Explore Our Blog

CAR T Cells Part I: Design, Production, and Toxicity

by Scientific Writing Team 

In 2017, the FDA approved two revolutionary drugs known as chimeric antigen receptor T (CAR T) cells. These drugs, Kymriah® (tisagenlecleucel) and Yescarta® (axicabtagene ciloleucel), which are both indicated for the treatment of types of B-cell cancers, are the culmination of decades of research.

The concept behind CAR T cells is deceivingly simple – use a patient’s own immune system to recognize and eliminate cancer cells. The genesis of this concept occurred in the nineteenth century, when the surgeon William Coley deliberately injected his patients with bacteria to provoke an immune response against their tumors. While cancer immunotherapy has become more sophisticated since Coley’s day, one commonality between Coley’s toxin and CAR T therapies is the T cell.

T Cells

T cells were named because they mature in the thymus and belong to a class of white blood cells known as lymphocytes. Functionally, T cells are the hub around which the adaptive immune system revolves. There are many different subsets of T cells, each with their own specific functions, but they almost all operate through the same kind of molecular machinery.

A T cell’s ability to recognize specific antigens is conferred by the T cell receptor (TCR). The extracellular region of the TCR recognizes specific foreign biomolecules, similar to how an antibody can recognize specific foreign proteins. Once the antigen is recognized, the intracellular region of the TCR generates a signal inside the cell. This signal then activates the T cell to do one of many jobs. These jobs can include recruiting neutrophils to ingest and dispose of large microbes, stimulating B cells to produce antibodies, or directly killing the cell attached to the antigen, just to name a few. Due to their range of abilities and their capacity to mobilize other arms of the immune system, T cells are an attractive choice for molecular engineers in the fight against cancer.

Chimeric Antigen Receptor Design

CAR T therapy starts with the design of the chimeric antigen receptor (CAR). As its name suggests, the CAR is a combination of protein domains from different sources. In general, these domains include 1) an extracellular region with a well-defined specificity for a particular surface antigen, 2) a hinge/linker region, and 3) an intracellular region capable of transmitting a signal inside the cell.

In the case of both Kymriah and Yescarta, the extracellular CAR region corresponds to an antibody fragment specific for the B-cell marker CD19. This confers targeting to B-cell cancers that overexpress the CD19 protein. Many CAR T therapies currently in clinical trials are using extracellular regions that have specificity for surface antigens expressed on other types of cancer.

Optimization of the other two CAR regions is an area of ongoing research. Several competing theories exist as to which domains to use, but more experimentation is required to understand the rules that govern these regions and to pick the domains that will work best for treating each type of cancer.

Gene Delivery

Once the CAR has been designed, the next step is to introduce the CAR-encoding gene into the T cells. Both Kymriah and Yescarta use autologous (i.e., derived from the patient who will be receiving the therapy) T cells as the basis of their therapy. The cells are extracted using a process called leukapheresis, in which white blood cells are removed from the patient’s blood while the remaining blood components (e.g., red blood cells, plasma) are returned to the circulation. T cells are isolated from the collected white blood cells and then infected with recombinant viruses engineered to carry a gene encoding the CAR. Genomic integration facilitates long-term expression of the CAR on the surface of the T cells. Although the site of integration is random (or semi-random) within the genome, there has been no evidence thus far of insertional mutagenesis among the hundreds of patients treated with CAR T therapies.

Non-viral methods for delivery of CAR genes are also being used in several clinical trials and may provide safety and manufacturing advantages over viral methods. These methods rely on a technique called electroporation, which uses a brief electric pulse to create transient pores in the cellular membrane. CAR genes encoded on either DNA plasmids or on in vitro transcribed messenger RNA (mRNA) can then be delivered into the cell via these pores. Whereas the plasmids can be integrated into the genome using enzymes called transposons, the mRNA-encoded CARs are expressed immediately, but their expression is short-lived.

