From Proof-of-Concept to a Global Drug Development Race

Bispecific antibodies — engineered molecules capable of simultaneously binding two distinct antigens or epitopes — have, as of 2022, produced more than 7 globally approved products and placed over 200 molecules in clinical or preclinical development. That progression from laboratory curiosity to competitive drug class has been rapid, spanning roughly 15 years of sustained platform engineering, and the pace of patent filing activity in 2022–2026 suggests the race is accelerating rather than plateauing.

The technology landscape spans two broad structural paradigms. The first is IgG-like full-length formats that retain Fc-mediated effector functions and favorable pharmacokinetics. The second is fragment-based Fc-free formats — scFv tandems, diabodies, BiTEs, and tandem Fabs — optimized for compact size and immune cell bridging. Each paradigm carries distinct engineering challenges: IgG-like formats must solve heavy-chain heterodimerization and light-chain mispairing, while fragment-based formats require separate half-life extension strategies to achieve therapeutic exposure.

This landscape is derived from 28 patent records and 48 literature sources spanning 2009 to 2026, covering assignees across the United States, Europe, China, South Korea, Japan, Israel, and Australia. It represents a snapshot of innovation signals within this dataset and should not be interpreted as a comprehensive view of the full industry. The dataset was analysed using PatSnap’s innovation intelligence platform, which aggregates global patent and literature data across more than 120 countries.

The timeline of innovation breaks into three distinct periods. The foundational period (2009–2013) established production paradigms: Biogen Idec’s stability-engineered IgG-like bsAbs targeting TRAIL-R2/LTβR (2009), Genentech’s knobs-into-holes strategy with a common light chain (2010), Eli Lilly’s LUZ-Y leucine zipper platform (2012), and Glenmark’s BEAT platform (2013). The expansion period (2014–2019) was catalysed by the approval of blinatumomab in 2014, triggering industrialisation: Roche’s CrossMAb review covering 4 clinical-stage molecules (2016), Pfizer’s common light chain synthetic libraries (2017), and AstraZeneca’s Fab-arm exchange combined with selective protein A purification (2019). The maturation and diversification period (2020–2026) is characterised by immune checkpoint combination targets, high-throughput single-cell screening, and computationally guided engineering — with Chinese biopharmaceutical companies dominating recent patent activity.

Four Engineering Clusters Defining the bispecific antibody Technology Landscape

The retrieved dataset resolves into four distinct engineering clusters, each addressing a specific technical bottleneck in bispecific antibody development. Together they map the full manufacturing and discovery workflow — from initial heterodimerization through to high-throughput candidate identification.

Cluster 1: Heavy Chain Heterodimerization

The most heavily developed cluster addresses ensuring two distinct heavy chains preferentially associate into a heterodimer rather than homodimers. Core mechanisms include CH3 domain steric complementarity (knobs-into-holes), electrostatic charge pairs, and leucine zipper appendages. Eli Lilly’s LUZ-Y platform (2012) appends a leucine zipper to the CH3 C-terminus, driving heterodimerization with only a single point mutation per heavy chain; the zipper is removed post-purification. Genentech’s extended knobs-into-holes strategy (2010) combined this steric approach with a common light chain to enable scalable full-length human bispecific production. F. Hoffmann-La Roche AG’s active European patent (December 2023) covers improved methods for modulating parameters to improve heteromultimeric protein assembly yield and efficiency.

Cluster 2: Light-Chain Pairing Solutions

A dedicated engineering cluster addresses the cognate light chain mispairing problem — the risk that each light chain will associate with the wrong heavy chain in an asymmetric bispecific. Roche’s CrossMAb technology (2016), demonstrated in 4 clinical candidates, uses immunoglobulin domain crossover (CH1/CL swap) to enforce correct pairing. Biogen’s EFab approach (2018) replaces CL/CH1 domains with IgE Cε2 domains and is described as robust across diverse antibody pairs. The University of Stuttgart’s eIg technology (2022) uses heterodimerizing IgE EHD2 domains (hetEHD2) with strategic cysteine mutations to force eFab heterodimerization without artificial disulfide bonds. Pfizer’s synthetic common light chain library (2017) generates bsIgGs with biophysical properties described as indistinguishable from conventional monoclonal antibodies.

