BiTE Pipeline in SCLC & Hematologic Tumors — PatSnap Eureka
Bispecific T-Cell Engager Pipeline in SCLC & Hematologic Tumors
DLL3, CD123, CD20, and PD-L1 are reshaping T-cell redirecting immunotherapy across small cell lung cancer and hematologic malignancies. Explore the patent landscape powering next-generation BiTE and trispecific engager formats.
Key Targets Driving the BiTE Pipeline
Two principal disease contexts — SCLC and hematologic malignancies — are addressed by distinct but mechanistically linked target landscapes spanning DLL3, CD123, CD20, and PD-L1.
DLL3 (Delta-Like Ligand 3)
Approximately 86% of SCLC tumors analyzed showed evidence of DLL3 expression by RNA-seq. Normal cell DLL3 expression is predominantly cytoplasmic and restricted to neurons, pancreatic islet cells, and pituitary cells — creating a favorable therapeutic window. DLL3 is also expressed across other tumors with neuroendocrine features, broadening the addressable patient population. Amgen specifies an IHC threshold of ≥25% DLL3+ cells or ≥2+ staining intensity for treatment eligibility.
Amgen WO + CN filings · IND-enabling criteria establishedCD123 (IL-3 Receptor Alpha Chain)
CD123 is broadly expressed on AML cells — including on leukemia stem cells — with expression increasing over time even in CD123-low cells, enabling myeloablative activity. Multiple Nottingham Trent University filings and a MacroGenics record converge on dual-affinity re-targeting (DART) diabody formats. CD123×CD3 DART formats are described with albumin-binding domain (ABD) or IgG Fc extensions for half-life optimization.
Phase 1 citations for flotetuzumab in patent background textCD20 × CD3 T-Cell Bispecific (TCB)
Roche's anti-CD20/anti-CD3 TCB is explicitly described in combination with an anti-PD1/anti-LAG3 bispecific antibody. Data show reduced T-cell internalization (sink effect), preferential binding to conventional T cells over Tregs, and increased tumor eradication in vivo. The architecture incorporates a P329G LALA Fc mutation to silence Fc-mediated NK effector functions and prevent off-target activation.
Active EP, SG, HU patents · Roche Glycart platformPD-L1 as BiTE Target & Co-stimulatory Inverter
University of Maryland patents describe a CD3×PD-L1 BiTE format that simultaneously activates CD8+ cytotoxic T cells via CD3 crosslinking and neutralizes PD-L1-mediated immune suppression, directing activated CTLs to PD-L1-expressing tumor cells. Separately, Regeneron's PD-L1×CD28 bispecific converts the inhibitory PD-L1:PD-1 axis into a co-stimulatory CD28 signal — a mechanistic inversion relevant to tumor-agnostic co-stimulatory strategies.
Univ. of Maryland CA active patents · Regeneron CN activeFive BiTE Formats Shaping the Pipeline
The retrieved patent dataset identifies five distinct bispecific or multispecific T-cell engager formats. The foundational modality — DLL3×CD3 bispecific antigen-binding molecules — is advanced by Amgen across WO and CN jurisdictions, explicitly positioning these as immune-activating successors to the failed ADC rovalpituzumab tesirine (Rova-T). Unlike ADCs, bispecific T-cell engagers recruit and activate cytotoxic T cells rather than relying solely on conjugated cytotoxins.
In hematologic malignancies, CD123×CD3 DART diabody formats — including flotetuzumab, JNJ-63709178, and APVO436/437 cited in patent background sections — represent the most densely documented modality. MacroGenics describes sequence-optimized CD123×CD3 DART-A with albumin-binding domain (ABD) or IgG Fc extensions to extend half-life, demonstrating T-cell activation and granzyme B/perforin upregulation in vitro at E:T ratios of 10:1, with in vivo antitumor activity at nanogram-per-kilogram dose levels.
The Roche T-cell bispecific (TCB) platform represents the broadest IP architecture in this dataset, with active patents across EP, SG, and HU jurisdictions. One Fab arm binds CD3 while another binds a tumor-associated antigen (FOLR1, CD20, CEA, or others). According to WIPO filing records, multi-jurisdiction active status is confirmed for this platform across European and Asian jurisdictions.
