Compendium

Immunoengineering and Cell Therapy

22 min read

Immunoengineering and Cell Therapy

A Tier 2 specialty bridging immunology, synthetic biology, and clinical translation. Immunoengineering treats the immune system as a programmable substrate: T cells reprogrammed with chimeric antigen receptors to recognize and kill cancer; NK cells, macrophages, and γδ-T cells armored against the immunosuppressive tumor microenvironment; antibodies redesigned as bispecific T-cell engagers; cytokines masked or split for tumor-restricted activity; checkpoint blockades to release exhausted T cells. The field’s commercial milestone was Kymriah (tisagenlecleucel; Novartis; FDA approval 30 Aug 2017) for CD19+ acute lymphoblastic leukemia — the first FDA-approved gene therapy of any kind and the first CAR-T. By 2026 the field counts six approved CAR-T products, three TCR-T and one TIL therapy, more than 15 bispecific antibodies, and a dozen checkpoint inhibitors generating combined annual revenue above $80B. This note covers CAR design, allogeneic vs autologous platforms, in vivo CAR generation, the broader cell-therapy modalities, bispecifics and trispecifics, checkpoint biology, engineered cytokines, microbiome immunotherapy, vaccine adjuvants, and the toxicity-management framework that keeps these therapies usable.

See also

CAR-T cell therapy — the approved landscape

CAR-T (chimeric antigen receptor T cell) therapy reprograms patient T cells ex vivo with a synthetic receptor combining an extracellular scFv (single-chain variable fragment from a monoclonal antibody), a hinge/transmembrane domain (CD8α or CD28), one or more intracellular costimulatory domains (CD28, 4-1BB/CD137, OX40, ICOS), and the CD3ζ ITAM signaling tail. Engagement of the scFv with tumor surface antigen triggers MHC-independent T-cell activation, proliferation, cytokine secretion, and target cell lysis.

CD19-directed (B-cell malignancies)

  • Kymriah (tisagenlecleucel; Novartis). FDA approved 30 Aug 2017 for pediatric/young-adult relapsed/refractory B-ALL; expanded May 2018 to adult r/r DLBCL; May 2022 to r/r follicular lymphoma. CD19-targeted scFv (FMC63 clone) + 4-1BB + CD3ζ; lentiviral delivery. List price $475k (B-ALL), $373k (DLBCL). Pivotal ELIANA trial — 81% complete remission (NCT02435849). Manufactured at Morris Plains NJ and Les Ulis France.
  • Yescarta (axicabtagene ciloleucel; Kite/Gilead). FDA approved 18 Oct 2017 for adult r/r DLBCL; April 2022 second-line DLBCL (ZUMA-7 NCT03391466 — superseded autoHSCT); March 2021 follicular. Same FMC63 scFv but CD28 + CD3ζ costimulation; retroviral. List price $373k. CD28 architecture gives faster expansion + earlier CRS than 4-1BB.
  • Tecartus (brexucabtagene autoleucel; Kite/Gilead). FDA approved 24 July 2020 for r/r mantle cell lymphoma; Oct 2021 adult r/r B-ALL. Same CAR construct as Yescarta but with an in-process T-cell enrichment step to remove circulating tumor cells.
  • Breyanzi (lisocabtagene maraleucel; Bristol-Myers Squibb / Juno). FDA approved 5 Feb 2021 for r/r LBCL; June 2022 second-line LBCL; March 2024 r/r CLL/SLL. 4-1BB + CD3ζ; defined CD4:CD8 1:1 ratio composition.
  • Aucatzyl (obecabtagene autoleucel, obe-cel; Autolus). FDA approved 8 Nov 2024 for adult r/r B-ALL. Fast off-rate CD19 binder (CAT-41BBz) reducing CRS severity; pivotal FELIX trial 76% CR/CRi.

BCMA-directed (multiple myeloma)

  • Abecma (idecabtagene vicleucel; BMS / 2seventy bio). FDA approved 26 Mar 2021 for r/r MM after 4+ prior lines; April 2024 third-line. Anti-BCMA scFv + 4-1BB + CD3ζ; lentiviral.
  • Carvykti (ciltacabtagene autoleucel, cilta-cel; Janssen / Legend Biotech). FDA approved 28 Feb 2022 for r/r MM after 4+ lines; April 2024 second-line (CARTITUDE-4 NCT04181827). Bispecific BCMA binder (two camelid VHH single-domain antibodies) gives high avidity. CARTITUDE-1 (NCT03548207) 98% ORR, 80% sCR.

Allogeneic vs autologous

Autologous CAR-T (every approved product through 2025) requires apheresis of patient T cells → 7-21 day GMP manufacturing → lymphodepletion (fludarabine 25-30 mg/m²/d + cyclophosphamide 300-500 mg/m²/d × 3 d) → reinfusion. Vein-to-vein time 3-6 weeks; cost-of-goods $50-150k; manufacturing failures 5-10%.