Cluster 3: Fragment-Based and Chemical Conjugation Formats

Fragment-based architectures — scFv tandems, BiTEs, and chemo-enzymatic conjugates — avoid chain pairing challenges entirely but require separate half-life extension strategies. Amgen Research (Munich)’s BiTE constructs (2018, Singapore patent) incorporate a specific Fc modality providing extended half-life and favorable pharmacokinetics. UCB Celltech’s Fab-dsFv albumin-binding format (2016) fuses an albumin-binding Fv domain to Fab constant domains with a stabilising inter-domain disulfide, achieving a serum half-life of 7.9 days in cynomolgus monkeys. Martin Luther University Halle-Wittenberg (2022) demonstrated chemo-enzymatic strategies using enhanced Trypsiligase variants (eTl) with click chemistry to covalently fuse Fab fragments into multiple bispecific formats from a single enzyme. University College London (2014) described site-specific bis-dibromomaleimide cross-linkers bridging disulfide bonds between scFv and Fab fragments with maintained binding activity.

Cluster 4: High-Throughput Discovery and Screening Platforms

High-throughput discovery addresses the combinatorial bottleneck in bispecific development: identifying the optimal binding-domain combination from large candidate pools. The University of California, Irvine single-cell-based functional screening pipeline (2021) achieves up to 1.5 million variant library cells per run and demonstrated isolation of rare clones at 0.008% abundance using a CD19×CD3 BiTE model. Merck KGaA’s intein-based combinatorial screening (2020) expands the dual-targeting antibody combinatorial space at industrial scale. According to WIPO, combinatorial biologics screening methodologies have been among the fastest-growing areas of biopharmaceutical patent activity in the 2020–2025 period.

“The UC Irvine single-cell pipeline achieves up to 1.5 million variant library cells per run — and can isolate rare functional clones at just 0.008% abundance.”

Where Bispecific Antibodies Are Being Applied: Clinical Domains

Oncology — specifically T-cell and NK-cell redirection — is the dominant application domain across retrieved results, accounting for the majority of clinical-stage molecules. However, the dataset reveals meaningful activity across five clinical domains, each with distinct target biology and engineering requirements.

Oncology: T-Cell and NK-Cell Redirection

Effector cell-redirecting bispecifics primarily target CD3 on T cells with a tumour-associated antigen on the other arm. Representative targets across the dataset include CD19, CD20, BCMA, EGFR, HER2, EpCAM, PSMA, CEA, and ErbB family members. Bristol-Myers Squibb’s active European patent for a BCMA×CD3 bispecific (2017) demonstrates greater than 70–85% tumour growth inhibition in multiple myeloma xenograft models at 1 mg/kg dosing. Excyte Biopharma’s BCMA×CD3 bsAb (2024, Australia, pending) bridges tumour cells and T cells while reducing ADCC effect for improved safety. As noted in a comprehensive review from Nature-indexed journals, T-cell engager architectures have become the dominant clinical format in haematological malignancy.

Oncology: Immune Checkpoint Modulation

A rapidly growing sub-domain combines PD-1/PD-L1 blockade with modulation of a second immune axis to overcome monotherapy resistance. RemeGen’s VHH-based PD-1/VEGF bsAb (2025, Israel, pending) exploits tumour-region VEGF enrichment for targeted delivery and dual immunosuppression relief. Akeso Biopharma’s CD73/PD-1 dual blockade (2022, Israel, pending) addresses adenosine-mediated immunosuppression in the tumour microenvironment. Betta Pharmaceuticals (2024) extends to TGF-β/GARP/PD-L1 combinations, and Harbour Biomed (2023) targets B7-H4/4-1BB co-stimulatory pathways — all aimed at cold tumours and resistance mechanisms not covered by PD-1 monotherapy. These combinations aim to address resistance mechanisms not covered by PD-1 monotherapy, a challenge well-documented in clinical literature indexed by NIH PubMed.

