BioGrade · expert reasoning data for biopharma agentsEnvironmentsBenchmarkCompare modelsData flywheelThesis
← TAA Surface-Target Prioritization
Discovery / Target Biology · Target biologist / antibody & cell-therapy discovery scientist (surface-antigen selection)

TAA Surface-Target Prioritization — agent trajectory for expert review

claude-sonnet-4-6 · 2026-06-04T05:51:17.390589+00:00 · 3 tool calls · 35.3s · 4851 tokens
run id: taa_prioritization_20260603_225041_01_evaluate-gpc3-as-a-tumor-associated-antigen-for

Query

Evaluate GPC3 as a tumor-associated antigen for hepatocellular carcinoma (LIHC).

Trajectory

Agent · turn 1

Sure! To evaluate GPC3 as a tumor-associated antigen for hepatocellular carcinoma (LIHC), I need to gather protein annotation, normal tissue expression, and tumor expression data. Let me fetch all of this in parallel.

⚙ calls uniprot_lookup(gene=GPC3)
⚙ calls gtex_expression(gene=GPC3)
⚙ calls tcga_expression(gene=GPC3)
↳ uniprot_lookup result
show data
{
  "source": "UniProtKB/Swiss-Prot",
  "endpoint": "https://rest.uniprot.org/uniprotkb/search",
  "organism": "Homo sapiens (9606)",
  "reviewed": true,
  "gene": "GPC3",
  "primaryAccession": "P51654",
  "proteinName": "Glypican-3",
  "subcellularLocations": [
    "Cell membrane"
  ],
  "transmembraneTopology": "GPI-anchor",
  "transmembraneCount": 0,
  "note": "Oncofetal; near-absent in adult normal tissue -> excellent specificity."
}
↳ gtex_expression result
show data
{
  "source": "GTEx v8",
  "endpoint": "https://gtexportal.org/api/v2/expression/medianGeneExpression",
  "datasetId": "gtex_v8",
  "gene": "GPC3",
  "unit": "TPM",
  "medianGeneExpression": {
    "Heart - Left Ventricle": 0.2,
    "Brain - Cortex": 0.3,
    "Lung": 1.0,
    "Liver": 0.3,
    "Kidney - Cortex": 0.5,
    "Colon - Transverse": 0.2,
    "Stomach": 0.3,
    "Pancreas": 0.2,
    "Nerve - Tibial": 0.3,
    "Whole Blood": 0.1,
    "Spleen": 0.2,
    "Skin - Sun Exposed": 0.5,
    "Esophagus - Mucosa": 0.3,
    "Thyroid": 0.5,
    "Testis": 2.0,
    "Ovary": 1.0,
    "Breast - Mammary Tissue": 0.5,
    "Prostate": 0.3
  }
}
↳ tcga_expression result
show data
{
  "source": "TCGA (UCSC Xena Toil recompute)",
  "endpoint": "https://api.gdc.cancer.gov/expression",
  "gene": "GPC3",
  "unit": "TPM",
  "medianTumorExpressionByCohort": {
    "BRCA": 2.0,
    "STAD": 3.0,
    "LUAD": 5.0,
    "LUSC": 8.0,
    "COAD": 1.0,
    "OV": 5.0,
    "PAAD": 2.0,
    "LIHC": 80.0,
    "HNSC": 3.0,
    "GBM": 2.0,
    "SCLC": 5.0,
    "DLBC": 0.5,
    "KIRC": 1.0,
    "PRAD": 1.0
  }
}
Agent · turn 2

Here is a comprehensive evaluation of GPC3 (Glypican-3) as a tumor-associated antigen (TAA) for hepatocellular carcinoma (LIHC):

🔬 GPC3 as a Tumor-Associated Antigen in Hepatocellular Carcinoma (LIHC)

1. 🧬 Protein Annotation (UniProtKB/Swiss-Prot · P51654)

Property Details
Full Name Glypican-3
Accession P51654
Subcellular Location Cell membrane
Membrane Topology GPI-anchored (extracellular, no transmembrane domains)
Transmembrane Domains 0
Key Note Oncofetal protein; near-absent in normal adult tissue → excellent specificity

Why this matters for a TAA:
- Being GPI-anchored on the extracellular face of the plasma membrane makes GPC3 highly accessible to antibodies, CAR-T cells, and bispecific engagers without the need for internalization.
- Its oncofetal nature means it is expressed during fetal development, silenced in healthy adult tissues, and re-expressed in tumors — a hallmark of an ideal TAA.

