Optimization Through Culture, Quality Control and Regulatory Standards
Viral vector products for cell and gene therapy (CGT), like other biologic products, can have complex development pathways. Development can be organized into a few stages, starting with discovery and design, during which different constructs are screened. Lead products then advance to in vitro and in vivo testing to demonstrate proof-of-concept, efficacy, and safety. The most promising candidates then proceed to clinical trials and ultimately to the delivery of a new therapy on the market.
Manufacturing and testing are a critical part of this journey, and the specific needs of a development program evolve over time. While smaller amounts of research-grade material are sufficient early on, there will be a need for progressively larger amounts of material that will be held to increasingly stringent quality requirements. It is important to begin with the end in mind and to plan manufacturing activities strategically to be in the best position for long-term success.
There are a number of considerations that can help streamline the pathway to the clinic. One of the first is which vector system will be used, with AAV and lentivirus being two of the most commonly used platforms. Depending on the vector system, there may be additional options. In the case of AAV, vector serotype is another early consideration. Next, the packaging plasmids must be sourced, either from off-the-shelf stocks (if available) or manufactured. The vector production method (adherent or suspension culture) will also need to be selected, and the required manufacturing scale will need to be identified. As discussed below, there is often a change in the manufacturing strategy along the way. To mitigate unnecessary disruption, another key consideration is to determine where the clinical trial(s) will be held, as different countries have different regulatory requirements. Lastly, it should also be decided as early as possible what analytical panel will be used for testing the final product for clinical release. Release assays may need to be developed and qualified or validated, depending on the clinical phase and trial sites. Also, the release panel will determine how much of the product will be needed for testing purposes, including stability, and thus how much will be available for clinical use.
The following sections take a look at some of these considerations in more depth.
Transitioning from adherent to suspension manufacturing
Production of research-grade AAV is most commonly done in adherent cells grown in tissue culture flasks, typically with the HEK293T cell line. Smaller amounts of vector may be produced in dishes or T-flasks, while roller bottles or stacked flasks may be used for larger amounts.
In contrast, larger-scale production, especially for clinical and commercial use, is most often done in suspension culture, using HEK293 cells grown in a bioreactor. Suspension culture is preferred for larger batches because it is easier to scale and thus more readily transitions to commercial production. Suspension culture in a bioreactor also offers better process control and lower contamination risk versus flasks and roller bottles. Furthermore, because suspension culture does not require serum, it is more aligned with regulatory guidance to avoid the use of animal-derived materials, as preferred by both the FDA and the EMA.
Most viral vector programs will, at some point, transition from adherent to suspension production. When that happens, it will be important to assess vector productivity in a suitable scale-down model. It might be assumed that the amount of vector generated from a flask will predict how much is generated from a bioreactor. However, this is usually not the case, as there are some key differences between small-scale research production and large-scale clinical/commercial production. Because they rely on different cell lines (HEK293T vs HEK293), there are different growth conditions (media, cell growth curves, etc.). There are also differences in the purification strategies. Small-scale purification usually includes just lysis and clarification, sometimes with an additional gradient ultracentrifugation step, while large-scale purification typically relies on chromatography. However, each system has advantages and disadvantages.
Adherent culture is relatively inexpensive and highly productive for a given volume, making it an economical way to generate vector material relatively quickly. However, adherent culture requires animal-derived materials and is more difficult to scale up. In contrast, suspension culture is easier to scale and avoids animal-derived materials but requires a higher initial investment of equipment (e.g., bioreactor) and thus more extensive training. However, it is possible to take advantage of the strengths of each system to employ both strategically during product development.
Because of the high relative yield and the faster turnaround time, adherent culture can be ideal for generating useable amounts of different constructs that can be screened for efficacy (e.g., proof-of-concept studies in vitro and efficacy in vivo). While adherent culture does not predict absolute productivity at larger scale in suspension, it does allow an assessment of relative productivity (construct vs. construct). Thus, a set of constructs can be produced relatively quickly and inexpensively so that a lead candidate can be identified. The quick turnaround allows for construct designs to be tested, modified, and re-tested. It is commonly known that the gene of interest (GOI) and overall construct design can impact productivity. In this way, several different GOI designs can be compared, as can different combinations of GOIs with different promoters, serotypes, etc. This saves considerable time and cost over making these comparisons at the bioreactor stage. Once the construct design has been finalized or narrowed down to a few candidates, it is then possible to optimize some manufacturing conditions at a small scale in shake flasks as needed, or an initial assessment may suggest proceeding directly to a small-scale bioreactor for both upstream and downstream optimization.
