Chimeric antigen receptor (CAR) T cells are engineered to express receptors that enable the targeting of speciﬁc proteins, namely those found on the surfaces of tumor cells. CAR T cells contain three distinct regions:
Comprised of three components; a signal peptide, an antigen-recognizing region, and a spacer. The ectodomain lies outside the cytoplasm, interacts with the extracellular space and is the part that initiates the CAR T cell’s immune response.
This membrane-spanning, hydrophobic, α-helical structure serves the purpose of stabilizing the CAR T cell receptor.
Predominantly comprised of three immunoreceptor tyrosine-based activation motifs (ITAMs) of the cytoplasmic portion of the CD3 ζ chain, as well as co-stimulatory regions. The endodomain is responsible for transmission of intracellular activation.
When the engineered receptor binds to the target protein on another cell’s surface, the endodomain clusters and transmits an activating signal to the T lymphocyte. In turn, this signaling triggers the T cell’s effector functions against the target cell, leading to the target cell’s destruction.
An advantage of tumor-speciﬁc T cells that are generated via this mechanism is that they respond to the antigen in a non-MHC-restricted manner. This feature allows the use of CAR T cells in patients with different haplotypes while circumventing tumor escape due to MHC down-regulation.
Furthermore, unlike conventional T cell receptors (TCRs), which are restricted to protein antigens, the range of antigens that can be targeted by single-chain, variable fragment (scFv)-based chimeric receptors also extends to non-classical T cell targets, such as carbohydrate tumor-associated antigens (TAAs).
CAR T-cell therapy has been studied most comprehensively in patients with B cell malignancies and has shown some encouraging early results with success rates ranging from 69% to 90% in pediatric patients with relapsed or refractory acute lymphoblastic leukemia (ALL).
To date, the FDA has approved two CAR T-cell therapies, Kymriah (Novartis) and Yescarta (Kite Pharma). However, with more than 380 CAR T cell-based clinical trials currently in progress, the potential for the approval of additional CAR T-cell therapies seems quite likely.
CAR Structure in More Speciﬁc Detail
THE ECTODOMAIN (EXTRACELLULAR)
Signal peptide: The signal peptide is designed to direct the newly formed, engineered protein into the endoplasmic reticulum. The signal peptide is cleaved from the engineered receptor at the cell’s surface.
The antigen recognition domain: The engineered antigen recognition element is composed of a chimeric protein, scFv, that is derived from the part of an antibody that speciﬁcally binds to the target protein and enables recognition of the human leukocyte antigen (HLA)-independent antigen. The extracellular recognition domains of a CAR T cell provide the targeting function by speciﬁcally recognizing the target antigen. The speciﬁcity of the scFv component is a crucial determinant, which affects the CAR T cell’s safety proﬁle; therefore, antibodies should be selected with extreme care in order to minimize the risk of nonspeciﬁc or off-target effects.
Hinge: The hinge, or spacer, links the scFv to the transmembrane domain and separates the binding moiety from the T cell membrane. Its sequence is usually taken from the hinges of IgG subclasses, IgD, and CD8. The hinge deﬁnes the distance of the scFv from the cell membrane and the epitope-dependent target. The hinge also dictates the ﬂexibility of the scFv, which can affect the performance of the CAR T cell.
THE TRANSMEMBRANE DOMAIN
The transmembrane (TM) domain is essential in the expression of the CARs on the cell surface. A variety of TM sequences were used for this purpose, including CD3ζ, CD4, CD8, and CD28 TM domains. Variable CAR expression is obtained using the different TM domains, with evidence suggesting high expression when the CD28 domain is used.
THE ENDODOMAIN (INTRACELLULAR) SIGNALING DOMAIN
The CAR intracellular signaling domain plays a crucial role in the CAR T cell activation, persistence, and effector functions. A variety of tumor antigens were targeted by CAR T cells, most of them against hematologic malignancies, with CD19 being the most frequently targeted tumor antigen. However, researchers are currently studying many other tumor antigens as possible targets for CAR T-cell therapy.