2. 📊 Normal Tissue Expression (GTEx v8 — Median TPM)

Tissue Median TPM
Whole Blood 0.1
Heart – Left Ventricle 0.2
Colon – Transverse 0.2
Pancreas 0.2
Spleen 0.2
Brain – Cortex 0.3
Liver ✅ 0.3
Stomach 0.3
Esophagus – Mucosa 0.3
Nerve – Tibial 0.3
Prostate 0.3
Kidney – Cortex 0.5
Skin – Sun Exposed 0.5
Thyroid 0.5
Breast – Mammary Tissue 0.5
Lung 1.0
Ovary 1.0
Testis 2.0

Key observations:
- Across virtually all normal somatic tissues, GPC3 expression is extremely low (≤ 1.0 TPM), with normal liver at only 0.3 TPM.
- The highest normal-tissue expression is in testis (2.0 TPM), which is an immunologically privileged site — this is a characteristic of many cancer-testis antigens and does not raise significant on-target/off-tumor safety concerns.
- There are no high-expression normal tissues (no values > 2.0 TPM) that would predict dose-limiting toxicity.

3. 🧪 Tumor Expression by TCGA Cohort (Median TPM)

TCGA Cohort Median TPM Relative Signal
LIHC (Liver HCC) 🎯 80.0 ⭐⭐⭐⭐⭐
LUSC (Lung Squamous) 8.0 ⭐⭐
OV (Ovarian) 5.0 ⭐
LUAD (Lung Adeno) 5.0 ⭐
SCLC (Small Cell Lung) 5.0 ⭐
STAD (Gastric) 3.0 ▪
HNSC (Head & Neck) 3.0 ▪
GBM (Glioblastoma) 2.0 ▪
BRCA (Breast) 2.0 ▪
PAAD (Pancreatic) 2.0 ▪
COAD (Colorectal) 1.0 ▪
KIRC (Kidney Clear Cell) 1.0 ▪
PRAD (Prostate) 1.0 ▪
DLBC (Diffuse Large B-cell) 0.5 ▪

Key observations:
- LIHC shows dramatically higher GPC3 expression (80.0 TPM) than any other tumor type in this panel.
- The tumor-to-normal liver ratio is ~267× (80.0 vs. 0.3 TPM), a remarkable therapeutic window.
- Expression in other tumor types is ≤ 8.0 TPM — suggesting GPC3-targeted therapies will have a strong LIHC-specific profile with manageable off-tumor signals.

4. ✅ Overall TAA Scorecard for GPC3 in LIHC

Criterion Assessment Score
Tumor overexpression 80 TPM in LIHC (median) ✅ Excellent
Normal tissue silencing ≤ 0.3 TPM in liver; ≤ 1.0 TPM in most tissues ✅ Excellent
Tumor-to-normal ratio ~267× in liver ✅ Outstanding
Membrane accessibility GPI-anchored, fully extracellular ✅ Excellent
On-target/off-tumor safety Only testis (2.0 TPM, immunoprivileged) ✅ Low risk
Tumor-selectivity across cancers Clear dominance in LIHC vs. all other cohorts ✅ Highly selective
Oncofetal biology Silenced in adult soma, re-expressed in HCC ✅ Ideal TAA profile

5. 💡 Summary & Therapeutic Implications

GPC3 is one of the strongest TAA candidates currently known for HCC, and the data here fully support its development:

  1. CAR-T / CAR-NK cells: The GPI-anchored extracellular presentation makes GPC3 a perfect CAR target. Multiple clinical trials (e.g., NCT02395250, NCT03884751) are already evaluating GPC3-CAR T-cells in HCC.
  2. Bispecific antibodies / ADCs: The high surface density in HCC (80 TPM) and near absence in normal liver (0.3 TPM) create a favorable pharmacological window.
  3. Diagnostic biomarker: The 267× tumor-to-normal ratio in liver reinforces GPC3's established use as a serum and IHC biomarker for HCC diagnosis.
  4. Safety profile: Testis-only meaningful normal expression is acceptable given immune privilege; no vital organs (heart, brain, lung, liver) show concerning expression levels.

