← Patent / Freedom-to-Operate Analysis
BD / IP · IP / patent analyst

Patent / Freedom-to-Operate Analysis — agent trajectory for expert review

claude-opus-4-8 · 2026-06-06T09:11:18.996372+00:00 · 7 tool calls · 44.1s · 13822 tokens
run id: fto_patent_20260606_020818_04_assess-fto-for-a-proposed-dll3-x-cd3-t-cell-enga

Query

Assess FTO for a proposed DLL3 x CD3 T-cell-engaging bispecific for small cell lung cancer. Search CN/US/WO, characterise claim overlap (composition vs. method), account for legal status and expiry, and render a CLEAR/WATCH/BLOCKED verdict with design-arounds.

Trajectory

Agent · turn 1

I'll assess freedom-to-operate for your proposed DLL3 × CD3 T-cell engager. Let me start by orienting to the corpus and searching for relevant families.

⚙ calls list_corpus()
⚙ calls search_patents(target=DLL3)
↳ list_corpus result
show data
{
  "n_families": 14,
  "targets": {
    "CLDN18.2": {
      "modalities": [
        "Antibody-drug conjugate",
        "Naked monoclonal antibody"
      ],
      "n_families": 6
    },
    "HER2": {
      "modalities": [
        "Antibody-drug conjugate",
        "Naked monoclonal antibody"
      ],
      "n_families": 4
    },
    "DLL3": {
      "modalities": [
        "Antibody-drug conjugate",
        "T-cell engager (bispecific)"
      ],
      "n_families": 4
    }
  },
  "assignees": [
    "AbbVie / Stemcentrx",
    "Amgen",
    "Astellas Pharma",
    "Boehringer Ingelheim",
    "Byondis",
    "Daiichi Sankyo",
    "Ganymed Pharmaceuticals",
    "Ganymed Pharmaceuticals / Astellas",
    "Genentech",
    "Harbour BioMed",
    "ImmunoGen",
    "Keymed Biosciences",
    "LaNova Medicines",
    "Sotio Biotech"
  ],
  "jurisdictions": [
    "CN",
    "US",
    "WO"
  ]
}
↳ search_patents result
show data
{
  "count": 4,
  "families": [
    {
      "publication_number": "US10000567B2",
      "jurisdiction": "US",
      "assignee": "Amgen",
      "title": "Bispecific T-cell engaging molecules binding DLL3 and CD3",
      "target": "DLL3",
      "modality": "T-cell engager (bispecific)",
      "priority_date": "2015-06-23",
      "expiry": "2036-06-23",
      "legal_status": "GRANTED",
      "claim_types": [
        "composition",
        "method"
      ],
      "keywords": [
        "DLL3",
        "CD3",
        "bispecific",
        "T-cell engager",
        "BiTE",
        "tarlatamab",
        "SCLC",
        "composition of matter",
        "half-life extended"
      ]
    },
    {
      "publication_number": "WO2021091960A1",
      "jurisdiction": "WO",
      "assignee": "Boehringer Ingelheim",
      "title": "DLL3/CD3 bispecific antibodies with an IgG-like format",
      "target": "DLL3",
      "modality": "T-cell engager (bispecific)",
      "priority_date": "2019-11-05",
      "expiry": "2040-11-05",
      "legal_status": "PENDING",
      "claim_types": [
        "composition"
      ],
      "keywords": [
        "DLL3",
        "CD3",
        "bispecific",
        "2+1 format",
        "knobs into holes",
        "pending",
        "IgG-like"
      ]
    },
    {
      "publication_number": "US9382323B2",
      "jurisdiction": "US",
      "assignee": "AbbVie / Stemcentrx",
      "title": "Anti-DLL3 antibody-drug conjugates (rovalpituzumab class)",
      "target": "DLL3",
      "modality": "Antibody-drug conjugate",
      "priority_date": "2013-08-28",
      "expiry": "2034-08-28",
      "legal_status": "GRANTED",
      "claim_types": [
        "composition"
      ],
      "keywords": [
        "DLL3",
        "ADC",
        "PBD",
        "rovalpituzumab",
        "cleavable linker",
        "composition of matter",
        "SCLC"
      ]
    },
    {
      "publication_number": "CN112457404A",
      "jurisdiction": "CN",
      "assignee": "Harbour BioMed",
      "title": "Heavy-chain-only DLL3 x CD3 bispecific for small cell lung cancer",
      "target": "DLL3",
      "modality": "T-cell engager (bispecific)",
      "priority_date": "2019-09-10",
      "expiry": "2039-09-10",
      "legal_status": "PENDING",
      "claim_types": [
        "method"
      ],
      "keywords": [
        "DLL3",
        "CD3",
        "bispecific",
        "heavy chain only",
        "step-up dosing",
        "method of treatment",
        "China",
        "pending"
      ]
    }
  ]
}
Agent · turn 2

The corpus has 4 DLL3 families, 3 of which are bispecific T-cell engagers directly on point. Let me pull the full claims for all four to analyze scope.

