August 25, 2022
The Future of Biomanufacturing
By Olivier Loeillot, Vice President Bioprocess, Cytiva
We are on the cusp of a new era in the life science industry. Between the technological advances happening in digitalization and the diverse pipeline of new molecules appearing, what we are doing today will shape a very different biomanufacturing landscape for tomorrow.
How might that look? Let’s start with the new modalities.
While the traditional monoclonal antibodies (mAbs) development and commercialization will remain strong and continue to grow, we’ve started to see new molecules hit the market. Viral vectors, Adeno-Associated Virus (AAV), Lipid Nano Particles (LNP), mRNA, and plasmid DNA are all part of a variety of new modalities offering to provide alternative therapies to the more traditional mAbs—offering in many cases, therapies where there previously were none. While it’s still too early to confirm which of these new molecules will become viable with established processes and platforms, we know that on the whole, they are expected to nearly double the biologics pipeline in less than a decade1.
This will pressure big pharma companies to shift focus from targeting many diseases in parallel, to focusing on a portfolio of therapies targeting one disease using a variety of different molecules. It will force them to choose between excelling in development or in manufacturing. Increasingly, biopharma companies will outsource manufacturing in favor of investing their resources to advance 10–15 different molecule types in parallel. This natural evolution will create a set of strong contract manufacturing organizations (CMOs or CDMOs) who will then specialize in producing those molecules.
There is another seismic shift taking place within manufacturing. Traditionally a Western dominated industry, biomanufacturing is becoming more global. Ten years ago, China had a very small part of the antibody pipeline. Today, Asia has a third of biomanufacturing, 40 percent of phase three antibody trials, and a rapidly growing biosimilar industry. In the future, the US will continue to be a driver of new therapies with an increasing share of new molecules originating in Asia; however, the majority of the manufacturing will be done in Asia or Europe.
That begs the question: how might this impact how manufacturing is done?
These new modalities lack the generic molecular handle that allowed the industry to develop a platform solution for mAbs. We are now dealing with a completely different situation. Take mRNA for example—tremendously successful with the COVID vaccines—the processes that we have today are not fully understood from a supplier perspective, look quite different depending on the customer, and are frankly, currently, not that good. One of the key reasons being, mRNA molecules are very different to what we have been working with up until now in terms of size, form, shape, sensitivity to enzymes that break them down, and how they’re actually made.
For decades, the industry has been focusing on “proteins” as the actual drug. While protein molecules are distinct from each other, as a group, they are much more similar to each other than mRNA molecules are to each other. It took decades to develop the processes and tools we have today. With new molecules, like mRNA, we will eventually be able to refine these processes and products offerings, but it cannot be done overnight. Customers are taking what we have on the shelf today—developed for protein-based drugs—and using it for mRNA. It works, but not as well as it could or should. Suppliers will need to come up with new products and new processes to make these new molecules faster and better. Faster in terms of bringing them to market, and better in terms of scaling out. The aim is to enable vastly smaller quantities than the drugs of the present. We will have smaller doses for a smaller patient population in shorter time commitments. mRNA is just one example. Gene therapy (AAVs) is a key component to enable that. The tools and processes that the industry has in place today are simply not good enough for the future especially if you must scale-out of small-scale manufacturing dynamically to geographically distributed sites. Over the next several years, we will need to define, develop, and deploy at scale a variety of new processes to meet the variety of requirements of new modalities.
Flexibility in manufacturing will become even more critical.
Manufacturers will need to scale up, scale down, or scale-out depending on the success or demand for the drug. And as we move toward personalized therapies, manufacturing lines will look different—you will see smaller and more nimble lines with five-liter bioreactors compared to the 500 liters we see today. There will be clinics to draw patients’ cells and walk it next door to a one-time manufacturing center where they will create a personalized medicine and have it back into the patient within days. There are even clinical studies happening this moment to produce antibodies in vivo. LNP deliver RNA to the liver where antibodies are produced, thus turning the human into the bioreactor.
Data can facilitate.
Digitalization will enable process optimization—driving down the cost of goods, speeding up the time to market, eliminating waste, and reducing the risks while improving reliability and predictability.
The main risks in biomanufacturing come from people having to get involved in the process at multiple steps. In highly manual plants where they also run paper-based batched records, the number of errors goes up significantly. You need more people to solve the lack of automation and the lack of connectedness. Paper records, manual processes, and human measurements and analysis will be replaced by automation. We are already seeing improvements in these areas with robotic aseptic filling workcells replacing gloved isolated manual filling of drug products. Machines are controlled by software, but not connected to each other. Chromatography process development is being done in silico. Soon, simulated, mechanistic modeling will move beyond chromatography. In fact, the entire field of in silico process development is progressing rapidly. With predictive simulation, you can dramatically accelerate the development of new therapeutics.
One of the most dramatic changes will be the broad adoption of cloud utilization. Data collection—from development and manufacturing to analysis of that data—will be in the cloud. Information will be more accessible, with people being able to access it from anywhere, anytime, on any device. Transferring processes from the development lab to manufacturing plant and from plant to plant will be faster and simpler. With augmented reality, manufacturers will be able to deploy a process from one part of the world to another much faster with less training and fewer errors. In less than 10 years, a cloud-based biomanufacturing industry will be the norm. And within the next 15-20 years, the goal will be hands-free biomanufacturing.
There will be obstacles to overcome.
Creating a “smart” supply chain will require standardization across the industry with equipment that can be plug and play regardless of the manufacturer. Aligning and deciding upon one industry standard and then implementing on current and future equipment is a long way off.
Collaboration between biomanufacturers and regulatory bodies will need to increase in order to lower the barriers to process improvement that regulations present, all the while maintaining and improving patient safety. We know from COVID that tremendous speed in drug development is possible. The challenge will be making this standard practice so the industry can easily iterate and continuously improve. Our industry has begun to plant the seeds of our future biomanufacturing landscape. Those of us who are fortunate enough to help create the future of biomanufacturing today will see a much different, much more robust, and a much more efficient industry tomorrow. There is much to be excited about.