Precision-Engineered CD3 T-Cell Engagers for Solid Tumours: Conditional Activation, Microenvironment Modulation, and Clinical Translation.

1. Introduction

The idea of redirecting T-cells to attack cancer is not new. It dates back to the late 1980s, when researchers first realised that bispecific antibodies could bridge tumour cells and immune cells without the usual MHC restrictions [ 1 ]. The concept is simple: one arm grabs a tumour antigen, the other grabs CD3 on a T cell, and the resulting immunological synapse triggers polyclonal killing regardless of TCR specificity [ 2 ]. For decades, this remained largely theoretical. Then came blinatumomab.

In B-cell acute lymphoblastic leukaemia, blinatumomab changed everything. Complete remission rates exceeding 40% in heavily pretreated patients transformed a research curiosity into standard therapy [ 3 , 4 , 5 , 6 ]. Mechanistically, synapse formation proceeds through sequential TAA anchoring, CD3ε capture, and lytic granule delivery—enabling individual T-cells to serially eliminate multiple targets [ 7 , 8 , 9 ]. This haematology success naturally raised an obvious question: could it work in solid tumours, where 90% of cancer deaths actually occur?

Initially, the answer was no. First-generation TCEs targeting CEA or EpCAM produced minimal responses in colorectal and epithelial cancers—often single-digit objective response rates—despite clear peripheral T-cell activation [ 10 , 11 , 12 , 13 , 14 ]. A 2022 meta-analysis of 17 monotherapy trials captured the disappointment: pooled ORR of just 9%, median PFS under two months [ 15 ]. Something about solid tumours was fundamentally different.

We now recognise three barriers that distinguish solid tumours from their haematologic counterparts. Antigen heterogeneity allows escape variants to emerge constantly, unlike the lineage-restricted CD19 or CD20 targets [ 16 ]. Physical barriers—dense extracellular matrix, abnormal vasculature, high interstitial pressure—prevent antibodies from reaching therapeutic concentrations [ 17 , 18 ]. And the tumour microenvironment itself is hostile territory, packed with myeloid-derived suppressor cells, regulatory T-cells, hypoxia-driven metabolic checkpoints, and microbiome-mediated immunosuppression that collectively exhaust infiltrating T-cells [ 19 , 20 , 21 , 22 , 23 ].

Engineering is now dismantling each barrier systematically. Conditionally activated masked TCEs restrict CD3 engagement to the tumour bed, widening the therapeutic window [ 24 ]. Trispecific formats add second antigens or costimulatory domains to outflank antigen escape and sustain T-cell fitness [ 20 , 25 , 26 ]. TME-directed strategies—stromal modulation, innate immune activation, oncolytic virotherapy, and macrophage repolarization—are increasingly paired with TCEs to unlock infiltration and counter local immunosuppression [ 27 , 28 ].

Clinical translation is accelerating. As of January 2026, over 50 phase I/II trials are running across lung, GI, breast, genitourinary, and CNS malignancies [ 29 ]. Tarlatamab has delivered the first phase III survival benefit for any TCE in solid tumours—40% response rates and 14.3-month median overall survival in extensive-stage small-cell lung cancer. Xaluritamig is approaching 43% responses in heavily pretreated metastatic castration-resistant prostate cancer. Step-up dosing and prophylactic cytokine blockade have tamed early safety concerns, dropping grade ≥ 3 CRS from 16–25% to 4–8% [ 9 ].

Yet critical gaps persist. We still lack an integrative framework connecting molecular design, TME modulation, and clinical outcomes. Evidence-based guidance for patient selection and rational combinations remains nascent.

For clarity, we use a simple generational framework to organise the evolution of T-cell engagers. TCE 1.0 refers to the original BiTE ® format—compact, Fc-free, with a short half-life requiring continuous infusion. Blinatumomab exemplifies this generation [ 3 , 4 , 5 , 6 ]. TCE 2.0 addressed the pharmacokinetic bottleneck by introducing Fc engineering and albumin-binding strategies, which extended half-life and enabled less frequent dosing. However, these modifications largely solved the delivery problem without addressing the core challenges of toxicity and limited solid tumour penetration. TCE 3.0—the focus of this review—represents a more fundamental shift, integrating conditional activation, multispecific targeting, and immunomodulatory payloads to tackle the three major barriers limiting efficacy in solid tumours: antigen heterogeneity, stromal exclusion, and T-cell exhaustion [ 19 , 20 , 21 , 22 , 23 , 24 , 25 , 26 , 27 , 28 , 30 , 31 ]. We emphasise that these generational labels are organisational tools for this review, not formal regulatory classifications.

This review synthesises the 2025–2026 clinical landscape of CD3-based TCEs in solid malignancies, defines dominant resistance mechanisms, evaluates emerging TME-targeted synergies, and proposes a biomarker-guided roadmap toward 2030 [ 32 ]. Our central argument is this: through systematic molecular engineering, biomarker-driven selection, and rational microenvironment modulation, TCEs are transitioning from experimental agents to adaptable platform therapies—and the coming decade will determine how far, how fast, and for how many patients this platform can ultimately reach.

The graphical abstract illustrates the biomarker-stratified deployment of next-generation CD3 T-cell engagers in solid tumours, highlighting advances in conditional activation, tumour microenvironment modulation, and orthogonal payload integration. Collectively, these strategies support improved response rates, reduced toxicity, and a forward-looking roadmap toward integration into standard-of-care treatment paradigms Figure 1 .

2. Methodology

A comprehensive search of PubMed/MEDLINE, Embase, Web of Science, and Cochrane Library was conducted from January 2018 to January 2026 to identify phase I–III trials evaluating CD3-based T-cell engagers in solid malignancies. Conference proceedings from major oncology meetings (ASCO, ESMO, AACR, SITC) and trial registries (ClinicalTrials.gov, EU Clinical Trials Register) were hand-searched. Reference lists of included studies and relevant reviews were screened for additional citations. Search strategies combined controlled vocabulary terms with keywords capturing bispecific antibody constructs, CD3-targeted mechanisms, and solid tumour indications. Only English-language publications were considered. Molecular engineering represents the foundational pillar of TCE platform development, directly addressing the safety, selectivity, and pharmacokinetic limitations that constrained first-generation agents.

