Next-generation immunotherapy candidates for Stage IV HGSOC with cold/immune-excluded tumor microenvironment.
This document is for RESEARCH AND EDUCATIONAL purposes only.
All treatment information requires validation by qualified oncologists.
See PatientOne disclaimers: ./README.md
PatientOne (PAT001-OVC-2025) presents a challenging case for immunotherapy:
| Factor | Status | Implication |
|---|---|---|
| BRCA1 germline | Pathogenic | Higher neoantigen load, but not sufficient alone |
| Platinum-resistant | Yes (8-month recurrence) | Limited standard options remain |
| TME phenotype | Cold / immune-excluded | Low CD8+ infiltration (~5-15 cells/mm²), thick stromal barrier |
| Checkpoint expression | Low PD-L1 in tumor core | Standard anti-PD-1 monotherapy unlikely effective |
| Proliferation | High (Ki67 45-55%) | Aggressive tumor requiring rapid intervention |
| Key mutations | TP53 R175H, PIK3CA E545K, PTEN LOH | PI3K/AKT/mTOR pathway activation |
Standard checkpoint inhibitors (anti-PD-1/PD-L1 monotherapy) show limited efficacy in cold/immune-excluded HGSOC. Next-generation approaches aim to overcome immune exclusion through:
- Bispecific antibodies — Simultaneously engage multiple targets to redirect immune cells
- Innate cell engagers — Activate macrophages/NK cells that are already present in tumor
- Bispecific T-cell engagers (BiTEs) — Force T-cell activation against tumor antigens in cold tumors
- Cancer vaccines — Generate new immune responses against tumor-associated antigens
- Engineered TIL therapy — Expand and enhance tumor-infiltrating lymphocytes ex vivo
| # | Therapy | Type | Developer | Mechanism | Phase | Key Result | PatientOne Relevance |
|---|---|---|---|---|---|---|---|
| 1 | Cadonilimab | PD-1 x CTLA-4 bispecific | Akeso | Dual checkpoint blockade with single molecule | II | 94.1% ORR with chemo (cervical) | Overcomes PD-L1 low limitation |
| 2 | NI-1801 | Mesothelin x CD47 ICE | Light Chain Bio | Activates macrophages against MSLN+ tumor | I | 75% 1-year OS platinum-resistant + pembro | Innate immune activation bypasses cold TME |
| 3 | Ubamatamab | MUC16 x CD3 BiTE | Regeneron | Forces T-cell killing of MUC16+ tumor cells | I/II | Activates T cells in cold tumors | MUC16 (CA-125) highly expressed in HGSOC |
| 4 | IGFBP-2 Vaccine | DNA plasmid vaccine | UW Medicine | Generates T-cell response against IGFBP2 | I (completed) | >50% alive at 8 years | Long-term immune memory |
| 5 | TIL Therapy | Engineered cell therapy | Multiple | Expand tumor-reactive T cells ex vivo | I/II | CD28/IL-15 methods improve persistence | Can rescue immune response from excluded cells |
Developer: Akeso Inc. Mechanism: Bispecific antibody simultaneously blocking PD-1 and CTLA-4, providing dual checkpoint inhibition with a single molecule. The bispecific format enables cis-binding on the same T cell, potentially more effective than combination of two separate antibodies.