Bottom line: GPC3 earns near-perfect marks as a TAA for LIHC — high tumor expression, near-absent normal expression, cell-surface accessibility, and exceptional liver-tumor specificity make it an ideal candidate for antibody-based, cellular, and targeted therapeutic modalities in hepatocellular carcinoma.

Final output

Here is a comprehensive evaluation of GPC3 (Glypican-3) as a tumor-associated antigen (TAA) for hepatocellular carcinoma (LIHC):

🔬 GPC3 as a Tumor-Associated Antigen in Hepatocellular Carcinoma (LIHC)

1. 🧬 Protein Annotation (UniProtKB/Swiss-Prot · P51654)

Property Details
Full Name Glypican-3
Accession P51654
Subcellular Location Cell membrane
Membrane Topology GPI-anchored (extracellular, no transmembrane domains)
Transmembrane Domains 0
Key Note Oncofetal protein; near-absent in normal adult tissue → excellent specificity

Why this matters for a TAA:
- Being GPI-anchored on the extracellular face of the plasma membrane makes GPC3 highly accessible to antibodies, CAR-T cells, and bispecific engagers without the need for internalization.
- Its oncofetal nature means it is expressed during fetal development, silenced in healthy adult tissues, and re-expressed in tumors — a hallmark of an ideal TAA.

2. 📊 Normal Tissue Expression (GTEx v8 — Median TPM)

Tissue Median TPM
Whole Blood 0.1
Heart – Left Ventricle 0.2
Colon – Transverse 0.2
Pancreas 0.2
Spleen 0.2
Brain – Cortex 0.3
Liver ✅ 0.3
Stomach 0.3
Esophagus – Mucosa 0.3
Nerve – Tibial 0.3
Prostate 0.3
Kidney – Cortex 0.5
Skin – Sun Exposed 0.5
Thyroid 0.5
Breast – Mammary Tissue 0.5
Lung 1.0
Ovary 1.0
Testis 2.0

Key observations:
- Across virtually all normal somatic tissues, GPC3 expression is extremely low (≤ 1.0 TPM), with normal liver at only 0.3 TPM.
- The highest normal-tissue expression is in testis (2.0 TPM), which is an immunologically privileged site — this is a characteristic of many cancer-testis antigens and does not raise significant on-target/off-tumor safety concerns.
- There are no high-expression normal tissues (no values > 2.0 TPM) that would predict dose-limiting toxicity.

3. 🧪 Tumor Expression by TCGA Cohort (Median TPM)

TCGA Cohort Median TPM Relative Signal
LIHC (Liver HCC) 🎯 80.0 ⭐⭐⭐⭐⭐
LUSC (Lung Squamous) 8.0 ⭐⭐
OV (Ovarian) 5.0 ⭐
LUAD (Lung Adeno) 5.0 ⭐
SCLC (Small Cell Lung) 5.0 ⭐
STAD (Gastric) 3.0 ▪
HNSC (Head & Neck) 3.0 ▪
GBM (Glioblastoma) 2.0 ▪
BRCA (Breast) 2.0 ▪
PAAD (Pancreatic) 2.0 ▪
COAD (Colorectal) 1.0 ▪
KIRC (Kidney Clear Cell) 1.0 ▪
PRAD (Prostate) 1.0 ▪
DLBC (Diffuse Large B-cell) 0.5 ▪

Key observations:
- LIHC shows dramatically higher GPC3 expression (80.0 TPM) than any other tumor type in this panel.
- The tumor-to-normal liver ratio is ~267× (80.0 vs. 0.3 TPM), a remarkable therapeutic window.
- Expression in other tumor types is ≤ 8.0 TPM — suggesting GPC3-targeted therapies will have a strong LIHC-specific profile with manageable off-tumor signals.

