Chimeric Antigen Receptor T-cell (CAR-T) therapy has produced some of the most striking responses ever seen in oncology — particularly in B-cell malignancies. Six FDA-approved CAR-T therapies are now standard care for relapsed or refractory leukemias, lymphomas, and multiple myeloma.
The basic concept
T cells normally recognize antigens presented on MHC by other cells. Cancer cells often evade T-cell recognition by downregulating MHC. CAR-T therapy bypasses this by giving T cells a synthetic receptor that recognizes a surface antigen directly — without MHC presentation.
What’s in a CAR
A chimeric antigen receptor is a single fusion protein with three main domains:
- Extracellular antigen-binding domain: Single-chain variable fragment (scFv) derived from an antibody that recognizes the target antigen (e.g., CD19 for B-cell cancers, BCMA for multiple myeloma)
- Hinge and transmembrane domains: Connect extracellular and intracellular components
- Intracellular signaling domains: Activate the T cell when the antigen is engaged. Modern (2nd/3rd-generation) CARs include CD3ζ plus one or two costimulatory domains (CD28, 4-1BB, OX40)
The manufacturing workflow
- Leukapheresis: T cells are collected from the patient’s blood
- T-cell selection and activation: CD3/CD28 beads activate the cells in vitro
- Gene transfer: The CAR construct is introduced via lentivirus, retrovirus, or non-viral methods (transposon, mRNA, CRISPR knock-in)
- Expansion: Cells are cultured for 7–14 days to reach therapeutic numbers (often hundreds of millions to billions)
- QC and release testing: Sterility, potency, identity, viability
- Patient lymphodepletion: Conditioning chemotherapy to make space and reduce regulatory T cells
- Infusion: CAR-T cells are infused back into the patient
The full process typically takes 3–4 weeks from collection to infusion.
Approved CAR-T therapies
| Brand | Target | Indication |
|---|---|---|
| Kymriah | CD19 | ALL, DLBCL, follicular lymphoma |
| Yescarta | CD19 | DLBCL, follicular, mantle cell |
| Tecartus | CD19 | Mantle cell, ALL |
| Breyanzi | CD19 | DLBCL, CLL |
| Abecma | BCMA | Multiple myeloma |
| Carvykti | BCMA | Multiple myeloma |
The major toxicities
- Cytokine release syndrome (CRS): Massive cytokine release as CAR-T cells engage targets. Ranges from fever to life-threatening shock. Managed with tocilizumab (IL-6 receptor blocker) and supportive care
- Immune effector cell-associated neurotoxicity syndrome (ICANS): Confusion, aphasia, seizures. Mechanism incompletely understood
- B-cell aplasia: CD19-CAR-T eliminates normal B cells along with malignant ones — patients require IVIG replacement
- Cytopenias and infections from prolonged immune dysfunction
Why CAR-T works for blood cancers but not solid tumors (yet)
- Antigen heterogeneity: Solid tumors have diverse antigen expression, leading to escape
- Immunosuppressive microenvironment: Tumor microenvironment dampens T-cell function
- Trafficking and persistence: CAR-T cells must reach the tumor and persist long enough to kill it
- On-target, off-tumor toxicity: Few solid tumor antigens are absent from normal tissue
Next-generation directions
- Allogeneic (“off-the-shelf”) CAR-T: Universal donor cells edited with CRISPR to remove TCR and HLA, eliminating the manufacturing wait
- Armored CARs: CAR-T cells that secrete cytokines (e.g., IL-12, IL-18) to remodel the tumor microenvironment
- Logic-gated CARs: AND-, OR-, and NOT-gated systems for solid tumor specificity
- CAR-NK and CAR-macrophage therapies: Alternative effector cells with different toxicity profiles
- In vivo CAR generation: mRNA-LNPs that turn T cells into CAR-T inside the patient, eliminating the manufacturing step entirely
CAR-T has already redefined what’s possible in cancer therapy. The next decade will determine whether the approach generalizes to solid tumors and to autoimmune disease, where early trials are showing intriguing results.
