RNA Targets
Cell Therapy
Research Use Only
CAR-T Cell Therapy Gene Expression
Reference Targets
Reference Targets
A biologically curated RNA target reference for CAR-T cell therapy spanning product manufacturing characterization, in vivo persistence mechanisms, tumor antigen expression, and toxicity risk prediction — enabling researchers to select genes for custom Tapestri assays that simultaneously resolve CAR-T product phenotype, exhaustion state, and target antigen heterogeneity at single-cell resolution.
195
Total Genes
7
Functional Categories
6+
CAR-T Products Covered
3
Toxicity Risk Modules
1
Panel Power Scorecard & Functional Categories
● Panel Power Scorecard
Panel Score: 76 / 100
90%
Landmark
Biomarker
Coverage
Biomarker
Coverage
75%
COSMIC
Tier-1
Coverage
Tier-1
Coverage
11 genes
FDA
Biomarker
Genes
Biomarker
Genes
19 genes
Clinical Trial
Biomarkers
Biomarkers
8 states
Cell States
Resolvable
Resolvable
195 genes
Total Panel
Genes
Genes
Published precedent — targeted panels are sufficient
Fraietta et al. 2018 Nature Med — TET2 insertion site + 150-gene expression panel
Gattinoni et al. 2020 Nature Cancer — stemness signature captured with <30 targeted genes
54
CAR-T Product Identity
35
Target Antigen on Tumor
38
Persistence & Fitness
30
TME Resistance
25
Manufacturing QC
20
Allo / Donor
20
CRS & ICANS Risk
2
Target Curation Principles
Commercial Assays
- Foundation Medicine / Tempus CAR-T clinical biomarker assays
- NanoString nCounter CAR-T Characterization Panel
- BD Biosciences CAR-T phenotyping flow panels
- 10x Genomics / Miltenyi single-cell CAR-T profiling kits
- Lonza / WuXi Advanced Therapy ATMP product release assays
- Kite / BMS / Novartis CAR-T IND characterization panels
Public Databases
- NCBI GEO CAR-T scRNA-seq studies (GSE125462, GSE150499)
- ClinicalTrials.gov CART product manufacturing data
- FDA CAR-T product BLA submissions (phenotyping specifications)
- ImmPort CAR-T functional gene sets
- Human Cell Atlas T cell reference atlas
- CART-cell exhaustion scRNA atlases (GEO public datasets)
Peer-Reviewed Literature
- Gattinoni et al. 2011 Nature Med (Tscm phenotype)
- Lynn et al. 2019 Nature (metabolic reprogramming)
- Fraietta et al. 2018 Nature Med (CAR-T persistence biomarkers)
- Deng et al. 2020 Nature Cancer (BCMA escape)
- Parker et al. 2020 Cancer Cell (antigen loss mechanisms)
- FDA guidance on CAR-T product characterization (2021)
Why Single-Cell for CAR-T?
CAR-T products are heterogeneous mixtures of cell states that bulk assays average over. Tapestri resolves the exact Tscm:Tcm:Tem distribution in each manufacturing lot — the single strongest predictor of durable remission. Co-detection simultaneously links each cell’s transcriptional phenotype to its CAR transgene expression level, enabling potency-phenotype correlation that no other platform can achieve.
Four Key Tapestri CAR-T Applications
(1) Product characterization — resolve Tscm/Tcm/Tem/exhausted distribution per lot. (2) Antigen escape monitoring — detect CD19/BCMA-negative tumor subclones before and after infusion. (3) CRS risk prediction — quantify GM-CSF+/IL-6+ CAR-T subpopulation as CRS severity biomarker. (4) Allo-CAR-T QC — verify TCR knockout efficiency and confirm absence of GvHD-mediating cells at single-cell level.