⚙ calls get_claims(publication_number=US10000567B2)
⚙ calls get_claims(publication_number=WO2021091960A1)
⚙ calls get_claims(publication_number=CN112457404A)
⚙ calls get_claims(publication_number=US9382323B2)
↳ get_claims result
show data
{
  "publication_number": "US10000567B2",
  "title": "Bispecific T-cell engaging molecules binding DLL3 and CD3",
  "legal_status": "GRANTED",
  "independent_claims": [
    {
      "claim_no": 1,
      "claim_type": "composition",
      "text": "A bispecific single-chain molecule comprising a first binding domain that specifically binds human DLL3 and a second binding domain that specifically binds human CD3 epsilon, optionally fused to an Fc domain extending half-life."
    },
    {
      "claim_no": 11,
      "claim_type": "method",
      "text": "A method of treating small cell lung cancer comprising administering the bispecific molecule of claim 1."
    }
  ]
}
↳ get_claims result
show data
{
  "publication_number": "WO2021091960A1",
  "title": "DLL3/CD3 bispecific antibodies with an IgG-like format",
  "legal_status": "PENDING",
  "independent_claims": [
    {
      "claim_no": 1,
      "claim_type": "composition",
      "text": "A bispecific antibody in a 2+1 IgG format comprising two binding sites for human DLL3 and one binding site for human CD3, with a knobs-into-holes heterodimeric Fc."
    }
  ]
}
↳ get_claims result
show data
{
  "publication_number": "CN112457404A",
  "title": "Heavy-chain-only DLL3 x CD3 bispecific for small cell lung cancer",
  "legal_status": "PENDING",
  "independent_claims": [
    {
      "claim_no": 1,
      "claim_type": "method",
      "text": "A method of treating relapsed small cell lung cancer comprising administering a heavy-chain-only DLL3 x CD3 bispecific antibody at a step-up dosing schedule to mitigate cytokine release syndrome."
    }
  ]
}
↳ get_claims result
show data
{
  "publication_number": "US9382323B2",
  "title": "Anti-DLL3 antibody-drug conjugates (rovalpituzumab class)",
  "legal_status": "GRANTED",
  "independent_claims": [
    {
      "claim_no": 1,
      "claim_type": "composition",
      "text": "An antibody-drug conjugate comprising an anti-DLL3 antibody conjugated via a cleavable linker to a pyrrolobenzodiazepine (PBD) dimer payload."
    }
  ]
}
Agent · turn 3

I have full claim scope for all four families. Let me pull the full record on the Amgen patent — the dominant threat — to confirm priority/expiry and any dependent specificity.

⚙ calls get_patent(publication_number=US10000567B2)
↳ get_patent result
show data
{
  "publication_number": "US10000567B2",
  "jurisdiction": "US",
  "assignee": "Amgen",
  "priority_date": "2015-06-23",
  "expiry": "2036-06-23",
  "legal_status": "GRANTED",
  "title": "Bispecific T-cell engaging molecules binding DLL3 and CD3",
  "target": "DLL3",
  "modality": "T-cell engager (bispecific)",
  "independent_claims": [
    {
      "claim_no": 1,
      "claim_type": "composition",
      "text": "A bispecific single-chain molecule comprising a first binding domain that specifically binds human DLL3 and a second binding domain that specifically binds human CD3 epsilon, optionally fused to an Fc domain extending half-life."
    },
    {
      "claim_no": 11,
      "claim_type": "method",
      "text": "A method of treating small cell lung cancer comprising administering the bispecific molecule of claim 1."
    }
  ],
  "keywords": [
    "DLL3",
    "CD3",
    "bispecific",
    "T-cell engager",
    "BiTE",
    "tarlatamab",
    "SCLC",
    "composition of matter",
    "half-life extended"
  ]
}
Agent · turn 4

FTO Assessment: DLL3 × CD3 T-Cell Engager for SCLC

Verdict: 🔴 BLOCKED (in the US, through 2036) — with conditional design-around paths


Corpus landscape

Four DLL3 families; three are directly relevant bispecific TCEs, one is an off-modality ADC.