Allogeneic (“off-the-shelf”) CAR-T uses healthy-donor T cells engineered to suppress GvHD and host rejection. Approaches:

  • TCR knockout (TRAC locus) via TALEN (Cellectis) or CRISPR (Allogene, CRISPR Therapeutics). Removes endogenous T-cell receptor → prevents GvHD.
  • CD52 knockout to enable alemtuzumab lymphodepletion.
  • B2M knockout to remove HLA class I → reduces host rejection but creates NK susceptibility; HLA-E knock-in to restore NK inhibitory signaling.
  • iPSC-derived T or NK cells (Fate Therapeutics, Century Therapeutics, Shoreline) — clonal master bank, infinite scale.

Allogene ALLO-501A (CD19 anti-leukemia), Caribou CB-010 (CD19, B2M-KO), CRISPR/Vertex CTX112 (CD19, next-gen), Wugen WU-CART-007 — in Phase 1/2 as of 2025.

CAR architecture generations

  • 1st gen. scFv + CD3ζ alone. Limited persistence.
  • 2nd gen. scFv + one costimulatory (CD28 or 4-1BB) + CD3ζ. All approved CD19 / BCMA CARs.
  • 3rd gen. scFv + two costimulatory (CD28 + 4-1BB, CD28 + OX40, etc.) + CD3ζ. Mixed clinical data.
  • 4th gen (“TRUCKs”). Adds inducible cytokine payload (IL-12, IL-15, IL-18) under NFAT promoter → secreted upon CAR engagement; recruits bystander immunity.
  • 5th gen / “armored” CARs. Combine TRUCK cytokines, dominant-negative TGF-β receptors (TGF-βRII-dn — Bluebird, Carisma), checkpoint disruptions (PD-1 knockdown by shRNA or knockout by CRISPR — PACE-CAR Carl June).

Logic-gated CARs

  • AND gates (synNotch). Wendell Lim UCSF. Antigen A engages a Notch-derived receptor → cleaves intracellular transcription factor → drives CAR expression against antigen B. Requires both antigens for activation; reduces on-target off-tumor toxicity. Arcus Biosciences, Cell Design Labs (acquired by Gilead 2017), Senti Bio.
  • OR gates. Tandem CAR (TanCAR) — two scFvs in series; either antigen triggers full signal. Useful for antigen-escape relapse (CD19-CD22 TanCAR — Lentigen / Miltenyi MB-CART19/22; NCT04303497). Dual CAR — two separate CARs co-expressed.
  • NOT gates / inhibitory CARs (iCAR). Daniel Powell UPenn. PD-1 or CTLA-4 intracellular domain fused to scFv against a healthy-tissue antigen → suppresses T cell when iCAR engaged. Prevents on-target toxicity against normal tissue sharing antigen with tumor.
  • Suicide / kill switches. iCasp9 inducible caspase-9 + AP1903 / rimiducin (10 mg IV) → dimerizes caspase-9 → 90% T-cell apoptosis within 30 min (Bellicum BPX-501; Magenta MGTA-117). HSV-TK + ganciclovir (classical, slower). CD20 / RQR8 + rituximab targeting epitope.

CRS and ICANS toxicity management

Cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS) are the dose-limiting toxicities. Graded per ASTCT 2019 consensus (Lee et al. Biol Blood Marrow Transplant):

  • CRS Grade 1. Fever ≥38 °C. Supportive + acetaminophen.
  • CRS Grade 2. Fever + hypotension (responsive to fluids) or hypoxia (low-flow O₂). Tocilizumab (anti-IL-6R mAb; Actemra; Roche) 8 mg/kg IV (max 800 mg), repeat q8h up to 4 doses. Add dexamethasone 10 mg q6h if no response in 24 h.
  • CRS Grade 3-4. Vasopressor-requiring hypotension; high-flow O₂ or ventilator. Tocilizumab + high-dose dexamethasone (10-20 mg IV q6h) or methylprednisolone 1-2 g/d. Anakinra (anti-IL-1 receptor; Kineret; Sobi) 100-200 mg SC q6-12h as steroid-sparing option, particularly for ICANS. Siltuximab (anti-IL-6 mAb; Sylvant; EUSA) 11 mg/kg as alternative when tocilizumab fails.
  • ICANS. Confusion, dysphasia, seizures, cerebral edema. ICE score (Immune Effector Cell-associated Encephalopathy) for grading. Tocilizumab does NOT cross BBB well — dexamethasone or methylprednisolone first-line. Anakinra crosses BBB at high SC dose (5-10 mg/kg/d).

Pretreatment risk stratification uses baseline tumor burden, LDH, ferritin, CRP. Prophylactic anakinra trials (NCT04148430) reducing severe CRS without compromising efficacy.

CAR-NK, γδ-T, iNKT, macrophage therapies

CAR-NK

NK cells are MHC-independent killers via KIR/NKG2A inhibitory + NKG2D/NCR activating receptor balance. CAR-NK advantages: no GvHD (no TCR diversity), shorter persistence (lower CRS risk), feasible from cord blood, iPSC, NK-92 cell lines.

  • Anti-CD19 cord-blood CAR-NK (MD Anderson; Rezvani Karim 2020 N Engl J Med) — 73% response in r/r CD19+ NHL/CLL; commercialized at Takeda TAK-007 (NCT04245722 Phase 2).
  • NKARTA NKX101 (NKG2D-CAR; r/r AML, MDS; NCT04623944).
  • Fate Therapeutics FT522 (iPSC-CAR19 with anti-rejection edits) — Phase 1 lymphoma.
  • Affimed AFM13 + NK (CD30-CD16A bispecific + NK cell adoptive transfer; pre-complexed; NCT04074746).