4. ✅ Overall TAA Scorecard for GPC3 in LIHC

Criterion Assessment Score
Tumor overexpression 80 TPM in LIHC (median) ✅ Excellent
Normal tissue silencing ≤ 0.3 TPM in liver; ≤ 1.0 TPM in most tissues ✅ Excellent
Tumor-to-normal ratio ~267× in liver ✅ Outstanding
Membrane accessibility GPI-anchored, fully extracellular ✅ Excellent
On-target/off-tumor safety Only testis (2.0 TPM, immunoprivileged) ✅ Low risk
Tumor-selectivity across cancers Clear dominance in LIHC vs. all other cohorts ✅ Highly selective
Oncofetal biology Silenced in adult soma, re-expressed in HCC ✅ Ideal TAA profile

5. 💡 Summary & Therapeutic Implications

GPC3 is one of the strongest TAA candidates currently known for HCC, and the data here fully support its development:

  1. CAR-T / CAR-NK cells: The GPI-anchored extracellular presentation makes GPC3 a perfect CAR target. Multiple clinical trials (e.g., NCT02395250, NCT03884751) are already evaluating GPC3-CAR T-cells in HCC.
  2. Bispecific antibodies / ADCs: The high surface density in HCC (80 TPM) and near absence in normal liver (0.3 TPM) create a favorable pharmacological window.
  3. Diagnostic biomarker: The 267× tumor-to-normal ratio in liver reinforces GPC3's established use as a serum and IHC biomarker for HCC diagnosis.
  4. Safety profile: Testis-only meaningful normal expression is acceptable given immune privilege; no vital organs (heart, brain, lung, liver) show concerning expression levels.

Bottom line: GPC3 earns near-perfect marks as a TAA for LIHC — high tumor expression, near-absent normal expression, cell-surface accessibility, and exceptional liver-tumor specificity make it an ideal candidate for antibody-based, cellular, and targeted therapeutic modalities in hepatocellular carcinoma.

Reference answer key (deterministic) — open after grading

Deterministic TAA scoring model (config.py / scoring.py): base = 100·(0.25·tumor_expression + 0.45·tumor/normal_ratio + 0.30·surface_quality), then ×safety multiplier (0.15-1.0, 70% vital-tissue expression + 30% breadth of normal expression). HARD GATE: no surface ECD -> SURFACE_INELIGIBLE, score capped at 15.0. Tiers: >=60 Tier 1, >=40 Tier 2, >=25 Tier 3, else Tier 4. A strong agent reaches the same gating and ordering from the raw tool data — especially rejecting intracellular antigens (TP53, GAPDH) regardless of tumor expression, and reading vital-tissue expression as a dose-limiting liability.

GeneCohortScore (0-100)TierTumor / mean-normal TPM (ratio)Flags
GPC3LIHC95.8Tier 1 - strong80.0 / 0.48 = 53.9x—

Expert grading

Surface-accessibility gating 1 (poor) 5 (excellent)
5: Treats a credible cell-surface extracellular domain as a hard gate: correctly deprioritizes intracellular antigens (TP53, GAPDH) as surface-ineligible for CAR-T/ADC/bispecific regardless of tumor expression, and notes they are only addressable via TCR-T/pMHC.
1: Recommends or fails to disqualify an intracellular antigen for a surface modality, or ignores localization/topology entirely and scores on tumor expression alone.
Tumor/normal specificity 1 (poor) 5 (excellent)
5: Judges the antigen on tumor expression RELATIVE to normal-tissue expression (tumor/normal ratio), not raw tumor TPM; recognizes that a high tumor TPM with high normal expression is a poor therapeutic window.
1: Equates high tumor expression with a good target, ignores normal-tissue baseline, or never computes/considers the tumor-vs-normal ratio.
Normal-tissue / vital-organ safety 1 (poor) 5 (excellent)
5: Flags expression in vital, dose-limiting tissues (heart, brain, lung, liver, etc.) as an on-target/off-tumor toxicity liability and weighs breadth of normal expression; treats vital-tissue expression as a dose-limiting gate, not a minor deduction.
1: Misses or downplays vital-tissue expression, treats safety as an afterthought, or never checks the GTEx normal profile against the vital-tissue list.
Integrated prioritization & tiering 1 (poor) 5 (excellent)
5: Synthesizes surface gate + specificity + safety into a clear, well-ordered prioritization (score/tier or ranked recommendation) that matches the evidence, with calibrated confidence and the right caveats; in comparisons, orders candidates correctly.
1: No clear prioritization, a ranking contradicted by its own evidence, or over/under-confidence inconsistent with the data.
Evidence faithfulness 1 (poor) 5 (excellent)
5: Every number and claim (TPM values, topology, ratios, flags) traces to the tool outputs; no fabricated expression values, variants, or programs.
1: Hallucinates TPM numbers, localization, or clinical programs, or contradicts the returned tool data.
Overall verdict
Your name / ID