Finally, it should be mentioned that while vector material from adherent systems can be used in vivo, or even in the clinic, since large-scale fixed bed bioreactors are a manufacturing option, it is advisable to perform toxicology studies with material that is representative of the eventual clinical batch. In other words, if it is known that the clinical batch will be produced in suspension, then the toxicology batch should generally be produced by the same system. Changing production methods from adherent to suspension would likely be considered a major process change by regulatory agencies, as documented in FDA’s 2020 Guidance on CMC Information for Gene Therapy IND Applications, and may prompt the need for comparability studies, since it has the potential to alter the product quality profile (and, as mentioned previously, involves the use of animal-derived materials). The concern is that if the product profile differs enough between a toxicology lot and a clinical lot, then it may be difficult to separate out any toxic effect of the product itself versus process-derived residuals. At a minimum, it would be important to look for differences in the rough residual profile (host cell DNA, host cell protein, etc.), the final achieved titer, and the final achieved percentage of full capsids (note that this can also vary by purification method irrespective of upstream culture method).
To summarize, adherent processes are very useful for early screening of constructs, but not as suitable for scaling up to large batch sizes. That is why it is often recommended to switch to a suspension method early in the development pathway, assuming that suspension will be used for the clinical material, to avoid comparability studies and/or the need to repeat a toxicology study.
Managing the “QC tax”
All cell and gene therapy products must be tested by an extensive panel of quality control (QC) tests to confirm safety, identity, purity, potency, and strength. CGT products tend to have relatively small batch sizes, compared to other types of drug products, and so the material is quite precious. In some cases, a single batch of AAV product might only be enough to treat 2-3 patients depending on the clinical indication and the dose.
Some QC tests require a set volume of test material amounting to several milliliters while others, like sterility, require a certain number of vials (10% if fewer than 100 vials, or 10 vials if 101-500 vials). Either way, considering that approximately 25 individual QC tests are typically needed to release a batch of vector product for clinical use, the total amount of material that can be required for QC release testing can be quite significant. Additionally, the amount of material required for other QC purposes can also be substantial. For stability studies, considerations such as the duration, the test panel, and the number of individual testing sites all impact the amount of product needed for testing. Other QC uses can include in-process testing and retained samples. Analytical development and qualification may be needed as well, although these activities should be done prior to GMP manufacturing, and some material may need to be held as reference material.
All told, about 20-25% of a product batch may be consumed for QC purposes, informally known as the “QC tax.” This “tax” is significant because it represents product that is no longer available for clinical use, which can add a risk of having insufficient clinical material to finish the trial and thus potentially delay treatment for some patients. This scenario adds complexity into planning sample allocation. As mentioned previously, there can be 25 individual release tests (or more). Without careful sample allocation, this could lead to well over 25 vials being used for release testing alone.
There are a few ways to mitigate the QC tax. These can include risk-based testing (reducing the frequency of non-critical in-process tests), sample optimization (minimizing sample volumes and dividing vials into multiple aliquots for testing, though this strategy must be tested during method validation), orthogonal testing (applying platform knowledge to reduce testing burden), and sample pooling (combining equivalent sample types). Another strategy is to use representative lots, such as an engineering lot, to perform assay qualifications and collect data from hold-time studies.
Stability testing can also be optimized. For example, for products manufactured with a platform process, stability data can sometimes be leveraged across programs. Additional strategies include the use of bracketing (testing at the extremes) and risk assessment to reduce time points. Taking a matrix approach to stability testing can also help. For example, if multiple batches of the same product have been manufactured, then design-of-experiment principles can be used to narrow down the test conditions on a particular batch. However, this plan must be documented ahead of time.
