CAR-T Cell Therapy: Structure, Manufacturing, and FDA-Approved Products

CAR-T cell therapy has achieved complete remissions in patients with refractory acute lymphoblastic leukemia (ALL), diffuse large B-cell lymphoma (DLBCL), and multiple myeloma — cancers that had no durable treatment options — by engineering a patient's own T cells to express a chimeric antigen receptor (CAR) that directly recognizes and destroys tumor cells. Six CAR-T products are now FDA-approved, and the field continues to expand into solid tumors and allogeneic manufacturing. For a broader overview of immunotherapy approaches including checkpoint inhibitors, see our cancer immunotherapy overview.

CAR Structure and Molecular Design

The chimeric antigen receptor (CAR) is a synthetic fusion protein that grafts antigen-recognition capability onto a T cell without requiring MHC-peptide presentation. Understanding its domain architecture explains why design choices downstream determine clinical efficacy and durability.

Extracellular antigen-binding domain. Most approved CARs use a single-chain variable fragment (scFv) derived from a monoclonal antibody, linking the variable heavy (VH) and variable light (VL) chains via a flexible (Gly4Ser)3 linker. Alternative binding modules include nanobodies (single-domain VHH camelid antibodies), receptor ligands, and DARPins, each with distinct epitope-access and stability profiles.

Hinge and spacer. The hinge connects the scFv to the transmembrane domain and controls how far the antigen-binding domain extends from the T cell surface. Common spacers include IgG4 CH2CH3 (long, flexible), CD8α hinge (short, rigid), and CD28 hinge. Spacer length is epitope-proximal — membrane-distal epitopes on short targets (e.g., CD19) work best with short spacers; membrane-proximal epitopes (e.g., on BCMA) prefer longer spacers.

Transmembrane domain. The TM domain anchors the CAR in the plasma membrane. CD28 TM domains promote receptor clustering and elevated tonic signaling; CD8α TM domains produce a quieter baseline. The TM choice influences exhaustion kinetics during prolonged stimulation.

Intracellular signaling domain. The CD3ζ chain carries three immunoreceptor tyrosine-based activation motifs (ITAMs) and is required for T cell activation. First-generation CARs carrying CD3ζ alone showed poor in vivo persistence.

Costimulatory domains and generations. Second-generation CARs add a single costimulatory domain — either CD28 or 4-1BB (CD137, TNFRSF9) — upstream of CD3ζ. CD28 costimulation drives rapid expansion and high early cytolytic activity but is more prone to T cell exhaustion. 4-1BB costimulation (used in tisagenlecleucel, ide-cel, and cilta-cel) drives slower but more sustained expansion and longer in vivo persistence, which is particularly important in multiple myeloma. Third-generation CARs combine two costimulatory domains (e.g., CD28 + 4-1BB) but have shown no consistent advantage over second-generation CARs in clinical trials. Fourth-generation "armored" CARs encode transgenic payloads — cytokines such as IL-12 or IL-15, or costimulatory ligands — that are secreted or displayed upon antigen engagement to remodel the tumor microenvironment.

The foundational CAR design framework was described by Sadelain et al., Nature Reviews Cancer, 2013.

CAR-T Manufacturing: From Leukapheresis to Infusion

Autologous CAR-T manufacturing is a six-step process that converts a patient's own blood cells into a personalized cellular medicine. Each step introduces variability that affects potency and persistence of the final product.