Clinical Evidence:
- Phase II cervical cancer: 94.1% ORR when combined with chemotherapy (practice-changing results)
- Phase II recurrent ovarian (NCT06560112): Recruiting, 172 patients, combination with AK112, estimated completion 2026
- Phase I/II ovarian + radiation (NCT06940921): Low-dose radiation + SBRT + cadonilimab for advanced ovarian with peritoneal metastases
- Approved: NMPA-approved (China) for cervical cancer (2022)
NCT IDs:
- NCT06560112 — Phase II, recurrent ovarian cancer, AK104 + AK112 + chemo (Recruiting)
- NCT06940921 — Phase I/II, LDRT + SBRT + cadonilimab, ovarian with peritoneal mets
Relevance to PatientOne:
- Dual checkpoint blockade may overcome low PD-L1 expression in cold TME
- CTLA-4 blockade can prime new T-cell responses (vs. PD-1 which reactivates existing)
- Combined with chemotherapy may convert cold → hot TME
- Platinum-resistant ovarian cancer is an active enrollment indication
Platform Tool Analysis Approach:
- mcp-spatialtools: Map PD-L1 (CD274), CTLA-4, PD-1 (PDCD1) spatial expression across tumor regions
- mcp-multiomics: Check CTLA-4 and PD-1 RNA/protein expression levels in resistant vs. sensitive samples
- mcp-perturbation (GEARS): Predict effect of dual PD-1/CTLA-4 blockade on immune activation genes
- ClinicalTrials.gov MCP (external): Search for recruiting cadonilimab + ovarian cancer trials
- PubMed MCP (external): Review latest cadonilimab efficacy data in gynecologic cancers
Developer: Light Chain Bioscience (Novimmune SA) Mechanism: Bispecific antibody targeting mesothelin (MSLN) on tumor cells and CD47 ("don't eat me" signal). By blocking CD47, NI-1801 enables macrophage-mediated phagocytosis of MSLN-expressing tumor cells. This innate cell engager (ICE) approach bypasses the need for T-cell infiltration — critical for cold tumors.
Clinical Evidence:
- Phase 1b (NCT05403554): NI-1801 + pembrolizumab in platinum-resistant ovarian cancer showed 75% 1-year overall survival rate (2025 data)
- Single agent and combination arms with paclitaxel also being evaluated
- Recruiting across epithelial ovarian, pancreatic, NSCLC, and TNBC
NCT IDs:
- NCT05403554 — Phase 1, NI-1801 as single agent, + anti-PD-1, + paclitaxel in MSLN+ cancers including ovarian (Recruiting, 70 patients, 7 sites)
Relevance to PatientOne:
- Activates innate immune system (macrophages) which ARE present in PatientOne's stroma
- Does not require T-cell infiltration — ideal for immune-excluded/cold TME
- Mesothelin frequently overexpressed in HGSOC
- Combination with pembrolizumab addresses both innate and adaptive immunity
- Platinum-resistant ovarian cancer is a primary study population
Platform Tool Analysis Approach:
- mcp-spatialtools: Map MSLN spatial expression (check if in Visium panel); map CD68 (macrophage marker) distribution
- mcp-multiomics: Check MSLN and CD47 RNA/protein levels across tumor samples
- mcp-deepcell: Quantify macrophage density and distribution from imaging (CD68+ cells)
- ClinicalTrials.gov MCP (external): Verify NCT05403554 enrollment status and eligibility criteria
- PubMed MCP (external): Search NI-1801 clinical data and mesothelin-targeting in ovarian cancer
Note: MSLN and CD47 may not be present in PatientOne's synthetic 31-gene Visium panel. If unavailable, check multi-omics RNA data or note as a gene panel limitation.
Developer: Regeneron Pharmaceuticals Mechanism: Bispecific antibody binding MUC16 (membrane-bound CA-125) on tumor cells and CD3 on T cells, forcing T-cell activation and tumor killing regardless of T-cell receptor specificity. This BiTE (bispecific T-cell engager) approach can activate T cells even in cold tumors where natural tumor recognition is absent.