Family Assignee Juris. Modality Claim types Status Expiry
US10000567B2 Amgen (tarlatamab) US DLL3×CD3 TCE Composition + Method GRANTED 2036-06-23
WO2021091960A1 Boehringer Ingelheim WO DLL3×CD3 2+1 IgG Composition PENDING 2040-11-05
CN112457404A Harbour BioMed CN DLL3×CD3 HCAb Method PENDING 2039-09-10
US9382323B2 AbbVie/Stemcentrx US DLL3 ADC Composition GRANTED 2034-08-28

Claim-by-claim overlap analysis

1. US10000567B2 (Amgen) — the blocking reference

2. WO2021091960A1 (Boehringer) — composition, narrower

3. CN112457404A (Harbour) — method, China

4. US9382323B2 (AbbVie/Stemcentrx) — off-modality


Composition vs. method synthesis


Why BLOCKED

The Amgen genus composition claim is granted, in-force, US, and written at a level of generality that an ordinary DLL3×CD3 single-chain bispecific cannot escape on format or sequence alone. Until June 23, 2036, US activity is exposed on both composition and method.


Design-around / risk-mitigation options

To escape Amgen claim 1 (the priority task):
1. Avoid the single-chain construct. Claim 1 recites a "bispecific single-chain molecule." A two-chain, full-length heterodimeric IgG-like or Fab-arm-exchange format may fall outside the literal claim language — requires careful claim-construction/file-history review of "single-chain" to confirm it isn't construed to read on the binding-domain arrangement generally.
2. Target CD3 via a non-epsilon epitope (e.g., CD3γ/δ) — claim 1 specifies "CD3 epsilon." A genuinely non-ε-binder would be outside, though most functional CD3 engagers hit ε; verify epitope experimentally.
3. Confirm validity exposure. Test whether the 2015 genus claim is vulnerable to written-description/enablement challenge (it claims any DLL3 + any CD3ε binder without sequence support) — potential invalidity/IPR lever rather than design-around.

To avoid Boehringer WO: use a 1+1 monovalent-DLL3 architecture or non-KiH heterodimerization (e.g., charge-pair, CrossMab without KiH). Monitor national-phase claims.

To avoid Harbour CN (if China commercialization intended): use conventional VH/VL (not heavy-chain-only) and/or a non–step-up dosing regimen; or limit China launch.

Jurisdiction/timing strategy:
- The Amgen barrier is US. Confirm whether Amgen has corresponding WO/EP/CN family members (not in this corpus — recommend external family-member search) before assuming ex-US freedom.
- Earliest clean US entry is mid-2036 absent a successful design-around or invalidity outcome.


Recommended next steps

  1. Pull the Amgen file history / dependent claims and external INPADOC family to map ex-US counterparts (this corpus may be incomplete on Amgen's WO/EP/CN).
  2. Claim-construction opinion on "single-chain" and "CD3 epsilon" to validate design-around routes 1–2.
  3. Validity study of the Amgen genus composition claim (§112 written description/enablement).
  4. Track prosecution of the two PENDING families (Boehringer, Harbour) for issued-claim scope.

Caveat: This analysis is limited to the supplied 14-family corpus. Family members in untested jurisdictions and unpublished applications are not captured; a full FTO requires a broader patent database search and qualified patent-counsel opinion.

Final output

FTO Assessment: DLL3 × CD3 T-Cell Engager for SCLC

Verdict: 🔴 BLOCKED (in the US, through 2036) — with conditional design-around paths


Corpus landscape

Four DLL3 families; three are directly relevant bispecific TCEs, one is an off-modality ADC.

Family Assignee Juris. Modality Claim types Status Expiry
US10000567B2 Amgen (tarlatamab) US DLL3×CD3 TCE Composition + Method GRANTED 2036-06-23
WO2021091960A1 Boehringer Ingelheim WO DLL3×CD3 2+1 IgG Composition PENDING 2040-11-05
CN112457404A Harbour BioMed CN DLL3×CD3 HCAb Method PENDING 2039-09-10
US9382323B2 AbbVie/Stemcentrx US DLL3 ADC Composition GRANTED 2034-08-28

Claim-by-claim overlap analysis

1. US10000567B2 (Amgen) — the blocking reference

2. WO2021091960A1 (Boehringer) — composition, narrower

3. CN112457404A (Harbour) — method, China

4. US9382323B2 (AbbVie/Stemcentrx) — off-modality


Composition vs. method synthesis


Why BLOCKED

The Amgen genus composition claim is granted, in-force, US, and written at a level of generality that an ordinary DLL3×CD3 single-chain bispecific cannot escape on format or sequence alone. Until June 23, 2036, US activity is exposed on both composition and method.


Design-around / risk-mitigation options

To escape Amgen claim 1 (the priority task):
1. Avoid the single-chain construct. Claim 1 recites a "bispecific single-chain molecule." A two-chain, full-length heterodimeric IgG-like or Fab-arm-exchange format may fall outside the literal claim language — requires careful claim-construction/file-history review of "single-chain" to confirm it isn't construed to read on the binding-domain arrangement generally.
2. Target CD3 via a non-epsilon epitope (e.g., CD3γ/δ) — claim 1 specifies "CD3 epsilon." A genuinely non-ε-binder would be outside, though most functional CD3 engagers hit ε; verify epitope experimentally.
3. Confirm validity exposure. Test whether the 2015 genus claim is vulnerable to written-description/enablement challenge (it claims any DLL3 + any CD3ε binder without sequence support) — potential invalidity/IPR lever rather than design-around.