1. Leukapheresis. Peripheral blood mononuclear cells (PBMCs) are collected from the patient by apheresis, typically yielding 5–15 × 10⁸ T cells. Product quality correlates with CD4:CD8 ratio and the proportion of naive/stem cell memory T cells in the starting material — heavily pre-treated patients with exhausted T cell repertoires present the greatest challenge.
2. T cell enrichment and activation. T cells are purified by immunomagnetic selection, then activated with anti-CD3/CD28 magnetic beads or soluble OKT3 + retronectin on fibronectin-coated vessels for 2–3 days. Activation upregulates retroviral/lentiviral entry receptors and induces proliferation.
3. Viral transduction. The CAR transgene is delivered by a replication-incompetent gammaretroviral (RV) or lentiviral (LV) vector at a multiplicity of infection (MOI) of approximately 5–10. Lentiviral vectors are preferred in current approved products (tisa-cel, liso-cel, ide-cel, cilta-cel) because semi-random integration into transcriptionally active regions and the absence of strong LTR enhancers reduce insertional mutagenesis risk relative to gammaretroviral vectors; they also transduce non-dividing cells more efficiently.
4. Expansion. Transduced T cells are expanded in a closed bioreactor system — G-Rex flasks for smaller batches, rocking-motion bioreactors (WAVE) for larger-scale manufacturing — for 7–10 days with IL-2 and/or IL-7/IL-15 supplementation.
5. Formulation and cryopreservation. Cells are harvested, washed to remove residual vector and cytokines, concentrated, and resuspended in a DMSO-based cryoprotectant medium for storage in liquid nitrogen. The final product can be shipped frozen to the treating center.
6. Lymphodepletion conditioning. Before infusion, the patient receives 3–5 days of cyclophosphamide + fludarabine to deplete regulatory T cells and create immunological space (homeostatic cytokines) that support engraftment and expansion of the infused CAR-T cells.

The total manufacturing timeline from leukapheresis to product release is typically 2–4 weeks. A detailed description of GMP-compliant manufacturing workflows appears in Levine et al., Human Gene Therapy, 2017.

Allogeneic ("off-the-shelf") CAR-T. Allogeneic products are manufactured from healthy donor T cells, offering immediate availability without per-patient manufacturing lead time. To prevent graft-versus-host disease (GvHD), the endogenous T cell receptor alpha constant (TRAC) gene is knocked out (typically by CRISPR-Cas9). Beta-2-microglobulin (B2M) knockout removes MHC class I expression to reduce host-versus-graft rejection. Several Phase I allogeneic programs are in clinical trials (e.g., CRISPR Therapeutics CTX110, Allogene ALLO-501A); as of 2026, no allogeneic CAR-T product has received FDA approval.

FDA-Approved CAR-T Products

The FDA has approved six autologous CAR-T products across B-cell malignancies and multiple myeloma. All six are CD3ζ-based second-generation CARs. The table below summarizes each product's target, indication, costimulatory domain, and approval year.

r/r = relapsed or refractory; ALL = acute lymphoblastic leukemia; DLBCL = diffuse large B-cell lymphoma; FL = follicular lymphoma; PMBCL = primary mediastinal B-cell lymphoma; MCL = mantle cell lymphoma; LBCL = large B-cell lymphoma; CLL/SLL = chronic lymphocytic leukemia/small lymphocytic lymphoma; MM = multiple myeloma. BCMA = B-cell maturation antigen (TNFRSF17/CD269), the primary target on malignant plasma cells in multiple myeloma.

Note that brexucabtagene autoleucel (Tecartus) shares its CD19-targeting construct with axicabtagene ciloleucel (Yescarta) but underwent a separate manufacturing validation for MCL and adult ALL because the starting T cell product differs in these indications.

CAR-T Toxicities and Clinical Management

The potency of CAR-T cells against target-bearing tumors is accompanied by two characteristic on-target toxicity syndromes. Understanding their distinct mechanisms drives different management algorithms.

Cytokine Release Syndrome (CRS)

CRS is the most common serious toxicity, occurring in 40–90% of patients to some degree. The mechanism: CAR-T cells engage antigen-positive tumor cells and release IFN-γ. IFN-γ activates macrophages, which secrete large quantities of IL-6, IL-1β, and TNF-α in a feed-forward loop, producing systemic inflammation. Clinical features include fever (onset within 1–14 days of infusion), hypotension, and hypoxia. Grading follows the ASTCT (American Society for Transplantation and Cellular Therapy) 2019 consensus scale: Grade 1 (fever alone) through Grade 4 (life-threatening organ dysfunction). Because IL-6 is the principal amplifier, tocilizumab (anti-IL-6 receptor monoclonal antibody) is first-line management for Grade ≥2 CRS. Corticosteroids are added for refractory or high-grade CRS. Early tocilizumab administration has been shown not to impair CAR-T efficacy.