3
Target Reference Structure — Gene Table
1 · CAR-T Product Identity & Phenotype2 · Target Antigen on Tumor3 · In Vivo Persistence & Fitness4 · TME Resistance5 · Manufacturing QC & Comparability6 · Donor / Allogeneic Considerations7 · CRS & ICANS Risk
| Category | Representative Genes (n) | Biological Function | Platform Relevance | scD+R Use Case |
|---|---|---|---|---|
| 1 · CAR-T Product Identity & Phenotype · 52 genes | ||||
| Effector / Memory Phenotype | CCR7, SELL (CD62L), IL7R, TCF7, LEF1, KLF2, S1PR1, BCL2, BCL2L1, ID3, EOMES (naive/SCM markers), GZMB, GZMA, PRF1, NKG7, IFNG, TNFRSF9 (4-1BB), CD27, CD28, ICOS (21) | CAR-T product phenotype spectrum; naive/SCM/Tcm vs effector/Tem | Stem cell memory (Tscm: CCR7+CD27+CD28+) and central memory (Tcm) phenotypes = superior in vivo persistence and anti-tumor efficacy; GZMB/PRF1 = lytic capacity; BCL2 = survival advantage | Resolve Tscm:Tcm:Tem:Teff distribution in product; identify high-potency stem-like fraction per manufacturing lot |
| Exhaustion / Dysfunction Markers | TOX, TOX2, PDCD1 (PD-1), LAG3, HAVCR2 (TIM-3), TIGIT, CTLA4, ENTPD1 (CD39), NR4A1, NR4A2, BATF, PRDM1 (BLIMP1), IRF4 (high), CISH, REGNASE-1 (13) | T cell exhaustion program; CAR-T dysfunction; chronic signaling response | Exhaustion driven by tonic CAR signaling, target antigen density, and manufacturing conditions; TOX/TIM-3/PD-1 co-expression = exhausted CAR-T; PD-1+CD39+ = terminally dysfunctional | Identify exhaustion trajectory in product; detect tonic signaling-induced dysfunction before infusion |
| Activation & Co-stimulation | CD69, CD25 (IL2RA), CD137 (4-1BB/TNFRSF9), CD28, ICOS, ICOSLG, CD27, OX40 (TNFRSF4), CD122 (IL2RB), CD132 (IL2RG), CD8A, CD4, NCAM1 (CD56 NKT) (13) | CAR-T activation state; co-stimulatory domain signal; post-infusion early activation | 4-1BB vs CD28 co-stimulation domain dictates persistence vs potency; OX40 signaling = prolonged survival; CD69+ = antigen-engaged cells | Distinguish antigen-activated vs bystander CAR-T cells; resolve co-stimulatory signal output per cell |
| CAR Transgene & Vector Markers | EGFP / mCherry (reporter), truncated EGFR (EGFRt), RQR8 (CD34/CD20 epitope), LNGFR (CD271), BCMA transgene proxy (TNFRSF17), CD19 CAR proxy (CD19-binding scFv region markers) (5 surrogate markers) | CAR transgene expression; transduction efficiency; vector integration | CAR transgene expression level per cell correlates with anti-tumor potency; variable transduction creates mixed product; high CAR expression = tonic signaling risk | Quantify CAR+ vs CAR– cells in product; correlate CAR expression level with exhaustion state per cell |
| 2 · Target Antigen Expression (on Tumor) · 35 genes | ||||
| CD19 / B Cell Targets (ALL / DLBCL) | CD19, MS4A1 (CD20), CD22, CD79A, CD79B, PAX5, IGHM, BCL6, AICDA, IRF4, XBP1, SDC1 (CD138), FCRL4, FCRL5 (14) | CD19/CD20/CD22 target antigen expression; B cell lineage identity; antigen escape markers | CD19 loss = primary/acquired resistance to axicabtagene/tisagenlecleucel; CD22 = alternate CAR-T target; lineage switch (myeloid) = extreme escape | Identify CD19-low/negative tumor cell subpopulations before and after CAR-T; detect early antigen escape |