With the continuous advancements in CAR T cell research, the composition of the endodomain has evolved to improve its functionality. The endodomain of the ﬁrst generation CAR T cells comprised only the CD3ζ chain that generates the primary signal for T cell activation; however, because ﬁrst generation CAR T cells were unable to produce sufﬁcient IL-2, a key cytokine required to propagate T cell activation, the cytokine had to be added externally. Therefore, in order to promote IL-2 production, various sequences from co-stimulatory receptors were added to the intracellular tail, such as CD28, 4-1BB (CD137), and OX40 (CD134). These dual-signal CARs were termed “second generation CARs”.
The third generation of CARs added another co-stimulatory region, combining three signaling domains, such as CD3ζ-CD28-4-1BB, or CD3ζ-CD28-OX40.
Fourth generation CARs, known as T cells redirected for universal cytokine killing (TRUCKs), were created by adding an IL-12 cassette to the base of the second generation endodomain to promote the secretion of IL-12 after the CAR T cell engages its target.
Differences in the intracellular co-stimulatory domain structure affect CAR T cell safety and activity by regulating cytokine production, expansion, cytotoxicity, and its persistence after administration to the patient.
CAR T Cells Immunotherapy
CAR T-cell therapy begins with harvesting the patient’s own T cells. Next, the T cells are activated and genetically modiﬁed by viral transduction in order to express the engineered CAR. The engineered cells are then expanded in vitro, subjected to quality control testing, formulated into the ﬁnished product, and ﬁnally, infused back into the patient, where they combat and destroy the cancer tumor cells.
CAR T cells use a combination of actions to attack cancer cells, including activating immune cells by secreting inﬂammatory cytokines that further promote inﬂammation and immune response, inducing cytotoxic activity by secreting cytotoxins that promote tumor cells apoptosis, and promoting the interleukin-induced development and division of immune cells.
Most of the clinical trials to date have targeted CD19, but many other antigens are now being studied as potential targets of CAR T-cell therapy. Thus far, most CAR T-cell therapy research and clinical trials have been directed against hematological malignancies.
However, recent CAR T-cell therapy is aiming for effective treatment of solid tumors. The major challenge for treatment of solid tumors is identifying speciﬁc antigens that can be targeted. Another important aspect is the efﬁcient direction of the CAR T cells to the target tumor and the persistence and survival of these cells in the tumor microenvironment.
T cell collection and enrichment: Peripheral blood mononuclear cells (PBMCs) are collected from the patient by a procedure called leukapheresis. The cells are then enriched for the CD4 and CD8 T cell subsets, which were found to be the most useful for this kind of therapy. The separation of the CD4 and CD8 cells can be performed by a variety of cell surface marker-based methods, including ﬂuorescence-activated cell sorting (FACS), Immuno-magnetic separation and magnetic activated cell sorting (MACS).
T cell activation: In order to be activated, T cells must become susceptible to viral transduction and expansion. Activation requires dual signaling and can be performed by antigen-presenting cells (APCs) or cell-based artiﬁcial APCs (aAPCs). Most commonly,
T cells are activated by cross-linking CD3 and CD28 by monoclonal antibodies in the presence of IL-2. The antibodies can be soluble, be bound to non-magnetic or magnetic beads, or coat the culture vessel.
Genetic modiﬁcation of the T cells: For an effective CAR T-cell therapy stable expression of the CAR is required. The transduction of the sequence coding for the engineered CAR can be achieved by several gene expression systems, including retroviral vectors, lentiviral vectors, and the transposon/transposase system. Another gene expression method is mRNA transfer, which enables cytoplasmic expression without genomic integration while avoiding long-term expression. Each of these systems has its own strong points, as well as weak points, regarding the simplicity of the manufacturing process and safety issues.
CAR T cell expansion: Effective therapy requires a large number of modiﬁed T cells. Various methods for the expansion of CAR T cells are used, employing various combinations of the cytokines IL-2, IL-7, IL-15, IL-21, and the small molecule TWS119.