3. Molecular Engineering: From BiTE ® to Conditionally Activated Prodrugs

3.1. From BiTE ® to Conditionally Activated Prodrug

The immunological synapse that drives TCE activity forms through a precise two-step process. High-affinity binding to a tumour-associated antigen anchors the TCE to the malignant cell surface, enabling diffusion-limited capture of CD3ε on nearby T-cells and lowering the TCR activation threshold by more than ten-fold [ 7 , 8 ]. This triggers actin cytoskeletal reorganisation, centrosome polarisation, and rapid lytic granule delivery—turning each surface-bound antigen into a surrogate activation signal. Because most CD3-based TCEs are monovalent for CD3 and dissociate after granule release, individual T-cells can serially kill multiple tumour targets. This explains why clinical activity persists even in partially exhausted T-cell compartments [ 9 ].

These constraints—dependence on sustained antigen expression and T-cell fitness—directly motivated the engineering innovations that follow.

The original bispecific T-cell engager (BiTE ® ) scaffold—comprising two single-chain variable fragments (scFvs) linked by a flexible Gly–Ser spacer—omitted the Fc region to avoid inadvertent systemic T-cell activation and enabled scalable expression in E. coli [ 3 , 33 ]. Subsequent engineering efforts focused on extending serum half-life through albumin fusion [ 34 ] or novel bispecific formats such as XmAb [ 35 ], addressing pharmacokinetic limitations. To further confine CD3 engagement to the tumour bed and expand the therapeutic index, conditionally activated or protease-activated prodrug designs were developed. In these constructs, a tumour-associated protease-cleavable peptide masks the CD3-binding domain until proteolytic removal within the tumour microenvironment (TME) [ 19 , 32 ].

Solid tumour-focused masked TCEs have entered clinical development, including JANX007 (anti-PSMA) [ 36 ], JANX008 (anti-EGFR) [ 37 ], and dual-masked PRO-XTEN™ formats such as VIR-5525 for PSMA [ 38 ]. Phase I updates reported between 2024 and 2025 demonstrate prostate-specific antigen (PSA50) declines in nearly all evaluable patients with metastatic castration-resistant prostate cancer treated with JANX007, accompanied by minimal cytokine release syndrome (≤grade 2). These findings correspond to ≥10-fold improvements in preclinical therapeutic index and ≥3-fold reductions in peripheral cytokine release [ 36 ].

3.2. Half-Life Extension Strategies

The short serum half-life of blinatumomab (approximately 1.5 h) necessitates continuous intravenous infusion [ 4 ], creating logistical challenges for patient care. To address this limitation, second-generation constructs have incorporated protein engineering strategies that dramatically extend circulating half-life. For example, xaluritamig (AMG 509) uses an Fc-silenced tail to achieve a half-life of approximately 3–7 days [ 39 ], enabling weekly or bi-weekly dosing schedules that improve patient convenience. This engineering approach has been validated clinically, with recent trials demonstrating significant improvements in solid tumour targeting [ 40 ]. Alternative strategies include fusion to albumin-binding domains [ 34 ] or incorporation of Fc-silencing mutations such as LALA or PGLALA [ 41 , 42 ], which prevent unwanted immune activation while preserving extended pharmacokinetics. Phase I studies have shown sustained CD3 receptor occupancy exceeding 90% at trough concentrations [ 39 ], without increasing the risk of grade ≥ 3 cytokine release syndrome compared to continuous-infusion regimens.

3.3. Trispecific and Multispecific Formats

Trispecific antibody designs represent an elegant solution to antigen escape, one of the key challenges limiting BiTE ® efficacy in solid tumours. By incorporating a third functional arm—whether a second tumour-associated antigen (such as PSMA paired with STEAP1), a conditional costimulatory domain (4-1BB or CD28), or an immune checkpoint-blocking module (PD-1 or LAG-3 scFv)—these next-generation constructs simultaneously engage multiple pathways [ 20 , 26 ]. The impact is substantial: preclinical models demonstrate up to five-fold greater T-cell expansion compared with parental BiTE ® formats [ 20 ], suggesting enhanced therapeutic potential. While still in early clinical development, multiple trispecific platforms are now advancing through phase I trials across various solid tumour indications, offering hope for more durable responses.

3.4. Affinity and Avidity Tuning

One of the most elegant strategies to widen the therapeutic window involves affinity tuning of the CD3-binding domain. Clinical candidates such as alnuctamab [ 43 ] and xaluritamig [ 44 ] demonstrate markedly improved safety profiles by reducing CD3-binding affinity (K_D ~ 50–100 nM) [ 45 ], which minimises off-target T-cell activation in the periphery while preserving robust immunologic synapse formation on tumour cells expressing dense antigen (≥10,000 copies per cell). The field is advancing rapidly—using structure-guided light-chain shuffling and in silico electrostatic steering, researchers have generated “finely tuned” affinity panels that are now entering first-in-human trials (NCT05987204) [ 46 ], offering the promise of potent anti-tumour activity with substantially reduced cytokine-mediated toxicity.

3.5. Payload-Armed TCEs

Next-generation T-cell engagers (TCEs) are now being designed to carry immunomodulatory payloads directly, broadening their functional potential beyond simple tumour-T-cell bridging. A particularly promising approach involves arming TCEs with IL-15 super-agonist complexes fused onto the BiTE ® backbone [ 47 , 48 , 49 ]. The IL-15 super-agonist ALT-803—an IL-15N72D variant linked to IL-15Rα-Fc—demonstrates remarkable potency, showing more than a 25-fold increase in biological activity compared with native IL-15 through improved receptor binding and enhanced in vivo stability [ 47 ]. This translates into meaningful biological effects: preclinical studies show that IL-15-armed TCEs substantially boost CD8 + T-cell proliferation, survival, and cytotoxic activity [ 47 , 48 ]. Particularly encouraging are findings from pancreatic ductal adenocarcinoma models, where IL-15 complexes promote the accumulation of rare TCF1+Slamf6+ stem-like CD8 + T-cells within the tumour microenvironment [ 48 ]—a population associated with sustained anti-tumour responses. While IL-15 has already been successfully incorporated into antibody-fusion platforms [ 49 ], integrating these payloads specifically into CD3-based TCEs for solid tumours represents an exciting new frontier in translational research, offering the potential to overcome T-cell exhaustion in immunologically “cold” tumours [ 47 , 48 , 49 ].

3.6. Manufacturing and Developability

Single-chain TCE formats face a significant manufacturability challenge: they are prone to aggregation at concentrations exceeding 50 mg/mL, complicating large-scale production. To address this limitation, several protein engineering strategies have been developed [ 50 ]. Knob-into-hole heterodimerization combined with charge-pair engineering suppresses light-chain mispairing, while glyco-engineered Pichia pastoris expression systems reduce high-mannose glycoforms and improve manufacturing yield to approximately 2.5 g/L. These advances have meaningful economic implications—cost-modelling analyses predict an approximately 40% reduction in cost of goods for weekly administered Fc-silenced TCEs compared with continuous-infusion blinatumomab, potentially enabling broader patient access to these transformative therapies.

