Future advances in vector manufacturing at scale 

Driving the next generation of manufacturing  

 Vector manufacture is a rapidly evolving field, and a number of advances will be key to the ability to deliver gene therapies at scale. The advances that have been made in the quarter-century I have been involved in the area are dramatic. When I produced my first adeno-associated virus (AAV) batch in the lab, it was a total yield of 1010 vector genomes (vg) in petri dishes. Fast forward to today, and in our GMP facility we’re up to 1017vector genomes produced from a single batch in a 500L bioreactor That’s a seven-log increase in vector production!    

But I still believe we can continue to improve – both in terms of yields and quality. One reason for this is to reduce the cost of goods. Vector-based products are always going to be expensive to make, but if we can increase the yields, the cost per patient should fall. Improving the quality of the products made is just as important, if not more so, because this could boost the efficacy of the vectors, and thus reduce the therapeutic doses required per patient. This, in turn, reduces the amount that will need to be manufactured.   

This could be achievable if the batch sizes are increased. Yes, this could be done by simply using a larger bioreactor, but this will never be enough. It will also be important to look at technology advances. This is something else that has progressed dramatically during my time in the industry, where the field has moved on significantly from the adherence in transfection that was routine back in 2000. For AAVs, I have seen baculoviruses, insect cells and HeLa cells, and even HSV as a helper virus.   

Embracing innovation  

New technologies are always emerging, and we need to be open to the possibilities they offer for improving productivity and increasing scale. Enhanced productivity may make it possible to improve yields without fundamentally changing the process, for example by moving towards suspension and transfection systems. If this proves feasible, then this is likely to be the easiest option.   

An alternative could be to use producer cell lines. This is already common for antibody manufacture, and may prove effective for vectors, too. It has scalability advantages as it is usually straightforward to simply move the process into a bigger bioreactor, and no transfection system is required. However, the downside for lentiviruses and AAVs is that every new lentivirus envelope or AAV serotype will require a new cell line to be developed. And currently the productivity of the cell lines is lower than transfection systems.   

But I’m always open to new manufacturing possibilities. Right now, the greatest needs lie in increasing yields and improving product quality, whether this is achieved by improving current technologies, or moving to new ones. Yes, at the outset there will be difficulties, as we found when we moved from adherent to suspension production. But the advantages can be significant.   

Looking ahead  

There are many opportunities for improvement. I am confident that in the next five to 10 years productivity could be improved by a further two or three logs, offering significant cost of goods advantages.