scientific and operational challenges are acknowledged early, they are manageable. When they are concealed or postponed, they have the potential to become crises.

Quality of Communication as a Risk Factor

Another dimension that distinguishes risk mitigating CDMOs is the quality of communication between CDMO and Biotech. Forward thinking CDMOs prioritize direct, consistent dialogue, clear client updates and meaningful technical discussions. Regular, science led interaction ensures there are no “unknown unknowns” accumulating in the background. Biotechs gain full visibility of progress, challenges, next steps and the rationale behind key decisions. This contributes not only to risk reduction but to trust, which is arguably the most valuable currency in outsourced development.

By viewing quality of communication as a risk factor, risk management becomes inseparable from service design. When communication is clear and teams are aligned, risk is controlled at its source. When they are absent, it doesn’t matter how many risk registers or mitigation plans are created, issues will seep into the gaps left behind.

Summary

Risk isn’t going anywhere, but some CDMOs are developing better mitigation strategies. The industry will continue to evolve and increase in complexity. With it, the expectations placed on CDMOs will rise. Whether a program succeeds or falters will increasingly depend on how well partners can anticipate challenges, maintain coherence across disciplines and communicate clearly with clients. Not every CDMO is equipped to do this. But those like Upperton that have invested in the resources to manage risk, will stand out in an inherently unpredictable industry.

The message is simple: risk is a variable shaped by design, behavior and environment. CDMOs that recognize this and embed risk mitigation into their operating models are the ones best positioned to support biotechs through the complexity of development and the pressures of commercialization. And in a sector where timelines matter, budgets matter and patient outcomes matter most of all, that difference is invaluable.

Lorna Patrick, Chief Operating Officer at Upperton has over 20 years’ experience in drug development, optimizing operational processes and change management across a variety of roles including manufacturing, analytical, project management, clinical operations, and regulatory affairs. She has previously held senior roles at large pharma and CDMO organizations across a number of functions.

Building analytical strategies from the outset of a cell therapy development program that are designed to exceed regulatory expectations simplifies decision-making, aligns early development with long-term commercialization goals, and reduces risk. Use of a design-of-experiment (DoE) approach to accelerate development of optimal, robust assays saves time, materials, and money while supporting improved decision-making. Approaching development as a series of milestones and stage gates allows for phase-appropriate analytical control, thereby mitigating quality and compliance risks as cell therapy programs move toward early clinical validation.  

Early investment in analytical development thus enables data-driven process changes and enhances understanding of both processes and products, ultimately guiding cell therapy programs throughout their entire lifecycles.

Benefit from a DoE Approach tO Analytical Method Development

Robust assays are a cornerstone of cell therapy program success across all phases from process development through commercialization and product release, making efficient and effective assay development essential. Achieving accurate and precise assessment of critical process parameters ensures therapeutic product consistency and effectiveness. 

Proper assay development, particularly for complex cell-based assays, can, however, involve significant investment in time and cost, which can be prohibitive for early-phase programs. Using a DoE approach can help minimize that investment by accelerating method development. Rather than exploring one factor at a time, as was traditionally done, DoE studies support the assessment of multiple parameters simultaneously, increasing resource efficiency and effectively reducing the cost and time required for analytical method development. DoE studies also lead to improved decision-making in the development of robust analytical methods for cell therapy programs, which, in turn, enable more-informed decision-making across the overall program.

Given the nature of cell therapies, critical quality attributes (CQAs) include, but are often not limited to, the purity and identity of the cell population (typically determined using flow cytometry). Because many different process parameters (e.g., multiplicity of infection (MOI), a given set of cytokines over a range of concentrations, expansion duration, etc.) can impact these CQAs and often influence one another, ensuring selected parameters support the target product profile (TPP) requires a deep understanding of both the product and process and relevant analysis conditions.

This knowledge is used to establish efficient DoE studies that will, when performed using material from a single donor, provide the data needed to evaluate the impact of multiple process parameters and support method development activities with less upfront experimentation. Large-scale runs using identified, optimized conditions and methods can then be performed with material from numerous donors to confirm the appropriateness of the defined process design space and analytical assays. 

Such an approach minimizes the experimentation, testing, time, and cost needed for both process and method development by reducing the impact of starting material variability while providing the necessary means to measure and understand it. It also results in the determination of an analytical control strategy comprising both release and characterization tests that ensures assessment of important potential changes and supports a robust IND-enabling package.

Reducing Risk Through Analytical Control

Development timelines are shrinking for all cell therapies, and particularly those with accelerated approval designations. Decisions made at the earliest stages of a project, including analytical method and control strategy choices, directly impact the feasibility of scaling to commercial production and lay the foundation for success—or create roadblocks—across the entire development lifecycle. They may also limit the data available to support future process changes, requiring costly comparability studies. 

Risk management should thus begin in early-stage development. Decisions are prioritized by approaching development as a series of milestones and stage gates. Mitigation strategies are then defined a priori as part of the product development lifecycle. This approach supports simplified decision-making, proactive risk management, and alignment of early development with long-term commercialization goals. Overall, risk-based stage gates enable teams to mitigate uncertainties early, optimize processes and analytical methods for scalability, and maintain regulatory compliance—all while keeping commercialization in focus.

Specifically, considering analytical methods, they should be optimized with every clinical batch to enhance understanding and product quality. Tests should therefore be refined, aligned with CQAs and critical process parameters (CPPs), and prepared for method validation in accordance with ICH guidelines. Meeting these milestones ensures the development of an adequate control strategy, thereby assuring product safety and efficacy and avoiding delays in product approval.

Using a phase-appropriate analytical control strategy, meanwhile, is essential for mitigating quality and compliance risks as cell therapy programs move toward early clinical validation. It allows developers to ensure product quality, begin understanding inevitable patient variability within a trial’s overall scheme, and establish an early path into the clinical trial setting.  

Furthermore, reliable analytics, coupled with a well-thought-out retention strategy, help define and enable the process changes needed to support later-stage clinical studies without significantly slowing overall development of a cell therapy program. As processes mature, analytical and characterization insights will efficiently guide the effort, making comparability studies much easier, more efficient, and more interpretable.

Accelerate the Path to IND with Early Investment in Analytics

Phase-appropriate method development, however, does not mean taking shortcuts. Early decisions—such as selecting raw materials, designing robust processes, and prioritizing analytical development—therefore shape the success of a product’s journey from research to commercialization. Well-constructed analytics are the foundation for any cell therapy program, as without trustworthy data, informed, defensible process changes cannot be made. Under the pressure of expedited timelines and tight budgets, however, deferring the implementation of system-suitability controls and method qualification often occurs, leading to the generation of unreliable data.

When analytics are unreliable, decisions lack a solid foundation, increasing the likelihood of optimizing for inappropriate endpoints or critical quality attributes. Because timely and robust process characterization is a critical foundation for meeting regulatory and safety requirements for commercialization, failing to thoroughly characterize a cell therapy process early sets off a chain reaction of technical and regulatory difficulties that are challenging to rectify. Without data-driven insight into how each unit operation behaves, critical parameters drift from loosely defined values in the lab to poorly controlled targets in GMP production. Unfortunately, repercussions generally grow with scale, such as quality teams being unable to defend product-release criteria.

The disconnect between early development and commercial requirements can jeopardize promising therapies without early alignment, necessitating costly, time-consuming method redevelopment, comparability studies, and IND revisions. The tension between speed and analytical robustness, in fact, sits at the heart of many Complete Response Letters (CRLs) that have recently delayed some cell therapies from reaching the market.

Investing early in robust analytical methods and a clear CMC roadmap enables teams to anticipate and mitigate risks while allowing for flexibility as knowledge deepens. Generation of the meaningful data needed to enable process characterization can, in fact, only be achieved by investing early in analytical development. This approach also avoids weaknesses in assay qualification, such as a lack of specificity, concerns about reproducibility, and poorly defined acceptance criteria, all of which are frequent sources of regulatory questions and clinical delays. As a result, early investment in robust analytical methods minimizes uncertainty and builds a foundation for scalability and product consistency.