4. Clinical Efficacy of CD3 T-Cell Engagers Across Solid Tumours

These engineering advances have now produced the first phase III-validated survival benefit for any TCE in solid tumours, establishing clinical proof that TCEs can function as platform therapies beyond the haematology setting.

4.1. Search Strategy and Trial Landscape

A systematic literature and trial database search (PubMed/MEDLINE, Embase, Web of Science, Cochrane Library, ClinicalTrials.gov, ASCO/ESMO/AACR; January 2018–January 2026) identified 52 phase I/II studies and three phase III trials evaluating CD3-based T-cell engagers (TCEs) in solid tumours. Only studies reporting objective response rates (ORR) according to RECIST or iRECIST criteria were included, and risk of bias was assessed using Joanna Briggs Institute appraisal tools [ 51 ]. Key phase III trials included the landmark tarlatamab studies in small-cell lung cancer [ 52 , 53 ], while phase I dose-escalation studies evaluated agents such as CDO-23 in non-small-cell lung cancer [ 54 ] and xaluritamig in metastatic castration-resistant prostate cancer [ 44 ]. The systematic search was conducted following PRISMA 2020 guidelines [ 55 ].

Together, these studies define the current clinical landscape of CD3-based T-cell engagers in solid tumours. The principal agents that have progressed to clinical evaluation, along with their target antigens and representative efficacy signals, are summarised in Table 1 .

4.2. Lung Cancers

In extensive-stage small-cell lung cancer (SCLC), the DLL3-targeted T-cell engager tarlatamab has demonstrated remarkable clinical activity. In a pivotal phase II study, tarlatamab achieved an objective response rate (ORR) of 40% and a median progression-free survival (mPFS) of 4.9 months, with grade ≥ 3 cytokine release syndrome limited to just 3% of patients through the use of step-up dosing strategies [ 52 ]. These encouraging results were confirmed and extended in the phase III DeLLphi-304 trial, which demonstrated a significant improvement in overall survival (median 13.6 vs. 8.3 months; hazard ratio 0.60, p < 0.001), leading to full FDA approval in November 2025 for second-line treatment of extensive-stage SCLC [ 53 ]. In parallel, early-phase dose-escalation studies are evaluating the EGFR-targeted bispecific antibody CDO-23 [ 54 ] and the DLL3-targeted trispecific construct HPN328 [ 56 ] in non-small-cell lung cancer (NSCLC), with both agents reporting preliminary signals of anti-tumour activity.

4.3. Genitourinary Cancers

In metastatic castration-resistant prostate cancer, PSMA represents a well-established therapeutic target with broad clinical validation across multiple treatment modalities [ 36 ]. Building on this biological rationale, JANX007—a conditionally masked PSMA × CD3 T-cell engager—has demonstrated encouraging early clinical activity in heavily pretreated patients. PSA50 declines were observed in approximately 73% of treated individuals at target doses of ≥2 mg, with an objective response rate of 30% among patients with measurable disease [ 36 ]. Cytokine release syndrome was predominantly low-grade, with 92% of events limited to Grade 1–2 severity; however, Grade ≥ 3 CRS occurred in approximately 8% of patients despite prophylactic tocilizumab administration, highlighting that conditional masking improves but does not fully eliminate immune toxicity [ 36 ]. In parallel, xaluritamig (STEAP1 × CD3) has shown robust anti-tumour activity, achieving an objective response rate of approximately 41% at optimised doses (≥0.75 mg), supporting STEAP1 as a clinically actionable target in this disease context [ 44 ].

4.4. Gastrointestinal Malignancies

Cibisatamab (RG7802) Monotherapy: The objective response rate (ORR) was 4% in patients with advanced CEA-positive solid tumours [ 11 ].

Cibisatamab + Atezolizumab Combination: The overall ORR for the combination across all evaluable patients was 7% [ 11 ]. Notably, in the specific cohort of patients with microsatellite stable colorectal carcinoma (MSS-CRC) receiving flat doses of both agents, the ORR was 14% [ 11 ].

4.5. Breast and Gynaecological Cancers

The clinical landscape for HER2-directed therapies in breast cancer highlights a critical mechanistic divide. In the HER2-low metastatic setting, the biparatopic antibody zanidatamab—which dual-targets distinct HER2 epitopes to maximise receptor blockade—has shown significant promise. When paired with chemotherapy, it has provided a much-needed therapeutic window for heavily pretreated patients, often extending the chemotherapy-free interval [ 57 ]. In contrast, HER2 × CD3 T-cell engagers are still finding their footing; their early clinical data—including HER2 × CD3 formats in HER2-low disease—currently demonstrate more modest activity than the robust responses seen with biparatopic antibodies and ADC platforms in the same setting [ 58 ].

A similar trend is visible in ovarian cancer, where targeting folate receptor-α (FRα) has become a cornerstone of treatment. However, the most striking results—with response rates reaching approximately one-third of patients—have been driven by next-generation ADCs like rinatabart sesutecan (PRO1184) in FRα-high tumours, rather than T-cell redirection [ 59 ]. This gap suggests that, in specific solid tumour environments, the direct “search and destroy” approach of cytotoxic payloads or receptor inhibition currently remains more effective than the complexities of CD3-based immune recruitment [ 59 ].

4.6. Central Nervous System and Other Solid Tumours

The central nervous system presents a uniquely hostile environment for T-cell engager therapy. The blood–brain barrier restricts systemic antibody penetration, immune privilege limits baseline T-cell trafficking, and local inflammation carries more severe consequences than in peripheral tissues. These constraints make locoregional delivery essential for CNS TCE development. In diffuse intrinsic pontine glioma (DIPG), where B7-H3 is broadly expressed, intratumoral delivery of B7-H3-directed immunotherapy has shown early signals. Partial responses on MRI with manageable safety profiles have been documented [ 60 ], with severe ICANS—the neurological toxicity most feared with brain-directed T-cell therapies—notably absent in initial reports. ICANS in the CNS context differs from systemic ICANS. Cerebral oedema and focal neurological deficits are the primary concerns, rather than the diffuse encephalopathy seen with systemic T-cell therapies [ 61 , 62 , 63 ]. The manageable profile observed likely reflects controlled, localised delivery generating a contained inflammatory response, combined with careful patient selection and intensive monitoring. For TCEs specifically—distinct from CAR-T-cells—the CNS remains largely unexplored. B7-H3-directed bispecific formats are in early evaluation for DIPG and other CNS malignancies [ 60 ], with locoregional delivery lessons directly informing development.

