Cell, Gene Therapy Manufacturing As Large-Scale Biology
Contributed Commentary By Mitch Finer
July 23, 2020 | The last three years we have witnessed an exponential expansion of the cell and gene therapy (CGT) space, both in the number of cell and gene therapies approved or in development, and the real and projected values of the market. This growth is rapidly changing the way we think about what medicines are.
More importantly, I believe it’s changing the way we must think about making medicines. Scaling a CGT product from the lab, where a few millions cells are produced, to manufacturing the billions of cells needed to treat patients comes with a unique array of complexities. These complexities require thinking about CGT manufacture as large-scale biology, a concept that has developed in my mind across my 35 years in the CGT space.
Large-scale biology has two dimensions. The “vertical” dimension encompasses the traditional challenges of making product quantities large enough for clinical testing and the market. The “horizontal” dimension encompasses the CGT-specific challenges of linking culture conditions to the product’s clinical performance.
In the vertical dimension, manufacturing is more challenging for CGT than for biologics because CGTs are “living drugs”. As such they are subject to many positive and negative influences—ex vivo in culture and in vivo as therapies—on how they differentiate, grow and function.
Culture conditions for a cell product don’t scale linearly between the lab and manufacturing facility, for various reasons. For example, cells can be cultured by suspension in solution or by adhesion on plates; but not all cell types can be cultured both ways. Cell types also vary in their proliferation rates and the number of expansions they can undergo without loss of quality.
Allogeneic and autologous cell products also have to be scaled differently, because differing quantities of each are needed for therapeutic purposes. Allogeneic products are manufactured by multiplying the volume of a bioreactor to yield a single, large batch of product that will treat multiple patients—an approach known as “scaling up”. Autologous products are manufactured by multiplying the number of bioreactors to produce small, individual batches of patient-specific products in parallel—an approach known as “scaling out”. Each approach has specific challenges.
Another complexity in vertical scaling of CGTs is that products often contain mixtures of cell subtypes. For example, a T cell-based therapy may contain both CD4 T cells and CD8 T cells, as well as T memory cells, and T regulatory cells (Tregs) and other specific subtypes of CD4 and CD8 T cells. The relative proportions of subtypes in the final product are subject to many factors and can influence the product’s clinical performance.
The horizontal dimension of large-scale biology involves determining how specific cell culture conditions influence the properties of product and translate into their clinical performance. This dimension requires studying a broad range of products; one product or even a few products will not reveal the complex associations between conditions and performance.
These associations are investigated by collecting a wide range of detailed analytics on each product, correlating those analytics with therapeutic efficacy, and pooling the data across multiple products. Those pooled data can be used to identify which properties—such as a marker on a certain cell type or a specific ratio of cell subtypes—are associated with good (or poor) clinical performance of a particular CGT product.
To take a concrete example: an ongoing question in CGT research concerns the optimum ratio of CD4 to CD8 T cells in a product. Some companies utilize whatever ratio of the cells their culture conditions yield. Others aim for a 1-to-1 ratio of the two cell types: they sort CD4 from CD8 T cells in the donor (or patient) samples, culture the cell types separately, then recombine them in equal numbers to yield the final product.
If we were to apply large-scale biology to this question by collecting broad and deep analytics on T cell products, we might find that the ratio of specific subsets of CD4 and CD8 T cells are key to a product’s optimum clinical performance, or that the optimum cell ratio varies according to product type.
At present, most CGT biotechs have neither the product number or range, nor the big data infrastructure, to gather the broad and deep analytics necessary to conduct large-scale biology. While CDMOs do handle a wide range of products, collecting detailed analytics on each one and applying those learnings to other products is not part of their business model.
Many CGT companies apply the small molecule approach to scaling: they wait to address manufacturing issues until their products are moving toward the clinic. Instead, CGT companies should begin considering the challenges of large-scale biology—vertical and horizontal— at the earliest stages of development.
The biology of CGT will continue to evolve as the space continues to grow, and success in the clinic will always depend on success in manufacturing. Companies must therefore balance the thinking that drives innovation against the thinking required to scale the biology, because CGT manufacturing is the far more complex piece of the CGT value chain.
Dr. Mitch Finer is the Chief Scientific Officer of ElevateBio and the President of ElevateBio BaseCamp. A globally recognized pioneer in cell and gene therapies, he is focused on pursuing and advancing development of transformative cellular, gene therapy and regenerative medicine products through strategic partnerships with leading academic researchers, medical centers and entrepreneurs. Dr. Finer received his Ph.D. in Biochemistry and Molecular Biology from Harvard University and a B.S. in Biochemistry and Microbiology from the University of California at Berkeley. He can be reached at firstname.lastname@example.org.