Bispecific Peptides in Cancer: Immune Cell Bridges
Peptide-Based Cancer Immunotherapy
19+ bispecifics approved
More than 19 bispecific immunotherapeutics have received global regulatory approval as of 2025, primarily for hematologic malignancies, with solid tumor indications expanding.
Molecular Cancer, 2025
Molecular Cancer, 2025
View as imageThe immune system's T cells can kill cancer cells, but only if they recognize them. Tumors evade recognition through multiple mechanisms: downregulating the surface molecules T cells detect, creating immunosuppressive microenvironments, and hiding behind "self" signals that inhibit immune attack. Bispecific molecules solve part of this problem by physically bridging T cells to tumor cells, bypassing the normal requirement for antigen recognition. While the bispecific field is dominated by antibody-based formats, peptide-based and peptide-dependent approaches occupy a distinct niche: they can access intracellular tumor antigens presented on MHC molecules, something conventional antibodies cannot do. This article examines how peptide components enable bispecific cancer therapies, the clinical evidence for tebentafusp as the first approved TCR-peptide bispecific, and where the field is heading. For the broader checkpoint inhibitor strategy, see the pillar article on peptide-based immune checkpoint inhibitors.
Key Takeaways
- Tebentafusp, the first approved ImmTAC bispecific, improved 1-year overall survival from 59% to 73% in metastatic uveal melanoma in a Phase III randomized trial (IMCgp100-202, NEJM, 2022).
- The TCR domain of tebentafusp recognizes a gp100 peptide presented on HLA-A*02:01, accessing an intracellular tumor antigen that conventional antibodies cannot reach.
- Bispecific T cell engagers (BiTEs) redirect T cells to tumor surfaces regardless of the T cell's native antigen specificity, eliminating the requirement for tumor-specific T cell priming.
- Motmaen et al. (2025) reviewed strategies for targeting peptide-MHC complexes with designed TCR-based and antibody-based constructs, establishing the structural basis for this approach.[1]
- Neoantigen peptide vaccines represent a complementary approach, priming the patient's own T cells against tumor-specific peptides rather than supplying an external bridging molecule (Zhang et al., 2025).[2]
- Peptide-drug conjugates use tumor-targeting peptides to deliver cytotoxic payloads directly to cancer cells, representing a parallel strategy to bispecific immune cell engagement.[3]
How bispecific molecules work in cancer
A bispecific molecule has two distinct binding domains. In the cancer immunotherapy context, one domain binds a component on the T cell surface (usually CD3, part of the T cell receptor complex) and the other binds a tumor-associated antigen. When both domains are engaged simultaneously, the bispecific physically bridges the T cell to the tumor cell, triggering T cell activation and killing of the tumor cell.
This bypass mechanism is powerful because it does not require the T cell to have been previously sensitized to the tumor antigen. Any T cell in the vicinity, including bystander T cells with irrelevant native specificity, can be redirected to kill the tumor cell. The clinical prototype is blinatumomab, a bispecific T cell engager (BiTE) that bridges CD3 on T cells to CD19 on B cell malignancies, approved by the FDA in 2014 for acute lymphoblastic leukemia.
The limitation of antibody-based bispecifics is that they can only target proteins expressed on the cell surface. Most tumor-specific proteins, including oncogenic drivers and mutated intracellular proteins, are not surface-exposed. They are instead processed by the proteasome and presented as short peptide fragments on MHC class I molecules. This is where peptide-based and TCR-based approaches gain their advantage.
Tebentafusp: the first peptide-MHC bispecific
Tebentafusp (brand name Kimmtrak) represents the most clinically advanced peptide-dependent bispecific. It belongs to the ImmTAC (Immune Mobilizing Monoclonal TCR Against Cancer) platform developed by Immunocore. The molecule consists of a soluble, affinity-enhanced T cell receptor (TCR) domain fused to an anti-CD3 single-chain variable fragment.