Quality control testing: The engineered CAR T cells should pass a set of quality control (QC) tests that ensure the quality of the cells prior to administering the CAR T cell infusion to a patient. These tests include safety assays for sterility, mycoplasma and endotoxin levels, purity, identity, and potency assays. The potency assays can be conducted via in vitro cytotoxic assay or determination of cytokines secretion (IFN-γ, TNF-α, IL-2, 10, and 21) in response to incubation with the target cells.
Several safety concerns have arisen with CAR T-cell therapy, the major ones being “on-target/off-tumor toxicity”, cytokine release syndrome (CRS), and neurotoxicity.
“On-Target/Off-Tumor Toxicity”: This phenomenon occurs when CAR T cells attack healthy tissues that also express the target antigen. For example, B cell aplasia may result when CAR T cells kill normal B cells expressing CD19. In order to avoid this phenomenon, the molecule chosen as the target of the engineered CAR T cells should be as tumor-speciﬁc as possible. Accomplishing this feature often entails reﬁning the afﬁnity and speciﬁcity of the CAR.
“Cytokine Release Syndrome” (CRS): This reaction occurs as the side-effect of a large number of activated T cells releasing pro-inﬂammatory cytokines, such as IL-6, TNF-α, and IFN-γ. The condition can be treated with steroids and cytokine-blocking drugs as well as by controlling the number of infused CAR T cells. Reducing the tumor burden is also helpful, since a direct correlation has been found between the severity of CRS and the tumor burden.
Neurotoxicity: Various neurological symptoms have been observed in patients undergoing CAR T-cell therapy. These include, among others, seizures, delirium, aphasia, and encephalopathy. These symptoms are usually short in duration, not severe, and reversible. However, rare instances of lethal cerebral edema have been reported in some clinical trials. Since the precise mechanism behind the neurotoxicity remains unclear, more research is needed in order to determine the pathophysiology and establish a course of prevention and treatment.
The Future of CAR T Cells Immunotherapy
Scientists are continuously exploring methods to improve the safety and efﬁcacy of CAR T-cell therapy. Additional research is critical to cope with various issues that limit the current use of CAR T-cell therapy, and innovative, next-generation CAR T-cell versions are certain to appear.
Fifth generation CAR T-cell therapy is currently underway. It is based on second generation CAR, with the addition of a cytoplasmic IL-2 receptor chain containing a binding site for the transcription factor STAT3. This new feature conveys an advantage by allowing for antigen binding that will provide all three signals required for a full T cell response.
A variety of strategies for improving the safety of CAR T-cell therapy are being investigated. Controlling the CAR T cells’ activity by incorporating activating or killing switches into the CAR T cells (which can be triggered when needed by small molecules, antibodies and even by ultrasound) can help to minimize the toxic side effects.
Split CAR T cells and inhibitory CAR (iCAR) are special CAR T cells that are speciﬁcally designed to address the “on-target/off-tumor” toxicities. In split CAR T cells, the co- stimulatory domains are separated and placed on two different CARs, each targeting a different tumor-associated antigen. Therefore, the CAR T cell can only be activated when both tumor-associated antigens are engaged; a circumstance not likely to occur in healthy tissues. The iCARs also harbor two CARs. One CAR is speciﬁc to a tumor-associated antigen with activation capacity while the other is speciﬁc to antigens expressed only on normal cells with an inhibitory signaling domain that eliminates the CAR T cell activation when it engages the target normal antigen.
Construction CAR T cells expressing receptors speciﬁc to chemokines, which are secreted either by the tumor cells or by their stromal cells, can ensure that these CAR T cells will travel to the exact location of the target tumor.
Other novel directions that have been explored “off-the-shelf (allogeneic)” CAR T cells that will not require engineering of patient-speciﬁc CAR T cells and in situ reprogramming of T cells, by administering nanoparticles carrying the DNA coding for the desired CAR.
Combining therapies, such as CAR T-cell and immune checkpoint therapies or treating the patient with a combination of several CAR T cells, each with different antigen speciﬁcities, might improve the efficacy of the therapy.