4.7. Cytokine Release Syndrome and Neurotoxicity

Across a pooled analysis of 1214 patients, grade ≥ 3 CRS occurred in 1–16% of cases when step-up dosing was employed, compared with 25–35% in regimens without step-up dosing; CRS in CD3-based bispecific therapy is mechanistically driven by abnormal macrophage activation and is amenable to cytokine-directed intervention [ 61 ]. Omission of step-up dosing shortened the median time to CRS onset from 14 h to 6 h [ 61 ]. Rates of ICANS remained low (≤4%). The use of prophylactic tocilizumab, with or without dexamethasone, further reduced grade ≥ 3 CRS to ≤3% [ 61 ].

4.8. Response Correlates and Emerging Biomarkers

Clinical responses were associated with tumour antigen density of at least 10,000 copies per cell or an immunohistochemistry (IHC) H-score ≥ 150 [ 25 ]. Baseline intratumoural CD3 + T-cell infiltration of ≥250 cells/mm 2 was associated with a twofold increase in ORR [ 25 ]. In addition, early peak levels of soluble IL-2 receptor on cycle 1, day 2 emerged as a predictor of grade ≥ 2 CRS.

Despite this clinical progress, resistance remains the dominant barrier to durable benefit. Each resistance mechanism identified here directly shapes the engineering solutions and combination strategies required to consolidate TCEs as platform therapies.

5. Resistance Mechanisms and Microenvironment Modulation

5.1. Resistance Mechanisms Revisited

For all the clinical momentum building around T-cell engagers, one sobering reality anchors the field: more than 70% of patients treated with CD3-based TCEs will experience disease progression within 12 months [ 64 ]. Response, it turns out, is only half the battle—durability is the other half, and it remains stubbornly elusive.

Three interconnected mechanisms explain why. Tumours lose or disguise their target antigens to evade detection. The microenvironment builds walls—physical, chemical, and immunological—that T-cells simply cannot penetrate. And the T-cells themselves burn out under the relentless pressure of continuous engagement. What makes this particularly challenging is that each mechanism amplifies the others; address one and the remaining two often compensate.

These are not abstract biological curiosities. They are the reason patients who initially respond eventually stop responding, and closing that gap is now the defining challenge of the field.

5.2. TME-Targeted Combination Strategies

Engineering can redesign the drug, but it cannot always reshape the tumour’s environment. That is where tumour microenvironment modulation comes in—dismantling the physical walls, immunological traps, and metabolic dead zones that prevent even the most sophisticated TCEs from reaching their targets.

5.2.1. Boosting Antigens with γ-Secretase Inhibitors

γ-Secretase sits at a critical regulatory node in tumour antigen biology. By cleaving NOTCH and other membrane-associated proteins, this protease complex actively suppresses surface expression of several T-cell engager targets, most notably DLL3 in small-cell lung cancer and BCMA in multiple myeloma [ 65 ]. Pharmacologic inhibition can markedly increase target density: preclinical studies demonstrate approximately 2.3-fold upregulation of DLL3 in SCLC xenograft models when γ-secretase is blocked, effectively expanding the pool of tumour cells vulnerable to redirected T-cell killing [ 66 ].

This pharmacologic manoeuvre holds particular appeal for tarlatamab, where response durability may be limited by antigen downregulation and lineage plasticity. While tarlatamab has established proof-of-concept with objective response rates approaching 40% and a statistically significant overall survival benefit in Phase III testing, the majority of patients ultimately progress—often through mechanisms that reduce effective DLL3 availability. The rationale for combining γ-secretase blockade with DLL3-directed therapy is therefore grounded in tumour biology rather than additive toxicity: by forcing malignant cells to retain higher antigen load, this approach could theoretically extend the window for productive immune synapse formation.

To date, this combination remains investigational. No published clinical trials have evaluated γ-secretase inhibitors in combination with tarlatamab in SCLC, though the mechanistic foundation supports prospective testing. In parallel, the γ-secretase inhibitor nirogacestat has advanced through clinical development for desmoid tumours at a dose of 150 mg orally twice daily [ 62 ], and preclinical exploration of BCMA upregulation prior to anti-BCMA T-cell engager therapy in multiple myeloma provides a translational roadmap [ 66 ]. Whether antigen-boosting strategies can convert partial responses into durable remissions, or merely delay inevitable escape, will require randomised evaluation.

5.2.2. STING Agonists–Innate Immune Ignition

The synthetic cyclic dinucleotide SB-11285, administered intravenously, activates the TBK1–IRF3 signalling axis, leading to induction of type I interferons and the chemokine CXCL10. In preclinical models of immunologically “cold” tumours, SB-11285 has enhanced anti-tumour immune activity, including increased CD8 + T-cell infiltration—though specific combination data with CEA-TCB remain emerging.

Clinically, SB-11285 has been evaluated in a Phase 1/1b study (e.g., NCT04096638 ) as monotherapy or in combination with atezolizumab in patients with advanced solid tumours, demonstrating acceptable tolerability with transient flu-like symptoms manageable via premedication. Early signals support further evaluation, with expansion cohorts ongoing.

5.2.3. Oncolytic Viruses–Stromal Remodelling

IV VSV-IFNβ selectively replicates in tumour cells and upregulates IFN-γ, MMP-2 and MMP-9. In preclinical desmoid-mouse models, pre-dosing with oncolytic VSV-IFNβ increased collagenase activity and intratumoural TCE concentration in fibrotic settings [ 67 ]. Early-phase trials of VSV-IFNβ variants have explored combinations with immunotherapies (including potential TCE synergies), showing safety without dose-limiting toxicity in select cohorts; biopsy assessments in related studies have indicated stromal changes, though specific collagen reduction metrics in TCE-primed contexts remain emerging [ 67 ].

5.2.4. CD40 Agonists–Macrophage Repolarisation

Fc-enhanced CD40 mAb APX005M (0.3 mg/kg IV day −2) shifts M2 → M1 macrophages (↑ iNOS/CD206 ratio in preclinical and early clinical settings) [ 68 ]. Sotigalimab has demonstrated activity in combinations for pancreatic adenocarcinoma and holds orphan designation for soft tissue sarcomas, with macrophage repolarization supporting enhanced immune responses. In sarcoma-related studies (including potential desmoid contexts), combinations with radiation or other agents have shown promising signals, and biomarkers like CD68 + iNOS + macrophages have been explored as predictors [ 69 ]; a parallel evaluation in pancreatic cancer continues [ 70 ].