For instance, early development release assays designed to establish identity, purity, potency, and safety are often insufficient to ensure a complete understanding of the product. It is therefore important to invest in characterization assays designed to interrogate additional aspects of the product. These assays include assessments of phenotype, metabolomics, transcriptomics, additional potency assays, and other product-specific assays.

These assays do not have specifications and, in addition to being used to increase product knowledge, may be employed to introduce new technology or testing techniques, confirm observations via parallel testing routes, or address stability concerns or other product-specific questions. They may eventually replace release tests as they are developed over time and elevated from characterization to release tests once data are compiled and specifications can be determined.

Indeed, refining analytical control strategies by tightening specifications and elevating critical characterization assays to release tests as data accumulates ensures consistent product quality. Specific actions include:

• Tightening test specifications as data on method performance and manufacturing history grows.
• Elevating informative characterization tests to release tests.
• Establishing reference standards for commercial production.
• Optimizing methods for the long-term analytical control strategy.
• Developing a coherent potency assurance strategy in line with FDA guidance.
• Preparing for method validation, conducted per ICH guidelines (e.g., ICH Q2R2).

Science-First Cell Therapy Development

Risk management in cell therapy development is about avoiding setbacks and creating the best possible foundation for success. A well-thought-out analytical strategy is one essential component for meeting rapidly evolving regulatory expectations and ensuring programs advance to commercialization. By leveraging early investment in analytics alongside a DoE approach, stage gates, and expert resources, innovators can confidently navigate the complexities of process and analytical method development and deliver high-quality, transformative therapies that improve patient outcomes globally. Working collaboratively, trusted contract development and manufacturing organization (CDMO) partners can support these efforts by reducing risk and creating a smoother path to commercialization.

Kincell Bio applies a science-first approach to analytical development, helping cell therapy developers establish robust testing strategies that support process understanding, product characterization, and regulatory readiness throughout clinical development. Our early-stage analytical strategies include:

• System-suitability controls and method qualification, avoiding future assay redesign and method comparability studies due to unreliable data generated from suboptimal analytical methods.
• Emphasis on precision and accuracy, particularly for methods such as cell count and viability, as this foundation supports dose escalation studies. 
• Building potency and characterization matrices aligned with the mechanism of action to connect quality with biology, account for variability, support comparability, and correlate with outcomes, guiding development decisions and building regulatory confidence for IND submissions.
• For allogeneic programs, donor qualification and correlation of donor attributes with potency and clinical outcomes.
• Understanding of the uniqueness of each cell therapy and the implications for model fitting/assay development.

By investing early in analytical development, Kincell Bio establishes testing solutions that support enhanced process and product understanding, enabling cell therapy programs to be guided throughout their entire life cycles. Using a DoE approach accelerates the development of optimal, robust assays, saving time, materials, and money while supporting improved decision-making. Using a stage-gate approach, meanwhile, allows for phase-appropriate analytical control, thereby mitigating quality and compliance risks as cell therapy programs move toward early clinical validation. Indeed, the analytical methods we develop enable accurate, precise assessment of CPPs and product CQAs, and they evolve with our clients’ programs, streamlining the development process while ensuring the safety, potency, and consistency of their drug substances and drug products.

Roger Herr leads analytical development at Kincell Bio, supporting scalable assay strategies for cell therapy programs from early development through commercialization. He previously held leadership roles at several biologics-focused organizations and has extensive experience in analytical development and assay design.

Patrick Kellish, Ph.D. is a Senior Scientist in Analytical Development at Kincell Bio, supporting analytical activities across cell therapy development and manufacturing. His work focuses on assays for cell identity, potency, safety, and product characterization.

Integrated formulation development services are particularly valuable for drug developers because they connect early formulation screening with downstream manufacturing and regulatory strategy. Experienced CDMOs combine analytical expertise, advanced stability and compatibility testing, and scalable process knowledge to identify drug product formulations that remain stable throughout processing, storage, and delivery. By evaluating factors such as excipient selection, container-closure systems, and manufacturability early in development, integrated formulation development support can reduce the risk of late-stage reformulation and accelerate timelines. 

Guided by widely adopted ICH guidelines, the implementation of Quality by Design (QbD) principles across the product lifecycle has become a foundational framework in modern pharmaceutical development. QbD is grounded in the idea that increased testing of a final product is an insufficient and inefficient means of enhancing product quality; instead, high quality should be built into a product at every step. This approach is routinely integrated into process development, analytical procedure design, and formulation composition, fostering a comprehensive understanding of manufacturing processes, analytical methods, and product formulations. Adopting QbD principles in formulation development workflows enables developers to systematically assess and understand how various attributes impact properties of the resulting product and guide formulation. 

Ultimately, formulation development is not a discrete phase but a continuous, knowledge-driven process that underpins product quality and performance throughout its lifecycle. This article explores the advantages of integrated, QbD-guided formulation development services and outlines key strategies for minimizing risk and maximizing success in biologic formulation development.

Beginning with the End in Mind

Establishing a Quality Target Product Profile (QTPP), derived from the program’s TPP, is a foundational step that anchors formulation development within a QbD framework. Rather than approaching formulation as a series of isolated experiments, QbD encourages developers to define success upfront: what the final drug product must achieve to meet patient, clinical, and commercial needs. The QTPP captures these expectations, outlining key characteristics such as intended indication, route of administration, dosage form, strength, stability requirements, and storage conditions. It also reflects practical considerations, including the target patient population (e.g., pediatric vs. adult), dosing frequency, and usability factors like injection volume and delivery device compatibility.

By beginning with this end-state vision, formulation scientists can make more informed and efficient decisions early in development. For biologics in particular, where molecular complexity and sensitivity introduce additional constraints, aligning formulation strategies with the QTPP helps ensure that stability, delivery, and manufacturability challenges are addressed proactively rather than reactively. From the QTPP, developers systematically identify critical quality attributes (CQAs), the physical, chemical, biological, or microbiological properties that must be controlled within defined limits to ensure product quality, safety, and efficacy. Examples of CQAs in biologic formulations may include aggregation levels, potency, purity, viscosity, and sterility. These attributes become the focal point of formulation development, guiding both experimental design and risk assessment.

Importantly, CQAs are not defined in isolation. They are directly linked to both the inherent properties of the drug substance and the intended clinical performance of the drug product. This connection enables a science- and risk-based approach to formulation design, where developers can prioritize which variables—such as pH, buffer composition, excipient type, or container-closure system—require tight control. In turn, this structured understanding lays the groundwork for identifying critical material attributes (CMAs) and critical process parameters (CPPs) in later stages of development.

Within an integrated formulation development model, the QTPP and its derived CQAs also serve as a unifying framework across functions. Analytical development, process development, and regulatory strategy can all align around the same product goals, reducing silos and enabling more efficient knowledge sharing. This alignment is particularly valuable when working with CDMOs, where cross-disciplinary expertise can be leveraged early to anticipate scale-up challenges, ensure analytical methods are fit for purpose, and support a smoother path to regulatory submission. 

Key Pillars of Formulation Development

Formulation development depends on a thorough evaluation of how drug substances and excipients interact across time and various conditions to create a product of reliable quality and performance. The strategic advantage of QbD lies in its emphasis on risk-based decision-making throughout formulation development. This is achieved by establishing statistically grounded relationships in which formulation attributes, including CMAs, directly influence product CQAs or other excipient performance-related properties as dependent variables. By embedding quality into the design process from the outset, QbD facilitates regulatory flexibility and ensures consistent product performance. Similarly, QbD principles extend to the analytical procedure lifecycle, supporting method robustness and continuous improvement through lifecycle management.