γδ-T cells

Vδ1+ or Vδ2+ γδ T cells recognize stress-induced ligands (MICA/B, BTN3A1 phosphoantigens) without HLA restriction. Allogeneic candidates: Adicet ADI-001 (anti-CD20 γδ-CAR; NCT04735471), GammaDelta Therapeutics (GSK acquired 2024).

iNKT cells

Vα24Jα18 invariant TCR + CD1d-glycolipid recognition. Athenex KUR-502 (anti-CD19 iNKT CAR), Appia Bio.

CAR-Macrophage (CAR-M)

CARMA (acquired by Carisma) CT-0508 — anti-HER2 CAR macrophage; adenoviral Ad5f35-transduced; Phase 1 in HER2+ solid tumors (NCT04660929). Macrophages enter solid tumors more readily than T cells.

TIL therapy (tumor-infiltrating lymphocytes)

  • Amtagvi (lifileucel; Iovance). FDA approved 16 Feb 2024 for advanced melanoma post-anti-PD-1. First TIL therapy approval. ~28% ORR in heavily pretreated patients. Manufacturing from tumor biopsy: cut tumor → enzymatic dissociation → IL-2 expansion 21-22 d → rapid expansion protocol (REP) with anti-CD3 + irradiated feeders + IL-2 → infusion + high-dose IL-2 bolus.

TCR-T

T-cell receptor (αβ heterodimer) recognizes intracellular antigens presented on HLA. Enables targeting of MAGE, NY-ESO-1, PRAME, HPV E6/E7, mutant KRAS.

  • Tecelra (afamitresgene autoleucel, afami-cel; Adaptimmune). FDA approved 1 Aug 2024 for advanced synovial sarcoma. Anti-MAGE-A4 / HLA-A*02:01 TCR-T. First TCR-T approval. Pivotal SPEARHEAD-1 NCT04044768 — 39% ORR.
  • Kimmtrak (tebentafusp; Immunocore). FDA approved 25 Jan 2022 for uveal melanoma. NOT a TCR-T — a soluble bispecific (anti-gp100/HLA-A*02:01 TCR fused to anti-CD3 scFv); ImmTAC platform. Distinct from cellular TCR-T but illustrative of the TCR-engagement modality.
  • Lete-cel (Adaptimmune anti-NY-ESO-1; sarcoma; Phase 2 NCT01343043).
  • TBio-4101 (TScan; TCR-T mutant KRAS).
  • Letetresgene autoleucel (lete-cel; GSK) — anti-NY-ESO-1; synovial sarcoma.

In vivo CAR generation

Manufacturing autologous CAR-T at vein-to-vein <1 day requires generating CARs inside the patient.

Targeted lipid nanoparticles + mRNA

Capstan Therapeutics (founded 2022; Drew Weissman, Carl June, Bruce Levine; acquired by AbbVie June 2025 $2.1B) — CD8-targeted LNP delivering CAR mRNA → transient CAR expression on circulating T cells. Pre-clinical demonstration in NHP and humanized mouse models (Rurik-Tombácz-Weissman-Epstein 2022 Science — anti-FAP CAR-T in vivo for cardiac fibrosis).

Targeted viral vectors

Umoja Biopharma VivoVec — lentivirus pseudotyped with anti-CD3 + anti-CD8 binders; transduces T cells in vivo. Interius BioTherapeutics INT2104 (NCT06639776 — first in vivo CAR-T trial 2024). EnaraBio, Sana Biotechnology, Tessera Therapeutics.

Non-viral genome editing in vivo

Beam Therapeutics + Tessera — base editing in vivo via LNP. Intellia NTLA-2002 demonstrated in vivo CRISPR feasibility in human liver; T cell tropism still emerging.

Bispecific antibodies — T-cell engagers and beyond

Bispecifics couple two binding specificities — usually one tumor antigen + one immune-cell activator (CD3 for T-cell engagement; CD16A for NK; FcγRIII for macrophage).

Approved T-cell engagers (BiTE / DuoBody / CrossMAb)

  • Blincyto (blinatumomab; Amgen). FDA approved 3 Dec 2014 for Ph− r/r B-ALL. CD19 × CD3 BiTE — 55 kDa single-chain tandem scFv; t½ ~2 h → continuous infusion. Pivotal TOWER NCT02013167. First approved BiTE; established platform validity.
  • Lunsumio (mosunetuzumab; Roche). FDA approved 22 Dec 2022 for r/r follicular lymphoma. CD20 × CD3 IgG1 bispecific (CrossMAb format).
  • Columvi (glofitamab; Roche). FDA approved 15 June 2023 for r/r DLBCL. CD20 × CD3 with 2:1 (CD20:CD3) format for higher avidity.
  • Epkinly (epcoritamab; AbbVie / Genmab). FDA approved 19 May 2023 for r/r DLBCL; subcutaneous.
  • Tecvayli (teclistamab; Janssen). FDA approved 25 Oct 2022 for r/r MM. BCMA × CD3.
  • Talvey (talquetamab; Janssen). FDA approved 9 Aug 2023 for r/r MM. GPRC5D × CD3 — first non-BCMA myeloma bispecific.
  • Elrexfio (elranatamab; Pfizer). FDA approved 14 Aug 2023 for r/r MM. BCMA × CD3.
  • Rozlytrek-bispecific Imdelltra (tarlatamab; Amgen). FDA approved 16 May 2024 for r/r small-cell lung cancer. DLL3 × CD3 BiTE — first solid-tumor T-cell engager.