5.2.5. Anti-VEGF/Angiogenesis Normalisation

Tumours often create chaotic, leaky blood vessels that raise pressure inside and starve the core of oxygen—making it tough for T-cells and T-cell engagers to reach the cancer cells deep inside [ 71 ]. Bevacizumab (15 mg/kg IV every 3 weeks) helps calm this chaos by normalising those vessels, improving blood flow, and opening up more perfused areas for better delivery. In mouse models of colorectal cancer, combining bevacizumab with CEA-targeted T-cell engagers boosted how much of the TCE could penetrate the tumour and led to stronger responses compared with the engager alone in related experiments [ 72 ]. Early human studies testing bevacizumab together with TCEs have shown that the combination is generally well tolerated in the first dose levels, with no serious limiting side effects reported so far, and more patients are being enrolled to gather further data [ 72 ].

5.2.6. TGF-β Blockade–ECM and T-Cell Exclusion

Many solid tumours exploit TGF-β signalling to recruit cancer-associated fibroblasts, which then weave dense extracellular matrix barriers that physically wall off T-cells from the tumour core [ 73 , 74 , 75 ]. Neutralising this pathway with agents such as fresolimumab—administered intravenously at doses up to 15 mg/kg in early-phase cancer trials [ 76 ] and 1–4 mg/kg every two weeks in subsequent studies [ 77 ]—offers a strategy to dampen TGF-β1, 2 and 3 activity and reopen the stromal gate. Preclinical work consistently shows that lifting this suppressive signal curbs fibroblast activation and restores T-cell infiltration [ 73 , 74 ]. In this context, three-dimensional colorectal cancer organoids have emerged as robust models for interrogating T-cell engagers such as CEA-TCB (cibisatamab), revealing how antigen density and microenvironment architecture shape therapeutic success [ 78 , 79 ]. Yet while the biological argument for pairing TGF-β blockade with T-cell engagers is compelling, prospective clinical validation in colorectal cancer remains to be established.

5.2.7. Metabolic Modulators–Lactate and Hypoxia

Tumours frequently exploit extreme hypoxia—regions where oxygen tensions plummet below 5 mmHg—to create havens of treatment resistance [ 80 ]. Hypoxia-activated prodrugs such as PR-104A turn this vulnerability into a therapeutic opportunity, remaining dormant in healthy tissues but unleashing cytotoxic DNA cross-linkers precisely where oxygen is scarce [ 81 ]. While this spatially restricted approach holds theoretical appeal for combining with T-cell engagers [ 82 ], the practical reality is more nuanced: disrupting hypoxic niches can inadvertently summon immunosuppressive granulocytic myeloid-derived suppressor cells [ 16 ], potentially undermining rather than enhancing T-cell infiltration. The optimal strategy for marrying metabolic targeting with T-cell engager therapy remains elusive [ 14 ].

5.2.8. Combination Sequencing and Biomarker-Guided Scheduling

Getting T-cell engagers to work is not about throwing more drugs at the problem—it is about getting the timing right. Gradually ramping up the dose rather than hitting hard from day one has already proven its worth, taking the edge off cytokine storms while keeping the anti-tumour punch intact [ 83 ]. But the real frontier lies in preparing the ground first: theoretically, boosting antigen levels, dismantling the stromal barriers that keep T-cells at bay, or normalising the metabolic wasteland could make tumours far more hospitable for T-cell engagement. Early attempts to widen the therapeutic window have explored pairing T-cell engagers with checkpoint inhibitors. In CEA-expressing solid tumours, the combination of cibisatamab with atezolizumab showed early hints of activity, built on the rationale that relieving PD-L1-mediated exhaustion might help sustain T-cell engager-driven responses [ 66 ]. But these remain just that—early hints, not definitive proof. The biomarker challenge is equally thorny. Unlike checkpoint inhibitors, where PD-L1 expression offers a crude guide, T-cell engagers demand more sophisticated readouts: target antigen density, immune effector function, and physical access to tumour cells. Mariathasan and colleagues revealed how TGF-β signalling in fibroblasts weaves dense extracellular matrix that physically walls T-cells off from the tumour core, blunting immunotherapy efficacy [ 84 ]. Breaking down these fortifications—whether through TGF-β blockade, vascular normalisation, or metabolic modulation—represents a logical next step, but one that remains clinically untested. Turning these scheduling concepts into prospective reality is now underway. The CO40939 study ( NCT03866239 ) is evaluating cibisatamab alongside atezolizumab in MSS colorectal cancer following obinutuzumab pretreatment, building on earlier phase I results [ 66 , 85 ]. Meanwhile, NCT04826003 is probing novel bispecific combinations in solid tumours [ 86 ], and the MajesTEC programme ( NCT05243797 , NCT05572515 , NCT05695508 ) is weaving teclistamab into various lines of myeloma therapy with adaptive dosing strategies [ 83 , 87 , 88 , 89 ]. Whether these sophisticated biomarker-guided approaches can reliably enhance T-cell engager efficacy awaits the verdict of prospective readouts. Collectively, these approaches highlight the importance of modifying the tumour microenvironment to overcome physical and immunological barriers to CD3-based T-cell engager activity. The principal tumour-extrinsic strategies currently under investigation, together with their mechanistic rationale, are summarised in Table 2 .

Biomarker-driven patient selection translates biological understanding into practical clinical tools, enabling precise deployment of TCE platform therapy to populations most likely to benefit.

5.3. Biomarker-Guided Patient Selection

Knowing which patients will respond is as important as having agents worth responding to. Biomarker-driven selection converts biological insight into clinical precision—identifying the right patient, at the right time, for the right platform.

5.3.1. Antigen Density Cut-Offs

Quantitative assessment of tumour antigen expression is increasingly used to enrich for responders to T-cell engagers. Higher DLL3 expression has been associated with improved activity of tarlatamab in exploratory analyses, although tarlatamab received FDA approval without requiring DLL3 testing, as clinical responses were observed across the spectrum of DLL3 expression, including in low or DLL3-negative tumours [ 90 , 91 ]. Similarly, PSMA PET imaging is used for patient selection in PSMA-directed therapies, with established SUV-based thresholds guiding patient eligibility for radioligand therapies, though analogous cut-offs for predicting T-cell engager response remain to be defined [ 92 ]. For CEA-targeted agents such as cibisatamab, RNA sequencing-based biomarker assessment using a CEACAM5 expression threshold of ≥1500 RPKM (reads per kilobase per million mapped reads) has been employed to enrich for clinical activity relative to low-expressing tumours [ 93 ]. Collectively, these antigen-based selection strategies are being refined within ongoing composite biomarker frameworks to support prospective patient selection [ 90 ].