In formulation development, the drug product comprises both the biologic drug substance and excipients. While excipients generally conform to compendial standards, they can significantly impact material attributes and overall product performance. When an excipient is identified as having CMAs, its acceptance criteria should be clearly defined and incorporated into the control strategy to ensure consistent product quality. In alignment with QbD principles, the International Pharmaceutical Excipients Council Federation emphasizes the importance of effective qualification strategies to manage variability during product formulation development. 

Although some level of inherent excipient variability is expected, the true extent is underestimated due to the use of composite or average results in Certificate of Analyses provided by excipient suppliers. This variability can have a detrimental impact on drug product robustness and performance. When considering excipients for formulation development, it’s beneficial to identify potential CMAs related to specific performance or functionality requirements, especially those characteristics not typically controlled or specified by suppliers.

Risk Assessment: Laying the Groundwork

With a clear understanding of the QTPP and associated CQAs, early risk assessment becomes a critical next step in guiding formulation development. At this stage, the emphasis is on speed and operational efficiency, with the goal of quickly identifying the variables most likely to impact product quality so that resources can be focused where they matter most. Rather than relying on trial-and-error experimentation, a structured, risk-based approach enables formulation scientists to prioritize key factors such as excipient compatibility, pH sensitivity, degradation pathways, and processing stresses that could affect stability, purity profiles, and overall performance.

One widely used tool in this phase is the Ishikawa, or fishbone, diagram, which helps systematically map the relationships between potential input variables and desired product outcomes. In this framework, the target attribute, such as a specific stability profile or key performance metrics, is positioned at the “head” of the fish, while major categories of contributing factors, including formulation components, process conditions, container systems, and environmental factors, form the primary “bones.” From there, additional branching reveals deeper, underlying causes within each category, enabling a more granular understanding of risk drivers. When applied to excipient selection, for example, this approach can highlight how factors like raw material variability, interaction with the drug substance, or impact on viscosity and aggregation may ultimately influence CQAs.

Complementing this qualitative risk mapping, high-throughput analytical and screening tools play an essential role in rapidly generating data to support early decision-making. Techniques such as accelerated stability studies, forced degradation, and miniaturized (scale-down) formulation screens can provide early insight into stability trends and functional performance across a wide design space. Together, these approaches help narrow down viable formulation prototypes and define a risk-informed scope for subsequent optimization studies. By front-loading risk assessment in this way, developers can design more focused and efficient experiments that build directly on early learnings, reducing development timelines while increasing confidence in the selected formulation strategy.

Completing the Picture

Following early risk assessment and preliminary screening, more in-depth Design of Experiment (DoE) studies provide a structured and data-driven approach to refining formulation and process variables. Building on the initial identification of high-risk factors, DoE enables systematic evaluation of how multiple inputs interact to influence CQAs. Rather than assessing one variable at a time, multifactorial experimental designs allow developers to uncover both main effects and complex interactions between formulation and process variables such as buffer composition, excipient concentration, and processing conditions. This approach improves efficiency while generating statistically robust datasets that deepen both formulation and process understanding. Within an integrated development model, these studies are closely aligned with analytical development and process scale-up considerations, ensuring that insights generated at the bench translate effectively to manufacturing and long-term product performance.

The insights gained from DoE studies form the foundation for defining a design space, a multidimensional region of input variables that has been demonstrated to consistently yield product meeting predefined quality criteria. Operating within this design space provides a high degree of assurance that product quality will be maintained, even in the presence of normal variability. Regulatory agencies recognize design space as part of a QbD element, allowing for greater flexibility in making adjustments within this established design space without the need for additional regulatory submissions. In the context of lifecycle management, this flexibility is particularly valuable. As products progress from early clinical development through commercialization and potential post-approval changes, a well-characterized design space enables more efficient scale-up, technology transfer, and continuous improvement without compromising quality or compliance.

As datasets grow in size and complexity, advanced data analytics and AI/ML tools are increasingly valuable in augmenting traditional DoE approaches. Machine learning algorithms can identify nonlinear relationships and higher-order interactions that may not be readily apparent through conventional statistical methods. These tools also support predictive modeling, enabling developers to simulate formulation performance across a broader parameter space and prioritize the most promising experimental conditions. When embedded within an integrated formulation development strategy, AI/ML can connect data generated across stages of development, from early screening through commercial manufacturing, to support adaptive learning and ongoing optimization.

Building a Foundation for Quality

By clearly defining what “quality” looks like from the outset and translating it into measurable CQAs, developers can apply QbD principles in a structured and proactive way—ultimately reducing development risk, minimizing late-stage changes, and accelerating the path to a robust, patient-ready biologic product. An experienced CDMO partner plays a critical role in realizing the full value of integrated, QbD-guided formulation development. CDMOs help ensure that decisions made early in development are informed by downstream considerations by bringing together multidisciplinary expertise in formulation science, analytical development, process engineering, and regulatory strategy. This holistic perspective reduces the likelihood of costly reformulation or process changes later in the lifecycle. In addition, established CDMOs offer access to advanced analytical platforms, high-throughput screening capabilities, and scalable manufacturing infrastructure, enabling seamless progression from early-stage development through clinical and commercial production. Their experience navigating regulatory expectations further supports the development of robust control strategies and well-documented design spaces, positioning programs for smoother approvals and long-term compliance.     

When guided by QbD principles and supported by integrated development strategies, formulation efforts can proactively address risk, enable flexibility, and support ongoing optimization as products evolve. In the complex landscape of biologics, this approach is essential for balancing speed to clinic with long-term robustness and scalability. By investing in thoughtful formulation design and development partnerships, drug developers can enhance the likelihood of delivering safe, effective, and commercially viable therapies.

Eliza Yeung has more than 25 years of experience in therapeutic drug development, spanning early discovery through late-stage development, across pharmaceutical, contract research, and CDMO organizations, including Novazyme, Genzyme, Analytical Research Laboratory, and, most recently, Cytovance Biologics.

The good news is that a shift is underway. Regulators, sponsors, and experienced CDMOs are increasingly recognizing that pragmatic, risk-based quality management isn’t a compromise—it’s a more sophisticated and ultimately more proactive approach than the compliance-theater model it replaces.

The Cost of Perfectionism

One of the most common patterns we observe in outsourced operations is what we call “empire building.” Rather than designing quality systems that are lean, efficient, and genuinely protective, organizations sometimes create complexity that sounds good from a podium but whose systems are so intricate that only their architects can navigate them. This provides a kind of job security and a perception of control, but it delivers very little return on investment and can actively impede manufacturing agility.

The zero-risk mentality also shows up in regulatory filings. Young companies, in particular, are understandably anxious to demonstrate rigor to agencies and often over-describe their manufacturing processes. What feels like thoroughness in the filing stage can later become an operational straitjacket. Locking in specific materials, equipment configurations, or procedural details at early development phases constrains the flexibility needed as a process scales toward commercial manufacturing, often without getting alignment from their CDMO. The big lesson, hard-learned by many, is not to be overly descriptive in filings—details that seem like due diligence at Phase 1 can require costly amendments at Phase 3.

Organizational dynamics in outsourced models exacerbate this tendency. With sponsors frequently operating as lean virtual companies—sometimes as few as 4 people managing a network of CDMOs and contract laboratories—the knowledge gap between sponsors and manufacturers can be significant. Bench scientists accustomed to the flexibility of a laboratory environment sometimes struggle to adapt to the discipline of commercial manufacturing, treating the manufacturing floor as an extension of their lab and making constant mid-process changes. The documentation burden this creates—and the deviations and investigations that follow—impose real costs on CDMOs that are already operating on tight margins, as well as delays in deliverables expected by the sponsor.

A Better Framework: Risk-Based Quality

The alternative is not less diligence—it is more strategically allocated diligence. A risk-based approach asks three fundamental questions for every quality activity: What is the return on doing this now? What is the patient and regulatory risk at this stage of the process? And what is the worst outcome if it is not done at this time? That framework focuses resources on the issues that genuinely matter to product quality and patient safety, while resisting the pull toward perfecting things that have minimal impact.

