Forecasting the Infrastructure of Cellular Therapies
If you have been keeping tabs on the biotech industry for the past couple years, you will have heard of cellular immunotherapy. In a nutshell, cellular therapy involves the extraction of living cells from blood or tissue which are then genetically modified to treat a specific disease. This approach has led to some truly extraordinary cancer therapies such as Novartis' Kymriah™ and Gilead/Kite Pharma’s Yescarta™ which recently received FDA approval. In brief, these therapies use cells from our body’s immune system which can be genetically reprogrammed to “seek and destroy” cancer cells circulating in the blood. While the concept of treating diseases with living cells is not new (Ex: tissue grafting or bone marrow transplants), the ability to use engineered living tissue represents a significant paradigm shift from traditional medicine.
Although they are transformative, these types of therapies face some difficult challenges: namely, they are really expensive and arduous to develop and manufacture. This is in part why they have such a high price point, $300k to $500k per treatment course, in which a major cost factor is the manufacturing. Unlike traditional drugs, the therapy is a living product which requires specialized production and supply chain considerations. Currently, most manufactures use traditional, large scale batch processes to filter, engineer, and grow the specialized cells. While this approach works in some cases, it is expensive to scale and difficult to adapt to produce more specialized types of therapies.
In this brief article, I wanted to share some thoughts and resources on what a future cell therapy manufacturing infrastructure could look like and point out some challenges that new biotechnologies need to solve to accelerate the industry. While this article is not exhaustive, my hope is that it can give insight and open dialogue about this emerging field.
Kinko's vs. Home Printers: Two different approaches to cell therapy manufacturing
Making cell therapies fundamentally breaks down into two approaches: production in large centralized facilities or production across distributed-local sites. In a way, this reflects a similar situation in the printing industry where you can either print from a Kinko's or from home on your personal printer. It is difficult to say which approach will be better as each has its own strengths and weaknesses. The centralized approach is easy to standardize and control, which is critical to ensure FDA compliance and a well characterized product, but it needs significant cap ex to build and maintain. Additionally, you have a logistics challenge because shipping living cells around the world is not the same as mailing documents. The distributed approach on the other hand requires much lower cost to build out and addresses the logistics problem as the therapy can be manufactured on site with the patient. However, this approach is difficult to standardize to ensure compliance and challenging to build out as a cohesive network.
An additional factor that can influence the centralized versus distributed approach is how the therapy is produced, specifically if it is made from the patient’s tissue (autologous) or from a donor (allogeneic). In general, a patient derived therapy lends itself to a distributed approach while a donor derived therapy will likely work best in a centralized approach where large batches can be produced at once. However, this is still debated in the community and there are successful examples of patient derived therapies being developed with a centralized approach and vice versa. In the near future, it is likely we will see both infrastructure approaches being adopted and built out which is ideal as patients will likely respond differently to each therapeutic approach. Just like Kinko's and home printers, sometimes you need to print a single document and sometimes you need to print a hundred flyers.
Whether centralized or distributed manufacturing is used, the need to reduce cost remains critical which is why new manufacturing technologies are so vital and disruptive. Automation is going to be a critical factor that can reduce cost where we will likely see new smart manufacturing technologies being integrated. Additionally, increasing the speed of batch production can reduce costs as well. New technologies that can more rapidly filter cells from tissue and perform genetic modifications can make a significant difference, especially if they are automated. Finally, reducing the cost of the materials for gene modification will be critical as well. Getting clinical grade materials to genetically modify the cells is extremely expensive and in itself is difficult to manufacture. This can be addressed by scaling production of the modification material or make the genetic modification process on the cells more efficient to reduce consumption. So, whether cell therapy adopts a centralized, distributed, or combined infrastructure I think the successful technologies will be the ones that can differentiate well into both approaches.
In a way, producing cell therapies is like forming an army where you have to assemble and train different types of people with varying backgrounds and abilities to play specific roles within the larger group. In an army you have units who specialize in infantry, cavalry, artillery, intelligence, etc. and each of those units require specialized training and equipment. In the same way, cell therapies use mixtures of different types of immune cells to attack a disease-causing target. New research is showing that specifically controlling the mixtures of cell types within the therapy leads to more targeted and effective outcomes. New manufacturing technologies that can accommodate these needs will be critical as the cell therapy industry matures and evolves. For example, tools that can rapidly purify specific types of cells from blood or tissue will be essential to building out these capabilities, similar to how an army recruiter matches potential soldiers with specific role. This presents a challenge but significant opportunity for new tech to balance precision with speed.
Getting out of the Lab and into the Clinic
In addition to industrial scale manufacturing, developing and translating new types of therapies out of the lab and into the clinic is critical but difficult. There is a reason why only a couple cellular immunotherapies have been approved by the FDA in the past year when it has been a research focus for over forty years. In a sense, each researcher has to be their own manufacturer and quality control division. At a research level, most of these production steps are done by hand with manual processes to isolate the cells from tissue then genetically modify them. However, some specialized tools are often needed to isolate and analyze cells throughout the process. These specialized tools are quite expensive and are rarely available for each researcher to purchase for their own lab. That is why most of them are purchased and shared in institutional core facilities, similar to the Kinko's model previously discussed, which can be rented by the hour. The main challenge with this model is the practical availability of the core facility resources and scheduling. In addition to doing the development work, researchers often find themselves juggling access to tissue, precisely arranging experimental workflow, and scheduling time at a busy core facility which is a lot handle. In performing these types of experiments first hand, I understand how frustrating, challenging, time consuming, and expensive it can get.
In these cases, it appears that a distributed model will be most effective in accelerating R&D for new therapies at research institutions and companies alike. At the R&D stage, flexibility and maximizing data turnout are the driving factors, so tools that extend or enable these capabilities are in demand. This can not only alleviate the mental load on the researchers but also give them more “shots on goal”, which will translate more ideas from the lab to the clinic. To be effective, these tools need to be accessible to the researcher price wise yet flexible enough to work with a researcher who job is to push the envelope. Here I think we will see several new disruptive technologies make their debut as fields like microfluidics and other micro-technologies continue to develop and mature.
I hope this article piqued your interest into the field of cell-based therapies. While not exhaustive, I wanted to provide some thoughts and resources into the promising future that these treatments hold and outline some thoughts on what the production infrastructure could look like in the near future. Below, I have listed a few articles if you want to research further. Happy reading!
Resources and Articles
Overview on Cell Therapy Manufacturing: https://www.frontiersin.org/articles/10.3389/fmed.2018.00150/full
Centralized vs Distributed Manufacturing:
Specific Cell Types in the Therapeutic Product
Micro-technologies in Cell Therapy