Non-T-cell-engager bispecifics

  • Vabysmo (faricimab; Roche). FDA approved 28 Jan 2022 for nAMD, DME. VEGF-A × Ang-2 — first ophthalmic bispecific.
  • Rybrevant (amivantamab; Janssen). FDA approved 21 May 2021 for EGFR Exon 20 insertion NSCLC. EGFR × MET.
  • Hemlibra (emicizumab; Roche). FDA approved 16 Nov 2017 for hemophilia A. FIX × FX — mimics factor VIII bridging activity.
  • Tepkinly / Bimzelx (bimekizumab; UCB). FDA approved 17 Oct 2023 for plaque psoriasis. IL-17A × IL-17F.
  • Opdualag (nivolumab + relatlimab; BMS). FDA approved 18 Mar 2022 for unresectable melanoma. NOT a bispecific — fixed-dose combination of anti-PD-1 + anti-LAG-3.

Immune checkpoint inhibitors

PD-1 / PD-L1

  • Keytruda (pembrolizumab; Merck). FDA approved 4 Sept 2014 for melanoma; now 40+ indications. Anti-PD-1 IgG4. $25B 2023 revenue — largest pharma product by sales.
  • Opdivo (nivolumab; BMS). FDA approved 22 Dec 2014. Anti-PD-1.
  • Tecentriq (atezolizumab; Roche). FDA 18 May 2016. Anti-PD-L1.
  • Imfinzi (durvalumab; AstraZeneca). FDA 1 May 2017. Anti-PD-L1.
  • Bavencio (avelumab; Pfizer / Merck KGaA). FDA 23 Mar 2017. Anti-PD-L1.
  • Libtayo (cemiplimab; Regeneron / Sanofi). FDA 28 Sept 2018. Anti-PD-1.
  • Jemperli (dostarlimab; GSK). FDA 22 Apr 2021. Anti-PD-1; dMMR endometrial.

CTLA-4

  • Yervoy (ipilimumab; BMS). FDA 25 Mar 2011 for metastatic melanoma. First approved checkpoint inhibitor.

LAG-3

  • Opdualag (relatlimab + nivolumab; BMS). First approved LAG-3 blocker, March 2022.

TIGIT, TIM-3, VISTA

Pipeline targets. Tiragolumab (Roche anti-TIGIT) failed primary endpoint in Phase 3 SKYSCRAPER lung trial 2022 — humbling signal that TIGIT is harder than PD-1.

Combination strategies

PD-1 + CTLA-4 (ipi + nivo) — CheckMate 067 melanoma OS benefit. PD-1 + chemo standard 1L NSCLC, gastric, esophageal. PD-1 + tyrosine kinase inhibitor (pembrolizumab + axitinib RCC; pembrolizumab + lenvatinib endometrial).

Cytokine engineering

Native cytokines (IL-2, IL-12, IL-15, IFN-α, TNF-α) have potent anti-tumor activity but systemic toxicity. Engineered variants:

IL-2

  • IL-2 muteins (“not-α”). Bind IL-2Rβγ (effector T, NK) preferentially over IL-2Rαβγ (Treg). Nektar NKTR-214 / bempegaldesleukin — PEG-IL2 — failed Phase 3 PIVOT-IO 001 (NCT03635983) 2022 in combo with nivolumab.
  • Synthorx / Sanofi THOR-707 (efineptakin alfa). Site-specifically PEGylated at non-α-binding face via UAA incorporation.
  • Asher Bio AB248 (CD8-cis IL-2) — IL-2 mutein fused to anti-CD8 single-domain antibody; CD8 T-cell selective. NCT05653882.
  • Medicenna MDNA11 (“super-IL-2”) — orthogonal IL-2 + IL-2Rβ pair.
  • Aldesleukin (Proleukin; original recombinant IL-2; FDA 1992). Still in use at high dose for melanoma and RCC despite extreme toxicity (capillary leak syndrome).

IL-12

Highest potency cytokine for IFN-γ induction → most toxic. Engineering:

  • Masked IL-12 (probody). CytomX / BMS — masking peptide cleaved by tumor proteases; restricts activity to TME.
  • Mural Oncology XmAb306 — IL-12 × anti-PD-1 fusion.
  • Werewolf WTX-330 — conditionally activated IL-12.
  • Intratumoral plasmid IL-12 (tavokinogene telseplasmid; OncoSec) — electroporation of plasmid encoding IL-12 directly into tumor; melanoma Phase 2.

IL-15 superagonists

  • N-803 (nogapendekin alfa inbakicept-pmln; Anktiva; ImmunityBio). FDA approved 22 Apr 2024 for BCG-unresponsive NMIBC bladder cancer. IL-15 + IL-15Rα-Fc complex (no IL-2 toxicity profile).
  • NIZ985 (Novartis). IL-15/IL-15Rα heterodimer.
  • HCW Biologics HCW9218 — IL-15/IL-21 bifunctional.

