Cell therapy

Cell therapy surpasses gene therapy in precision and durability of therapeutic effect because it engineers autologous (or, increasingly, allogeneic but universally edited) cells ex vivo.

Our ability to engineer large synthetic programs into cells ex vivo is vastly greater than our ability to deliver large multi-kb upgrades through in vivo gene therapy. In addition, you can get quality assurance and validation while in vitro before insertion into human. Or, alternatively, you can also test in vitro, in vivo in animals, etc.

The other advantage of cell therapy over gene therapy is with respect to the immune system. Cell therapy enjoys a friendlier immune profile than gene therapy because it can be engineered ex vivo to cloak or actively modulate immune reactivity (like through HLA matching, universal “off-the-shelf” hypo-immunogenic edits, or other techniques) before the patient ever sees the cells. By contrast, gene therapies must be delivered in vivo by viral (or other nano-object) vectors that instantly present foreign capsid or trans-gene products to the host immune system, which then neutralizes the payload and creates pre-existing or induced immunity that blocks re-dosing. Moreover, any transduced cell becomes a perpetual antigen source, risking cytotoxic T-cell and antibody responses against the corrected tissue, while administrated cells can be equipped with suicide switches or transient immune privilege, allowing tight control of immune engagement.

If an engineered cell needs to secrete a compound for which the immune system is not tolerant, then several things can be done (regardless of whether you are using cell therapy or gene therapy). One of the simplest ways to avoid raising an immune response to a secreted protein is to re-design the molecule so that it is no longer immunologically "foreign".

  1. Molecular humanization: turn a non-human protein into one that looks as if it had always been part of the human proteome as far as the immune system is concerned. For example, "complementarity-determing regions" CDR loop grafting, surface-residues humanization (mutate the positions of the solvent-exposed residues unrelated to the active site such that you instead use the most common residues found in orthologous human proteins, thereby erasing potentially immunogenic epitopes), or "humaneering" algorithms where you align hundreds of human homologues and build a consensus human sequence for the most exposed loops and exchange non-conservative positions to this humaneered pattern, keeping the enzyme or receptor core untouched.

  2. Various camouflage strategies where the immune system is made to believe that the new protein is something that it already tolerates. For example, fusion to an extremely abundant, immunologically inert scaffold such as human serum albumin (HSA). For examply, glycan shielding by introducing new N-linked glycosylation sites onto the exposed, potentially immunogenic faces of the protein, and we hope that those bulky, endogenous sugar trees block direct B-cell receptor contact and reduce proteasomal processing into antigenic peptides. For example, self-marker conjugation where the protein is fused to the Fc region of any human IgG subclass such that the Fc portion signals "self".

  3. Molecular chimeras: very similar to "humanization" as above, but you specifically take an already-tolerated human protein domain (such as a ubiquitous helicase) and embed the cataylic residues of (for example) a minimal Cas nickase inside a region that normally has a different function.

A practical gene-therapy pipeline would often combine these layers: first humanize the drug molecule as far as possible, then fuse the remainder to an innocuous carrier such as albumin or Fc, and finally use codon-optimized, tissue-specific expression cassettes to drop the overall neo-antigen load still further.

immunosuppression is another option that seems to work.

Some other things that can be done:

  1. Immunoprivileged sequestration: express the therapeutic protein within an immune-privileged anatomical compartment such as the anterior chamber of the eye. The physical barrier lowers T-cell priming and antibody neutralization while still allowing efflux of the secreted molecule into the systemic circulation at therapeutically relevant concentrations.

  2. Local immune “stealth” engineering: Use CRISPR or rAAV knock-in to delete or silence MHC-I/II genes in the graft (B2M, CIITA), or co-express PD-L1, CD47, CD200, and HLA-G together with “don’t-eat-me” proteins. Add ectopic B7-H4, VISTA, IL-10, TGF-β, and IDO to create a tolerogenic cytokine niche right at the implantation site.

  3. Secretion of “self” variants: Re-encode the therapeutic peptide into the genomic context of a naturally circulating human protein that circulates at high abundance (e.g., albumin, transferrin, or IgG Fc domains). Fusion to the native sequence keeps plasma antigen load low and prevents recognition as non-self.

  4. Codelivery of local tolerogens: Engineer cells to co-secrete PD-L1-Fc, HLA-E/Fc, or CD200L-Fc fusion proteins, or release biodegradable micro-particles containing rapamycin, dexamethasone, or anti-CD3 scFv to convert the pericellular milieu into an active site of immune tolerance rather than inflammation.