Anti-Angiogenesis

F. Hoffmann-La Roche AG established an early dominant position in VEGF/ANG-2 co-targeting through multiple filings across IL and SG jurisdictions from 2011 to 2017, representing a sustained multi-year IP strategy. The first-generation CrossMAb-based bispecific blocking both VEGF-A and angiopoietin-2 for tumour angiogenesis was filed in Israel in 2011. Boehringer Ingelheim International filed a VHH-based bispecific targeting VEGF and Dll4 with PEGylation or anti-serum albumin VHH for half-life extension (2013, Israel).

Autoimmune, Hematology, and Infectious Disease

Beyond oncology, the dataset captures three further application areas. In autoimmune disease, Ono Pharmaceutical’s active Singapore patent (January 2026) for a PD-1/CD19 bsAb simultaneously allows PD-1/PD-L1 interaction while suppressing B-cell activity — a mechanistically novel concept for autoimmune conditions. In haematology, Chugai Seiyaku’s European patent (2020) covers a Factor IX/IXa × Factor X bsAb that functionally substitutes for coagulation factor VIII using a common L-chain approach with CDR shuffling. In infectious disease, the University of Texas Medical Branch (2022) demonstrated an IgG-(scFv)₂ bispecific design (14-H-06) that outperforms cocktail and CrossMAb counterparts against SARS-CoV-2 variants including Omicron via higher inter-spike crosslinking.

Geographic and Assignee Landscape: Who Holds the bispecific antibody IP

F. Hoffmann-La Roche AG is the single most prolific patent assignee in this dataset, with 9 patent records spanning VEGF/ANG-2, CCL2, and bispecific assembly methods, filed across IL, SG, and EP jurisdictions between 2011 and 2023. However, the most significant structural shift in the filing landscape is the emergence of Chinese biopharmaceutical companies as the dominant recent cohort.

Chinese assignees — RemeGen Co., Ltd. (IL, 2024–2025), Akeso Biopharma, Inc. (IL, 2022), Betta Pharmaceuticals Co., Ltd. (IL, 2024), Harbour Biomed US, Inc. (IL, 2023), and Excyte Biopharma Ltd (AU, 2024) — collectively represent the most active recent filing group in the 2022–2025 window. Critically, filings in AU, IL, and EP jurisdictions indicate intent for international market protection, not domestic-only filing. This pattern aligns with broader trends documented by WIPO in its annual IP statistics, which have tracked Chinese PCT application volumes surpassing the United States in multiple technology categories.

Jurisdiction analysis reveals that IL (Israel) is the most frequently appearing patent jurisdiction in this dataset, with 18+ filings — likely reflecting PCT/Israel national phase entries as a proxy for broad international filing strategies. EP (European Patent Office) active grants appear for Roche (assembly methods, 2023), Bristol-Myers Squibb (BCMA×CD3), and Ab Studio Inc. (differential-affinity bsAbs, 2025). SG (Singapore) appears for Amgen, Numab, and Ono. AU (Australia) for RemeGen and Excyte. This distribution reflects a global filing strategy among top assignees rather than concentration in a single jurisdiction. The European Patent Office has noted the biopharmaceuticals sector as one of the highest-volume technology areas for PCT national phase entries in recent annual reports.

Five Emerging Directions Shaping the Next Generation of bsAbs

Based on the most recent filings and publications (2022–2026) in this dataset, five distinct directions are identifiable as defining the next generation of bispecific antibody engineering.

1. VHH-Based Architectures for Tumour Microenvironment Targeting

RemeGen’s 2024–2025 IL filings describe a VHH (single-domain antibody from camelid-derived frameworks) targeting VEGF, combined with an anti-PD-1 arm, to achieve tumour-region enrichment through the high-VEGF microenvironment. The highly humanised VHH confers thermal stability and expression efficiency advantages over conventional Fab-based arms.

