Decision spine
Frames the controlling thesis, evidence status, buyer implications, and research-use limits.
Oncology
Zemi Dossier
Research-use only
Next-Generation Immuno-Oncology Platforms
A decision package on engineering therapeutic index across ADCs, bispecifics, immune-cell engagement, tumor microenvironment strategy, and resistance.
Decision this dossier helps answer
Which next-generation IO programs improve therapeutic index rather than simply adding potency, complexity, or combination burden?
Problem this dossier solves
The investable question is therapeutic index, not modality novelty.
The dossier evaluates next-generation immuno-oncology by asking whether a platform widens the treatment window or merely intensifies known failure modes. It ties mechanism, patient selection, resistance, toxicity, and combination strategy to buyer action.
What is moving: current field evidence and platform activity
What is unresolved: decision-grade proof, endpoint logic, and failure modes
What Zemi solves: evidence-gated next-study strategy
Full decision package
109-page research report plus 37-sheet Evidence & Decision Workbook
Buy now
$3,500$5,500Launch price
Controlling thesis
The investable question is therapeutic index, not modality novelty.
The dossier evaluates next-generation immuno-oncology by asking whether a platform widens the treatment window or merely intensifies known failure modes. It ties mechanism, patient selection, resistance, toxicity, and combination strategy to buyer action.
A Zemi Dossier turns uncertain emerging research into a decision package: what is known, what is inferred, what is hypothetical, what would falsify the thesis, and what should be built, funded, avoided, monitored, or validated next.
Research report preview
First 30 pages of the actual Next-Generation Immuno-Oncology Platforms research report.
The public page shows the real document style, density, evidence posture, and decision framing before a buyer requests access to the full Zemi Dossier package.
Research report
The research report is the argument. It is not the whole product.
The PDF research report is the structured argument: field status, evidence maturity, buyer implications, falsifiable hypotheses, power-calculated next studies, strategic recommendations, limitations, and traceability posture.
Inspectable infrastructure
Pairs the report with claim ledgers, evidence tables, power rows, audit surfaces, and decision tools.
Executable next studies
Links hypotheses to endpoint logic, assumptions, sample-size logic, budget, and timeline planning.
Research report table of contents
Public TOC generated from the source report.
The full research report is delivered as a licensed PDF. This table shows the decision structure without publishing the full product.
| Section | What it covers | Buyer value |
|---|---|---|
| Executive Summary | Condenses the dossier thesis, evidence posture, buyer implications, and next-study priorities. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Decision Brief | Turns the report into immediate buyer triage: what matters now, what is still uncertain, and what to do next. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Scope and Evidence Integrity | Defines the evidence window, source boundaries, research-use limits, and labeling discipline. | Lets the buyer audit support strength before relying on the package. |
| Denominator Definitions | Defines the populations, measurements, or comparison bases needed to interpret evidence without overgeneralizing. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Decision Snapshot Overview | Maps findings to buyer actions: build, fund, avoid, monitor, or validate next. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Buyer Decision Cheat Sheet | Maps findings to buyer actions: build, fund, avoid, monitor, or validate next. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Buy Build Wait Avoid | Maps findings to buyer actions: build, fund, avoid, monitor, or validate next. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Source Quality Dashboard | Shows how strong the source base is and where conclusions depend on indirect or thinner support. | Lets the buyer audit support strength before relying on the package. |
| Traceability Certificate | Links claims to source support so users can separate what is known, inferred, and hypothetical. | Lets the buyer audit support strength before relying on the package. |
| Therapeutic Index Engineering Model | Defines the governing argument and the evidence gates that would support or falsify it. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| Tumor Immune Geometry Classifier | Operationalizes the thesis as a reusable decision surface for comparing programs or evidence states. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| Current Signal Update 2025 2026 | Frames the field's June 2026 state, active programs, research fronts, and timing signals. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Cross Modality Convergence Map | Shows how separate research signals converge, diverge, or interact across domains, mechanisms, or study layers. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Cross Domain Synthesis | Shows how separate research signals converge, diverge, or interact across domains, mechanisms, or study layers. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Evidence Maturity and Platform Matrix | Links claims to source support so users can separate what is known, inferred, and hypothetical. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| Competitive and Asset Landscape | Frames the field's June 2026 state, active programs, research fronts, and timing signals. | Connects research claims to practical diligence and monitoring decisions. |
