You’ve developed a candidate viral vaccine or viral vector product, and initial preclinical data are showing excellent efficacy. Now you’re ready to scale up for IND application and testing in human subjects…or are you? Whether you are developing a viral vaccine, cancer vaccine, oncolytic virus, gene therapy or other viral vector application, the choices you make at this critical stage can greatly influence the success or failure of your project. In this article, we explore common pain points and some of the most dangerous pitfalls that process development teams encounter when scaling viral vector and vaccine production for clinical trials and beyond.
Viral vaccine and viral vector manufacturing: the high price of failure
The viral vaccine and viral vector industry is notoriously challenging, with long candidate development times and high attrition rates. Compared to traditional pharmaceuticals and recombinant protein therapeutics, the added complexity of viral biology compounds the difficulty of developing a well-characterized and robust manufacturing process.
Historically, developing and licensing a vaccine takes from 10-14 years, with only 6% of candidates progressing from the preclinical phase to market [1-3]. Given that the average cost of moving a single vaccine candidate through to the end of phase 2a clinical trials is between $31m and $68m, the price of failure is high . In many cases, the underlying causes of this high attrition can be traced directly or indirectly to decisions made in the early development phases.
Dialing up production from lab-scale to clinical trial levels may seem like a straightforward exercise, but in reality the process can be complex, time-consuming, and expensive. Unexpected problems introduced by poor design choices, process changes, and unpredictable biology can seriously delay your project or derail it altogether.
So, what’s the good news? With careful planning, informed choices, and intelligent design from the very beginning, you can de-risk your project and possibly even accelerate it in the process. But before we get into that, let’s dig a little deeper into why development and scale-up of these types of products are so risky.
Why is scale-up of viral vectors so challenging?
Biological entities like viruses are inherently complex and difficult to control. Rather than consisting of a single, well-defined chemical or biomolecule, viral particles are relatively large multi-component structures. This means that compared to other therapeutics like recombinant proteins, for example, the production process is often more complicated, and there are many more elements that need to be characterized, optimized, and controlled to ensure the structural and functional integrity of the final product.
During upstream processing (USP) for viral vector products, the vector, production cell line, and culture system together comprise an even more complex system, with many interdependencies that need to be considered holistically. In order to optimize for safety and productivity, as well as to ensure that critical parameters are maintained during scale-up and clinical production, this whole system needs to be thoroughly studied and understood.
Any changes to viral seeds or cell banks, raw materials, culture parameters or other upstream processing steps can have a profound impact on the downstream process. It’s therefore essential to develop a robust model of your process, so that you can optimize and scale-up your process in a controlled manner.
Top 5 pitfalls in scale-up for viral vector manufacturing
The transition from the initial lab-scale process to a final commercial process needs to be planned carefully from the beginning to avoid surprises later on, after significantly more time and money has been invested in development.
With this in mind, here are some of the most dangerous rocks to avoid as you navigate the treacherous waters of viral particle scale-up:
Insufficient system or process knowledge.
After years of painstaking research and successful completion of proof-of-concept studies, it might seem reasonable to assume that you already know everything you need to know about your product, and the process you have designed may have worked well to support preclinical animal studies.
However, as you progress to human clinical trials, and later into commercial production, your vector will need to be manufactured at scales that are multiple orders of magnitude larger than those required for animal studies. At these scales, the influence of variables that were paid only limited attention during preclinical investigations often becomes more apparent. For example, slight variations in the timings of infection or harvest could lead to varying levels of inhibitory metabolites in the culture medium, which in turn limit your potential to achieve the best possible yields or virus quality at production scale. If these metabolites have not previously been profiled, this could cause an unexpected development bottleneck.
As this scenario illustrates, without sufficient knowledge of all the relevant process parameters and their interactions, development complications and delays are almost unavoidable. It can be virtually impossible to keep your process under control, predict how process changes will affect the product CQAs (critical quality attributes), and optimize for factors such as cost-efficiency, performance and yield.
Many critical performance indicators and interdependencies are not obvious, and don’t become apparent until you deliberately go looking for them in a systematic way, by applying quality by design (QbD) principles throughout the whole development phase.