Other engineered cytokines

  • Vibostolimab (Merck MK-7684; anti-TIGIT). Combo with pembrolizumab — Phase 3 KeyVibe-007 lung 2024 data mixed.
  • PEG-IFNα-2a (Pegasys; Roche) — HCV, polycythemia vera.
  • Romiplostim (Nplate; Amgen) — peptibody TPO mimetic.

Microbiome immunotherapy

Gut microbiome modulates response to immunotherapy. Akkermansia muciniphila, Faecalibacterium prausnitzii, Bacteroides fragilis correlate with anti-PD-1 response (Gopalakrishnan-Wargo 2018 Science; Routy-Zitvogel 2018 Science).

Fecal microbiota transplant (FMT)

  • Rebyota (Ferring). FDA approved 30 Nov 2022 for r/r C. difficile. First approved FMT product.
  • Vowst (Seres SER-109). FDA approved 26 Apr 2023 for r/r C. difficile. Oral spore-based formulation.

Live biotherapeutics for immuno-oncology

  • Seres SER-401, SER-301 — checkpoint-responder microbiome restoration.
  • Vedanta VE800 — defined consortium for anti-PD-1 augmentation.
  • 4D Pharma MRx0518 — Enterococcus gallinarum monoclonal; combo with anti-PD-1 (NCT03637803). Company entered administration 2023 — illustrates field’s commercial fragility.

Engineered probiotics (synthetic biology x microbiome)

Synlogic SYNB1934 (engineered E. coli Nissle for PKU; collaboration with Roche); SYNB1353 (homocystinuria); SYNB8802 (hyperoxaluria). Genome Compiler. ZBiotics (S. boulardii consumer probiotic, not therapeutic).

Vaccine adjuvants

Cross-reference virology-and-vaccine-platforms.

Aluminum salts (Alum)

Aluminum hydroxide (Alhydrogel; Croda), aluminum phosphate. Workhorse adjuvant since 1926 (Glenny). Mechanism: depot effect + NLRP3 inflammasome activation. In most pediatric vaccines.

Oil-in-water emulsions

  • MF59 (Novartis/Seqirus). Squalene + Tween 80 + Span 85. In FLUAD (seasonal flu, elderly).
  • AS03 (GSK). Squalene + DL-α-tocopherol + Tween 80. In Pandemrix (H1N1 2009; controversial narcolepsy signal in Scandinavia); Arepanrix.
  • AF03 (Sanofi). Squalene + sorbitan-based.

TLR agonists

  • MPL (monophosphoryl lipid A). TLR4 agonist. Detoxified LPS from Salmonella minnesota R595.
  • AS04 (GSK). Alum + MPL. In Cervarix (HPV), Fendrix (HepB for renal patients), Mosquirix (RTS,S malaria).
  • AS01 (GSK). Liposome + MPL + QS-21 (saponin from Quillaja saponaria). In Shingrix (zoster; 97% efficacy in adults ≥50), Mosquirix (malaria), Arexvy (RSV).
  • AS02 (GSK). Emulsion + MPL + QS-21. Historic malaria use.
  • CpG (oligodeoxynucleotide). TLR9 agonist. CpG 1018 (Dynavax). In Heplisav-B (HepB; FDA 2017), Vaxelis. Also as standalone in cancer vaccines.
  • Poly-IC, Poly-ICLC (Hiltonol; Oncovir). TLR3 + MDA5. Investigational in cancer vaccines and HIV.

Saponins

  • QS-21 (Quillaja saponin fraction 21). Component of AS01 and Matrix-M (Novavax; in Nuvaxovid COVID and R21/Matrix-M malaria — WHO recommendation 2023).

STING agonists

cGAS-STING activators (cyclic dinucleotides; ADU-S100; MK-1454; SB-11285). High potency intratumoral. Tumor-targeted ENPP1 inhibitors emerging.

Cationic lipid nanoparticles

mRNA vaccine LNPs (ALC-0315 ionizable lipid in BNT162b2; SM-102 in mRNA-1273) themselves have adjuvant activity through TLR-independent inflammation.

Solid tumor challenges

Hematologic CAR-T has succeeded; solid tumors remain hard. Barriers + engineering responses:

  1. Antigen heterogeneity. No universal solid-tumor antigen comparable to CD19. Targets in trial: HER2, EGFR, mesothelin, GD2, claudin-18.2, B7-H3, GPC3, PSMA, FAP, MUC1, CEA, MAGE-A4. Dual + AND-gated CARs.
  2. Tumor microenvironment. TGF-β, adenosine, MDSCs, Tregs, hypoxia, dense ECM. Armored CARs (TGF-βRII-dn; CD39/CD73 blockade; adenosine A2AR antagonists).
  3. Trafficking. T cells must extravasate from blood, penetrate stroma. Chemokine receptor engineering (CCR2, CXCR2). Local administration (intratumoral, intracavitary, intracerebroventricular for CNS tumors).
  4. Persistence. Solid-tumor CAR-T often exhausts. Armored cytokine (IL-7, IL-15, IL-18); central memory T cell enrichment; less exhausted iPSC-T or naive-T enriched.
  5. On-target off-tumor toxicity. Mesothelin CARs in pancreatic / mesothelioma — pulmonary toxicity. HER2-CAR T cells caused fatal pulmonary edema (Morgan 2010 Mol Ther). Logic gates and suicide switches.