| Key Findings | Summarizes the most decision-relevant findings and their implications for the dossier thesis. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Biomarker Control | Explains which biological signals matter for interpreting the field and discriminating between programs. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Immune Geometry | Summarizes a report component that ties field evidence to the dossier's decision logic. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Mechanism-Specific Combination | Explains which biological signals matter for interpreting the field and discriminating between programs. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Tumor Selectivity | Explains which biological signals matter for interpreting the field and discriminating between programs. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Spatial / Conditional Activation | Explains which biological signals matter for interpreting the field and discriminating between programs. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Resistance Anticipation | Explains which biological signals matter for interpreting the field and discriminating between programs. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Evidentiary Standard | Shows how strong the source base is and where conclusions depend on indirect or thinner support. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Quantitative Evidence | Links claims to source support so users can separate what is known, inferred, and hypothetical. | Lets the buyer audit support strength before relying on the package. |
| Key Findings Quant Synthesis | Summarizes the most decision-relevant findings and their implications for the dossier thesis. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Mechanistic Deep Dive | Explains which biological signals matter for interpreting the field and discriminating between programs. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Tumor Setting Playbook | Operationalizes the thesis as a reusable decision surface for comparing programs or evidence states. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Safety Risk and Failure Modes | Identifies translational, regulatory, safety, or delivery constraints that shape whether the thesis can be tested. | Shows where the thesis could fail or require more validation before action. |
| Manufacturing Cmc and Feasibility | Identifies translational, regulatory, safety, or delivery constraints that shape whether the thesis can be tested. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Clinical and Translational Implications | Identifies translational, regulatory, safety, or delivery constraints that shape whether the thesis can be tested. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Non Obvious Early Opportunities | Connects research maturity to competitive position, program strategy, and diligence watchpoints. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Near Discovery Watch List | Connects research maturity to competitive position, program strategy, and diligence watchpoints. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| Prioritized Hypotheses | Translates the thesis into falsifiable studies, endpoints, assumptions, and sample-size logic. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
| F1 (Flagship, paradigm-level; classifier validation cohort) — Spatial Activation × Tumor Selectivity — the masking-threshold conservation | Operationalizes the thesis as a reusable decision surface for comparing programs or evidence states. | Gives the buyer a reusable way to compare programs, risks, and evidence states. |
| F2 (Flagship, paradigm-level) — Resistance Anticipation × Biomarker Control — myeloid density, not antigen density, as the cold-tumor TCE biomarker | Explains which biological signals matter for interpreting the field and discriminating between programs. | Helps the buyer inspect the argument, evidence posture, and decision logic before relying on the package. |
Evidence & Decision Workbook
The workbook is the decision engine.
A locked public preview of selected workbook-derived sheets: sheet structure, decision modules, power rows, traceability, and audit posture. Licensed buyers receive the full workbook files.
Read-only workbook preview with clipped values and no formulas. Licensed buyers receive the full Evidence & Decision Workbook with editable assumptions and working files.
Workbook inventory
Delivered workbook sheets.
Full inventories belong on dossier pages, not the homepage. Sheet names below are extracted from the actual workbook file.
| # | Workbook sheet | What it does |
|---|---|---|
| 1 | Workbook Guide | Navigation layer for using the workbook, understanding sheet roles, and following the decision package. Public preview exposes Sheet, Contents fields and clipped entries such as Claim Ledger, 90 customer-facing claims, full §32., Evidence Table. |
| 2 | Claim Ledger | Claim-by-claim ledger showing verdicts, source support, inference status, and decision relevance. Public preview exposes claim_id, claim_text, report_section, report_page fields and clipped entries such as K001, In the POLARIX 5-year update,., Quantitative. |