Practically, this means distinguishing between quality activities that must remain tightly controlled in-house and those that can be efficiently outsourced to specialists. Core QA functions—batch release, deviation management, change control—should remain internal to the manufacturing site. These activities require access to real-time operational information and site-specific institutional knowledge that cannot be effectively managed remotely. Document control, from SOPs to batch records, similarly demands on-site, easily controlled availability.

At the same time, certain QA activities are well-suited to outsourcing. Supplier audits, for instance, can be effectively and efficiently handled by qualified consultants or third parties. For QC testing, the calculus is more nuanced. Routine in-process control testing—endotoxin assays, identity tests, the analyses needed to progress a batch—should generally remain in-house for speed and control. Complex characterization and final release testing, however, can and often should be handled by specialist contract testing laboratories that have invested deeply in particular analytical capabilities.

The critical caveat when outsourcing any testing function is time. The moment you hand a sample to an external laboratory, you lose direct control of your timeline. Without robust service-level agreements that specify turnaround commitments and escalation pathways, you may find yourself waiting on results while a batch sits in limbo. That risk must be actively managed.

“A CDMO with excellent manufacturing execution and strong in-process control, paired with a carefully selected external testing partner, may ultimately serve your program better than a facility trying to do everything under one roof.”

Rethinking the One-Stop-Shop Model

The “one-stop-shop” CDMO model, which began evolving in the 2000s and 2010s, is undergoing a quiet reassessment. Comprehensive in-house QC capabilities were once a standard selling point—and for some programs, they remain the right choice. But the true cost of full-service QC is often understated. The visible return from testing services looks attractive; what it doesn’t capture is the ongoing cost of errors, rework, out-of-specification investigations, write-ups, and audits. These hidden costs, combined with the burden of maintaining and retaining specialized laboratory staff, are prompting CDMOs to reconsider where their investments are best deployed.

For sponsors, this evolving landscape means being more intentional about what you actually need from a CDMO versus what looks reassuring in a capabilities presentation. A CDMO with excellent manufacturing execution and strong in-process control, paired with a carefully selected external testing partner, may ultimately serve your program better than a facility trying to do everything under one roof.

The Regulatory Relationship

Perhaps no symptom of zero-risk thinking is more visible than the anxiety that surrounds FDA inspections. A Form 483—the FDA’s notice of inspectional observations—is widely treated as a harbinger of catastrophe. It should not be. The agency reviews operations across a broad spectrum of the industry, from leading-edge facilities to organizations still running on decades-old systems. Observations on a 483 represent the agency’s informed perspective on where a quality system can improve. Engaging with them constructively, with genuine corrective action rather than defensive minimization, is both the professionally appropriate response and a practical opportunity to modernize systems that may genuinely have become liabilities.

The same spirit should govern the broader sponsor-CDMO-agency relationship. Regulatory agencies are not adversaries to be managed; they share the fundamental goal of ensuring safe and effective products reach patients. Companies that engage early and transparently—bringing regulators into their development thinking rather than presenting them with finished filings—consistently fare better than those who treat regulatory strategy as damage control.

What the Industry’s Own Research Confirms

The tensions we observe in client engagements are documented at scale by industry research. Cytiva’s 2025 Global Biopharma Index—a biennial survey of 1,250 senior biopharma executives across 22 countries, published in October 2025—found that 40% of respondents believe cost-cutting is actively compromising product quality at their organizations, and 36% report that process changes are driven more by financial pressure than quality rationale. The overall industry resilience score fell to 5.96 out of 10 in 2025, down from 6.08 in 2023 and 6.60 in 2021—a trajectory that reflects the cumulative strain of short-term priorities overriding long-term operational discipline.

This data does not indict risk-based quality management—it indicts the absence of it. The organizations driving these metrics are not the ones making principled, quality-rational decisions about where to allocate resources. They are cutting costs without a framework, responding to financial pressure rather than quality rationale. That is precisely the failure mode a risk-based approach is designed to prevent. Bureaucratic compliance theater and undisciplined cost-cutting are two sides of the same coin: both substitute the appearance of rigor for its substance.

The regulatory response to that declining performance has been direct. FDA’s FY2024 quality report documented 989 drug inspections—a 27% year-over-year increase—with 561 Form 483s issued to drug program firms and over 62% of inspections targeting foreign manufacturing sites, the highest foreign-inspection share on record. In the first half of FY2025 alone, warning letters increased 73% compared to the same period in FY2024. And in June 2025, the FDA launched an internal AI system called ‘Elsa’ that analyzes 483 observation histories, adverse events, and CAPA records to prioritize high-risk facilities for inspection algorithmically. Quality systems that exist primarily on paper will increasingly find themselves at the top of that list.

The ISPE Global Pharmaceutical Innovation Survey adds a further dimension: 48% of respondents identified regulatory challenges as the most significant or significantly greater barrier to developing innovative manufacturing technology. The irony is that notable organizations that over-specify their regulatory filings and build bureaucratic quality systems around compliance theater find themselves less equipped, not more, to respond to the agency’s actual expectations for quality maturity and continuous improvement.

Raising the Bar on Human Factors

People ultimately execute quality systems—and that human dimension is too often an afterthought in quality design. Manufacturing is behaviorally driven. Operators must take steps, attach connections, follow sequences, and make judgment calls under time pressure. A quality system that ignores that reality, that piles procedures on top of procedures without asking whether they are operationally feasible, will fail not through any gap in documentation but through the accumulated weight of the unworkable.

The most effective quality cultures we have observed treat their manufacturing teams as partners in quality—training them to understand why procedures exist, not just how to follow them, and building feedback mechanisms that surface operational friction before it becomes a compliance problem.

The goal, ultimately, is not 100% on every dimension. The real objective is conformance to requirements, with genuine criticality placed where it belongs—on the things that protect patients and product integrity. That is a higher bar than it sounds. And it is a far more sophisticated ambition than simply doing more of everything.

Lisa Cozza and Scott Myers are principals at Tunnell Consulting, specializing in quality management, regulatory compliance, and outsourced manufacturing strategy for biopharmaceutical companies.

NIST Technical Note 1297, 1994 edition¹ references the combined standard uncertainty of measurement U_c which combines the individual standard uncertainties whether arising from randomized or systematic (non-random) sources (within the NIST Technical note this is referred to as the law of propagation of uncertainty). Taking such an approach the relationship between uncertainty of measurement and standard error of mean and bias uncertainty for an analytical method can be represented as follows:

Standard error of the mean (SEM) = StdDev (SD) for Sample Measurement / SQRT n (where n = sample size)

Standard combined uncertainty U_c = SQRT (SEM² + b²) where b is the uncertainty associated with bias measurement. 

Expanded uncertainty U = k U_c where k refers to a coverage factor  

When considering the above, it should be recognized that commonly the uncertainty of measurement is predominately driven by the SEM. As for the bias, it is only the uncertainty associated with the bias measurement that contributes U_c (if the accounts for any systematic, bias effect) If the method does not account for the bias, then U will need to consider the bias value along with the associated uncertainty of bias measurement. 

If one attempted to reduce the SEM by increasing the sample size (n) then the natural question is how many preparations/measurements should be made for my analytical method? The answer to such a question should be driven by the requirements of the analytical method which can be defined within the analytical methods target profile (ATP). A key component of the ATP is the uncertainty of measurement (U) where the ATP should define the requirements of U (which will consider the specification that the sample is being tested against). 