An emerging frontier is the trispecific engager format from Zymeworks BC, which simultaneously achieves T-cell and target antigen engagement while incorporating a PD-1-blocking domain — showing better potency across cell lines with varying PD-L1/mesothelin surface expression levels versus bispecific formats in vitro. IP strategists should monitor this architecture relative to competitive patent landscapes for freedom-to-operate analysis.
Patent Data Signals Across the BiTE Landscape
Key quantitative signals extracted from the PatSnap Eureka patent dataset, covering assignee IP activity, target prevalence, and combination strategy frequency.
Assignee IP Activity by BiTE Modality
Roche holds the broadest multi-jurisdiction portfolio; Nottingham Trent University leads CD123×CD3 records in this dataset.
Combination Strategy Distribution in Retrieved Records
PD-1/PD-L1 axis blockade is the most consistently recurring combination partner across the dataset, appearing in both separate co-administration and single-molecule formats.
DLL3 Expression & SCLC Clinical Context — Key Quantitative Signals
Data points extracted from Amgen WO/CN patent specifications and Legend Biotech CN filing, reflecting the clinical rationale for DLL3-targeted T-cell engager therapy in SCLC.
Six Combination Strategies Identified in the Dataset
Retrieved patent records signal active pursuit of combination strategies that pair T-cell engagers with checkpoint inhibitors, co-stimulatory agonists, and molecular subtyping frameworks.
BiTE + PD-1/PD-L1 Axis Blockade
The most consistently recurring combination across the dataset. F. Hoffmann-La Roche AG records explicitly claim combination therapy of T-cell activating bispecific molecules with PD-1 axis binding antagonists (with optional TIM3 antagonist). T-cell bispecific molecules may upregulate PD-1/LAG-3 on activated T cells, creating a mechanistic combination rationale. The University of Maryland's CD3×PD-L1 BiTE encodes this combination in a single molecule rather than as a co-administration strategy.
Anti-CD20/CD3 TCB + Anti-PD1/LAG3 Bispecific
Roche's 2023 IL record specifically describes this triplet combination for CD20-expressing malignancies, with mechanistic data showing the anti-PD1/LAG3 bispecific's "reduced sink effect" (less internalization by T cells) provides additive benefit over anti-PD1 alone when combined with CD20-TCB. This is the clearest multi-agent combination signal in the B-cell malignancy segment of the dataset.
DLL3-BiTE + SCLC Molecular Subtype Selection
Signals from University of Texas Board of Regents (2023 US filing) and Genentech records suggest that SCLC subtype classification (SCLC-A/-N/-P/-I) may be used to match targeting agents to the appropriate cell-surface antigen profile. MICA and TMEM87A are identified as subtype-specific surface proteins alongside DLL3. Genentech's classification patents describe immune signatures (CD8A, GZMB, PRF1, CXCL9/10) that stratify ES-SCLC patients likely to respond to PD-1 axis blockade.
BiTE + 4-1BB Agonism (Tumor-Localized Co-stimulation)
Regeneron's PSMA×CD3 + anti-4-1BB records describe localized co-stimulatory enhancement. In vivo data cited within CN active records show that 5 mg/kg PSMAxCD3 + 25 mg/kg anti-4-1BB produced significantly superior tumor-free survival versus either agent alone. This co-stimulatory combination strategy is likely generalizable to SCLC and hematologic BiTE programs where co-stimulatory signals are insufficient in the tumor microenvironment.
Key Patent Holders in the BiTE Pipeline
Commercial IP activity spans US, European, Israeli, Chinese, and Singaporean jurisdictions. The innovation landscape is predominantly patent-driven across academic and biopharma assignees.