| 3 | Evidence Table | Evidence-verification surface for checking sources, support quality, and the audit trail. Public preview exposes identifier, title, authors, journal fields and clipped entries such as PMID 39651791, Blinatumomab in Standard-Risk B-Cell., Gupta S, Rau RE,. |
| 4 | Source-Support Verdicts | Independent check on whether key claims are actually supported by cited sources. Public preview exposes Claim ID, Claim (short), Primary, Source title fields and clipped entries such as K001, In the POLARIX 5-year update,., PMID 40991874. |
| 5 | Source Quality Summary | Summary of source strength, gaps, and areas where conclusions depend on weaker evidence. Public preview exposes Claims audited (source run), 255 fields and clipped entries such as Customer-facing (kept), Excluded, 165 (89 off-topic + 7 structural +.. |
| 6 | Platform Scorecard | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Mechanism-specific combination, 3, 2, 3, 3 fields from this sheet. |
| 7 | Five-Lever Scorecard | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Lever, Question it answers, What it buys, What it costs (the trade) fields and clipped entries such as Tumor Selectivity, Does it discriminate. |
| 8 | Asset Sponsor Tracker | Tracks programs, trials, sponsors, or competitive context for diligence and monitoring. Public preview exposes Construct, Sponsor, Target / modality, Format / DAR / valency fields and clipped entries such as Tarlatamab (Imdelltra), Amgen, DLL3 × CD3 TCE. |
| 9 | Trial Catalyst Tracker | Tracks programs, trials, sponsors, or competitive context for diligence and monitoring. Public preview exposes Trial ID, Title, Phase, Status fields and clipped entries such as NCT02465060, Molecular Analysis for Therapy., PHASE2. |
| 10 | Safety-Signal Tracker | Tracks biomarkers or readouts that determine whether the thesis is moving toward testable evidence. Public preview exposes Construct, Modality, CRS / ILD, Neuro / other fields and clipped entries such as Tarlatamab, DLL3 TCE, CRS 53%. |
| 11 | Resistance Tracker | Tracks programs, trials, sponsors, or competitive context for diligence and monitoring. Public preview exposes Resistance route, Modality hit, Mechanism, Design answer fields and clipped entries such as Antigen loss / down-regulation, TCE, ADC,. |
| 12 | Biomarker-Endpoint Map | Tracks biomarkers or readouts that determine whether the thesis is moving toward testable evidence. Public preview exposes Biomarker, Construct, Role, Endpoint / effect fields and clipped entries such as DLL3 (IHC), Tarlatamab, DLL3 ADCs, Patient selection. |
| 13 | Regulatory Tracker | Tracks regulatory, endpoint, reimbursement, or qualification requirements that shape testing strategy. Public preview exposes Agent, Regulatory status, Indication, Evidentiary basis / economics fields and clipped entries such as Tarlatamab, Full FDA approval. |
| 14 | Figure Source Map | Traceability surface linking claims, figures, calculations, and workbook outputs back to sources. Public preview exposes figure_id, figure_title, figure_family, section_anchor fields and clipped entries such as FIG-01, The five modalities are five., Central. |
| 15 | Figure Attempt Ledger | Figure and numeric reproduction log for validating calculations, chart inputs, or displayed values. Public preview exposes figure_id, file, backend, status fields and clipped entries such as FIG-01, fig_hero_system.jpg, generative. |
| 16 | Hypotheses Ledger | Maps hypotheses to evidence state, falsification logic, endpoints, and next-study design. Public preview exposes ID, Tier, Paradigm, Axis pair fields and clipped entries such as F1, Flagship, Spatial Activation × Tumor.. |
| 17 | Most Valuable Next Studies | Prioritizes next studies that could change the decision, including endpoints and evidence gates. Public preview exposes #, study_title, research_question, hypothesis_tested fields and clipped entries such as Masked vs unmasked therapeutic., Is spatial. |
| 18 | Validation Falsification Gates | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes gate_id, gate_type, statement, which_readout fields and clipped entries such as VG1, CONFIRMS, 'Safer' next-gen constructs trade.. |
| 19 | Verification & Search Log | Release-discipline surface for revisions, unresolved issues, audit findings, and readiness tracking. Public preview exposes Every database, query, and date.. fields from this sheet. |
| 20 | Excluded Claims | Disposition log for rejected, bounded, or quarantined claims so they do not leak into the decision thesis. Public preview exposes ID, Set, Claim (short), Primary fields and clipped entries such as B001, beyond_capacity, Excluded 187 — Immune checkpoint.. |
| 21 | Salvage Ledger | Disposition log for rejected, bounded, or quarantined claims so they do not leak into the decision thesis. Public preview exposes Source element, Decision, What changed fields and clipped entries such as Five-axis 'triangle' organizing idea, Re-position,. |
| 22 | Evidence Maturity Matrix | Evidence-verification surface for checking sources, support quality, and the audit trail. Public preview exposes Modality class, Highest evidence tier, Lever coverage (paid), Binding failure mode fields and clipped entries such as ADC, clean over-expressed. |
| 23 | Power Calculations | Power-calculation layer for assumptions, endpoint logic, computed sample sizes, and sensitivity checks. Public preview exposes hypothesis_id, study_id, endpoint, test_type fields and clipped entries such as F1, NS-F1, confirmed response, unmasked vs.. |