Ex vivo Activation and Expansion

In addition to transduction with the CAR gene, the T cells must also be activated and expanded before returning them to the patient. This process mimics the natural progression of T cell development and allows for the generation of large numbers of functional CAR T cells.

Naïve T cells typically require two signals to become activated. In the body, these signals are delivered by antigen-presenting cells (APCs). For the production of CAR T cells, however, it is more efficient to use artificial APCs, which may be derived from engineered cell lines or may even be non-cellular, such as magnetic beads that are coated with antibodies.

At this stage, the CAR T cells can also be induced to develop into specialized T cell subtypes by treating them with particular cytokines (cytokines are a class of small proteins secreted by immune cells and are important for cellular signaling). This is relevant because a growing body of evidence suggests that certain subtypes of T cells (or defined mixtures of subtypes) have superior tumor-killing properties.

Finally, the CAR T population must be expanded to ensure that a sufficient dose is available for injection into the patient. Expansion is performed in large scale bioreactors and takes more than a week. In the end, the CAR T cells are concentrated to an infusible volume, frozen, and transported back to the clinic for injection into the patient.

One intriguing alternative to the process described above is the use of “off-the-shelf” CAR T cells. These CAR T cells are initially taken from donors who are genetically similar (but not identical) to the patient who will be receiving the therapy. The donor T cells are transduced with the CAR gene just as before, but they are also modified to eliminate expression of immunogenic proteins, such as their native TCRs. This is accomplished with gene-editing tools that can selectively inactivate particular genes within the genome. Deletion of these genes makes the “off-the-shelf” CAR Ts compatible with a patient’s immune system and reduces the risk of graft versus host disease, in which the donated T cells could attack the recipient’s healthy cells. “Off-the-shelf” CAR Ts may also shorten the time between diagnosis and treatment, as these cells would ideally be prepared in advance and stored. Such an approach may also reduce the cost of the treatment, since the process of T cell harvesting and transduction would not need to be repeated for each patient.

Toxicity of CAR T Cells

Since CAR T therapeutics have entered the clinic, several toxic side effects have been identified. “On-target/off-tumor” toxicity was an expected side effect of CAR T therapy from the outset. As its name implies, “on-target/off-tumor” toxicity occurs when the CAR T cells attack non-tumor cells that express the intended target antigen. For example, Kymriah and Yescarta both target normal B cells as well as B cell-derived cancer cells because both of these cell types express CD19 (even though normal B cells generally express it at lower levels). As a consequence, the patient is unable to make antibodies and becomes more susceptible to infection by microorganisms. The newest generation of CAR T cells are being designed with clever safety strategies to minimize the level of “on-target/off-tumor” toxicity, which will hopefully help avoid this side effect in the future.

Cytokine release syndrome (CRS) is the other main toxicity associated with CAR T therapy and results from the overproduction of inflammatory cytokines. These cytokines, such as TNF-α and IL-6, are produced when CAR T cells proliferate in vivo, indirectly activating macrophages and other cells of the immune system. Depending on the level of cytokine production, the clinical manifestations of CRS range from flu-like symptoms to life-threatening shock. To help mitigate this risk, drugs such as Actemra® (tocilizumab), an IL-6 receptor antagonist, have been repurposed to treat CRS when it occurs as a result of CAR T therapy.


The arrival of CAR T therapy in the clinic represents a significant turning point in the era of personalized medicine. Unlike past attempts at similar technologies, such as Provenge® (sipuleucel-T), the clinical superiority of CAR T therapy (90% response rates by some estimates) makes it impossible to ignore. With dozens of experimental CAR T therapies in clinical trials at this time, CAR T cells promise to be part of the cancer treatment landscape for decades to come. While their complexity challenges our notions of what a drug is and their production can be daunting, the success of these novel therapies and the ingenuity behind them inspire us to keep pushing the boundaries of drug design and development.

Nuventra has extensive experience with immunotherapies and cellular therapies, including CAR T.

Contact a consultant to discuss your program.