(493b) Towards Cost-Effective Downstream Processes for the Purification of Therapeutic Viruses

Viruses are an important new class of biotherapeutics that have been gaining significant interest for their applications as vectors in gene therapy and vaccines and as tumor-killing agents in novel cancer immunotherapies. While these new therapies present great promise in clinical trials, the large-scale purification of viruses is still hindered by low overall yields that are typically below 20%. Host-cell DNA is a particularly concerning type of impurity for its potential oncogenic activity and therefore the U.S. Food and Drug Administration (FDA) requires gene therapy products to contain no more than 10 ng of host-cell DNA per dose and no DNA fragments greater than 200 base-pairs in length. As with other types of biotherapeutics, virus downstream processes involve a series of many unit operations targeted to removing process- and product-related impurities. Given the broad variety of commercially available process equipment and materials, it is a challenge to select the optimal conditions that would maximize product recovery and purity while minimizing production costs.

To address this challenge for the purification of adenoviruses, we used an integrated approach to evaluate the effect of an enzymatic DNA digestion step on the performance of the subsequent DNA removal step using the Sartobind Q anion-exchange membrane for chromatography.^a A series of optimization experiments were performed in high-throughput format using 96-well filter plates to evaluate the effects of enzyme type (Benzonase or Denarase), concentration, and time in the digestion step performed before chromatography. Overall, both types of enzymes performed similarly and there was a strong direct correlation between enzyme concentration and digestion time in reducing DNA levels. Interestingly, almost complete DNA reduction was achieved with 24-hour long digestion times independent of enzyme concentration; however, this long incubation period also caused inactivation of more than 99% of infectious virus particles and therefore is not a viable option. Next, a selected set of digestion conditions using Denarase were applied to prepare lysate material for runs using the laterally-fed membrane chromatography (LFMC) device – a novel format for membrane chromatography that enables high-resolution separations by providing uniform flow distribution and minimal void volume within the device. Denarase was selected because it is a considerably cheaper alternative to Benzonase. By implementing the changes defined after accounting for the interactions between the DNA digestion and membrane chromatography steps, the process was scaled up by a factor of 10 and the amount of residual host-cell DNA per dose was reduced by a factor of 80 with a virus recovery of 73%.

These results, along with further data from the literature, are currently being used in a study that is focused on assessing the economic feasibility of the adenovirus purification process. Both upstream and downstream processes were modelled using BioSolve Process (Biopharm Services). The upstream process starts with suspension cell cultures for virus propagation in stirred tank bioreactors, which is then followed by cell harvesting and cell disruption. The downstream process consists of a clarification step for the removal of host-cell debris, a host-cell DNA digestion step, and then chromatographic purification and polishing for the removal of host-cell protein and DNA, followed by concentration, formulation, and sterile filtration steps. Several production scenarios are presently being analysed to identify key cost drivers and to quantify their contribution to the cost of goods of adenovirus production. The first goal of the study is to quantify the costs associated with the use of Benzonase or Denarase in the DNA digestion step by applying the collection of experimental data obtained in the integrated study described above. Membrane chromatography presents a series of practical advantages over resin-based chromatography for the removal of host-cell proteins and DNA from large target biomolecules such as viruses, including more accessible binding sites and high binding capacity, besides allowing high flow rates. Therefore, the second goal of the study is to evaluate how both materials impact the overall production costs. Finally, a third analysis is focused on the upstream process to determine the impact of using different initial bioreactor cell densities on the final yield and costs. With BioSolve, each scenario is analyzed for a range of process scales from tens to thousands of liters to reflect clinical, pre-clinical and commercial scale process volumes.

Considering practical limitations in a comprehensive analysis of the economic aspects of virus manufacturing will help address critically outstanding questions in the field and will guide a more rational selection of process conditions. Such an approach will enable the development of cost-effective ways to produce therapeutic viruses at various scales of manufacturing, ultimately benefiting patients to have fast and affordable access to new treatment options.

^aReference: Kawka, K., et al. "Integrated development of enzymatic DNA digestion and membrane chromatography processes for the purification of therapeutic adenoviruses." Separation and Purification Technology 254 (2021): 117503.