It should be noted that all of these strategies must include regulatory involvement. It is best to discuss these strategies with regulatory advisors ahead of testing to align on strategy. In some cases, with a proper risk analysis in place, it may be possible to reduce the number of QC tests performed, but this can’t be assumed. Regulatory agencies may be open to suggestions on strategy, but this must be discussed. Ultimately, there is no single strategy that works in every case. Planning for the QC tax depends not only on the test panel, but also on product concentration, patient dose/fill volume, and clinical trial design. The goal is to manage product consumption by strategically using the material.
Considering US vs. EU standards
In terms of overall requirements, the US and EU have some similarities in the regulations governing CGT products. However, each jurisdiction has its own set of specific guidelines.
In the US, manufacturers are expected to follow US Pharmacopeia (USP) and the FDA 21 CFR Part 211 guidelines on current good manufacturing practices. In addition, the FDA has released a series of guidance documents on specific topics within the CGT field, such as the 2020 guidance on CMC Information for Human Gene Therapy IND applications, among others. Collectively, these guidelines provide a framework for manufacturers to follow for process documentation and product testing.
The EU has a similar set of standards. Manufacturers are expected to follow the European Pharmacopeia (Ph EU), EU GMP Annex 1, and the European Medicines Agency (EMA) guidelines on Advanced Therapy Medicinal Products (ATMPs), which include specific documents on gene therapies, cell therapies, vaccines, and other products.
Although US and EU guidelines have some broad similarities, there are some key differences between them such that products manufactured to meet US guidelines may not be immediately approved for use in the EU (and vice versa). One of the key differences between FDA and EU standards is the EU requirement for QP (Qualified Person) involvement during manufacturing. Other differences between the US and EU include raw material qualification standards, standards for environmental monitoring, batch record content and formatting, and quality system expectations. In the EU, release tests must be performed with methods validated to EU standards. EU manufacturing also requires a risk management plan and documentation of compliance with Annex 1.
Despite these differences, it is sometimes the case that a company may want to expand into other geographic areas after initially targeting only one. For example, it is possible to obtain EU approval for material manufactured to US standards, but a significant amount of retrospective compliance activities are required. As a first step, a QP will need to get involved to conduct an audit. Following risk assessments and a gap analysis, discrepancies in the requirements between locations may be identified, including:
· Facility-related gaps, including cleanroom classification and documentation, that the contamination control strategy aligns with Annex 1. HVAC and environmental monitoring data may need to be re-analyzed and aligned with EU alert and action levels. EMPQ may need to be reperformed.
· Analytical methods will need to be validated per EU standards (validation reviews), potentially resulting in full re-validations.
· Sterility assurance strategies will need to be enhanced, and media fill protocols may need to be revised to align with Annex 1 expectations.
· Raw material documentation may also need to be enhanced (for example, TSE/BSE statements may be needed), and retrospective vendor audits may be needed. A full risk management plan will also need to be developed.
· Product labeling may need to be redesigned to align with EU language and safety requirements. EU importation will need to be arranged, and testing must be done in EU-certified labs.
Altogether, this retrospective work can add considerable cost and time to a project. It is therefore always advisable to identify target markets early, so that manufacturing can be performed in a compliant way. Having early involvement of a QP will save time, as will batch documentation designed for global GMP compliance, ICH-compliant testing method validations, raw materials selected for both US and EU acceptability, quality systems and environmental monitoring alignment, and an early risk assessment and contamination control strategy plan. While it is true that data from US trials can be leveraged in applications to the EU, perhaps strengthening the application scientifically, from a cost perspective, it is best to plan on EU submission up front to avoid retrospective compliance activities.
Finally, while this discussion is framed from the perspective of taking a US-manufactured product to the EU, there may also be situations in which a product manufactured to EU standards is submitted for US approval. In this scenario, because of the QP involvement required by the EU, many of the discrepancies between EU and US standards would likely be addressed early. However, even in these cases, some level of retrospective compliance activities may still be required.
Conclusions
In gene therapy, there are certain best practices that will streamline the pathway to the clinic. However, it is often the case that plans and strategies will change over time, in response to various factors, such as biological reasons, budget, changing regulations, etc.
By staying flexible, programs can be kept on track, in spite of course corrections that might occur along the way.
Explore concept-to-cure CGT CDMO solutions: https://bit.ly/4obADKG
Sponsored content is written and provided to BioSpace by the advertiser. It is published with the advertiser’s approval without contribution from BioSpace’s editorial and insights teams.