The TCR domain recognizes a peptide from gp100 (glycoprotein 100, also known as PMEL), a melanocyte differentiation antigen, presented on HLA-A*02:01 MHC class I molecules. gp100 is an intracellular protein. It reaches the cell surface only as a processed peptide fragment displayed in the MHC groove, invisible to conventional antibodies but recognizable by TCR-based constructs.
In the Phase III IMCgp100-202 trial, tebentafusp improved 1-year overall survival from 59% (investigator's choice of pembrolizumab, ipilimumab, or dacarbazine) to 73% in HLA-A*02:01-positive patients with previously untreated metastatic uveal melanoma. This was the first randomized trial to demonstrate an overall survival benefit for any systemic therapy in metastatic uveal melanoma, a cancer notoriously resistant to checkpoint inhibitors. The FDA approved tebentafusp in January 2022.
The mechanism has a notable feature: tebentafusp produced lower objective response rates than might be expected from the survival benefit. Many patients showed disease stabilization or even initial tumor growth before eventual benefit. This pattern suggests the mechanism involves sustained immune engagement rather than rapid tumor debulking, consistent with the biology of T cell redirection.
The HLA restriction is a practical limitation. Tebentafusp only works in patients expressing HLA-A*02:01, which occurs in approximately 50% of Caucasian populations but at lower frequencies in other ethnic groups. This restriction is inherent to the peptide-MHC targeting strategy: the TCR domain recognizes a specific peptide in the context of a specific MHC allele.
Peptide-MHC targeting: accessing the intracellular proteome
The strategic value of peptide-MHC targeting extends well beyond tebentafusp. Roughly 90% of the human proteome is intracellular and thus invisible to conventional antibody therapies. MHC class I molecules present peptide fragments from these intracellular proteins on the cell surface, creating a molecular barcode that reflects the cell's internal state. Tumor cells present peptides from mutated proteins (neoantigens), overexpressed proteins, and reactivated developmental antigens that distinguish them from normal cells.
Motmaen et al. (2025) reviewed the structural principles for designing TCR-based and antibody-based molecules that target specific peptide-MHC complexes. The challenge is specificity: the peptide-MHC surface is relatively flat compared to the deep binding pockets of typical drug targets, making it difficult to achieve high-affinity binding without cross-reactivity to similar peptide-MHC complexes on normal cells. Tebentafusp's TCR domain was engineered through affinity maturation to achieve picomolar binding to the gp100-HLA-A*02:01 complex while maintaining acceptable selectivity.[1]
This approach opens the door to targeting oncogenic driver mutations that are entirely intracellular: KRAS G12D, p53 hotspot mutations, and fusion proteins that drive specific cancers. Several TCR-based bispecifics targeting these neoantigens are in preclinical and early clinical development. For more on how the peptide-MHC presentation system works, see how peptide-MHC complexes present cancer targets to T cells.
Neoantigen peptide vaccines: priming from within
A complementary strategy to external bispecific bridging is training the patient's own T cells to recognize tumor-specific peptides. Neoantigen peptide vaccines identify mutations unique to a patient's tumor, synthesize peptide fragments containing those mutations, and administer them to stimulate a targeted T cell response.
Zhang et al. (2025) reported results from a Phase II randomized trial of individualized neoantigen peptide vaccines in cancer patients. The approach involves whole-exome sequencing of the patient's tumor, computational prediction of which mutant peptides will bind the patient's specific MHC alleles, synthesis of those peptides, and vaccination combined with adjuvant therapy.[2] Truex et al. (2020) demonstrated that automated flow synthesis could produce clinical-grade neoantigen peptides rapidly enough to support personalized vaccination timelines.[4]
The distinction between neoantigen vaccines and bispecific peptides is mechanistic: vaccines prime the patient's endogenous T cells over days to weeks, while bispecific molecules redirect existing T cells within hours. Vaccines produce a durable memory response; bispecifics require continuous administration. In theory, these approaches could be combined: a neoantigen vaccine primes tumor-specific T cells, and a bispecific molecule enhances their engagement with tumor cells. For deeper coverage of the vaccine approach, see personalized cancer vaccines: how neoantigen peptides target your tumor.