5.3.2. T-Cell Infiltration and Spatial Immunophenotyping

The immune landscape inside a tumour before treatment begins offers valuable clues about who will respond to T-cell engagers. In extensive-stage small-cell lung cancer, not all tumours behave the same way. Those classified as the SCLC-I subtype—where CD8 + T-cells are already engaging with cancer cells and immune checkpoints are active—hold up better against immunotherapy than their neuroendocrine counterparts (SCLC-N, SCLC-P, SCLC-A). Patients with this inflamed phenotype typically see their disease progress after about 5–6 months, whereas those with neuroendocrine subtypes face progression in just 2–3 months [ 93 , 94 ].

The story in uveal melanoma is more nuanced. Here, tebentafusp delivered a striking survival benefit compared to standard care—cutting the risk of death by nearly half and helping 73% of patients reach the one-year mark versus 59% with conventional treatment [ 95 ]. Yet peeking inside the tumours revealed something curious: while patients with high levels of the target antigen gp100 mounted a vigorous immune response—with T-cell infiltration surging two- to three-fold within just over two weeks—this early fireworks display did not ultimately determine who lived longer [ 96 ]. Tebentafusp’s ability to serially kill tumour cells, even when antigen levels are modest, uncouples initial immune activation from long-term survival. These observations point to two practical insights for developing T-cell engagers. First, the degree of T-cell infiltration already presents at baseline—something pathologists can gauge with routine CD3 or CD8 staining—helps identify patients primed for response. Second, tracking how T-cell numbers evolve early during treatment may serve as a real-time measure of whether the drug is engaging its target, regardless of whether that immediate immune burst translates into lasting clinical benefit. The ongoing ATOM trial and other prospective studies are now putting these principles to the test across solid tumour types.

5.3.3. Soluble Immune Profiles and Dynamic Monitoring

The bloodstream offers a real-time window into how T-cell engagers are performing. When these drugs switch the immune system into high gear, molecular signals spill into circulation well before symptoms appear. Soluble interleukin-2 receptor-α—essentially a footprint left by activated T-cells—and interleukin-6, the chemical messenger that drives cytokine release syndrome, follow different rhythms in the days after treatment starts [ 71 ]. In children and adults with B-cell acute lymphoblastic leukaemia receiving blinatumomab, doctors noticed that rising sIL-2Rα levels flagged those at risk for severe immune reactions. The sickest patients showed dramatic spikes in this marker, distinguishing them from those with milder responses [ 71 ]. The story with teclistamab in multiple myeloma proved more encouraging. By carefully stepping up the dose rather than hitting patients hard from day one, investigators kept severe cytokine release syndrome below 1% while still achieving responses in nearly two-thirds of patients—a remarkable balance of safety and efficacy [ 32 ]. Tracking these markers over time—watching how they rise and fall—lets clinicians separate patients who need aggressive monitoring from those who can be managed more conservatively. Consensus guidelines now incorporate these dynamic signals into standardised toxicity grading, moving the field toward predictive rather than reactive management [ 65 ]. Taken together, these findings highlight the need to move beyond single biomarkers toward integrated, biology-informed models for patient stratification and treatment optimisation. The key tumour- and immune-derived parameters that underpin such multiparametric approaches, together with their proposed clinical utility, are summarised in Table 3 .

5.4. Resistance Mechanisms and Adaptive Strategies

Resistance to T-cell engagers rarely announces itself dramatically—it accumulates quietly through small biological shifts that individually seem manageable but collectively tip the balance from response to progression. Three dominant mechanisms drive this process, and their interdependence is what makes them so challenging: address one and the others invariably compensate. What follows is not a catalogue of failure modes but a precise biological map—and increasingly, a design brief for the next generation of TCE platforms.

5.4.1. Primary Resistance: When Tumours Fail to Engage

The first and most fundamental escape mechanism is invisibility—shedding, downregulating, or simply never expressing enough target antigen to sustain productive immune engagement. This is not a rare event. It is the dominant mode of clinical failure across every solid tumour TCE indication evaluated. The most formidable barrier is simply missing the target. In metastatic uveal melanoma, tebentafusp produces objective shrinkage in approximately 8% of cases, with disease stabilisation in an additional 43% [ 97 , 98 ]. Long-term follow-up reveals three-year overall survival of 26% versus approximately 18% historically [ 98 ], establishing proof-of-concept for redirected T-cell killing in this otherwise treatment-refractory malignancy [ 90 ]. In small-cell lung cancer, the challenge differs. DLL3 is present on 85–94% of tumours, yet tarlatamab delivers median progression-free survival of just 4.9 months, with most patients progressing by six months [ 9 ]. The disconnect between antigen presence and therapeutic response points to microenvironmental barriers—stromal exclusion, insufficient T-cell infiltrate, or compensatory signalling—that blunt T-cell engager activity even when targets are theoretically available [ 51 ]. Among patients whose tumours have undergone lineage plasticity—such as EGFR-mutant adenocarcinomas transformed to small-cell histology—outcomes remain dismal. Historical data with platinum-etoposide chemotherapy shows median PFS of 3–4 months [ 99 ], and emerging real-world experience with tarlatamab suggests even more limited benefit, with median time on treatment of 1.5 months [ 100 ].

5.4.2. Acquired Resistance: Evolution Under Pressure

For patients who initially respond, the second failure mode emerges: tumour adaptation. Preclinical models of HER2-directed T-cell engagers reveal acquired resistance through JAK2 down-modulation, crippling tumour-intrinsic interferon-γ signalling without altering HER2 expression itself [ 101 ]. This mechanism mirrors JAK1/2 loss-of-function mutations observed in resistance to checkpoint inhibitors, suggesting convergent evolutionary pathways across immunotherapy modalities [ 102 ]. In multiple myeloma, where BCMA-directed T-cell engagers have transformed relapse management, biallelic antigen loss now accounts for approximately 30% of treatment failures. Pre-existing heterozygous deletions—present in 13–15% of patients for GPRC5D and 3–8% for BCMA—provide the substrate for second-hit evolution under therapeutic pressure. Soluble antigen shedding compounds the problem, with circulating BCMA rising 2.5–4-fold at progression, acting as a pharmacological sink that sequesters therapeutic binding. The same pharmacological trap affects CEA-targeted agents, with an additional twist. Soluble CEA in bloodstream acts as a decoy—intercepting cibisatamab before tumour engagement and reducing drug available for cytotoxicity [ 78 ]. Combined with patchy antigen expression, this sink effect creates a dual barrier that dose escalation cannot overcome. Active strategies include γ-secretase-mediated antigen upregulation, decoy-resistant construct engineering, and dual-antigen co-targeting to reduce single-target dependence [ 82 , 83 ].