To avoid Boehringer WO: use a 1+1 monovalent-DLL3 architecture or non-KiH heterodimerization (e.g., charge-pair, CrossMab without KiH). Monitor national-phase claims.

To avoid Harbour CN (if China commercialization intended): use conventional VH/VL (not heavy-chain-only) and/or a non–step-up dosing regimen; or limit China launch.

Jurisdiction/timing strategy:
- The Amgen barrier is US. Confirm whether Amgen has corresponding WO/EP/CN family members (not in this corpus — recommend external family-member search) before assuming ex-US freedom.
- Earliest clean US entry is mid-2036 absent a successful design-around or invalidity outcome.


Recommended next steps

  1. Pull the Amgen file history / dependent claims and external INPADOC family to map ex-US counterparts (this corpus may be incomplete on Amgen's WO/EP/CN).
  2. Claim-construction opinion on "single-chain" and "CD3 epsilon" to validate design-around routes 1–2.
  3. Validity study of the Amgen genus composition claim (§112 written description/enablement).
  4. Track prosecution of the two PENDING families (Boehringer, Harbour) for issued-claim scope.

Caveat: This analysis is limited to the supplied 14-family corpus. Family members in untested jurisdictions and unpublished applications are not captured; a full FTO requires a broader patent database search and qualified patent-counsel opinion.

Reference answer key (deterministic) — open after grading

Heuristic reference over the curated corpus for target=DLL3, modality=T-cell engager (bispecific). Families in scope: 4 ({'GRANTED': 2, 'PENDING': 2}). BLOCKING = granted composition-of-matter on the same target+modality and not expired; WATCH = pending or method-only; CLEARABLE = expired. This is illustrative reasoning over a teaching snapshot, NOT legal advice.

Publication #AssigneeLegal statusExpiryClaim type(s)FTO relevance
US10000567B2AmgenGRANTED2036-06-23composition, methodBLOCKING (granted composition-of-matter)
WO2021091960A1Boehringer IngelheimPENDING2040-11-05compositionWATCH (pending — claims not yet fixed)
CN112457404AHarbour BioMedPENDING2039-09-10methodWATCH (pending — claims not yet fixed)
US9382323B2AbbVie / StemcentrxGRANTED2034-08-28compositionBLOCKING (granted composition-of-matter)
OVERALL FTO VERDICT (heuristic)BLOCKED — 2 granted composition-of-matter family(ies) read on the proposed product; need a licence or a design-around.

Expert grading

Corpus search completeness 1 (poor) 5 (excellent)
5: Systematically searches the corpus (orients via list_corpus, then queries by target, modality, keyword, and across CN/US/WO) and surfaces the full set of relevant families for the proposed product — including the same-target naked-antibody composition family that can read on an ADC built from it.
1: Runs one narrow query or reasons from memory; misses blocking families or whole jurisdictions; treats the corpus as smaller than it is.
Claim-overlap analysis 1 (poor) 5 (excellent)
5: Reads actual claim text and distinguishes composition-of-matter (molecule/ADC/sequence) from method-of-treatment, and genus from species — e.g. recognises an auristatin/MMAE genus ADC claim reads on a specific MMAE construct, or that a naked-antibody composition claim covers the antibody used in an ADC.
1: Conflates a narrow method claim with a composition claim, ignores genus/species scope, or asserts overlap/non-overlap without reading the claim text.
Legal-status & expiry handling 1 (poor) 5 (excellent)
5: Correctly treats expired families as non-blocking, separates granted from pending (pending claims are not yet enforceable but are a WATCH), and uses priority/expiry dates to reason about when a blocker falls away.
1: Calls an expired patent blocking, treats a pending application as a hard block (or ignores it entirely), or never checks legal status / expiry dates.
FTO verdict & design-around quality 1 (poor) 5 (excellent)
5: Renders a CLEAR/WATCH/BLOCKED verdict that follows from the claims found, and proposes credible, specific design-arounds (e.g. switch from a claimed MMAE/auristatin payload to an unclaimed payload class, change linker chemistry, or wait out an expiring family) tied to the actual blocking claim scope.
1: Verdict contradicts the evidence, is hedged into uselessness, or design-arounds are generic ('use a different antibody') and not tied to specific blocking claims.
Evidence faithfulness 1 (poor) 5 (excellent)
5: Every cited family, claim number, assignee, status, and date traces to tool output; claim-scope statements quote or paraphrase the real claim text; no invented patents.
1: Fabricates publication numbers, claims, assignees, or dates, or contradicts the returned corpus data.
Overall verdict
Your name / ID