For the bias component of (U) this should be ascertained and quantified through method development (as per ICHQ14) where Analytical Quality by Design (AQbD) will identify/characterize those method parameters/settings that impact bias along with defining the elements of the Analytical Control Strategy (ACS) to reduce their influence to an acceptable level. Examples of such measures would be optimizing sample preparation conditions to ensure quantitative recovery of the analyte, minimizing matrix influence etc. Analytical development should define the ACS where Analytical Procedure Performance Qualification (APPQ) will confirm the ACS suitability (by testing the ACS against protocol defined acceptance criteria for the various analytical method attributes) and ongoing suitability of the ACS demonstrated through continuous performance verification (CPV). Analytical development along with APPQ and CPV represents Analytical Lifecycle Management (as per USP<1220>) which also includes change management where any change to an analytical method would require assessment of the potential impact to effectiveness of the ACS (recognizing those elements of the ACS that are established conditions (EC) necessitating regulatory notification for any change). For example, a change to an analytical method’s reference standard lot number would commonly be associated with a bridging/comparability study to assess any impact to the ACS. 

In situations where the ACS has minimized any influence of bias, which could be addressed through a correction within the method (when bias has been demonstrated to be consistent through the method’s reporting range), the expanded measurement of uncertainty primary contributor is the method’s randomized error. Analytical method development would need to identify those impactful method settings/parameters which impact the analytical method’s randomized error and define the associated ACS to minimize their influence. This would include establishing the number of sample preparations and measurements to achieve an acceptable level of SEM along with those other elements of the ACS to control systematic and random errors such as those associated with instrument performance, operator technique, environmental fluctuations and sample preparation variability. An example of an element within an ACS to mitigate the impact of error of measurement associated with instrument performance, sample preparation etc., would be the incorporation of an internal standard into a method. Statistical tools such as ANOVA can be utilized to partition the total variability associated with an analytical method into components, allowing an understanding of what components have the greatest contribution to the pooled sample standard deviation, recognizing that method precision is normally evaluated at three levels: repeatability, intermediate precision and reproducibility.

So, returning to the question of how many sample preparations/measurements are needed for an analytical method is obviously case dependent. Ultimately, the aim is to drive the analytical method’s U to a level that meets the requirements of the ATP, which in turn is driven by the Quality Target Product Profile (QTPP). Increasing sample size (n) is one approach to reduce SEM, however, it must be recognized that there are limitations/practical considerations as you are reducing the SEM by the SQRT of n and that you would also want to reduce the StdDev associated with the Sample Measurement. With that in mind, Analytical Lifecycle Management (ALM) needs to achieve a complete understanding of the analytical method bias and SD influences to ensure the analytical control strategy is targeting those that are most impactful to achieving the ATP requirements. The ALM program needs to be risk based where, via knowledge management, an ACS is afforded which focuses on the most significant method bias/SD influencers. For example, it would be inappropriate to address an unacceptable level of SEM by increasing the number of sample preparations/measurements to an impractical level when a more significant and effective reduction could be afforded by incorporating an internal standard or increasing sample load/volume to reduce the SD. 

A component of ALM is to continually monitor the ongoing effectiveness of the ACS through CPV which can include a periodic holistic assessment of those investigations that were associated with the execution of the test procedure. Such test method investigations can be associated with those that were initiated due to exceeding test method defined criteria (e.g., agreement between multiple sample preparations). Such test method criteria will be established through method development and PPQ, that when exceeded, reflect an atypical outcome necessitating an investigation. In addition to test method defined criteria there will commonly be method system suitability criteria (from the analysis of reference standard and/or system suitability solutions) which together represent a key component of the method’s ACS. Any resulting investigation will include a laboratory component, but for scenarios where the sample agreement criteria was exceeded, a manufacturing component of the investigation could be required (if a laboratory root cause was ruled out indicating a potential manufacturing cause). This will be particularly important if one of the sample preparations is Out of Specification (OOS), which will be discussed in the next paragraph. The goal of such investigations is to determine if the root cause is associated with the respective control strategies: for the analytical method, the ACS, but also the manufacturing control strategy (if it was suspected that the lack of agreement between multiple preparations was deemed reflective of the quality of the sample that was tested). Ultimately, the aim through CPV is to ensure an effective control strategy is in operation that delivers against the requirements of the ATP and QTPP. 

A dilemma that faces a QC laboratory is “if my analytical method requires multiple preparations/multiple measurements (to minimize error of measurement), what action do I take if one of those preparations/results are outside of the specification (OOS) but then the reported result (which is based upon an average) is within specification?” Within the FDA’s May 2022 OOS guidance (Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production),  there is reference to two different scenarios, where the first addresses averaging results from multiple sample preparations from the original submitted analytical sample and then the second scenario addressing averaging results from the same final sample preparation. For the former scenario the OOS guidance states the following:

“….where a series of assay results (intended to produce a single reportable result) are required by the test procedure and some of the individual results are OOS, some are within specification, and all are within the known variability of the method, the passing results are no more likely to represent the true value for the sample than the OOS results.  For this reason, a firm should err on the side of caution and treat the average of these values as an OOS result, even if that average is within specification.”²  

An important point to note with the above excerpt (and as discussed earlier), there needs to be an understanding of the known variability of the method and that this is reflected in the analytical control strategy such that if the results are not within the expected variability, an investigation should still be initiated (even if all are within specification) and that would be addressed via a test method (sample preparation agreement) criterion. The resulting investigation will focus on determining if the cause for exceeding the test method criterion reflects either (or both) an issue with the effectiveness of the analytical or process control strategy. If an OOS result is also associated with exceeding the sample agreement criterion, the investigation should determine if the cause for exceeding the agreement is also associated with the cause for the OOS. 

For the latter scenario the OOS guidance states:

“….an HPLC test method may specify both acceptance criteria for variability and that a single reportable result be determined by averaging the peak response from a number of consecutive, replicate injections from the same test vial.  In these cases, and given the acceptance criteria for variability are met, the result of any individual replicate in and of itself should not cause the reportable result to be OOS.”² 

The above excerpt is stating that the method is averaging the injection responses to derive a reported result for that sample preparation, and that if the effectiveness of the analytical control strategy has been demonstrated, then the reported result is valid. So again, in such situations the analytical control strategy should include a criterion for the expected variability of multiple measurements from the same sample preparation as determined through method development/PPQ. If this is exceeded, then an investigation will be initiated. If the variability criteria were met but one of the injections responses represented an OOS result, then it is highly likely that the reported result is out of trend (OOT) as the U for that result would overlap with the regulatory specification reflecting a false negative/positive risk which would necessitate an investigation. This will be discussed further below.

It is common to ask what is considered an acceptable level for U. Acceptability is normally based on assessing the level of U in relation to the specification range for that analyte and determining the % value of U in relation to that specification range. Commonly a value of 12.5% of the specification range is used as a rule of thumb, but caution should be applied with taking such an approach. 

U = (Q^max – Q^min) / 8 where Q^max and Q^min represent the material specification range

When defining the acceptability of the method’s U, one must reflect on the risk of making an incorrect quality decision from the testing of the manufacturing sample due to the method’s U. As such, when establishing the U in the context of the ATP, there needs to be an assessment of the risk that the analytical method will generate a false positive or negative result during production support. This goes back to the earlier comment of understanding the production process capability. In Figure 1 below, scenarios 2 and 3 represent a false positive/negative risk.³

To assess such a risk, one needs to understand the method’s U, but as discussed earlier, there also needs to be an understanding of the expected capability of the manufacturing process towards that attribute. For those manufacturing processes that are less capable (where the quality of the generated material rides close to the specification), there is more emphasis on developing analytical methods with lower U value to minimize the risk that when testing the manufacturing sample, the generated result’s U value overlaps with the specification (representing a false positive/negative risk). To minimize the risk of making an incorrect quality decision, i.e., incorrect batch disposition decision, companies will commonly implement internal specifications (such as OOT limits that were mentioned earlier) which will prompt further action such as initiating an investigation. This is represented below in Figure 2 as the transition zones.⁴