  5. Suicide-switch insurance: Embed inducible drug-responsive (iCasp9, truncated EGFR antibody recognition) or small-molecule regulated (HCV-NS3, FKBP-casp9) suicide mechanisms into the cell product so that if robust rejection is detected despite the above strategies, the graft can be promptly ablated with minimal systemic toxicity.

  6. Transient expression via mRNA delivered by electroporation or otherwise: Limit the duration of exposure to the antigenic construct by using non-integrating payloads that decay over days-to-weeks—long enough for acute therapeutic benefit, short enough to avert chronic immune recognition.

  7. Antigen-camouflage glyco-shielding: Glyco-engineer the cells to overexpress sialyl-transferases (ST3G4/6Gal1) and CD47 so their glycocalyx hides or degrades nascent neo-peptides, and simultaneously blocks the uptake by antigen-presenting cells.

  8. Tolerogenic vaccination before grafting: Administer a “prime-and-tolerize” protocol where the patient receives tolerogenic dendritic cells prepared with the therapeutic protein (or key peptide epitopes) plus rapamycin and IL-10 in order to drive antigen-specific Treg expansion prior to cell implantation.

  9. HLA-matched “off-the-shelf” universal lines: For allogeneic secreting cells, derive clonal hypo-immunogenic master banks using CRISPR-mediated triple KO (B2M, CIITA, CD155) plus transgenic over-expression of NK-inhibitory ligands (HLA-E, non-polymorphic β2m fusion, CMV UL-18 domain). This minimizes both CD8+ T- and NK-mediated rejection even when robust antigen is released in the long term.

These strategies are modular and combinations might be able to achieve layered immunity control while still permitting sustained systemic secretion of the therapeutic compound.

CAR T cell therapy

In vivo CAR T cell generation to treat cancer and autoimmune disease -- a gene-delivery system to generate CAR-T cells in vivo by dosing of a CD8-targeted lipid nanoparticle carrying anti-CD19 CAR mRNA.

various autologous stem cell therapy methods

xenotransplantation, transplants, implants

Cellular implants are another method of delivering gene therapy. Of particular interest is that this method does not really require DNA/mRNA uptake by other cells. You insert new cells and then those new cells do a new job. Ideally these are immunosilent or immunocompetent cells.

other cell therapy

trogocytosis: Programmable macromolecule delivery via engineered trogocytosis

Engineered bacteria launch and control an oncolytic virus

More cell therapy

immunocompetence/MHC engineering, CAR innovations, trogocytosis & endocytosis/trafficking, synthetic immune-system engineering, and alternative effector cell therapies.

Cell therapy overviews

  • Big-picture state of the field & where cell therapies are going (engineering lenses): Bashor et al., Nat Rev Drug Discov (2022). bashorlab.rice.edu
  • Perspective on “living drugs” & core design trade-offs: Irvine, Science (2022). Science
  • Broad 2025 overview of adoptive cell therapy & genetic enhancements (solid tumors focus): Albarrán-Fernández et al., Nat Commun (2025). Nature
  • Current challenges & directions in CAR-T (concise 2025 review): Zugasti et al., Signal Transduct Target Ther (2025). Nature
  • Universal donor cells & immune cloaking (review): Simpson, Front Bioeng Biotechnol (2023). PMC
  • Hypoimmune iPSC products surviving long-term in NHPs: Hu et al., Nat Biotechnol (2024). Nature
  • Trogocytosis driving antigen escape in CAR-T: Hamieh et al., Nature (2019). Nature
  • Mechanistic update on CAR trogocytosis (reverse transfer, 2023): Zhai et al., Signal Transduct Target Ther (2023). Nature
  • Receptor dynamics/trafficking knob for CAR function (internalization/recycling): Xie et al., Front Immunol (2025). Frontiers
  • Synthetic biology for specificity/safety control in engineered immune cells: Zhu et al., Cell Mol Immunol (2024). Nature

immunocompetence & MHC engineering (toward universal/off-the-shelf cells)

  • Review of strategies (B2M/CIITA KO; HLA-E/-G “camouflage”; CD47; kill-switches): Simpson (2023). PMC
  • Hypoimmune (B2M−/−, CIITA−/−, CD47+) iPSCs/islets persist in fully allogeneic primates (proof of immunocompetence in vivo): Hu et al. (2024). Nature
  • Cell-therapy–oriented summary of immune-evasive editing & translational hurdles: Deuse & Schrepfer, Cell Stem Cell (2025). Cell
  • Practical review of HLA editing (B2M/CIITA disruption; HLA-E rescue to deter NK “missing-self”): Choi et al., Stem Cell Res Ther (2025). BioMed Central
  • Perspective on allogeneic CAR platforms (TRAC KO, HLA camouflage, IL-15 armoring): Fang et al. (2025). PMC
  • Foundational hypoimmune hPSC editing targeting adaptive & innate arms: Han et al., PNAS (2019). PNAS

