Gene therapy manufacturing is currently too slow and expensive, and this project aims to fix the production line itself. The core problem is that making the viral vectors needed to deliver gene therapies—specifically adeno-associated viruses (AAV)—is inefficient, driving up costs and keeping treatments from patients. This project will build a single, scalable bioprocessing system that works across three different volumes (10 mL, 100 mL, and 1000 mL), validated against a standard 5L bioreactor. It will also develop new chromatography columns designed to capture more AAV particles, and a predictive model that uses real-time data to forecast yields at different scales. If successful, the system would let scientists run far fewer and smaller experiments to develop clinical materials, slashing both time and cost. The environmental impact is also significant: higher AAV yields mean less raw material, energy, and water per dose, reducing the manufacturing footprint and waste. This is not a new therapy itself, but a fundamental improvement to the invisible industrial machinery that makes those therapies possible—a more sustainable, scalable platform for producing the growing number of gene therapies in development.
View original technical description
Patients suffering from genetic conditions continue to wait for the promise of cell and gene therapies to be realised. The major bottleneck for access to therapies is the cost of goods, largely comprising manufacturing costs arising from process inefficiencies. Viral vectors, particularly adeno-associated viruses (AAV), are pivotal in the production of cell and gene therapies, with their importance expected to grow as the field shifts toward i_n vivo_ therapies. However, achieving high-yield AAV production at commercial scales remains a significant challenge. This project aims to address this issue by developing an innovative bioprocessing system capable of upstream and downstream processing across three log scales (10 mL, 100 mL, 1000 mL) validated against a 5L bioreactor. The integrated downstream processing will employ novel AAV-specific chromatography columns designed to maximise vector recovery. The system will measure viral titres derived from HEK293-T cells to ensure flexibility and robustness across different production platforms. A key component of this work will be the development of a novel downstream serotype specific capture agent and predictive model that integrates real-time data from the bioprocessing system with historical data, providing insights into expected yields at various scales. This model will enable process development scientists to significantly reduce the number and size of experiments required, thereby accelerating the development of clinical trial material and full-scale manufacturing and significantly improving sustainability. In addition to optimising production efficiency, this project will also develop a sustainability model, quantifying the environmental benefits of this scalable, streamlined bioprocessing strategy Improvements in AAV yields arising from this innovative project will significantly enhance sustainability in gene therapy manufacturing by reducing the resource requirements for large-scale production. Increasing AAV yields will deliver a corresponding reduction in manufacturing footprint with less raw material, energy and water needed to produce the same amount of therapeutic product. This reduction in resources will not only lower production costs but also minimize environmental impact, as less waste is generated during the process. Additionally, the project's integration of real-time process control and predictive modelling will optimize production efficiency, further contributing to a streamlined and more sustainable biomanufacturing workflow. These gains will help establish a scalable and environmentally responsible platform for viral vector production, supporting the growing demand for gene therapies in a way that aligns with global sustainability goals.
Plain English summaries and category classifications on this site are generated by AI and may not perfectly reflect the original research.
Is something wrong? Let us know