Putting too much faith in “plug and play” manufacturing platforms
A recent and highly enabling trend in the viral vector market is the use of platform manufacturing processes to simplify and accelerate the development process—particularly when generating material for phase I clinical trials. Such platforms use prefabricated processes (USP, DSP and non-product specific assays) that have previously been developed for a particular vector backbone. Instead of developing the process from scratch, your vector backbone is simply plugged into a platform process that has been designed for a similar vector. The majority of effort can then be focused on confirming that the process yields material of sufficient quantity and quality for phase I testing.
In many cases, a platform process can cut development time down by several weeks or even months, and is the best option to reach the phase I milestone as quickly as possible. Nevertheless, it is important to recognize that the platform approach is no substitute for true process development capabilities and expertise.
Given the inherent complexity of viral vectors and the current state of the art when it comes to standardization of platform technologies, there remains a very real chance that a particular platform will be unsuitable for your viral vector. In this case, it is vital to have the necessary process development capability on hand to keep your project back on track. Even in cases where the platform approach does prove successful, further process development is always needed in order to progress to phase II studies. This means that if either you or your development partner lacks the requisite capabilities in-house, valuable time can be wasted and additional costs incurred to transfer the technology to a partner with the right process development and manufacturing capabilities. At minimum, your new partner will need to carry out a process confirmation run, as well as additional work to implement and qualify the necessary analytical assays.
Failure to design for scalability.
Designing for scalability goes hand-in-hand with modern QbD strategies. Since process scale-up can lead to many unexpected problems and bottlenecks in development, it’s important to design your process with the end goal in mind. While this concept may seem obvious, the importance of designing for scalability is often underappreciated. The final scale requirements impose various constraints and have far ranging effects on many factors, such as choice of equipment, cell line requirements, raw materials, cost of goods (COG), and even end-product formulation and stability.
Skipping or postponing steps in the development process may sometimes seem expedient, but in the long run they can cause more problems than they solve. For example, the majority of viral vector processes are initially developed in adherent cell cultures that are propagated in T-flasks. In such cases, the quickest way to produce enough material for phase I trials may be to take the traditional approach of expanding the surface area and number of flasks (scale-out), rather than going down the more time-consuming route of transitioning to a fixed-bed bioreactor or microcarrier culture system (scale-up).
While this shortcut may get you to clinical trials faster, ultimately this approach is more costly, difficult to control, labor-intensive, and takes up more lab space. Switching to a bioreactor format, such as a fixed-bed or microcarrier format, could help overcome these problems. However, if you do this at a later phase in development, it is a process change that can potentially lead to unpredictable changes in the product CQAs. As a result, additional studies will be necessary to demonstrate comparability of product efficacy and safety.
Similarly, it may be tempting to postpone in-depth product characterization until the later stages of development. However, if this can be achieved using material produced under scaled-down conditions that adequately mimic the final process, you will have more time to de-risk your process and identify the most cost-efficient solutions.
GMP compliance expectations become increasingly stringent across the product development stages. No matter how carefully you plan, process changes may be needed during clinical development scale-up and optimization. These changes will have regulatory implications that you need to bear in mind.
When planning to launch in different regions, it’s also important to be aware of any differences in local regulatory requirements and guidance that could affect your product or process design. In addition, documentation and process materials that worked at research stage may no longer be adequate to ensure compliance. In a nutshell, regulatory awareness and design for cGMP compliance is essential for success, and should be accounted for as early as possible in the development process.
From problems to solutions
In this article, we’ve touched on some of the biggest sources of project delays and failures in viral vaccine and viral vector production, but there’s a lot more to learn, and of course it’s not all doom and gloom. In upcoming articles, we’ll turn our attention to the practical steps and considerations that will help you bypass these problems and get your product to market sooner.
Curious to learn more? Subscribe to the blog series, and never miss an article.
- “This is how much it costs to develop a vaccine.” The Cost of Things. MarketWatch, 1 October 2020.
- D’amore T and Yang Y-P. Advances and Challenges in Vaccine Development and Manufacture. BioProcess International (2019) Volume 17, September issue.
- Pronker ES et al. Risk in vaccine research and development quantified. PLOS One (2013) 8(3): e57755.
- Gouglas D, et al. Estimating the cost of vaccine development against epidemic infectious diseases: a cost minimisation study. The Lancet (2018) 6(12): E1386-1396.