CAR-T design innovations (emerging techniques)

  • State-of-the-art in CAR-T (solid tumors & AML), with engineering levers (costims, armored, TRUCKs): Zugasti et al. (2025). Nature
  • “Proximal-signaling” CARs (e.g., ZAP-70 CARs) for alternative activation logic: Tousley et al., Nat Biomed Eng (2023). PubMed
  • Adapter/switchable CARs (retargetable; dosing of activity): Nixdorf et al., Leukemia (2023). Nature
  • Covalent “SNAP-CAR / SNAP-synNotch” receptors programmable by tagged antibodies: Ruffo et al., Nat Commun (2023). Nature
  • Direct in-vivo CAR-T engineering: Short & Rafiq, Trends Biotechnol (2024). ScienceDirect

trogocytosis (antigen transfer) & endocytosis/trafficking

other immune focused trogocytosis articles:

  • Antigen transfer from tumor→T cell drives antigen-low escape, fratricide, exhaustion (seminal): Hamieh et al., Nature (2019). Nature
  • Reverse direction: tumors trogocytose CAR molecules, depleting CARs & masking antigens: Zhai et al., Signal Transduct Target Ther (2023). Nature
  • Focused review of trogocytosis in CAR immunotherapy (mechanisms & countermeasures): Chen et al., Front Immunol (2024). PMC
  • News & perspectives on the “trogocytosis front” (2025 update): Barbera, Trends Immunol (2025). Cell
  • CAR receptor internalization/recycling as a tunable axis (less lysosomal routing → less exhaustion): Xie et al., Front Immunol (2025). Frontiers
  • Review on CAR surface density/dynamics affecting endocytosis & signaling (2025): Hinckley-Boned et al. (2025). PMC
  • Biophysics advance: CAR phase separation tunes sensitivity/persistence: Xu et al., Immunity (2024). Cell

synthetic immune-system engineering

  • How synthetic biology sharpens specificity & control in cell therapies: Zhu et al., Cell Mol Immunol (2024). Nature
  • Survey of programmable/synthetic receptors beyond classical CARs: Teng et al. (2024). Nature
  • SynNotch→CAR two-step circuits for multi-antigen precision (review): Shirzadian et al., Front Immunol (2025). PMC
  • synNotch as a general engineered receptor platform (mechanisms & therapeutic cell types): Piraner et al., Nature (2025). Nature
  • Tissue-sensing synNotch circuits that deliver therapeutic payloads in vivo: Simic et al. (UCSF/Lim Lab preclinical). limlab.ucsf.edu

beyond αβ T cells: CAR-NK, CAR-macrophage, CAR-Treg

  • CAR-T vs CAR-NK: side-by-side engineering & clinical progress (solid tumor emphasis): Peng et al., Cell Mol Immunol (2024). Nature
  • CAR-NK directions & advantages (safety, innate recognition): Zhong et al., Cell Death Discov (2024). Nature
  • CAR-macrophages—mechanisms, early trials, and solid-tumor rationale: Lu et al., Biomarker Res (2024); plus Morva et al., Front Immunol (2025). BioMed Central, Frontiers
  • Engineered Tregs (CAR-Treg) for tolerance/autoimmunity: Bittner et al., Trends Immunol (2023); Bulliard et al., Front Immunol (2024). Cell, PMC
  • Disease-specific CAR-Treg example (MS model, MOG-CAR-Tregs): Frikeche et al., J Neuroinflammation (2024). BioMed Central

more

  • Historical review and approvals up to 2023: Mitra et al., J Transl Med (2023). PMC
  • Monitoring/persistence & antigen loss tech landscape: Chen & Ma, Am J Pathol (2024). ScienceDirect

MHC editing strategies

goal: better resist the host immune system, like in allogeneic cell therapies or universal donor stem cells.

MHC (HLA) Editing Techniques

See gene editing for more gene editing techniques.

Synthetic immune checkpoint engagers protect HLA-deficient iPSCs and derivatives from innate immune cell cytotoxicity

Direct MHC Disruption

Knocking out MHC class I and/or II expression in therapeutic cells reduces visibility to CD8⁺ T cells. Since MHC-I isn't essential for cell survival, many cancers naturally employ this as an immune-evasion tactic ScienceDirect, Frontiers.