Clinical Evidence:
- Phase I/II (NCT03564340): 890-patient study of ubamatamab alone and with cemiplimab in recurrent ovarian cancer, recruiting since 2018
- Phase I/II (NCT04590326): REGN5668 (MUC16xCD28) + ubamatamab + cemiplimab combinations, 612 patients
- Phase II (NCT06787612): Multi-arm study in platinum-resistant ovarian cancer, 220 patients, combinations with bevacizumab, cemiplimab, PLD (Recruiting 2025)
- Demonstrated T-cell activation in MUC16+ cold tumors in early data
NCT IDs:
- NCT03564340 — Phase 1/2, ubamatamab ± cemiplimab, recurrent ovarian (Recruiting, 890 patients, 51 sites)
- NCT04590326 — Phase 1/2, MUC16xCD28 + ubamatamab combos, MUC16+ malignancies (Recruiting, 612 patients)
- NCT06787612 — Phase 2, ubamatamab ± agents, platinum-resistant ovarian (Recruiting, 220 patients)
- NCT06444880 — Phase 2, ubamatamab in SMARCB1-deficient MUC16+ malignancies
Relevance to PatientOne:
- MUC16 (CA-125) is the most established ovarian cancer biomarker — PatientOne has elevated CA-125
- BiTE mechanism forces T-cell activation in cold tumors regardless of pre-existing immune infiltration
- Platinum-resistant ovarian cancer is a primary study population (NCT06787612)
- Combination with checkpoint inhibitors may provide synergistic benefit
- Largest clinical program of any next-gen immunotherapy for ovarian cancer
Platform Tool Analysis Approach:
- mcp-spatialtools: Map MUC16 spatial expression if available in Visium panel; check CD3D, CD3E expression
- mcp-multiomics: Quantify MUC16 RNA/protein levels in tumor vs. stroma
- mcp-perturbation (GEARS): Predict T-cell activation gene changes upon MUC16-CD3 engagement
- mcp-mockepic: Retrieve CA-125 (MUC16) trends from clinical data as proxy for target expression
- ClinicalTrials.gov MCP (external): Check eligibility and recruiting sites for NCT03564340, NCT06787612
Developer: UW Medicine (University of Washington) Mechanism: DNA plasmid vaccine encoding IGFBP-2 (insulin-like growth factor binding protein 2), a protein overexpressed in many ovarian cancers. The vaccine generates de novo T-cell responses against IGFBP-2-expressing tumor cells, creating long-term immune memory.
Clinical Evidence:
- Phase I (NCT01322802, Completed): 25 patients with advanced ovarian cancer. >50% of vaccinated patients alive at 8+ years — remarkable long-term survival
- Phase II (NCT03029611, Terminated): Concurrent IGFBP-2 vaccination + neoadjuvant chemo in Stage III/IV ovarian cancer. Terminated after 11 patients (enrollment challenges, not safety)
- Durable immune responses detected in Phase I responders
NCT IDs:
- NCT01322802 — Phase I, IGFBP-2 vaccine in advanced ovarian cancer (Completed, 25 patients)
- NCT03029611 — Phase II, IGFBP-2 vaccine + neoadjuvant chemo (Terminated, 11 patients)
Relevance to PatientOne:
- Vaccine approach generates new T-cell responses — does not depend on existing infiltration
- Long-term immune memory may provide durable protection against recurrence
- IGFBP-2 overexpression common in HGSOC
- Could be combined with checkpoint inhibitors to enhance vaccine-generated T-cell activity
- Particularly relevant post-debulking surgery as maintenance/adjuvant strategy
Platform Tool Analysis Approach:
- mcp-multiomics: Check IGFBP2 RNA and protein expression levels in PatientOne tumor samples
- mcp-spatialtools: Map IGFBP2 spatial expression if available in Visium panel
- mcp-genomic-results (fgbio): Check for IGFBP2 genomic alterations (amplification, overexpression)
- ClinicalTrials.gov MCP (external): Search for next-generation IGFBP-2 vaccine trials
- PubMed MCP (external): Review long-term follow-up data from NCT01322802
Note: IGFBP2 may not be in PatientOne's synthetic 31-gene Visium panel. Check multi-omics RNA data for expression levels.
Developer: Multiple academic centers and companies (NCI, Iovance Biotherapeutics, Leiden UMC) Mechanism: Tumor-infiltrating lymphocytes (TILs) are extracted from a tumor biopsy, expanded ex vivo to billions of cells, and re-infused after lymphodepletion. Modern engineered TIL approaches (2025+) use CD28 co-stimulatory domains and IL-15 cytokine support to improve T-cell persistence and anti-tumor activity, even from cold tumors with limited initial infiltration.