5.4.3. The Microenvironment Strikes Back

The tumour-intrinsic alterations, the ecosystem turns hostile. PD-L1 upregulation—observed in 40–60% of patients following T-cell engager exposure—provides adaptive dampening of effector function [ 103 ]. Myeloid-derived suppressor cells and M2-polarised macrophages infiltrate tumours post-treatment, establishing physical and chemical barriers to T-cell access [ 103 ]. Exhaustion markers (PD-1, TIM-3, LAG-3) accumulate on engaged T-cells by the second to third week of therapy, coinciding with pharmacodynamic decline and presaging clinical progression [ 76 ].

T-cell exhaustion is particularly insidious because it unfolds in a biphasic manner. Early phases are marked by vigorous T-cell infiltration and intense cytokine release—clinically observed as cytokine release syndrome (CRS). Although this surge is often interpreted as evidence of durable therapeutic activity, it frequently represents the onset of progressive functional decline rather than sustained immune control [ 33 , 65 ]. Continuous CD3 engagement drives transcriptional reprogramming characterised by induction of TOX and NR4A family factors, accumulation of inhibitory receptors, and a gradual shift toward terminally exhausted T-cell states [ 90 , 91 , 92 ].

A central determinant of long-term functionality is the survival of TCF1 + progenitor-exhausted T-cells. Once this stem-like compartment is depleted, restoration of effective anti-tumour immunity becomes difficult [ 82 , 93 ]. The choice of costimulatory domain therefore has profound consequences. CD28 promotes rapid activation but can accelerate terminal differentiation, whereas 4-1BB supports metabolic persistence and memory formation. However, excessive tonic signalling through 4-1BB may itself promote exhaustion via NF-κB-mediated transcriptional remodelling [ 91 ].

This delicate balance creates a narrow therapeutic window that requires careful engineering. Preserving the TCF1 + progenitor pool has therefore become a key design objective. Strategies such as IL-15 super-agonist fusions that expand stem-like T-cell subsets [ 94 ], treatment-free intervals that allow functional recovery [ 84 ], and incorporation of checkpoint blockade within trispecific architectures represent emerging structural solutions to sustain T-cell functionality during therapy [ 82 , 93 ].

5.4.4. Rational Counter-Strategies

The precision with which resistance mechanisms have been characterised is, paradoxically, a source of optimism. Each failure mode points directly to a rational countermeasure—and the field is moving rapidly from identifying patterns to building against them. These resistance patterns suggest specific antidotes. Gamma-secretase inhibitors upregulate DLL3 expression 2–3-fold in preclinical models, providing mechanistic rationale for combination or sequencing with tarlatamab [ 104 ]. Checkpoint co-blockade—targeting LAG-3 or TIM-3 alongside PD-1—aims to reverse exhaustion; early-phase trials pairing tebentafusp with LAG-3 inhibitors are now exploring this mechanistic hypothesis [ 105 ]. For antigen-loss escape in myeloma, sequential targeting—switching between BCMA and GPRC5D directed T-cell engagers upon biallelic deletion—has demonstrated clinical feasibility, extending the window of T-cell redirection [ 106 , 107 ]. What emerges from this analysis is not a picture of intractable biological complexity but a set of clearly defined engineering and clinical targets. The principal resistance mechanisms shaping TCE response and durability, together with their direct implications for next-generation platform design, are summarised in Table 4 .

6. Discussion

The evidence reviewed here demonstrates a coherent trajectory: TCEs have transitioned from empirical biologics to rationally designed platform therapies with validated clinical impact in solid tumours.

6.1. From Proof-of-Concept to Platform Therapy in Solid Tumours

The clinical trajectory of CD3-based T-cell engagers (TCEs) in solid tumours reflects a decisive shift from early empirical exploration to rational, mechanism-guided platform development. Initial programmes targeting carcinoembryonic antigen (CEA) and epithelial cell adhesion molecule (EpCAM) yielded objective response rates below 10% in colorectal and epithelial malignancies, even with checkpoint inhibitor combination [ 10 , 11 , 12 , 13 , 14 ]. These modest outcomes, in stark contrast to the transformative efficacy of blinatumomab in haematological malignancies [ 3 , 4 , 5 , 6 ], underscored the biological constraints of solid tumour architecture: antigen heterogeneity, stromal barriers, and immunosuppressive microenvironments. Contemporary third-generation TCEs address these constraints through systematic redesign. Conditional activation mechanisms, half-life extension, structured step-up dosing, and biomarker-enriched selection have collectively enabled objective response rates of 35–41% in defined solid tumour contexts [ 9 , 28 , 32 ]. The phase III DeLLphi-304 trial establishes definitive clinical validation: tarlatamab demonstrated superior overall survival versus chemotherapy in second-line DLL3-positive small-cell lung cancer (SCLC; median 13.6 vs. 8.3 months; hazard ratio 0.60; 95% CI, 0.47–0.77; p < 0.001), achieving full FDA approval in November 2025 [ 42 , 43 , 100 ]. This represents the first phase III-validated survival benefit for any TCE in solid tumours and positions DLL3 as a therapeutically actionable lineage antigen despite historical immunotherapy resistance. In metastatic castration-resistant prostate cancer (mCRPC), the STEAP1-targeted TCE xaluritamig demonstrated dose-dependent efficacy with objective response rates exceeding 40% at optimised doses, supported by substantial PSA declines [ 44 ]. The conditionally masked PSMA-targeting construct JANX007 exemplifies tumour-restricted activation, achieving high biochemical response rates with favourable safety profiles [ 36 ]. Together with tebentafusp—the first TCE approved for any solid tumour (metastatic uveal melanoma, January 2022), which demonstrated 49% reduction in mortality risk despite modest radiographic response [ 85 , 90 ]—these data establish that TCE clinical value extends beyond conventional cytoreduction. Multidimensional biomarker strategies now guide patient selection. Integration of tumour antigen density (≥10,000 copies per cell or immunohistochemistry H-score ≥ 150), baseline intratumoural CD3 + T-cell infiltration (≥250 cells/mm 2 ), and dynamic soluble immune kinetics (soluble IL-2 receptor-α) outperforms single-parameter prediction, with composite models achieving area under the curve values of approximately 0.83 [ 71 , 72 , 73 , 94 , 95 ]. Safety engineering has progressed commensurately: step-up dosing protocols have reduced grade ≥ 3 cytokine release syndrome from 16–25%, historically, to 1–4%, enabling outpatient administration [ 7 , 32 , 70 ]. T-cell engagers have reached an inflection point in solid tumour oncology. What began as cautious early-phase exploration has rapidly matured—tarlatamab’s FDA approval in 2025 marked the first definitive proof that these agents can extend survival in aggressive epithelial malignancies. Table 5 depicts the momentum, tracking the pipeline’s evolution from standalone TCEs toward smarter combinations that pair checkpoint inhibition, conditional activation, and targeted remodelling of the tumour microenvironment.

