Abstract:
The packaging efficiency (full to total capsid ratio, F/T ratio) for AAV is one of the critical quality attributes for the manufacturing and commercialization of AAV-based gene therapy. According to available literature and our data, production of AAV using the triple transfection method resulted in approximately 5 to 50 % of full AAV capsid in upstream crude harvest depending on serotype or gene of interest (GOI). The low quality of upstream crude harvest poses challenges for downstream purification. Previously, the lack of a reliable and cost-effective analytical tool to measure the F/T ratio on upstream crude harvest provided a major gap for upstream process development. ELISA-based approach coupled with genomic titer analysis is usually the method to estimate the F/T ratio for upstream crude harvest or other in-process samples. However, the data obtained by this approach are often unreliable and ambiguous due to low accuracy and high assay variations. Other more reliable analytical tools for F/T analysis, such as AUC (Analytical Ultracentrifugation) or negative staining TEM (Transmission Electron Microscopy), and cryo-EM (Cryogenic electron microscopy) are difficult to apply to upstream crude harvest due to low titer and high impurities in these samples while the cost for these assays is very high. Recently, several new analytical methods have been developed that can provide accurate and cost-efficient ways to analyze the F/T ratio for upstream crude samples. In this study, by using a Design of Experiment (DOE) approach, we sought to evaluate the impact of several production process parameters on AAV titer and packaging efficiency using the Ambr 250® High-Throughput Platform. We utilized our proprietary AAV packaging plasmids and AAV8-GFP as the study model. Titer and the F/T ratio on upstream crude samples were analyzed using the ddPCR titer assay and SEC-UV-MALS (Size-Exclusion Chromatography coupled with Ultraviolet and Multi-Angle Light Scattering detectors) analysis, respectively. After conducting two rounds of DOEs, we were able to significantly increase the full percentage of AAV-GFP in crude harvest by approximately 3-fold from the baseline. Additionally, we identified several production process parameters that played important roles in packaging efficiency for AAV8-GFP and built a strong model based on the DOE to predict an optimal condition for maximizing packaging efficiency for AAV8-GFP. In summary, this study demonstrates that we can significantly increase the AAV packaging efficiency at the upstream production stage, and a similar approach is ready to be implemented into upstream process development for different AAV serotypes and client-specific GOIs.
Author: Tsai-Yu Chen, PhD, Senior Scientist, Process Development