Clinical Evidence:
- Phase II NCI (NCT01174121): TIL + pembrolizumab in metastatic ovarian cancer, 332-patient multi-cancer study (Recruiting, estimated completion 2028)
- Phase I/II Denmark (NCT03287674, Completed): TIL + ipilimumab/nivolumab in metastatic ovarian, 7 patients
- Phase I/II Leiden (NCT04072263): Adoptive T-cell therapy in recurrent ovarian cancer, 12 patients
- Phase Ib Toronto (NCT03158935, Completed): TIL + pembrolizumab in advanced ovarian and melanoma
- 2025 advances: CD28/IL-15 engineering methods improve TIL expansion from cold tumors and in vivo persistence
NCT IDs:
- NCT01174121 — Phase II, TIL + pembrolizumab, metastatic ovarian and other cancers (Recruiting, NCI, 332 patients)
- NCT03287674 — Phase I/II, TIL + ipilimumab/nivolumab, metastatic ovarian (Completed)
- NCT04072263 — Phase I/II, adoptive T-cell therapy, recurrent ovarian (Leiden UMC)
- NCT03158935 — Phase Ib, TIL + pembrolizumab, advanced ovarian (Completed)
Relevance to PatientOne:
- Can recover tumor-reactive T cells even from immune-excluded tumors
- PatientOne has CD8+ T cells at tumor periphery/stroma — these can be harvested for TIL expansion
- Engineered TIL with CD28/IL-15 improvements may overcome exhaustion markers
- Combination with checkpoint inhibitors prevents re-exhaustion after infusion
- NCI Phase II (NCT01174121) is actively recruiting for metastatic ovarian cancer
Platform Tool Analysis Approach:
- mcp-spatialtools: Map CD8A, CD8B, GZMB (granzyme B) expression at tumor periphery — identifies TIL harvest site
- mcp-deepcell: Quantify CD8+ cell density in stroma vs. tumor; assess TIL harvest feasibility
- mcp-quantum-celltype-fidelity: Assess T-cell exhaustion markers (PDCD1, LAG3, HAVCR2/TIM3, TIGIT) to predict TIL quality
- mcp-perturbation (GEARS): Predict gene expression changes after TIL + checkpoint combination
- ClinicalTrials.gov MCP (external): Check NCT01174121 eligibility criteria and recruiting sites
| Tool | Type | Server | Immunotherapy Use |
|---|---|---|---|
| Spatial expression mapping | Built-in | mcp-spatialtools | Map immune markers, checkpoint ligands, target antigens across tissue |
| Perturbation prediction | Built-in | mcp-perturbation (GEARS) | Predict treatment response, immune activation |
| Cell type fidelity | Built-in | mcp-quantum-celltype-fidelity | Assess T-cell states, exhaustion, activation |
| Multi-omics integration | Built-in | mcp-multiomics | Check target RNA/protein expression levels |
| Cell segmentation | Built-in | mcp-deepcell | Quantify immune cell density and distribution |
| Image analysis | Built-in | mcp-openimagedata | Analyze IF/MxIF immune staining |
| Genomic analysis | Built-in | mcp-fgbio | Check genomic alterations in immunotherapy targets |
| Clinical data | Built-in | mcp-mockepic | Retrieve biomarkers (CA-125/MUC16), treatment history |
| TCGA comparison | Built-in | mcp-mocktcga | Compare immune profiles to TCGA ovarian cohort |
| Clinical trial search | External MCP | ClinicalTrials.gov MCP | Search recruiting trials, check eligibility |
| Literature search | External MCP | PubMed MCP | Find clinical evidence, review articles |
| Mutation analysis | Future work | cBioPortal MCP | Cross-reference mutations across cancer cohorts |
| Reasoning chains | Future work | SequentialThinking MCP | Multi-step treatment decision reasoning |
Built-in servers are part of the Precision Medicine MCP platform and available in Claude Desktop when configured.