 It is recognized that QC tests a submitted analytical sample as it is not practical to test the parent batch/lot in its entirety, and as such, one of the primary requirements of the submitted sample is to represent (as closely as possible) the parent batch/lot (for the attributes that are to be tested). With that in mind, there needs to be an understanding of how that attribute is dispersed through the parent batch and thus the ease of obtaining a representative sample for that attribute (and any associated risk with the sampling approach). If it is an attribute where there is a risk of it being heterogeneously distributed, for example, an impurity that is formed during lyophilization, then obtaining a representative sample is more of a challenge versus sampling for a homogeneous attribute. In recognition of that risk, when sampling for a heterogeneous attribute, one can employ stratified sampling along with an increased sample size (versus sampling of a homogenous attribute). Such sampling approach needs to be reflected in the associated analytical method, for example, where one would commonly test the representative sample from each stratified sampling point to assess the uniformity of distribution. This becomes particularly relevant for the testing of PPQ batches where it is recognized that a higher level of sampling commonly occurs (versus routine production), where one of the goals is to establish the capability of the associated process control strategy. After the PPQ batches, it may be possible to justify a lower level of sampling where, for example, testing is limited to the worse case sampling sites, and the focus of that testing is to confirm the ongoing suitability of the process control strategy. A word of caution regarding the testing of composite samples—this needs to be justified with consideration of the nature of the attribute which includes demonstrating that the attribute is homogenously distributed. Obviously, if the sampling plan is to demonstrate UOD (per USP<905>), then this is not congruent with testing a composite. If, through PV, it has been demonstrated that the attribute is homogeneously distributed (and this is also scientifically justified), then the sampling plan will focus on obtaining a representative sample and testing of a composite sample maybe justified. 

Conclusion

Generating quantitative sample test data with an acceptable level of U is a fundamental requirement of the ATP (for a quantitative method) and should be the goal of the method’s ACS where there is an understanding of all the sources of random and non-randomized error (and those controls to mitigate their effect). When defining the acceptability of the method’s U it is recommended that there is consideration for the risk of generating false positive / negative data (whereby there is an awareness of the capability of the associated manufacturing process).

References

1. NIST; Barry N. Taylor and Chris E. Kuyatt; “NIST Technical Note 1297, 1994 Edition, Guidelines for Evaluating and Expressing the Uncertainty of NIST Measurement Results”; September 1994;  https://emtoolbox.nist.gov/Publications/NISTTechnicalNote1297s.pdf

2.   FDA; “Investigating Out-of-Specification (OOS) Test Results for Pharmaceutical Production – Guidance for Industry”; May 2022 (Revision 1); https://www.fda.gov/media/158416/download

3. Eurachem/CITAC Guide; “Use of Uncertainty Information in Compliance Assessment”; Second Edition; 2021; https://www.eurachem.org/images/stories/Guides/pdf/MUC2021_P1_EN.pdf

4. Burgess, Christopher et al; Pharmacopeial Forum; “Fitness for use: Decision rules and target measurement uncertainty”; 42(2); January 2016; https://www.researchgate.net/publication/298822306_Fitness_for_use_Decision_rules_and_target_measurement_uncertainty

Paul Mason, Ph.D., is an Executive Director at Lachman Consultant Services, Inc. with more than 20 years of pharmaceutical industry experience. His background spans Quality Control, Analytical Development, CMC submissions, and scientific support for complex FDA review issues across sterile parenteral, API, and oral solid dosage forms.

Cell Line Considerations

The choice of a producer cell line (mammalian vs. insect) and culture format (adherent vs. suspension) underlies any viral vector production system. Insect cells were initially favored for Adeno-Associated Virus (AAV) production due to scalability, cost-effectiveness and perceived advantages in product quality. More recently, advances in process optimization and successful scale up have addressed concerns around mammalian HEK293 systems, shifting focus back to this cell line for its now-demonstrated flexibility, scalability, robustness and product quality.

For HEK293 systems, several factors drive selection. First, suspension cultures are generally preferred for scalability and compatibility with chemically defined, serum-free media. Second, monoclonal lines with documented single-cell origins are increasingly favored over polyclonal populations for consistent, high-titer production. Third, while HEK293T cells have been used clinically, concerns about the T antigen are shifting preference toward lines without the T antigen.

Cell Banking   

Large-scale cGMP production of HEK293 cell banks is critical but often overlooked. Reliable post-thaw recovery is essential to maintain production timelines, yet HEK293 cells are more sensitive to banking conditions than CHO cells, requiring tailored best practices to reduce risk, especially when making cell banks at larger scales.  Aside from the preparation of the cells themselves, large-scale freezing conditions, like controlled-rate freezing, must be optimized specifically for HEK293 cells, as standard protocols for other cell lines often do not translate effectively to HEK293 cells.  

PRO TIP:  Successful cell banking starts with healthy, log-phase cells grown under well-characterized culture conditions, such as optimal growth vessels, culture volumes, shaking speeds and shaking orbits.  In all instances, cells scaled up for cell bank production should grow identically in the larger-scale vessels as in the smaller shake flasks used for routine cell culture maintenance.  Any increase in doubling times or decrease in viability during the scale up process may lead to inferior cell performance.  Materials compatibility is also key: DMSO (5–10%) is commonly used for cryoprotection, but prolonged exposure at larger scales increases the risk of incompatibility. Glass pipettes should be used for transferring concentrated DMSO and DMSO-compatible containers like polypropylene for freeze media preparation. (can cite if needed: https://tools.thermofisher.com/content/sfs/brochures/D20480.pdf)  

Considerations for Obtaining Cell Density and Viability Assessments Post-Thaw

Accurate viability assessment immediately post-thaw can be challenging at low cell densities. At this stage, certain components found broadly in culture media may interact with trypan blue, forming trypan blue aggregates that may be miscounted as dead cells on some cell counters.  In instances where production batch records require cell viability post-thaw, this simple issue could put an entire production run at risk.   

PRO TIP:  Dilute a small sample of cells directly from the thawed freeze vial into phosphate buffered saline (PBS) solution before counting (e.g., 50 µL of cells from the freeze vial into 450 µL PBS) to prevent trypan blue aggregation and improve accuracy. Once cells recover and reach standard densities, this step is no longer necessary.

Scaling from Shake Flasks to Bioreactors

Overall, production cell lines should grow as well, or better, in large-scale bioreactors as in shake flask scale during the seed train and production phases. 

PRO TIP:  If cell growth is reduced or inconsistent compared to small-scale shake flask cultures, prioritize optimizing core bioreactor conditions like mixing and gassing first before attempting to address the issues with supplements that could affect the upstream or downstream processes in unknown ways, such as anti-clumping agents and surfactants. Optimizing cell health is essential for consistent transfection and high viral vector yield across scales.  

Optimization of Critical Quality Attributes

Using AAV as an example, once cell culture conditions are established, process optimization is key to achieving high yields and product quality. Development starts with DNA design—selecting optimal capsids and genomes, refining promoters, codon usage and minimizing sequence liabilities to ensure strong, safe expression.  Capsid choice is critical, as it determines tissue targeting and efficiency. Beyond natural serotypes, engineered capsids—via rational design, directed evolution and computational methods—expand performance options. After selecting DNA and capsid, production is optimized using design-of-experiments (DOE) to tune the amounts of DNA, transfection conditions and plasmid ratios. 

PRO TIP:  Because maximizing for titer alone can reduce product quality, optimization must balance multiple attributes, including titer, full capsid percentage, collateral DNA packaging and potency. To accelerate AAV development and identify optimal vector design and production conditions, use automated liquid handlers and statistical modeling in scale-down models to run large DOE studies, enabling evaluation of multiple variables simultaneously to define processes that maximize both titer and vector quality.

Plasmid DNA Technologies

High-quality plasmid DNA is foundational to robust viral vector production. Impurities, inaccurate concentration, poor topology or lot-to-lot variability can negatively affect transfection efficiency, yield and final product quality. Alternative DNA technologies such as nanoplasmids, doggybone DNA and minicircles are attracting interest because they may reduce bacterial backbone content, improve expression and lower the risk of unwanted sequence carryover. However, adoption depends on whether these benefits outweigh the added costs and operational and regulatory complexities compared to traditional plasmids. 