Immune Effector Cell-Associated Neurotoxicity Syndrome (ICANS)

ICANS (formerly called CAR-T-related encephalopathy syndrome, CRES) is mechanistically distinct from CRS. Activated monocytes and T cells disrupt the blood-brain barrier, allowing inflammatory cytokines — particularly IL-6, IL-1β, and IFN-γ — to enter the CNS. Clinical manifestations range from mild confusion and aphasia (Grade 1–2) to seizures, cerebral edema, and life-threatening encephalopathy (Grade 4–5). Management is primarily corticosteroid-based (dexamethasone), and tocilizumab is less effective for ICANS because IL-6 signaling in the CNS is driven by trans-signaling via soluble IL-6R rather than membrane IL-6R. ICANS and CRS can co-occur but their trajectories differ — CRS typically precedes ICANS onset. The ASTCT 2019 consensus grading and management recommendations are described in Neelapu et al., Nature Reviews Clinical Oncology, 2018.

Other Toxicities

Prolonged cytopenias arise from lymphodepletion conditioning combined with on-target depletion of normal B cells (B cell aplasia from CD19-targeting CARs) or hematopoietic suppression. B cell aplasia from CD19-targeting products requires immunoglobulin replacement therapy. Patients are also at elevated infection risk during the aplasia window, particularly from bacterial, fungal, and viral pathogens. Monitoring involves serial complete blood counts, immunoglobulin levels, and cytokine panels (IL-6, IFN-γ, ferritin) post-infusion.

Next-Generation CAR-T Engineering

Single-antigen CAR-T products face a fundamental limitation: antigen escape, in which tumor cells that downregulate or lose the target antigen evade killing. The next generation of CAR-T designs addresses this and other barriers through combinatorial engineering.

Dual-targeting CARs. Bicistronic constructs encode two CARs on a single vector — for example, a tandem CD19+CD22 CAR for B-ALL or a BCMA+CD38 combination for myeloma. Targeting two antigens simultaneously raises the threshold for escape because both antigens must be lost.

Logic-gated CARs. AND-gated CARs require co-expression of two antigens on the target cell for full activation, improving solid-tumor selectivity. Inhibitory CARs (iCARs) expressing an inhibitory receptor (e.g., PD-1 or CTLA-4 intracellular domain) that recognizes a healthy-tissue antigen suppress activation when off-target cells are encountered, acting as a safety switch.

Armored CARs (TRUCKs). "T cells redirected for universal cytokine killing" (TRUCKs) express an inducible cytokine cassette — typically IL-12 or IL-15 — that is released upon antigen engagement. Constitutive IL-12 secretion enhances macrophage recruitment and NK cell activity in the tumor microenvironment, partially countering TGF-β-mediated immunosuppression.

In vivo CAR-T generation. The most disruptive emerging concept is delivery of the CAR gene directly into the patient's T cells in vivo using ionizable lipid nanoparticles (LNPs) loaded with CAR-encoding mRNA or non-integrating DNA. This approach would eliminate the need for ex vivo manufacturing entirely. Several programs are in early preclinical to Phase I stages.

Solid tumor barriers. CAR-T cells remain largely ineffective in solid tumors owing to antigen heterogeneity, poor trafficking to the tumor bed, and the immunosuppressive tumor microenvironment (TGF-β, adenosine, PD-L1 expression). Next-generation armored and logic-gated designs are specifically aimed at overcoming these barriers.

BioHippo Research Tools for CAR-T Studies

Supporting CAR-T cell therapy research requires reagents across multiple assay types — from cytokine quantification to target-antigen detection to effector function analysis. BioHippo stocks validated tools for each stage.

Target antigen quantification. BCMA (TNFRSF17/CD269) ELISA kits are essential for monitoring soluble BCMA shedding, which correlates with disease burden in multiple myeloma and can affect cilta-cel and ide-cel efficacy. BioHippo offers the Human TNFRSF17/BCMA PicoKine® ELISA Kit (sensitivity <10 pg/mL, sandwich format) and the Human BCMA EZ-Set™ DIY Antibody Pairs for high-throughput custom panel development. Browse all ELISA kits →

Cytokine monitoring (CRS panel). Post-infusion cytokine monitoring is standard of care. BioHippo's Human IL-6 ELISA Kit EZ-Set™ and Human IFN-γ PicoKine® ELISA Kit (sensitivity <2 pg/mL) are suitable for serum and plasma samples in the CRS monitoring window.