External MCP tools (ClinicalTrials.gov, PubMed) are available as separate MCP servers in Claude Desktop and Claude Code — they are not bundled with the platform but can be added to claude_desktop_config.json.
Future work tools (cBioPortal, SequentialThinking) are planned but not yet implemented.
| Therapy | Cold TME | BRCA1 Mutation | Platinum Resistance | Target Expression | Overall Fit |
|---|---|---|---|---|---|
| Cadonilimab | ★★★ Dual checkpoint may convert cold→hot | ★★ BRCA1 increases neoantigens | ★★★ Active in resistant disease | ★★ Requires PD-1/CTLA-4 on T cells | High |
| NI-1801 | ★★★ Bypasses T-cell requirement (innate) | ★ Independent of BRCA | ★★★ Active in platinum-resistant | ★★★ MSLN common in HGSOC | High |
| Ubamatamab | ★★★ Forces T-cell activation in cold tumors | ★ Independent of BRCA | ★★★ Phase 2 in platinum-resistant | ★★★ MUC16/CA-125 highly expressed | Very High |
| IGFBP-2 Vaccine | ★★ Generates new responses, slow onset | ★ Independent of BRCA | ★ Better as maintenance | ★★ IGFBP2 common in HGSOC | Moderate |
| TIL Therapy | ★★★ Recovers T cells from periphery | ★★ BRCA1 tumors have more neoantigens | ★★ Independent of prior therapy | ★★★ CD8+ present at periphery | High |
- Ubamatamab — Strongest evidence (largest trials), directly targets CA-125/MUC16, designed for platinum-resistant ovarian
- NI-1801 — Uniquely addresses cold TME via innate immunity, promising early survival data
- Cadonilimab — Dual checkpoint may overcome PD-L1-low limitation, active ovarian trials
- TIL Therapy — Personalized approach, NCI trial recruiting, requires specialized centers
- IGFBP-2 Vaccine — Long-term potential, better suited as maintenance after initial response
| NCT ID | Therapy | Phase | Status |
|---|---|---|---|
| NCT06560112 | Cadonilimab + AK112, recurrent ovarian | Phase II | Recruiting |
| NCT06940921 | Cadonilimab + radiation, advanced ovarian | Phase I/II | Active |
| NCT05403554 | NI-1801 ± pembrolizumab ± paclitaxel, MSLN+ cancers | Phase 1 | Recruiting |
| NCT03564340 | Ubamatamab ± cemiplimab, recurrent ovarian | Phase 1/2 | Recruiting |
| NCT04590326 | REGN5668 + ubamatamab combos, MUC16+ cancers | Phase 1/2 | Recruiting |
| NCT06787612 | Ubamatamab multi-arm, platinum-resistant ovarian | Phase 2 | Recruiting |
| NCT01322802 | IGFBP-2 vaccine, advanced ovarian | Phase I | Completed |
| NCT03029611 | IGFBP-2 vaccine + neoadjuvant chemo, ovarian | Phase II | Terminated |
| NCT01174121 | TIL + pembrolizumab, metastatic ovarian | Phase II | Recruiting |
| NCT03287674 | TIL + ipilimumab/nivolumab, metastatic ovarian | Phase I/II | Completed |
- PatientOne Quick Start Guide — Complete patient overview and workflow
- Data Modes Guide — DRY_RUN vs. Actual Data configuration
- TEST 5: Integration — Integrated analysis including immunotherapy recommendations
- Prompt Library: Clinical & Genomic — Prompt 17: Immunotherapy targets
- Prompt Library: Workflows — Section 5b: Immunotherapy research workflow
- Prompt Library README — Full prompt index
Last Updated: 2026-02-13 NCT IDs Verified: Via ClinicalTrials.gov MCP (February 2026) Data: All analysis approaches reference PatientOne synthetic data (PAT001-OVC-2025)