Indicators of a Successful Transfection

Successful transfection can be first detected by characteristic changes in cell growth and viability. Within the first 24 hours post-transfection, cells in suspension culture typically undergo one division, facilitating nuclear entry of plasmid DNA. This is often followed by a reduction in growth rate and a decline in viability, driven by the metabolic burden and cellular stress associated with transfection and viral replication processes. 

PRO TIP:  Depending on the system and production time, cell viabilities in the range of ~60–80% are commonly observed at harvest. Conversely, continued rapid cell division with minimal or no loss of cell viability is a strong indicator of poor or failed transfection.

PRO TIP:  A more direct assessment of transfection efficiency can be obtained by monitoring expression of the transfer gene in producer cells, when applicable. This is typically evaluated at early time points like 18 hours post-transfection to distinguish true transgene expression from later auto-transduction effects. Depending on the system, 60–80% of cells may be transgene-positive. However, the percentage of positive cells alone is not sufficient as the overall expression level across the population—such as mean fluorescence intensity (MFI) in fluorescent-based methods—is often a more representative indicator of transfection efficiency.

Considerations for Optimizing Plasmid DNA Complexation at Scale

Successful large-scale transfection requires understanding and careful control of key complexation parameters, including reagent stability, order of addition, mixing behavior and complexation kinetics. While these variables can be tightly managed at a small scale, larger volumes inherently introduce longer addition, incubation and mixing times. Because failures at scale are disproportionately costly, it is essential to assess process robustness by systematically evaluating extended holding times during each step of complex formation.  Small-scale studies should be performed to define the design space of critical steps—such as reagent dilution and DNA complexation.  

Mixing is a key consideration and becomes increasingly complex at larger volumes. Insufficient mixing can result in incomplete or heterogeneous complex formation, leading to variability in particle size and charge. Conversely, excessive mixing can either inhibit or accelerate complexation kinetics (depending on the reagent used), producing particles that are less efficiently taken up by cells. 

PRO TIP: Mock large-scale complexations can be performed using dyes to visualize mixing efficiency and identify any unexpected flow patterns that lead to incomplete plasmid DNA complexation. 

Scalable Purification: Large Scale AAV Purification

AAV purification involves seven key steps: harvest, clarification, concentration, affinity purification, full capsid enrichment, buffer exchange and fill–finish. During harvest, cells are lysed and treated with nucleases to release AAV particles and reduce residual DNA. The lysate is clarified to remove debris, typically via centrifugation or depth filtration, both of which are scalable through continuous or staged systems.

A concentration step is often performed before affinity capture to improve efficiency, given the high binding capacity of AAV resins. Tangential flow filtration (TFF) is widely used due to its scalability and recovery. Affinity chromatography serves as a robust primary purification step, with columns scalable across a range of sizes, while membrane-based methods offer faster processing but limited scalability.

Empty and full capsids are typically separated using chromatography, especially anion exchange resins, membranes or monoliths. Resins provide the greatest scalability, whereas membranes and monoliths enable higher flow rates but have format constraints. Finally, TFF enables buffer exchange and concentration, followed by sterile filtration and aseptic filling.

PRO TIP: The pre-affinity concentration step is mainly used to reduce loading time rather than being strictly required. With high-flow affinity resins and membrane-based capture technologies, clarified harvest can often be loaded directly onto the affinity step. However, feasibility should be evaluated based on feed volume, impurity load and capacity to ensure consistent performance at scale.

Large Scale Lentiviral Vector (LV) Purification

LV purification is more challenging than AAV, with approaches including ion exchange, size exclusion and affinity methods. Due to the vector’s fragility, some processes rely on minimal purification with clarification and concentration. Unlike AAV, lysis is not required since LV particles are secreted into the supernatant, though nuclease treatment is used to reduce residual DNA.

Clarification removes cells and debris, typically via depth filtration, sometimes preceded by low-speed centrifugation. Depth filters are favored for scalability and gentle handling. Because LV titers are relatively low, a concentration step is often needed, with TFF commonly used; however, shear and membrane conditions must be optimized to preserve infectivity.

When included, chromatography typically uses membrane- or monolith-based ion exchange methods, as high flow rates reduce the risk of irreversible binding and may eliminate the need for prior concentration. Size-exclusion chromatography can also be used, while affinity methods are emerging but not yet widely adopted. Buffer exchange and formulation are usually performed with TFF, followed by sterile filtration and aseptic filling with attention to filter compatibility.

PRO TIP: Due to LV’s large size and envelope sensitivity, measurable losses can occur during sterile filtration. In addition, LV is prone to degradation and loss of infectivity during extended hold times. To mitigate these losses, processes should be designed for continuous operation where possible, minimizing pauses or overnight holds. Sterile filtration is best limited to a single final step, with the addition of a large pore size prefilter to preserve vector yield and potency.

Analytics

Viral vector process development requires robust analytical assays to measure key quality attributes. For AAV, genome titer is assessed by quantitative PCR (qPCR), digital/droplet digital PCR (d/ddPCR), with d/ddPCR offering higher precision and qPCR enabling higher throughput. Empty-to-full capsid ratios are increasingly measured by mass photometry, while impurities like residual DNA are analyzed using qPCR or d/ddPCR or next-generation sequencing. Additional methods, like CE-SDS, LC-MS and SEC, assess capsid identity, purity and aggregation, alongside cell-based assays for functional infectivity, though these do not fully predict in vivo potency.

PRO TIP:  AAV qPCR or d/ddPCR titer assays typically include a nuclease step to remove non-encapsidated DNA. When working with crude lysates, media components can inhibit nuclease activity, leading to incomplete digestion and artificially inflated titers. To improve accuracy, dilute lysates (e.g., 1:10) in PBS + 0.001% Pluronic F-68 before the nuclease step to ensure more complete removal of plasmid DNA.

For LV, process development focuses on maximizing yield while preserving functional titer and minimizing impurities. As with AAV, consistent product quality depends on monitoring multiple attributes using robust analytical and cell-based assays. Historically, the predominance of ex vivo applications reduced pressure to develop highly discriminating assays. However, as in vivo use expands, improved analytics are needed to distinguish functional lentiviral particles from extracellular vesicles and enable more precise product characterization.

Viral vector manufacturing continues to evolve as the industry moves toward commercial scale. Consistent, high-quality production requires integrated process design, robust scale-up strategies and strong analytical control. Advances in automation, high-throughput tools and data-driven strategies are improving efficiency and reproducibility. Continued innovation will be essential to reduce costs and expand patient access, enabling scalable production of next-generation therapies.

Jonathan Zmuda, Ph.D. is a Senior Director of Cell Biology R&D at Thermo Fisher Scientific located in Frederick, MD (USA).  Jon leads a team of scientists dedicated to developing new products for protein expression, viral vector production and nucleic acid delivery.  Dr. Zmuda received his Ph.D. in Cell Biology from the University of Maryland, College Park and his BSc degree from Dickinson College in Carlisle, PA.

With contributions from Arjen Van den Berg, Emily Jackson-Holmes, Kenneth Thompson, and Nils Williston.

Yet as companies with packaging interests set short- and long-term sustainability goals, they are finding secondary packaging provides a wide pathway to those objectives.

In the following piece are the unique perspectives of three industry veterans. Peter Belden is Chief Commercial Officer at Tjoapack. Just recently, Tjoapack was acquired by Alcami.

Adding to Belden’s thoughts is John Jansen, Head of Global Marketing for CurTec. Jansen authored a previous piece for Contract Pharma, “The Crucial Role of Secondary Packaging in Protecting Product Integrity,” in April 2026.