Cytotoxicity markers. Perforin and granzyme B are the primary effector molecules released by activated CAR-T cells upon target engagement. BioHippo carries anti-Perforin antibodies validated for ELISA and flow cytometry. Browse all primary antibodies →

CAR-T model cell lines. Raji cells (Burkitt lymphoma, CD19+) and MM.1S cells (multiple myeloma, BCMA+) are the standard in vitro target models for CD19- and BCMA-directed CAR-T killing assays respectively. BioHippo's Raji cell line and MM.1S cell line are available as cryopreserved stocks.

BCMA recombinant proteins. BioHippo's Recombinant Human CD269/TNFRSF17/BCMA Protein is available as His-tagged formats from E. coli and mammalian expression for use as ELISA standard, immunogen, or binding-assay ligand.

Frequently Asked Questions about CAR-T Cell Therapy

What is CAR-T cell therapy?

CAR-T cell therapy is a form of adoptive cell therapy in which a patient's own T cells are genetically engineered to express a chimeric antigen receptor (CAR) — a synthetic surface protein that directs the T cell to find and kill tumor cells bearing a specific antigen. Unlike conventional T cell activation, CARs bypass the requirement for MHC-peptide antigen presentation, allowing direct recognition of native surface proteins on cancer cells. The engineered cells are expanded in the laboratory and re-infused into the patient after lymphodepleting chemotherapy.

How is CAR-T manufactured?

CAR-T manufacturing proceeds in six steps: (1) leukapheresis to collect T cells from the patient's blood; (2) T cell activation using anti-CD3/CD28 beads or OKT3; (3) CAR gene delivery via a replication-incompetent lentiviral or retroviral vector; (4) ex vivo expansion in a bioreactor for 7–10 days; (5) formulation and cryopreservation; and (6) patient lymphodepletion with cyclophosphamide + fludarabine before infusion. The total process takes 2–4 weeks from leukapheresis to product release.

What cancers can be treated with CAR-T therapy?

All six FDA-approved CAR-T products target blood cancers. CD19-directed CARs (tisagenlecleucel, axicabtagene ciloleucel, lisocabtagene maraleucel, brexucabtagene autoleucel) treat relapsed or refractory B-cell ALL, diffuse large B-cell lymphoma, follicular lymphoma, and mantle cell lymphoma. BCMA-directed CARs (idecabtagene vicleucel, ciltacabtagene autoleucel) treat relapsed or refractory multiple myeloma. Solid tumors remain an active area of research with no approved products yet.

What are the side effects of CAR-T therapy?

The two most important toxicities are cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS). CRS presents as fever, hypotension, and hypoxia within days of infusion and is managed with the anti-IL-6R antibody tocilizumab. ICANS causes encephalopathy, confusion, and aphasia and is managed with corticosteroids. Additional risks include prolonged cytopenia, B cell aplasia requiring immunoglobulin replacement, and opportunistic infections during the aplasia window. Most CRS and ICANS events are Grade 1–2 and resolve with management; Grade 3–4 events occur in approximately 20–30% of patients depending on tumor burden and the specific CAR-T product.

What is the difference between autologous and allogeneic CAR-T?

Autologous CAR-T cells are manufactured from the patient's own T cells, eliminating the risk of graft-versus-host disease (GvHD) but requiring 2–4 weeks of per-patient manufacturing. All six currently FDA-approved CAR-T products are autologous. Allogeneic ("off-the-shelf") CAR-T cells are manufactured from healthy donor T cells, offering immediate availability. To prevent GvHD, donor T cells undergo CRISPR-mediated knockout of the endogenous T cell receptor (TRAC locus). To reduce rejection by the host immune system, beta-2-microglobulin is knocked out to eliminate MHC class I surface expression. Multiple allogeneic programs are in Phase I trials; none are FDA-approved as of 2026.