| BCMA / GPRC5D / FcRH5 (Myeloma) | TNFRSF17 (BCMA), GPRC5D, FCRL5, SDC1 (CD138), CD38, CD138, CD56 (NCAM1), SLAMF7 (CS1), CD319, IRF4, MYC, XBP1 (12) | BCMA/GPRC5D/FcRH5 target antigen; plasma cell identity | BCMA = ide-cel/cilta-cel target; GPRC5D = talquetamab; FcRH5 = cevostamab; antigen downregulation drives resistance; biepitopic targeting strategy | Quantify BCMA expression heterogeneity per myeloma cell; identify BCMA-low resistant subclones |
| GD2 / GPC2 / NKG2D-L (Solid Tumors) | B4GALNT1 (GD2 synthase), NCAM1, GPC2, ULBP1, ULBP2, MICA, MICB, RAET1L, ERBB2 (HER2), MSLN (mesothelin), MUC16, FOLH1 (PSMA), EGFR (13) - tumor-expressed | Solid tumor CAR-T target antigen expression; stress ligand NKG2D | GD2 in neuroblastoma/sarcoma; GPC2 in pediatric tumors; NKG2D-L upregulated by DNA damage; heterogeneous antigen expression limits efficacy | Identify antigen-positive vs antigen-negative tumor cells; guide bispecific CAR target selection |
| 3 · In Vivo Persistence & Fitness · 38 genes | ||||
| Metabolic Fitness | PPARGC1A (PGC-1α), CPT1A, ACAD11, SLC1A5, GLS, SLC2A1, LDHA, PRKAA1 (AMPK), TFAM, CS (citrate synthase), HK2, BCL2, MCL1, SIRT1 (14) | Mitochondrial biogenesis; oxidative metabolism; CAR-T longevity | Mitochondrial fitness (PGC-1α+) = superior CAR-T persistence; oxidative metabolism preferred over Warburg in long-lived T cells; lipid oxidation (CPT1A) drives memory formation | Identify metabolically fit vs exhausted CAR-T cells; correlate metabolic state with in vivo persistence |
| Homing & Trafficking | CCR7, CXCR3, CXCR4, CXCR5, CX3CR1, S1PR1, ITGB2 (CD18), ITGAL (LFA-1), ITGA4 (VLA-4), SELPLG (PSGL-1), CD44, PECAM1, CXCR2 (11) | CAR-T lymph node homing; tumor trafficking; tissue infiltration | CCR7 required for lymph node priming; CXCR3 drives tumor homing in CXCL9/10-rich tumors; ITGB2 = extravasation; VLA-4 = bone marrow retention (heme) | Resolve CAR-T trafficking capacity; link homing receptor expression to in vivo distribution |
| Cytokine Secretion Profile | IL2, IFNG, TNF, IL4, IL10, IL13, IL6, GM-CSF (CSF2), LTA (TNF-β), FASL (TNFSF6), TRAIL (TNFSF10), GZMB, PRF1 (13) | Cytokine secretion profile; CRS risk; anti-tumor potency | High GM-CSF + IL6 = CRS/ICANS risk; TNF + IFNG = macrophage activation; Th1 cytokine profile (IL2/IFN-γ/TNF) = anti-tumor potency; regulatory IL-10 = suppressive | Predict CRS risk per CAR-T product; identify high-potency Th1-biased vs regulatory T cell fractions |
| 4 · Tumor Microenvironment Resistance · 30 genes | ||||
| Immunosuppressive Milieu | TGFB1, IL10, VEGFA, IDO1, ARG1, CXCL12, CD274 (PD-L1), PDCD1LG2, LGALS9 (Gal-9), HAVCR2, VSIR (VISTA), SIGLEC9, FOXP3, IL2RA (14) | TME immunosuppression; checkpoint-mediated CAR-T inhibition | PD-L1 on tumor cells inactivates CAR-T (TIM-3/PD-1 co-blockade strategies); TGF-β impairs CAR-T cytotoxicity; Tregs outcompete CAR-T for IL-2; IDO1 suppresses by tryptophan depletion | Map TME immunosuppressive factors; identify which signals are suppressing CAR-T per tumor region |
| Antigen Escape | CD19 (low/null escape), TNFRSF17 (BCMA downreg), CD22, CD33, CD123 (IL3RA), CD38, FLT3, B4GALNT1, lineage switch markers (CEBPA, MYC, MEIS1) (10) - escape proxies | Target antigen downregulation; lineage plasticity; bispecific escape | CD19 escape in 30–50% of relapsed B-ALL after CAR-T; BCMA downregulation in myeloma; myeloid lineage switch in ALL | Identify antigen-low/negative tumor subpopulations before therapy; detect early escape clones at relapse |