| Assignee | Primary Target / Modality | Jurisdictions | Development Signal | Key Differentiator |
|---|---|---|---|---|
| Amgen Inc. | DLL3 × CD3 bispecific (SCLC) | WO, CN | IND-Enabling | IHC patient selection criteria (≥25% DLL3+); explicit clinical trial design signals |
| Nottingham Trent University | CD123 × CD3 DART (AML/MDS) | IL, CN, SG | Phase 1 Cited | Three records; flotetuzumab-class Phase 1 citations; companion biomarker strategy for HMA-refractory enrichment |
| F. Hoffmann-La Roche AG / Roche Glycart AG | TCB platform (CD20, FOLR1, CEA) | EP, SG, HU, IL | Active IP | Broadest multi-jurisdiction portfolio; P329G LALA Fc silencing; explicit combination claims (PD-1, LAG-3, TIM-3, ICOS) |
| MacroGenics | CD123 × CD3 monovalent DART | CN | Phase 1 Cited | Earliest CD123×CD3 record in dataset (2016); ABD/Fc half-life extension architecture; nanogram/kg in vivo activity |
| University of Maryland, Baltimore County | CD3 × PD-L1 BiTE fusion | CA (×2) | Preclinical | Academic-origin IP; single-molecule dual function (T-cell activation + checkpoint neutralization); both CA records active |
| Zymeworks BC | Trispecific engager (TAA+CD3+PD-1) | CN | Emerging | 2024 filing; first trispecific signal in dataset; better potency vs. bispecific in PD-L1/MSLN cell lines |
| Regeneron Pharmaceuticals | PSMA×CD3 + 4-1BB; PD-L1×CD28 | IL, JP, CN | Active IP | Co-stimulatory inversion strategy (PD-L1→CD28); 4-1BB combination in vivo data; tumor-agnostic co-stimulatory signals |
Monitor BiTE Assignee Filings in Real Time
Set alerts for continuation filings from Amgen, Roche, and Zymeworks across all BiTE-relevant jurisdictions via PatSnap Analytics.
What This Patent Landscape Means for R&D Teams
DLL3 is the most IP-active SCLC BiTE target in this dataset, with Amgen holding multiple-jurisdiction filings incorporating explicit clinical patient-selection criteria. The abandoned Rova-T program has not diminished DLL3 as a target — retrieved records position bispecific T-cell engagers as the mechanistically superior successor modality by engaging the immune system rather than relying on cytotoxin delivery. According to NIH clinical trial registries, DLL3-targeted programs continue to advance in SCLC.
CD123×CD3 bispecific diabodies face a crowded competitive landscape in AML/MDS, with multiple assignees (Nottingham Trent University, MacroGenics, and cited Phase 1 molecules) converging on the same target pair. Differentiation signals in retrieved data include: half-life extension architecture (ABD vs. Fc fusions), companion biomarker strategies for HMA-refractory patient enrichment, and DART versus scFv-based diabody scaffolds. Teams should review the PatSnap customer case studies for examples of competitive differentiation in crowded oncology target spaces.
SCLC molecular subtyping (SCLC-A/-N/-P/-I) is emerging as the precision oncology framework for matching targeting agents to patient populations, as reflected in the University of Texas Board of Regents filing. Academic and biopharma teams developing DLL3-BiTEs may need to engage with subtype stratification strategies to identify which SCLC subtypes express DLL3 at threshold levels and which express alternative surface antigens (MICA, TMEM87A) that could serve as alternative or complementary BiTE targets.
Trispecific engager formats represent an emerging architecture that could displace conventional BiTE + checkpoint inhibitor co-administration strategies. The Zymeworks BC 2024 record is the earliest signal of this in the dataset; freedom-to-operate analysis relative to TeneBio's multispecific heavy-chain UniAb platform and Roche's TCB scaffold is relevant. The EPO and WIPO filing landscapes for trispecific formats should be monitored for continuation activity. For API-based monitoring, PatSnap Open API enables programmatic access to emerging filing signals.
Bispecific T-Cell Engager Pipeline — Key Questions Answered
Multiple records converge on delta-like ligand 3 (DLL3) as the dominant tumor-associated antigen for T-cell engager strategies in SCLC. Approximately 86% of SCLC tumors analyzed showed evidence of DLL3 expression by RNA-seq, while normal cell DLL3 expression is predominantly cytoplasmic and restricted to neurons, pancreatic islet cells, and pituitary cells — a favorable therapeutic window.
Multiple retrieved patents converge on CD123 as the optimal AML/MDS target because: (i) it is expressed broadly on AML blasts and leukemia-initiating cells; (ii) expression increases over time even in initially CD123-low cells; (iii) CD123 targeting enables myeloablative potential relevant to post-transplant settings.