Solid-tumor CAR-T programs

  • GD2 CAR-T for neuroblastoma (Carl June UPenn; Crystal Mackall Stanford NCT00085930 → NCT04539366).
  • Claudin-18.2 CAR-T (CARsgen CT041; Chinese approval anticipated 2025).
  • Mesothelin-CAR (MCY-M11 mRNA CAR-NK; iMatrix CARS).
  • B7-H3 CAR-T (Seattle Children’s; pediatric DIPG, NCT04185038).
  • Tagraxofusp-erzs (Elzonris; Stemline / Menarini). FDA 21 Dec 2018 for BPDCN. CD123-targeted diphtheria-toxin fusion. Not a CAR-T but illustrative of CD123 targeting.

Manufacturing and scale-up

Closed-system manufacturing

  • Miltenyi Biotec CliniMACS Prodigy. Bench-top closed automated CAR-T manufacturing — selection, activation, transduction, expansion, formulation. 7-14 day cycle. Widely deployed in academic centers; standard for Chinese CAR-T.
  • Lonza Cocoon. Cassette-based.
  • Cytiva Sefia. Single-use bag-based.
  • Ori Biotech IRO. Multi-modal.

Viral vector manufacturing

Lentivirus (LV) and γ-retrovirus production at HEK293T scale-up. Suspension HEK293 (Cevec CAP-T; VivaBioCell; HEK293T-suspension adapted) replacing adherent culture for scalability. Yields 10⁷-10⁹ TU/mL. Per-dose viral cost $5-30k.

CAR-T cost-of-goods (COGS)

  • Apheresis + viral vector + manufacturing: $50-150k
  • Lymphodepletion + infusion + ICU: $100-200k (CRS management)
  • Total list price: $373k-$475k (US wholesale acquisition)

Allogeneic projected COGS $10-30k per dose at scale → potential price disruption if outcomes match autologous.

Diagnostics for cell therapy

Companion diagnostics

CD19 IHC, BCMA flow / IHC, MSI / dMMR (Jemperli), TMB (pembrolizumab tissue-agnostic 2020), PD-L1 IHC (22C3 pharmDx for Keytruda; SP142 for Tecentriq; 28-8 for Opdivo).

Persistence monitoring

  • Vector copy number (VCN) by qPCR — CD19 CAR persistence at ≥1 copy/100 cells for >6 mo correlates with durable remission.
  • MRD (minimal residual disease) by flow (10⁻⁴) or NGS (Adaptive Biotechnologies ClonoSEQ FDA 2018; 10⁻⁶).

Patient selection

LDH, ferritin, CRP at infusion. High disease burden + high inflammatory markers → CRS risk. Bridging therapy (chemo, RT, brexucabtagene’s tumor-debulking step) reduces severe CRS.

Regulatory framework

FDA OTAT / Office of Therapeutic Products

Reorganized 2023 → OTP (Office of Therapeutic Products) within CBER. Pre-IND → Phase 1/2 → Phase 2 pivotal → BLA. Accelerated approval for unmet need; RMAT (Regenerative Medicine Advanced Therapy) designation (21st Century Cures 2016) for cell + gene therapies.

EMA ATMP

Advanced Therapy Medicinal Product regulation; CAT (Committee for Advanced Therapies). Hospital exemption allows custom non-routine cell therapy in specialist centers.

China NMPA

Faster CAR-T approval timeline than US — Carvykti precursor LCAR-B38M approved in China before FDA. Relmacabtagene autoleucel (relma-cel; JW Therapeutics CD19 CAR-T) approved Sept 2021.

Long-term safety follow-up

FDA requires 15-year follow-up post-CAR-T for integration-site oncogenesis monitoring. In Nov 2023 FDA issued safety communication on T-cell malignancies post-CAR-T (~30 cases reported); pivotal investigation ongoing 2024-2026. Boxed warning added to all CAR-T labels April 2024.

Antibody-drug conjugates (ADCs)

ADCs combine antibody specificity with a cytotoxic payload — a third immune-engineering modality alongside CAR-T and bispecifics, generating >$15B annual revenue 2024.

