Implementing Quality by Design for AAV Products

In this podcast, we talked with Parth Trivedi, Business Development Manager, Pall Corporation, about the importance of implementing a Quality by Design strategy for AAV product manufacturing and specific key steps for successful assessment.

We began by talking about the importance of Quality by Design (QbD) in AAV product manufacture and how this pertains to the regulatory landscape. Parth explained that there have been several recent regulatory approvals of gene therapy products, but in 2020 there were also regulatory setbacks. These setbacks mostly involved lack of sufficient data in the chemistry, manufacturing, and controls documentation or CMC. This led Pall to create a framework for QbD assessment and implementation for AAV based products.

Parth pointed out that for over a decade the FDA has advocated for a QbD approach in pharmaceutical manufacturing and there is good documentation and regulatory guidance around this approach. QbD is heavily based on prior knowledge and detailed understanding of both the product and the process variables. Most of the industry’s experience in QbD has been in traditional drugs and now we need to learn and apply these principles to gene therapy products.

Implementing Quality by Design

I then asked Parth if he could talk a bit more about what companies should consider before implementing a QbD approach. He said that QbD relies on prior knowledge and detailed understanding of the product and the process, so prior to implementation, it is important to collect and generate information data, analyze it and interpret any process changes that have an impact on the product.

Thus, it is important to understand the quality target product profile (QTPP) and define it, focusing on specifications around the safety, purity, and efficacy of the drug product.

A Framework for QbD Assessment for AAV Products

I then asked Parth if he could discuss a recent white paper that Pall released, “Quality by Design (QbD) for Adeno-Associated Virus (AAV) – A Framework for a QbD Assessment for AAV Products Within the Chemistry Manufacturing and Controls (CMC) Documentation.” I asked him to walk listeners through the four steps that the Pall team identified in creating a risk- and science-based QbD assessment.

Identification of CQAs based on QTPP and Risk Assessment

Parth explained that in this step it is important to focus on the empirical view, starting with a deep understanding of product knowledge and process knowledge. The product knowledge is where you should focus on the QTPP and the critical quality attributes (CQAs) and the risk associated with it. He went on to say that how we identify CQAs depends on the drug substance or drug product’s physical, chemical, biological, and related characteristics, which will eventually impact the quality, purity, activity, efficacy, and safety of the drug. Another important consideration is impurity profiling, there are three categories an impurity can be profiled into – product related impurities, process related impurities and adventitious agents.

There are variations in the impurities, especially with upstream vs. the downstream. Specifically focusing on the AVV process on the upstream side, there are different processes that create different impurities, for instance – transfection process vs infection, adherent vs suspension, and mammalian cell vs. insect cell culture types. These parameters and the method of manufacturing will have its own impact on the process related impurities and product related impurities.

From there, it is important to know what the target product profile is and what the quality attributes are. Examples of common quality attributes are non-infectious AAVs, empty capsids, aggregated AAVs, and encapsulated host cell DNA. The next step is to ask which of these are critical attributes and this can be determined using a risk assessment guided by the ICH guidelines. Essentially looking at each attribute and assessing the risk associated with each for impact on process and product. These are then ranked with risk rankings dependent on the likelihood of the risk and the severity to human health if it were to occur. From there, the rankings can determine whether it is a quality attribute and also a critical quality attribute.

Identification of CPPs and CMAs based on risk assessment

Next, Parth discussed the next step and how it connects with critical quality attributes. He explained that critical quality attributes are related to impurities – product related, process related, and adventitious agents. If we look at process related impurities, we need to understand what are the process parameters that coincide with that impurity because each process step will have certain parameters and multiple variables.

For example, at depth filtration or a clarification stage, you are not only trying to remove the cells and cell debris, but also trying to achieve the transmission of the product to the next stage. If it is a tangential flow filtration, you will also be focusing on flux to achieve process performance and at the same time whether the product is impacted by this step, so looking at shear rates. These parameters are identified and need to be evaluated to see if they are critical and thus create a critical process parameter (CPP).