Conversely, this tactic risks triggering natural killer (NK) cells, which attack cells lacking MHC-I—termed “missing self.”

Stealth Transgenes

“Stealth” transgene strategies target both MHC class I and II to evade host immune detection. Early studies show promise in engineered T cells successfully avoiding cellular immune responses Journal of Infection and Chemotherapy.

Viral-Strategy Mimicry

Many viruses have evolved to inhibit MHC antigen presentation by:

  • Blocking peptide entry via TAP inhibitors
  • Retaining MHC-I in the endoplasmic reticulum (ER) or Golgi apparatus
  • Directing MHC for degradation via ER-associated pathways Frontiers.

Therapeutic cells imght be able to mimic immune stealth we see in some viruses by co-opting these mechanisms with targeted gene edits, like by modifying antigen-processing pathways.

Immune-Modulating Receptor Engineering

Alloimmune Defense Receptor (ADR)

In alloimmune defense receptor (ADR) strategies, therapeutic T cells are empowered to eliminate activated host T and NK cells while sparing resting immune cells. By targeting the avictation marker 4‑1BB, ADR-expressing T cells resist rejection and persist in vivo when co-expressed with CAR constructs PMC.

Stem Cells & Allogeneic Grafts: Toward Universal Donors

Gene editing of pluripotent stem cells (iPSCs) to make "stealthy" universal donor cells for cell replacement therapies that need to avoid immune attack Nature. Strategies include MHC knockout combined with immune shielding techniques to evade T and NK cells.

Epigenetic & Transcriptional Modulation

Reactivating MHC or Downregulating Detection Pathways*

  • Tumors often downregulate MHC via epigenetic suppression or via downregulating key transcription factors like NLRC5 Frontiers.
  • CRISPR–dCas9 epigenetic tools can potentially be used to fine-tune gene expression, such as either dampening antigen presentation in therapeutic cells or, in broader contexts, restoring it during immunotherapy ScienceDirect.

Emerging strategies for immmune system evasion

  • Partial MHC Retention + Non-classical HLA: Completely removing classical MHC risks NK detection. A nuanced approach might knock down classical MHC while expressing non-classical or inhibitory HLA molecules to suppress NK activity, akin to tumor evasion tactics Wikipedia.
  • NK Checkpoint Modulation: Therapeutic cells might express ligands that bind NK inhibitory receptors (like HLA-E to interact with KIRs), or modulate KIR pathways to escape NK-mediated killing Nature.
  • Local Immune Suppression: Engineering cells to secrete immunosuppressive cytokines (e.g., TGF-β) selectively could locally dampen the host immune response while minimizing systemic impact.
  • Mapping Virus-Derived Peptides: Incorporating viral peptide sequences that interfere with MHC pathways—as viruses do—for controlled immune evasion, though highly speculative and risky.
  • Combining ADR and MHC Editing: Pairing MHC disruption with ADR-type systems might create "immune-evasive but immune-persistent" cells that both hide and defend.

Summary table

Strategy Description Pros Cons / Risks
MHC (HLA) knockout Remove MHC-I/II to evade T-cell detection Strong stealth NK-mediated rejection
Stealth transgenes Modulate both MHC classes Broader evasion Complex engineering
Viral mimicry (TAP, retention, etc.) Disrupt antigen presentation akin to viral proteins Highly efficient stealth Potential dysfunction, off target
ADR (alloimmune defense receptor) Kill activated host T/NK cells via 4‑1BB targeting Active defense, persistence Target specificity, immune dysregulation
iPSC “stealth” editing Develop universal donor cells Wide therapeutic potential Needs multi-path evasion strategies
Epigenetic modulation (CRISPR-dCas9) Fine-tune MHC-related genes Reversible, tunable Off-target effects, stability issues
NK checkpoint modulation Express NK inhibitory signals (e.g., HLA-E) or target KIRs Evade NK scrutiny Over-inhibition risks
Local immunosuppression Local cytokine release (e.g., TGF-β) Context-specific tolerance Risk of local immune suppression
Combined ADR + MHC editing Hide and defend simultaneously Maximal persistence tactics Highest complexity and risk

MHC and HLA references

Multi-layered approach for immune evasion or resistance:

  1. Hide — Edit out or modify MHC molecules.
  2. Defend — Engineer immune-targeting receptors like ADR.
  3. Suppress — Modulate local or systemic immunity.
  4. Balance — Evade both T cells and NK cells by blending classical and non-classical signals.
  5. Leverage Epigenetics — Use precision control over immune-related gene expression.
  6. Combine — Stack strategies for robust, long-lived therapeutic cells.