| 24 | TIG_Input_Template | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Evidence completeness (0-1), Fraction of the five proxies., ____ fields and clipped entries such as Escape velocity (fast/slow/unknown), Measured. |
| 25 | TIG_Feature_Score | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Lever (output class), Meaning (3 = severe deficit), Deficit 0-3 (ENTER), Weight (expert-init) fields and clipped entries such as. |
| 26 | TIG_Class_Output | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Evidence completeness (ENTER 0-1), 0.8 fields and clipped entries such as Source directness (ENTER., Adjacent, Escape velocity (ENTER.. |
| 27 | TIG_Decision_Rules | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Binding lever (class), Recommended strategy /., Primary functional endpoint, Required safety readout fields and clipped entries such as. |
| 28 | TIG_Escape_Rules | Workbook surface supporting a specific evidence, decision, study-planning, or release-control workflow. Public preview exposes Escape velocity (entry form), Trispecific recommendation fields and clipped entries such as fast, Trispecific / dual-antigen. |
| 29 | TIG_Minimum_Assay_Panel | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Assay / data item, Lever, Tier fields and clipped entries such as Tumor-to-normal antigen expression., Antigen selectivity, Required. |
| 30 | TIG_Archetype_Cards | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Archetype, Profile (deficits S/SA/G/R/B), Dominant lever, Second-highest class fields and clipped entries such as Broad solid-tumor TCE,. |
| 31 | TIG_Validation_Log | Executable decision surface for classifying programs, risks, gates, or evidence states. Public preview exposes Linked flagship / validation., F1 (masked-vs-unmasked window. fields and clipped entries such as Held-out performance target, held-out AUC >= 0.80,. |
| 32 | TIG_Traceability_Map | Traceability surface linking claims, figures, calculations, and workbook outputs back to sources. Public preview exposes TIG element, Maps to finding, Evidence ID, Figure fields and clipped entries such as Antigen-selectivity lever, finding-biomarker-wins,. |
| 33 | Traceability Certificate | Traceability surface linking claims, figures, calculations, and workbook outputs back to sources. Public preview exposes claim_id, report_section, report_page, claim_text fields and clipped entries such as K001, Quantitative Evidence, § Quantitative Evidence. |
| 34 | Score Basis | Explains how scores or ratings were assigned so the decision instrument can be inspected. Public preview exposes Platform Scorecard (5 levers), 0-4 per lever, node-fit/maturity/safety/durabili. fields and clipped entries such as Evidence Maturity Matrix,. |
| 35 | Competitive Landscape | Tracks programs, trials, sponsors, or competitive context for diligence and monitoring. Public preview exposes company, asset_platform, modality, target_indication fields and clipped entries such as Amgen, Tarlatamab (Imdelltra), DLL3 x CD3 TCE. |
| 36 | Version Registry | Release-discipline surface for revisions, unresolved issues, audit findings, and readiness tracking. Public preview exposes v2.1 (REV-2 remediation), June 2026, Counts reconciled from one., Claude (Opus 4.8) fields and clipped entries such as v2.0 (premium. |
| 37 | Audit Issues Log | Release-discipline surface for revisions, unresolved issues, audit findings, and readiness tracking. Public preview exposes #, Issue, Status, Resolution fields and clipped entries such as AI-1, Counts reconcile from one source?, RESOLVED. |
Power-calculated next studies
Power calculations are the execution bridge.
Power calculations are planning-grade research calculations for endpoint logic, assumptions, sample size, budget, and timeline. They do not prove hypotheses or constitute clinical protocol.
| Hypothesis | Primary endpoint | Test | Effect | α | Power | Computed n | Design n |
|---|---|---|---|---|---|---|---|
| F1 | confirmed response, unmasked vs masked at matched exposure | two-proportion (pooled) | 0.5 | 0.05 | 0.80 | 170 | 170 |
| F2 | myeloid-density AUC for TCE response vs antigen density | AUC (Hanley-McNeil, external) | 0.72 | 0.05 | 0.80 | 52 | 150 |
| F3 | response in heterogeneous tumors, cleavable vs non-cleavable | two-proportion (pooled); design n x4 for | 0.45 | 0.05 | 0.80 | 89 | 360 |
| F4 | normal-mimic cytotoxicity, logic-gated bystander vs logic-ga | two-sample t (Cohen d) | 0.8 | 0.05 | 0.80 | 25 | 26 |
| F5 | masked-construct response, protease-high vs protease-low | two-proportion (pooled) | 0.5 | 0.05 | 0.80 | 58 | 120 |
Release audit discipline
Audit posture is part of the product.
Public pages do not publish exact audit scores. They should state the release discipline, revision posture, and limitations without implying peer review or validated clinical guidance.
Research-use decision package
Not medical advice, clinical guidance, investment advice, or a clinical protocol.
Known / inferred / hypothetical
The package separates cited evidence from inference and frontier hypothesis generation.
Audit-gated and revision-controlled
Adversarial audit checks source support, formulas, figures, limitations, and release readiness.
Buy this dossier
Full decision package — 109-page report + 37-sheet workbook
Buy now
- Decision research report (109 pages, PDF)
- Evidence & Decision Workbook (37 sheets, .xlsx) — claims, source-support verdicts, power calculations, traceability, and audit log
- Live, editable power-calculated next studies
$3,500$5,500Launch price