Approved ADCs (selected)

  • Mylotarg (gemtuzumab ozogamicin; Pfizer). FDA 17 May 2000 (withdrawn 2010 — relaunched 1 Sept 2017 with lower dose). CD33-calicheamicin for AML.
  • Adcetris (brentuximab vedotin; Seagen/Pfizer + Takeda). FDA 19 Aug 2011. CD30-MMAE for Hodgkin lymphoma and CD30+ T-cell lymphomas.
  • Kadcyla (ado-trastuzumab emtansine, T-DM1; Roche). FDA 22 Feb 2013. HER2-DM1 for HER2+ breast cancer.
  • Besponsa (inotuzumab ozogamicin; Pfizer). FDA 17 Aug 2017. CD22-calicheamicin for B-ALL.
  • Polivy (polatuzumab vedotin; Roche). FDA 10 June 2019. CD79b-MMAE for DLBCL.
  • Padcev (enfortumab vedotin; Astellas + Seagen). FDA 18 Dec 2019. Nectin-4-MMAE for urothelial cancer.
  • Trodelvy (sacituzumab govitecan; Gilead-Immunomedics). FDA 22 Apr 2020. TROP2-SN38 (topoisomerase I inhibitor) for TNBC + urothelial.
  • Enhertu (trastuzumab deruxtecan, T-DXd; Daiichi Sankyo + AstraZeneca). FDA 20 Dec 2019, expanded HER2-low Aug 2022 (DESTINY-Breast04). HER2-deruxtecan (topo I); changed HER2-low breast cancer standard of care.
  • Tivdak (tisotumab vedotin; Seagen + Genmab). FDA 20 Sept 2021. Tissue factor-MMAE for cervical cancer.
  • Elahere (mirvetuximab soravtansine; ImmunoGen). FDA 14 Nov 2022. Folate receptor α-DM4 for FRα+ ovarian.

Payload classes

  • Microtubule inhibitors. MMAE (monomethyl auristatin E), MMAF, DM1, DM4 (maytansinoid).
  • DNA crosslinkers. Calicheamicin, PBD (pyrrolobenzodiazepine).
  • Topoisomerase I inhibitors. SN-38 (irinotecan metabolite), deruxtecan (DXd, exatecan derivative).
  • Topoisomerase II inhibitors. Anthracyclines (PNU-159682 in Daiichi pipeline).
  • RNA polymerase II inhibitors. Amanitin.

Linker chemistry

  • Cleavable. Cathepsin B-cleavable Val-Cit-PABC (in Adcetris, Padcev, Polivy, Tivdak); acid-sensitive hydrazone (Mylotarg, Besponsa); reducible disulfide.
  • Non-cleavable. Thioether (Kadcyla — releases lysine-DM1 metabolite after antibody catabolism).
  • DAR (drug-to-antibody ratio). 4-8 typical for cleavable MMAE / deruxtecan; 3-4 for DM1.

Site-specific conjugation

Cysteine engineering (THIOMABs; Junutula-Genentech 2008 Nat Biotechnol), unnatural amino acid sites (Sutro Biopharma XpressCF), enzymatic (Sortase A — Pfizer; transglutaminase — Innate Pharma; SMARTag — Catalent), glycan-remodeling (Synaffix GlycoConnect).

Engineered antibody formats beyond bispecifics

  • DARPins (Molecular Partners + Novartis). Designed Ankyrin Repeat Proteins; small (14-21 kDa), thermostable.
  • Nanobodies (camelid VHH; Ablynx/Sanofi). Caplacizumab (Cablivi; FDA 6 Feb 2019 for aTTP) — first nanobody approval.
  • Bicycle peptides (Bicycle Therapeutics). Constrained bicyclic peptides for tumor targeting.
  • Centyrins, Affibodies, Avimers, Fynomers, Knottins. Alternative scaffolds.
  • Megabodies (Steyaert). Nanobody fused to scaffold protein for cryo-EM rigidification.

Synthetic biology in immunoengineering

  • synNotch receptors (Lim UCSF). Engineered transmembrane Notch with custom extracellular binder + intracellular transcription factor → antigen-A → expression of payload (CAR for antigen B, cytokine, suicide gene).
  • MESA receptors (Leonard Northwestern). Modular Extracellular Sensor Architecture; ligand-induced split-TF.
  • GEMS receptors. Generalized Extracellular Molecule Sensor.
  • iSNAP, SUPRA-CAR, SparX. Universal CAR + soluble antigen-specific adaptor — same CAR-T cell can be retargeted by infusing a different adaptor.
  • TaC (tag-CAR) systems. anti-FITC or anti-PNE CAR + FITC/PNE-tagged antibody — modularity at the level of soluble adapter.

Manufacturing economics for cell therapy

Autologous COGS breakdown

  • Apheresis (Spectra Optia, Terumo BCT). $3-5k/procedure.
  • GMP viral vector (lentivirus typically). $5-30k/dose — drives demand for non-viral edits (electroporation + RNP) which Casgevy demonstrated at clinical scale.
  • Cell processing (Miltenyi CliniMACS Prodigy, Lonza Cocoon, Cytiva Sefia). Labor + reagents $10-30k; 7-21 d cycle.
  • Release testing. Sterility USP <71>, mycoplasma USP <63>, endotoxin USP <85>, identity (flow CD3/CD19), viability >70%, potency (cytokine or cytotoxicity), VCN <5/cell, RCL screen; $15-30k QC.
  • Cryopreservation + logistics. Liquid N2 dry shipper (Cryoport, World Courier).
  • Lymphodepletion + infusion + ICU. $50-200k hospital costs.

Allogeneic projected COGS

$10-30k/dose at scale (100s-1000s doses per master bank). Master cell bank from one healthy donor → TCR/HLA edits → banked, thawed and re-infused on demand. Re-dosing typically needed since allogeneic cells expire faster than autologous.