2. Dual Immune Checkpoint Co-targeting Beyond PD-1/PD-L1

Recent filings extend beyond established PD-1 combinations to include TGF-β/GARP/PD-L1 combinations (Betta Pharmaceuticals, 2024), B7-H4/4-1BB co-stimulatory targeting (Harbour Biomed, 2023), and CD73/PD-1 adenosine pathway blockade (Akeso Biopharma, 2022). These combinations aim to address cold tumours and resistance mechanisms not covered by PD-1 monotherapy.

3. Autoimmune Disease as an Emerging Indication Class

PD-1/CD19 bispecifics for autoimmune disease represent a non-oncology frontier. Ono Pharmaceutical’s active Singapore patent (January 2026) for a PD-1/CD19 bsAb simultaneously allows PD-1/PD-L1 interaction while targeting B-cells — a mechanistically novel therapeutic concept for autoimmune conditions distinct from any oncology application.

4. Computational and AI-Assisted Bispecific Engineering

Sanofi’s work on integrated computational antibody engineering platforms for SARS-CoV-2 neutraliser design (2021) demonstrates that in silico structure-based optimisation is being embedded into the bispecific engineering workflow. This approach accelerates rational CDR optimisation and specificity prediction without exhaustive empirical screening, and is broadly applicable beyond viral neutralisation. Roche Munich’s 2020 publication established that bispecific antibody geometry — beyond domain composition — critically determines T-cell synapse architecture and effector function, driving format engineering toward precise spatial control of antigen-binding arms.

5. Format Geometry as a Functional Design Variable

Roche Munich (2020) established that bispecific antibody geometry — beyond domain composition — critically determines T-cell synapse architecture and effector function. This insight is driving format engineering beyond simple domain fusion toward precise spatial control of antigen-binding arms, representing a shift from empirical to structure-guided design that complements AI-assisted approaches.

Strategic Implications for R&D and IP Teams

The bispecific antibody IP landscape presents five actionable strategic implications for R&D organisations, IP counsel, and business development teams operating in this space.

• Chinese biopharmaceutical IP buildout requires immediate competitive monitoring. Among retrieved results, Chinese assignees account for a disproportionate share of filings from 2022–2025, with patents filed in AU, IL, and EP jurisdictions — indicating intent for international market protection. Western R&D teams should conduct freedom-to-operate analysis on immune checkpoint co-targeting and VHH-based bispecific architectures. PatSnap’s patent analytics platform enables systematic monitoring of assignee filing activity across jurisdictions.
• Light-chain pairing IP remains a crowded but essential space. CrossMAb (Roche), EFab (Biogen), eIg (University of Stuttgart), and common light chain (Pfizer) technologies each hold positions in the pairing solution landscape. The 2022–2024 appearance of novel domain-substitution variants (hetEHD2, 2xVH) suggests design-around space still exists.
• High-throughput single-cell bispecific screening is a differentiated capability. The UC Irvine single-cell pipeline achieving 1.5 million variant library cell throughput per run, Merck KGaA’s intein-based combinatorial screening, and Amgen’s high-throughput cloning and purification protocol collectively define a new standard. R&D organisations lacking equivalent internal throughput will face discovery velocity disadvantages in multi-target combination optimisation.
• Solid tumour access remains the critical unmet technical challenge. While haematological malignancy bsAbs have achieved clinical validation, multiple literature sources from 2021–2023 confirm no bsAb has been approved for solid tumours. Engineering approaches addressing tumour penetration, antigen density thresholds, and cytokine release syndrome mitigation represent high-value white space.
• Infectious disease and non-oncology applications are underexploited relative to their scientific validation. Bispecific broadly neutralising antibody constructs for HIV and SARS-CoV-2, bsAbs for haemophilia A (Chugai’s Factor IX/X approach), autoimmune disease (Ono’s PD-1/CD19), and inflammatory disease (Roche’s CCL2, Numab’s IL23R) collectively represent a diversification opportunity that current clinical pipeline concentration in oncology may be obscuring.

“No bispecific antibody has been approved for solid tumours as of 2021–2023 publications in this dataset — engineering approaches addressing tumour penetration and cytokine release syndrome mitigation represent high-value white space.”