(716g) A Comparative Study of High and Low MOI Approaches for Adenovirus Manufacturing in HEK 293 Cells

Adenoviruses (Ads) are critical for oncolytic virotherapy and vaccine delivery. Current upstream process (USP) manufacturing of these therapeutics is focused on maximizing viral yield while ensuring quality and cost efficiency. Conventional approaches utilize a high multiplicity of infection (MOI) of 5-10 infectious units per cell (IFU/cell) over a 2-day culture period. High-MOI processes pose a challenge in large scale manufacturing as it requires a large amount of initial virus feedstock. Studies have shown that low-MOI (MOI < 1) processes are effective at creating Ad in T-flasks with a significantly (order of magnitude in volume) lower virus feedstock input. Recent efforts have shown promising results at the bioreactor level; however, these findings were for just one specific cell and virus combination. In this study, we aimed to explore whether similar low-MOI processes could be successfully applied to a related cell line and a different Ad serotype.

To evaluate both MOI strategies, HEK 293 suspension cultures were carried out across three vessel types: T-150 flasks, 500 mL shaker flasks, and 2L UniVessel® SU Culture Vessels with the Biostat ® B-DCU. Cultures were infected with Ad5-GFP at either a high MOI (5 IFU/cell over 2 days) or a low MOI (0.1 IFU/cell over 5 days). Samples collected throughout the experiment, and the final cell lysate were analyzed for infectious titre (Incucyte), viral genome quantification (qPCR), metabolite profile (FLEX2), and capsid titre determination (Octet). In the T-150 flask experiments, the high MOI condition yielded approximately 2.5x more virus than the low MOI condition, which contradict the trends reported in literature. A similar outcome was observed in the shake flasks. Based on these results, the subsequent bench scale experiment in the 2L UniVessel® was conducted using the high-MOI process, resulting in an 8x increase in viral yield compared to the T-150 flasks and a 3x increase versus the shaker flask. While these yields show an improvement, they remain below expected levels, indicating that further optimization is needed. Future work will focus on improving scalability by addressing factors such as controller optimization, inoculation conditions, and infection kinetics.