Unlike ADCs, bispecific T-cell engagers recruit and activate cytotoxic T cells rather than relying solely on conjugated cytotoxins. Amgen records explicitly position the DLL3×CD3 bispecific format as an immune-activating alternative to the failed ADC rovalpituzumab tesirine (Rova-T), arguing that bispecific T-cell engagers mobilize endogenous T cells independent of payload delivery mechanism.
Retrieved results signal several combination strategies: BiTE + PD-1/PD-L1 axis blockade (the most consistently recurring combination); anti-CD20/CD3 TCB + anti-PD1/LAG3 bispecific; DLL3-BiTE + SCLC subtype selection; trispecific engager formats that build checkpoint blockade into the engager scaffold itself; and BiTE + 4-1BB agonism for tumor-localized co-stimulation.
Amgen records specify an IHC-based patient selection threshold of ≥25% DLL3-expressing SCLC cells or ≥2+ staining intensity as criteria for treatment eligibility — an important translational signal reflecting clinical trial design.
Zymeworks BC's 2024 trispecific format (targeting a tumor antigen + CD3 + PD-1 domain) represents an emerging post-BiTE generation that builds checkpoint blockade into the engager scaffold itself, potentially eliminating the need for co-administration of separate checkpoint inhibitors. In vitro data comparing bispecific versus trispecific formats showed the trispecific format demonstrated better potency across cell lines with varying PD-L1/mesothelin surface expression levels.
Still have questions about the BiTE patent landscape? Let PatSnap Eureka answer them for you.
Ask PatSnap Eureka About BiTE Patents →Accelerate Your BiTE & Oncology Pipeline Intelligence
Join 18,000+ innovators already using PatSnap Eureka to map patent landscapes, identify FTO risks, and track emerging modalities in SCLC and hematologic malignancies.
References
- Methods for treating DLL3-expressing cancer — Amgen Inc., 2024, WO [Patent]
- Methods for treating DLL3-expressing cancer (CN) — Amgen Inc., 2025, CN [Patent]
- Anti-DLL3 chimeric antigen receptor and its uses — Legend Biotech (Nanjing Legend Biotech Co., Ltd.), 2023, CN [Patent]
- Bispecific CD123 × CD3 diabodies for the treatment of hematologic malignancies — NanoString Technologies, Inc., 2021, SG [Patent]
- Use of bispecific CD123 × CD3 diabodies for the treatment of hematologic malignancies — Nottingham Trent University, 2023, IL [Patent]
- Bispecific monovalent diabodies capable of binding CD123 and CD3 and their uses — MacroGenics, 2016, CN [Patent]
- Bispecific T cell activating antigen binding molecules — Roche Glycart AG, 2018, HU [Patent]
- Combination therapy of T cell activating bispecific antigen binding molecules and PD-1 axis binding antagonists — F. Hoffmann-La Roche AG, 2022, EP [Patent]
- Combination therapy employing a PD1-LAG3 bispecific antibody and a CD20 T cell bispecific antibody — F. Hoffmann-La Roche AG, 2023, IL [Patent]
- A recombinant bispecific polypeptide for coordinately activating tumor-reactive T-cells and neutralizing immune suppression (2017) — University of Maryland, Baltimore County, 2017, CA [Patent]
- A recombinant bispecific polypeptide for coordinately activating tumor-reactive T-cells and neutralizing immune suppression (2026) — University of Maryland, Baltimore County, 2026, CA [Patent]
- Immunomodulatory trispecific T-cell engager fusion proteins — Zymeworks BC, 2024, CN [Patent]
- Use of a bispecific antigen-binding molecule that binds PSMA and CD3 in combination with 4-1BB costimulation — Regeneron Pharmaceuticals, Inc., 2025, JP [Patent]
- NIH National Institutes of Health — Clinical trial and cancer biology reference resource
- WIPO — World Intellectual Property Organization — International patent filing data and PCT records
- EPO — European Patent Office — European patent register and legal status data
All data and statistics on this page are sourced from the references above and from PatSnap's proprietary innovation intelligence platform. This report is derived from a limited set of patent and literature records retrieved across targeted searches and represents a snapshot of innovation signals within this dataset only.
PatSnap Eureka searches patents and research to answer instantly.