For the white paper, the team identified the variables with a critical effect on a critical quality attribute. These that can be controlled in a manufacturing suite have been defined as critical process parameters (CPPs) and the materials which are going to impact all critical quality attributes, which will typically be controlled outside of the manufacturing suite are considered critical material attributes (CMAs).

CPP and CMA’s can be identified using different strategies, primarily based on data trending, thus data generation and data analysis is quite critical. Identification is also based on the rationale that there is a scientific understanding, there is a certain amount of this information available through prior knowledge, publications or something that has been developed during the developmental stages. There are of also CMAs based on the materials and the raw materials that create an impact on the quality of the product. It is important to identify CPPs and CMAs as early as possible, in the preclinical stages ideally.

Establishing the NOR and PAR for the design space

After understanding and identifying critical process parameters, Parth explains that the next stage would be to identify and establish the Normal Operating Ranges (NOR) and the Proven Acceptable Ranges (PAR). To do this, you must look at the significant data generation and understanding of the product and process at each stage of the drug development and manufacturing. For example, if you are looking at downstream, and specifically development for an ultrafiltration/diafiltration step. If the normal operating range is 10 to 15 PSI, would it be acceptable if there was a change in those pressures? What if the manufacturing system ends up operating at 18 PSI, which is out of the NOR? If you have generated enough data to create a design space, you will know what the maximum acceptable or PAR is and so if there are any variations that happen during the development stage or later in the manufacturing stage, you will know the impact on the product. The PARs are there so if the process is operating outside of NOR, the PARs would provide an understanding of the impact on the product and it would have been previously decided whether it would impact product safety, purity, or efficacy. Normal operating ranges are a subset or part of proven acceptable ranges, and NORs are recommended during the tech transfer and the actual manufacturing process.

Defining the control and testing strategy to ensure consistent operation

Next, Parth discussed the last step outlined in the assessment framework for QbD development. Control and testing strategy plays a very strong part in QbD development because these ultimately ensure the quality and safety of the product. It also ensures the quality and safety for the input material. Once controls are in place it is easy to maintain control of what happens during manufacturing and in the manufacturing suite, as well as maintaining control of the materials coming in from the outside. However, these control and testing strategies are fully dependent on the product and process parameters that have been developed. These control testing strategies will need to be scaled up in parallel along with the product and process parameters for when the process is tech transferred and taken to manufacturing scale. Additionally, it is important to test during the process or monitoring as well as during the lot release.

Key takeaways for upstream and downstream QbD assessment?

The authors of the white paper agree that the upstream production of AAV is quite well understood, including its cell culture operating parameters and the industry is continuing to make great advances in this area. For downstream production, depending on the type of cell culture and the process that has been developed there will be a few variations, but overall there is still quite a good understanding of downstream processes for AAV. There are also some areas for further understanding. For example, on the chromatographic stages, it would be helpful to understand how the chromatography step could be improved to reduce any adventitious viruses and to separate empty from full capsids in a simpler, more effective way than what we are doing now.

Advice for Implementing a QbD Approach

Last, I asked Parth if he had any additional advice for companies interested in implementing this program. AAV manufacturing is still a newer field of biopharmaceuticals and some of the design spaces are still not fully understood, so some of the steps may need additional development and sophistication around the process for full optimization. Overall, the ideas and implementation around QbD is something that we do have the framework to implement and is supported by the regulatory bodies. Pall is available to help their customers with implementation of these QbD strategies and the paper has good information about the QbD approach and provides a great deal of support on how to establish QbD strategies for AAV production.

For more information, please see Quality by Design (QbD) for Adeno-Associated Virus (AAV) – A Framework for a QbD Assessment for AAV Products Within the Chemistry Manufacturing and Controls (CMC) Documentation

Pin It on Pinterest