Bispecific Antibodies in Cancer: How They Work and Which Cancers They Treat

Standard monoclonal antibodies are precision instruments — they bind to one target and block it, activate an immune response against it, or deliver a payload to cells bearing it. Bispecific antibodies take this further by simultaneously engaging two distinct targets, enabling biological effects that neither arm alone could achieve. Since 2021, bispecific antibodies have transformed the treatment landscape for multiple myeloma, lymphoma, leukemia, and lung cancer, with a pipeline of hundreds more in clinical trials spanning nearly every solid tumor type.

This article explains the engineering behind bispecific antibodies, the major classes and how they differ mechanistically, the FDA-approved products currently available, the unique toxicities they produce, and the competitive landscape relative to other immuno-oncology approaches.

The Engineering Challenge: Building an Antibody with Two Specifities

A conventional antibody has two identical antigen-binding arms (Fab regions) that both recognize the same epitope. The core engineering challenge in building a bispecific is creating a single molecule where each arm binds a different target — without the two arms impairing each other's function or the molecule forming unwanted aggregates during manufacturing. Over 100 distinct bispecific formats have been described, reflecting decades of engineering innovation to solve this problem.

The most influential early format was the BiTE (Bispecific T-cell Engager), developed by Amgen. BiTEs are small, flexible molecules containing just two single-chain variable fragments (scFvs) — the antigen-binding portions of two different antibodies — connected by a short linker peptide. They lack the Fc region of a conventional antibody, making them very small (approximately 55 kDa versus ~150 kDa for a full antibody) and enabling them to act as molecular bridges between T cells and tumor cells. The trade-off is a very short serum half-life (hours), requiring continuous intravenous infusion — a significant practical burden that later-generation IgG-based bispecifics have largely overcome.

Full-length IgG-format bispecifics retain the Fc region, providing longer half-lives suitable for weekly or less frequent subcutaneous or intravenous dosing. These include the CrossMAb technology (Roche/Genentech), the DuoBody platform (Genmab), and various knobs-into-holes and charge pair engineering approaches that promote correct heavy-chain pairing. The resulting molecules can be produced at scale using standard antibody manufacturing infrastructure — a major advantage over CAR-T therapy's patient-specific manufacturing.

Major Classes of Bispecific Antibodies in Oncology

Class 1: T-Cell Engagers (TCEs)

T-cell engaging bispecifics are the largest and most clinically mature class. They bind simultaneously to a tumor-associated antigen on cancer cells and to CD3ε on T cells, physically bridging the two cell types and forcing T cells into direct contact with tumor cells. This proximity triggers T-cell activation and cytotoxic killing of the tumor cell, regardless of whether the T cell's native T-cell receptor recognizes the tumor. This is mechanistically similar to CAR-T therapy — both harness cytotoxic T cells — but without the need for genetic engineering of patient cells.

The prototypical T-cell engager is blinatumomab (Blincyto), a CD19×CD3 BiTE approved in 2014 for B-cell acute lymphoblastic leukemia — the first FDA-approved bispecific antibody. In recent years, IgG-format T-cell engagers targeting BCMA, CD20, FcRH5, and DLL3 have received approval, with significantly improved half-lives enabling less frequent dosing.

Class 2: NK Cell Engagers

Natural killer (NK) cell-engaging bispecifics bind a tumor antigen on one arm and an NK cell-activating receptor — typically CD16 (FcγRIIIa) or NKG2D — on the other. NK cells do not require antigen presentation via MHC molecules, potentially making them effective in MHC-negative tumors where T-cell based therapies may fail. NK cell engagers are less developed clinically than T-cell engagers, but several are in Phase 1–2 trials.

Class 3: Dual Checkpoint Blockade

Rather than directing immune cells to kill tumors, dual checkpoint bispecifics block two immunosuppressive pathways simultaneously with a single molecule. The approved example is cadonilimab (PD-1×CTLA-4) in China, and in the US, Opdualag (nivolumab + relatlimab) is technically a combination rather than a single bispecific molecule, but illustrates the concept. Multiple PD-1×LAG-3, PD-L1×TIM-3, and PD-1×TIGIT bispecifics are in late-stage trials. The rationale is that simultaneous blockade of two checkpoint pathways may overcome resistance to single-agent checkpoint inhibition.

Class 4: Tumor-Targeting Bispecifics (Non-Immune Effector)

Some bispecifics target two tumor-associated antigens rather than engaging immune cells. The rationale is that dual antigen targeting can improve tumor selectivity (requiring both targets to be co-expressed on the same cell for binding), reduce on-target off-tumor toxicity, and prevent antigen-loss escape. Examples include zenocutuzumab (Bizengri), which targets HER2 and HER3 to block NRG1 signaling, and amivantamab (Rybrevant), which targets EGFR and MET.