Tumor-targeting peptides as delivery vehicles
A separate application of bispecific peptide design uses one peptide domain to target the tumor and another to deliver a therapeutic payload. Peptide-drug conjugates link a tumor-homing peptide (which binds a receptor overexpressed on tumor cells) to a cytotoxic drug, concentrating the toxicity at the tumor while sparing normal tissue.
Ma et al. (2017) reviewed the peptide-drug conjugate design strategy, describing how short tumor-targeting peptides (RGD motifs, somatostatin analogs, GnRH-derived sequences) can serve as delivery vectors for chemotherapy agents that would otherwise cause intolerable systemic toxicity.[3] Melphalan flufenamide (melflufen, brand name Pepaxto) received accelerated FDA approval in 2021 as a peptide-drug conjugate for heavily pretreated multiple myeloma, though it was subsequently withdrawn from the US market in 2022 after a confirmatory trial showed inferior overall survival compared to standard care.[5]
Jeong et al. (2023) designed a melittin-derived peptide-drug conjugate (M-DM1) that combined a tumor-targeting antimicrobial peptide with the cytotoxic drug DM1. The conjugate inhibited tumor progression and induced effector T cell infiltration in mouse models, demonstrating that peptide-drug conjugates can have immunomodulatory effects beyond direct cytotoxicity.[6]
Challenges and open questions
Several obstacles stand between current bispecific peptide approaches and broad clinical impact.
Cytokine release syndrome (CRS). Bispecific T cell engagers can trigger massive, simultaneous T cell activation, releasing inflammatory cytokines at levels that cause systemic toxicity. CRS is a dose-limiting adverse event for blinatumomab and has been observed with tebentafusp. Managing the therapeutic window between efficacy and CRS severity remains an active challenge.
Solid tumor penetration. Most approved bispecific T cell engagers target hematologic malignancies, where tumor cells circulate in the blood and are easily accessible. Solid tumors present physical barriers: dense stroma, high interstitial pressure, and immunosuppressive microenvironments that inhibit T cell infiltration and function. Peptide-based constructs are smaller than full antibodies and may penetrate tumors more effectively, but this advantage has not been definitively demonstrated in clinical trials.
HLA restriction. TCR-based bispecifics like tebentafusp recognize peptides only in the context of specific MHC alleles. Each new target peptide-MHC combination requires a new TCR engineering campaign. This limits the population eligible for any single product and increases development costs.
T cell exhaustion. Chronic stimulation of redirected T cells in the tumor microenvironment can induce exhaustion, a state of reduced effector function characterized by upregulation of inhibitory receptors PD-1, LAG-3, and TIM-3. Combining bispecific T cell engagers with checkpoint inhibitors may address this, but the combination increases both efficacy and toxicity risks.
Manufacturing complexity. Bispecific constructs are more complex to manufacture than conventional antibodies or peptides. TCR-based molecules require careful quality control to ensure the engineered affinity does not introduce cross-reactivity to normal tissue peptide-MHC complexes, a safety concern that requires extensive screening.
The field is moving toward addressing these limitations through next-generation formats: conditionally active bispecifics that only engage in the tumor microenvironment, half-life-extended constructs that reduce dosing frequency, and combination strategies with checkpoint inhibitors or neoantigen vaccines.
The Bottom Line
Bispecific peptides and peptide-dependent bispecific molecules represent a distinct approach to cancer immunotherapy that accesses the intracellular proteome through peptide-MHC targeting. Tebentafusp demonstrated a survival benefit in metastatic uveal melanoma by redirecting T cells to an intracellular tumor antigen, a first for the TCR-bispecific platform. Neoantigen peptide vaccines take a complementary approach by training the patient's own T cells. Peptide-drug conjugates use tumor-targeting peptides for cytotoxic delivery. Each strategy faces specific limitations: HLA restriction for TCR-based approaches, manufacturing timeline for personalized vaccines, and toxicity for drug conjugates. The common thread is that peptide components enable targeting of cancer biology that antibody-only approaches cannot reach.