Hospital reimbursement

CMS Inpatient PPS DRG 018 (CAR-T cell immunotherapy). New Technology Add-on Payment (NTAP) covers 65-80% of incremental cost first 2-3 years post-FDA approval. Insufficient at smaller hospitals — concentration at high-volume centers.

Centralized vs decentralized manufacturing

  • Centralized. Novartis Morris Plains NJ, Kite El Segundo CA, BMS Bothell WA, Janssen Raritan NJ + Ghent Belgium. Higher throughput; vein-to-vein 21-35 days.
  • Decentralized / point-of-care. Miltenyi CliniMACS Prodigy at hospital, Lonza Cocoon, Cellares Cell Shuttle — same-hospital manufacturing; vein-to-vein 7-14 days. China deploys widely; FDA expected to expand decentralized authorization 2025-2026.

Translational pharmacology

Pharmacokinetics

CAR-T expansion (peak Day 7-14 post-infusion) and persistence (>6 mo associated with durable remission). Measured by qPCR vector copy number or flow CAR enumeration. Bispecifics (Blincyto) — short half-life ~2 h → continuous infusion; Lunsumio IgG1 — weeks half-life with step-up. Checkpoint inhibitors — 2-3 wk half-life; q3w or q6w.

Biomarkers and predictive signatures

  • CAR-T. Peak expansion AUC, persistence, B-cell aplasia (CD19), serum-free light chains (BCMA).
  • PD-1/L1. PD-L1 IHC (22C3, SP142, 28-8 clones), TMB (≥10 mut/Mb tissue-agnostic pembrolizumab), MSI/dMMR (Jemperli), tertiary lymphoid structures (research-only).
  • Bispecifics. Cytokine flux (IL-6, IL-10, IFN-γ) post-dose; pre-emptive tocilizumab in high-risk patients.

Quantitative systems pharmacology (QSP)

ODE models of CAR-T expansion + target killing + immune feedback (Hosseini-Macallan AstraZeneca; Stein-Boyer Novartis). Inform dose selection, lymphodepletion choice, combination scheduling. Open-source SimBiology and proprietary Certara QSP toolkits.

Selected leading research groups (2024-2026)

  • Carl June (UPenn). CD19 CAR-T originator (Kymriah lineage), iCasp9 + dnTGFβRII armored CARs, PCSK9 in vivo CAR pre-clinical.
  • Michel Sadelain (MSKCC). CAR architecture (Sleeping Beauty + 4-1BB); founded Mnemo Therapeutics.
  • Steven Rosenberg (NCI). TIL pioneer; lifileucel + neoantigen TCR-T NCI translational pipeline.
  • Crystal Mackall (Stanford). GD2 CAR-T for pediatric solid tumors + DIPG (NCT04196413).
  • James Allison (MDACC). CTLA-4 founding (Yervoy); Nobel Medicine 2018.
  • Tasuku Honjo (Kyoto). PD-1 founding; Nobel Medicine 2018.
  • Drew Pardoll (Johns Hopkins). Checkpoint translational; Bloomberg-Kimmel Institute.
  • Wendell Lim (UCSF). synNotch, logic-gated cell engineering.
  • Marcela Maus (MGH). Solid tumor CAR-T (CARv3-TEAM-E EGFRvIII).
  • Katy Rezvani (MDACC). Cord-blood CAR-NK platform → TAK-007.
  • Stephan Grupp (CHOP). Pediatric CAR-T clinical lead — ELIANA / Kymriah.
  • Renier Brentjens (Roswell Park). CD19 CAR + IL-12 TRUCK pioneer.

Adjacent

Further reading

  • Sadelain, M., Rivière, I., Riddell, S. — “Therapeutic T cell engineering” Nature 2017, 545:423.
  • June, C.H., Sadelain, M. — “Chimeric antigen receptor therapy” N Engl J Med 2018, 379:64.
  • Larson, R.C., Maus, M.V. — “Recent advances and discoveries in the mechanisms and functions of CAR T cells” Nat Rev Cancer 2021, 21:145.
  • Lim, W.A., June, C.H. — “The principles of engineering immune cells to treat cancer” Cell 2017, 168:724.
  • Janeway, C.A., Travers, P., Walport, M., Shlomchik, M. — Janeway’s Immunobiology 10th ed., Garland 2022.
  • Murphy, K., Weaver, C. — Janeway’s Immunobiology 9th-10th editions — checkpoint and T-cell biology background.
  • Murray, P.R., Rosenthal, K.S., Pfaller, M.A. — Medical Microbiology 9th ed., Elsevier 2021 — microbiome immune-modulation chapters.
  • Stryer, L., Berg, J.M., Tymoczko, J.L., Gatto, G.J. — Biochemistry 9th ed., WH Freeman 2019 — cytokine and antibody biochemistry background.
  • Mellman, I., Chen, D.S., Powles, T., Turley, S.J. — “The cancer-immunity cycle: indication, genotype, and immunotype” Immunity 2023, 56:2188.
  • Lee, D.W., Santomasso, B.D., Locke, F.L., et al. — “ASTCT Consensus Grading for Cytokine Release Syndrome and Neurologic Toxicity Associated with Immune Effector Cells” Biol Blood Marrow Transplant 2019, 25:625.

Graph View