FDA-Approved Bispecific Antibodies in Oncology (2026)

How T-Cell Engagers Trigger Immune Killing

The mechanism of T-cell engaging bispecifics is elegant in its directness. When a T-cell engager binds simultaneously to CD3 on a T cell and a tumor antigen on a cancer cell, it creates an immunological synapse — the structured interface at which T cells normally kill their targets. The proximity of the two cells, and the cross-linking of CD3 signaling machinery, triggers T-cell activation, proliferation, and the release of perforin and granzymes — cytotoxic proteins that create pores in the tumor cell membrane and activate apoptosis (programmed cell death).

Critically, this process is MHC-independent. Normal T-cell activation requires the T-cell receptor to recognize a tumor-derived peptide presented in the groove of MHC class I molecules on the cancer cell surface. Many cancers evade immune recognition by downregulating MHC class I expression. T-cell engagers bypass this evasion mechanism entirely — as long as the tumor antigen remains expressed on the surface, the T-cell engager can redirect T cells to kill.

The potency of T-cell engagers is remarkable. A single T cell engaged by a bispecific can kill multiple tumor cells sequentially — a property called serial killing — before moving on to engage another tumor cell. This allows a small number of T cells to eliminate a large tumor burden. In vitro studies show efficacy at picomolar concentrations — some T-cell engagers are active at concentrations as low as 10 picograms per milliliter.

Toxicities: CRS, ICANS, and Infections

Like CAR-T therapy, T-cell engaging bispecifics cause cytokine release syndrome (CRS) through the same mechanism — widespread T-cell activation and cytokine release. However, CRS from bispecific antibodies is generally less severe than from CAR-T therapy and tends to occur primarily with the first few doses, particularly during "step-up" dosing schedules. Step-up dosing — starting with a very low "priming" dose and gradually increasing to the target therapeutic dose over 1–3 cycles — has become standard practice to reduce first-dose CRS severity. Hospitalization for the first one or two doses is required at most centers.

ICANS (immune effector cell-associated neurotoxicity syndrome) also occurs with bispecific antibodies, though generally at lower rates and severity than with CAR-T. It is managed similarly — with corticosteroids as first-line treatment. Infections are a major practical concern with prolonged bispecific therapy, particularly for myeloma patients who are already profoundly immunocompromised. BCMA-targeting bispecifics cause immunoglobulin depletion analogous to B-cell aplasia from CD19-targeting CAR-T, requiring prophylactic immunoglobulin replacement, antiviral prophylaxis (acyclovir), and in many centers, prophylactic antibiotics and antifungals.

Unique to GPRC5D-targeting bispecifics like talquetamab are on-target off-tumor effects related to GPRC5D expression in skin and oral mucosa. Skin toxicities (rash, nail changes) and dysgeusia (altered taste) are class effects of talquetamab seen in the majority of patients, though they are generally manageable with topical corticosteroids and rarely lead to treatment discontinuation.

Bispecifics vs. CAR-T: How Do They Compare?

Both CAR-T therapy and T-cell engaging bispecifics harness cytotoxic T cells to eliminate cancer, but they differ fundamentally in how they achieve this and in their practical profiles.

In clinical practice, bispecific antibodies are increasingly being used earlier in the treatment algorithm — not just as rescue therapy — and are being studied in combination with each other and with other agents to improve depth and duration of response. The field is also investigating whether bispecifics can serve as a bridge to CAR-T for patients who need immediate treatment while awaiting manufacturing, or as maintenance therapy after CAR-T to sustain remissions.

The Pipeline: What's Coming Next

More than 200 bispecific antibody programs are in clinical development as of 2026. In solid tumors, which have proven far more resistant to T-cell engagers than blood cancers, novel approaches are being tested to overcome the immunosuppressive tumor microenvironment. Bispecifics that simultaneously engage T cells and block checkpoint pathways (e.g., EGFR×PD-L1 or EGFRvIII×CD3) are in Phase 1–2 trials for glioblastoma, lung cancer, and colorectal cancer.

In multiple myeloma, sequential and combination bispecific approaches — such as anti-BCMA and anti-GPRC5D bispecifics together — are being investigated based on the hypothesis that dual targeting can prevent escape through antigen downregulation. Early data suggest this combination is feasible with manageable toxicity and potentially deeper responses than either agent alone.