Also joining the discussion is Brooke Marshall, Vice President of Business Development for Pyramid Pharma Services. Most recently, Pyramid and Phio Pharmaceuticals entered a drug product manufacturing agreement centered around skin cancer treatment.

After identifying these subject matter experts, Contract Pharma posed three essential questions. The first gathered perspective on how long secondary packaging has played second banana to primary, and if both forms are adopting sustainable practices at the same rate. The second asked to identify the main challenges related to scalability, and whether sustainability and reproducibility are compatible.

Finally, we asked for a look ahead. Many companies have set sustainability goals whose anticipated fulfillment is fast approaching in the year 2030. Other goals have a target of the year 2050. By projecting out current trends by years, even decades, a picture of secondary packaging’s role in sustainable processes becomes complete.

Secondary Packaging in the Driver’s Seat

The experts agree that primary packaging has historically received the lion’s share of attention when it comes to sustainability. The reasons for this are obvious: its direct impacts on product and patient safety, as well as regulatory compliance. In contrast, Belden notes secondary packaging has often been pigeonholed as a later-stage opportunity. Jansen goes a step further: He calls it a “missed opportunity.”

For example, Jansen says, CurTec’s ECO LITE drums are made from a 30–70% blend of plant- and fossil-based plastics. They are also 100% recyclable. Their simplified, fully circular design leads the charge in secondary packaging sustainability.

“While primary packaging faces slower adaptation due to rigorous stability testing, secondary packaging offers a quick win for CO₂ reduction,” Jansen says.

For this reason and many others, secondary packaging is increasingly a major vector for sustainable practices. Our sources cite its advantages in reducing material use and waste as well as improving recyclability and logistics. Also, they say, secondary packaging can engender smarter solutions along the supply chain, and in traceability. Plus, comparatively lower risk exists versus primary packaging.

“This shift is also driven by practicality,” Marshall says. “Secondary packaging often offers a more accessible path to achieving near-term sustainability gains. Changes can typically be implemented with fewer regulatory hurdles compared to primary packaging.”

According to Marshall, this allows organizations to pilot innovations, adopt standardized formats, and scale improvements more quickly.

“Early adoption of standardization in materials, formats, and labeling can help companies avoid more complex, costly transitions later,” Marshall says.

Notably, say the experts, both primary and secondary packaging are increasingly moving toward smarter, more sustainable processes. But they do find that this is not happening at equal rates.

“Primary packaging remains more constrained by material and regulatory requirements,” Belden says. “Secondary packaging has generally moved faster in adopting smarter and more sustainable processes.”

As Marshall puts it, this makes secondary packaging an increasingly strategic focus for organizations looking to delivery measurable sustainability progress.

Roadblocks in Scalability and Reproducibility

However, Marshall adds, scalability presents a unique challenge. Sustainability and reproducibility are not competing priorities, necessarily. However, their designs should have intention to scale together.

This requires concentration on standardization, validation, and support from reliable supply chains, according to Belden.

“The main challenge is scaling across different SKUs, formats, batch sizes, and market requirements without adding complexity or compromising control,” Belden says.

That added complexity can include such factors as cost pressures and supplier capabilities, Marshall says, to equipment compatibility and evolving compliance requirements. Pilot programs can demonstrate value effectively, Marshall contends, but they only go so far.

“Platform-based approaches, aligned specifications, and strong supplier partnerships are essential to ensure consistency,” Marshall says. “Without them, variability can increase and undermine both efficiency and sustainability goals.”

Other risks are never too far away, either. This particularly applies to the supply chain. Here, Marshall says early introduction of risk assessments is the key to defining and maintaining reproducibility.

“Proactively identifying potential sources of variability, whether in materials, processes, or logistics, helps avoid retrofitted solutions later,” Marshall says. This is especially true when dealing with globally distributed materials, and in prevention of waste and other inefficiencies.

Belden concurs.

“Sustainability helps reproducibility when it is built into the operating model,” Belden argues. “But it becomes a risk when material choices or regional waste systems are not yet mature.”

As for CurTec, Jansen says its line of drums was in part the product of an investment in hybrid manufacturing equipment. In other words, this technology allows for precise control of material distribution during blow molding—ultimately guarding against weak spots.

“Reducing weight without compromising on performance and safety is a real challenge in heavily regulated industries. So we’re very proud to have found a way,” Jansen says. “Every unit meets the same high standards for strength and safety.”

“Smart solutions must fit within highly regulated environments and remain efficient at commercial scale,” Belden concludes. “Process consistency, modular automation, and well-qualified materials are essential.”

Too Early to Look Ahead?

It is now the second quarter of 2026. By the end of the third quarter, we will be in the final third of the 2020s. That is absolutely not too early, the experts say, to realistically evaluate if sustainability goals set for 2030 are attainable.

“If every company in the sector really challenges their R&D departments to meet 2030 Packaging and Packaging Waste Regulation requirements, not in the future but right now, as we did at CurTec, then there should be no problem reaching these sustainability goals,” says Jansen.

Progress is evident in several areas, notably alternative materials and material optimization, but also automation and incorporating smart features. Jansen mentioned CurTec’s exploration of integrating recycled content alongside bio-based and virgin feedstocks.

But while these endeavors may set the industry on a short-term course for 2030 goals, 2050 still feels far away. Progress up to now, Marshall says, has been uneven. Regional variability is one factor to overcome.

What will fundamentally determine the achievability of long-term goals, according to Marshall, are the decisions being made in the present.

“Choices related to material types, recyclability pathways, and compatibility with existing and future infrastructure will either enable or constrain future progress,” Marshall says. “As a result, early alignment on sustainable materials, supported by standardization and scalable design principles, is critical.”

Belden noted that change, however quickly it is sought, tends to happen gradually.

“Every shift in materials or processes must be validated and implemented carefully,” Belden states. “It is possible to assess momentum today, but long-term impact will depend on how consistently the industry can scale proven solutions over time.”

Latest News Roundup

All of this expert insight isn’t just talk. Innovations and investments impacting packaging continue to back up prevailing opinions on industry trends.

In March 2026, Colbert Packaging announced the installation of new die-cutting equipment at its Kenosha, Wisc. facility. The additional die cutter was expected to increase Colbert’s efficiency in folding carton production capabilities.

The same month, Schreiner MediPharm said it has developed a transparent deep-freeze seal. A tear in this film is irreversible, and—depending on cardboard material—may also cause folding box fibers to tear. Future addition of security features is an option for manufacturers.

Meanwhile, at INTERPHEX 2026 in New York City in April, Verista showcased its COUNTQ and KITQ integrated offerings. The company said these are designed to increase control in component counting and reconciliation processes in packaging workflows. This not only reduces manual effort, but also errors.

Similarly, IL Group took the occasion of the trade show to launch its Single Minute Changeover Labeler Series Automatic Labelers. IL Group said these are designed to integrate seamlessly into modern pharmaceutical packaging lines. This system’s consistency and repeatability align with the greater smart tech and sustainability objectives of this industry sector.

Conclusion

As artificial intelligence and automation continue to evolve in the pharmaceutical industry, smart technologies are increasingly in fashion. At the same time, companies are assessing both their short- and long-term sustainability objectives. While mid-century goals may not be easily forecast, the race to a more sustainable decade in the 2030s is on.

Pharmaceutical packaging can be a pathway to helping these companies practice what they preach in terms of sustainability. However, primary packaging carries inherent roadblocks due to the exacting nature of product and patient safety, plus regulatory compliance.

By contrast, secondary packaging, while still being held to stringent standards, is one more layer removed from a drug product. The materials that comprise this facet of the packaging process can lend themselves more easily to recyclability.

How to amplify secondary packaging as a beacon of sustainability—while incorporating technologies eliminating human error—is now the question. Its sustainable potential exceeds that of primary packaging, even though the latter has traditionally garnered more attention.

Companies’ abilities to adapt may well determine where their goals stand in 2030, 2050, and beyond.