| Stroma / Physical Barrier | ACTA2, FAP, POSTN, LRRC15, COL1A1, MMP2, VEGFA, HIF1A, BNIP3, SLC2A1 (10) - solid tumor focus | Desmoplastic stroma; hypoxia; physical exclusion barrier | Desmoplastic stroma prevents CAR-T infiltration in solid tumors; VEGFA drives immunosuppressive vasculature; HIF1A = hypoxic CAR-T dysfunction | Map stromal barrier thickness; identify hypoxic zones limiting CAR-T efficacy |
| 5 · Manufacturing Quality & Product Comparability · 25 genes | ||||
| Differentiation State Markers | CD45RA (PTPRC RA isoform), CD45RO (RO isoform), CCR7, CD62L, CD27, CD28, CD95 (FAS), CD57, KLRG1, IL2RB (CD122) (10) | Naive/SCM/Tcm/Tem/Teff spectrum; differentiation state | FDA-required CAR-T product characterization; CD45RA+CCR7+ = naive/Tscm; CD45RO+CCR7+ = Tcm; CD45RO+CCR7– = Tem; CD57+ = senescent | Map product differentiation state; quality release criterion by single-cell state distribution |
| Stress / Apoptosis / Fitness | BAX, BCL2, BCL2L1, MCL1, BCL2L11 (BIM), CASP3, CASP7, HSP70 (HSPA1A), HSP90AA1, HSPA5, DDIT3 (CHOP), ATF4, CDKN1A (p21), CDKN2A (15) | Apoptosis sensitivity; stress response; replicative capacity | High BIM/BAX = apoptosis-primed (poor persistence); BCL2 overexpression = survival advantage; HSPA5/ATF4 = ER stress from manufacturing conditions | Assess apoptotic priming per cell; identify stressed subpopulations from ex vivo expansion |
| 6 · Donor & Allogeneic CAR-T Considerations · 20 genes | ||||
| GvHD Risk / HLA | TRAC (TCR alpha const.), TRBC1, TRBC2, HLA-A, HLA-B, HLA-C, B2M, CIITA, PDCD1, LAG3, NKG2D, CD38 (12) | αβTCR expression; HLA matching; GvHD risk | TRAC knockout (CRISPR) in allo-CAR-T eliminates GvHD risk; HLA mismatch = host rejection; B2M KO = stealth from host NK; allogeneic CAR-T persistence challenge | Monitor TCR knockout efficiency per cell; identify TCR+ contaminating cells in allo product |
| NK-CAR & Universal CAR Markers | NCAM1 (CD56), KLRD1, NKG7, NCR1, KLRK1 (NKG2D), FCGR3A, PRF1, GZMB, IL15, IL15RA, CD244 (2B4), DNAM1 (CD226) (8) - NK-CAR focus | NK-derived CAR product identity; innate-like anti-tumor function | NK-CAR-T (FT500/CYNK-001): no GvHD; off-the-shelf; NKG2D-CAR provides MHC-independent killing; IL-15 armored NK-CAR-T improves persistence | Characterize NK-CAR product identity; compare NK vs αβT vs γδT CAR product composition |
| 7 · CRS & Neurotoxicity (ICANS) Risk · 20 genes | ||||
| CRS Risk Cytokines | IL6, GM-CSF (CSF2), IL1B, TNF, IFNG, IL2, MCP-1 (CCL2), CXCL10, IL6R, IL6ST (gp130), IL1R1, TNFRSF1A, IL6 amplification loop (13) | Macrophage activation; CRS inflammatory cascade; ICANS pathway | High GM-CSF+IL6 = severe CRS predictor; macrophage IL-6 amplification loop drives grade 3–4 CRS; CSF2 = tocilizumab-refractory CRS risk | Identify GM-CSF-secreting CAR-T subpopulation; predict patient CRS risk from product cytokine profile |
| Endothelial / BBB Markers | ICAM1, VCAM1, SELE (E-selectin), CXCL10, IL6, ANGPT2, VWF, PECAM1, S100B, ENO2 (NSE) (10) - peripheral proxy | Endothelial activation; BBB disruption; ICANS pathophysiology | ICAM1/VCAM1 upregulation = endothelial activation; ANGPT2 = barrier disruption; CSF S100B/NSE = neurotoxicity severity markers | Identify endothelial activation-associated gene expression in tumor vasculature/stroma |
Total: 195 genesCat 1: 54 · Cat 2: 39 · Cat 3: 40 · Cat 4: 35 · Cat 5: 25 · Cat 6: 24 · Cat 7: 23
ⓘ Select genes appear in more than one functional category reflecting their multi-role biology. The total above counts unique genes; per-category counts include all category-relevant entries.