Viral vectors are vehicles for delivery of therapeutic DNA in cell and gene therapies. With over 1,000 cell and gene therapy (CGT) clinical trials underway globally, there is a growing need to address challenges in viral vector manufacturing – both upstream and downstream. In a previous post, we outlined the key steps in downstream processing (DSP) for the manufacturing of lentiviral vectors (LVV). Achieving high concentration without aggregation is a significant technical bottleneck in manufacturing LVV. In this post, we will explore the concept of viral aggregation and its influence in every step of DSP for LVV.
What is Viral Aggregation?
Viral particles tend to form aggregates in what can be defined as a survival mechanism that helps viruses resist environmental stress and withstand degradation by disinfectants. It is a complex process influenced by cell type and cell-related impurities, the type of virus, biochemical properties of the virus (e.g. virus size and shape, isoelectric point, etc.), physiochemical factors (e.g. pH, ionic strength) as well as operational factors (e.g. process temperature). Although researchers began studying viral aggregation in the 1950s, the phenomenon continues to be studied and viewed as one of the main challenges in viral vector manufacturing.
How does Viral Aggregation affect DSP?
DSP is a series of unit operations aimed at achieving the highest possible recovery of concentrated and purified LVV. Viral aggregation occurs due to electrostatic and hydrophobic interactions and aggregated virus particles can cause complications in DSP, leading to significant vector losses and decreased yields during the membrane-based processes (e.g. filtration) as well as purification processes (e.g. chromatography). Furthermore, aggregates impact readouts of virus infectivity that tell us about the quality of the final product. Therefore, it is essential to consider the impact of viral aggregates at every unit operation in DSP. The goal is to optimize DSP to prevent or minimize aggregation of viral particles.
Strategies to Mitigate Negative Effects of Viral Aggregates on LVV Recovery
Critical process parameters and the respective operating ranges for each unit operation in LVV DSP could be identified to minimize viral aggregation. The first step in doing so is to understand the properties of the viral vector you are working with. The type of viral particle, its surface structure, hydrophobicity, size and shape, and total charge all affect the degree of aggregation.
During membrane-based processes, such as ultrafiltration or sterile filtration, viral particles may bind (specifically or non-specifically) to the surface of the filter causing membrane fouling and decreasing viral vector passage through the filter. When it comes to sterile filtration, the strict requirement for using a 0.2 µm pore size filter contributes to huge viral vector loss, sometimes 40 per cent or even more. Although optimization of sterile filtration to control interactions between the virus filter can be accomplished by optimizing load, flow rate and filter surface area, as well as examining different filter materials and pre-conditioning options, the presence of viral aggregates makes it increasingly challenging.
Alternatively, adding a diafiltration unit, to buffer exchange the virus solution to a proper formulation buffer, provides an opportunity to control virus-virus interactions and reduce aggregation prior to sterile filtration. Optimization of ionic strength, pH and osmolarity of the formulation buffer are also instrumental in reducing viral aggregation. To implement all these strategies, a proper analytical method to assess virus aggregation is inevitable.
Performing analytical tests to determine size distributions throughout DSP is an important strategy to determine aggregation state and mitigate the effects of this phenomenon. Early work in this area focused on repurposed size-exclusion chromatography, typically used for protein purification (e.g. monoclonal antibodies). This technique is ineffective for analyzing viral particles because of the significant size difference between viruses and proteins; generally, virus particles are 10 times larger than proteins. Alternative approaches to analyze viral aggregation include dynamic light scattering (DLS), nanoparticle tracking analysis (NTA), tunable resistive pulse sensing (TRPS), and flow cytometry (FC) based methods.
Generally, DLS is the first to-go option for determining virus size distribution, but the resolution of the method is very low when it comes to unpurified virus solution due to the presence of host-cell related contaminants. Alternative techniques are NTA, TRPS and FC based methods. The TRPS method is advantageous as it provides a readout of concentration of virus particles at different size distributions, where a percentage of viral aggregates could be calculated. However, DLS, NTA, and TRPS methods could handle a single sample measurement at the same time. Recently, the FC method has been highlighted because it provides a high-throughput platform to study aggregation behaviour of viral vectors. The main challenge is that only a few commercially available FC instruments are able to detect particles at the nano-size level.
In summary, innovation in process optimization and analytics will be key to develop robust, scalable and cost-effective processes for LVV manufacturing. Working with a team skilled in developing custom analytics for viral aggregate sizing and testing their application in LVV DSP workflows will help improve your process, decrease costs and advance the development of new CGTs.
For custom solutions to your LVV product, contact us here.