6.2. Resistance as the Central Determinant of Durability

The resistance mechanisms detailed in Section 5.4 are not simply biological footnotes to an otherwise encouraging clinical story. They are the central determinant of whether T-cell engagers become genuinely durable platform therapies or remain agents of impressive but transient responses—and the field knows it. What the past two years have taught us about resistance is ultimately more instructive than the response rates themselves. Antigen loss tells us we need multispecific constructs and antigen-upregulation strategies built in from the start rather than bolted on after patients stop responding. Microenvironmental exclusion tells us that vascular normalisation, stromal remodelling, and conditional activation are not optional combination partners but biological necessities for reaching the patients most in need. T-cell exhaustion tells us that costimulatory engineering, IL-15 fusions, and treatment scheduling that actively protects the TCF1 + progenitor pool is as important to platform durability as the initial response rate. The most important conceptual shift here is this: resistance biology is no longer functioning as an obstacle to TCE development. It is functioning as a design specification. Each mechanism that limits durability has become a precise engineering target, and the next generation of TCE platforms is being built directly against those targets—not around them. Understanding resistance is only clinically useful if it translates into better decisions at the bedside—and that requires equally precise tools for identifying which patients will benefit. Multiparametric biomarker models integrating antigen density, intratumoural CD3 + T-cell infiltration, myeloid burden, and soluble IL-2 receptor kinetics now outperform any single metric in predicting who will respond and who will not, achieving composite AUC values of approximately 0.83 versus 0.64 for individual markers [ 70 ]. Prospective validation of these models across multiple phase II studies is currently underway [ 95 ]—and their maturation will be as consequential to TCE platform therapy as the drugs themselves. Knowing what drives resistance is only half the answer—the other half is doing something about it. Section 6.3 examines how that knowledge is now being translated into rational combinations and next-generation engineering strategies that are actively reshaping the TCE development landscape.

6.3. Rational Combinations and Next-Generation Engineering

Recognition of these resistance axes has catalysed development of tumour microenvironment priming strategies and multifunctional TCE architectures. Microenvironment modulation demonstrates encouraging early signals. In the myeloma setting, γ-secretase inhibition has demonstrated substantial target antigen upregulation—with BCMA surface density increasing by a median of 33-fold (range 8.7–157) in patients receiving CAR-T therapy [ 65 ]—establishing proof-of-concept for pharmacological antigen enhancement. Whether analogous upregulation of DLL3 can be achieved in SCLC through γ-secretase inhibition without compromising safety remains an open and clinically important question that warrants dedicated preclinical and early-phase clinical investigation STING agonists activate TBK1–IRF3 signalling, inducing type I interferons and CXCL10 to enhance CD8 + T-cell infiltration. CD40 agonists (sotigalimab) repolarise suppressive macrophages, increasing iNOS/CD206 ratio toward pro-inflammatory M1 phenotypes. TGF-β blockade (fresolimumab) dismantles stromal barriers erected by cancer-associated fibroblasts [ 73 ], while anti-angiogenic strategies normalise vasculature, improving TCE penetration [ 88 , 90 ]. Orthogonal payload integration expands functional scope. Trispecific constructs incorporating IL-15 super-agonists, checkpoint blockade fragments (anti-PD-1, anti-LAG-3), or 4-1BB costimulatory domains expand intratumoural TCF1 + CD8 + populations and attenuate exhaustion markers in preclinical models [ 20 , 26 , 47 ]. IL-15-armed TCEs approximately double stem-like T-cell density; first-in-human studies are planned for 2026 [ 50 , 51 ]. Hypoxia-activated and protease-cleavable prodrug designs improve tumour selectivity, with HIF-1α degron constructs achieving 12-fold tumour-to-blood selectivity in glioma models [ 80 ].

6.4. TCEs as Combinatorial Partners with CAR-T and CAR-NK Cell Therapies

Despite the transformative impact of chimeric antigen receptor (CAR)-T-cell therapy in haematological malignancies, its translation to solid tumours has been limited by physical stromal barriers, antigen heterogeneity, and the rapid exhaustion of infused cells within hostile microenvironments. CD3-based T-cell engagers, by contrast, mobilise the endogenous polyclonal T-cell repertoire rather than relying on a numerically constrained infused product—a mechanistic distinction that makes these two platforms complementary rather than competing [ 13 , 64 ]. Their convergence represents one of the most strategically important directions in solid tumour immunotherapy today.

6.4.1. TCE–CAR-T Combinations: Three Principles

The rationale for pairing TCEs with CAR-T-cells rests on three interdependent principles. First, a short course of TCE therapy can serve as an immune primer—expanding tumour-infiltrating T-cells, upregulating inflammatory chemokines, and reducing suppressive myeloid burden—before CAR-T infusion, converting cold tumours into landscapes more hospitable to adoptive cellular products. Second, because bispecific TCEs engage a different tumour antigen or activate bystander endogenous T-cells independently of the CAR construct, they can re-engage antigen-loss escape variants that have evaded single-antigen CAR-T killing—one of the dominant failure modes of that platform [ 82 ]. Third, soluble circulating TCEs reach anatomically inaccessible tumour niches where infused CAR-T cells, limited by trafficking and persistence, may never arrive.

Preclinical evidence in multiple myeloma provides direct proof-of-concept. Sequential exposure to BCMA-directed TCEs followed by BCMA-targeted CAR-T cells sustained anti-tumour activity even as biallelic antigen loss began to emerge under TCE selection pressure [ 65 ]. The same biology underlies the clinical rationale for γ-secretase inhibitor priming: by upregulating BCMA surface density by a median of 33-fold, this pharmacological manoeuvre simultaneously sensitises tumour cells to both TCE-mediated and CAR-T-mediated killing, and has already entered clinical evaluation in the myeloma setting [ 93 ]—a paradigm with direct translational relevance to solid tumour antigens such as DLL3 and STEAP1.

6.4.2. CAR-NK Cells: A Mechanistically Orthogonal Partner

CAR-NK cell platforms introduce a different and practically compelling dimension. Allogeneic CAR-NK cells—derived from peripheral blood, umbilical cord blood, or induced pluripotent stem cells—can be manufactured off-the-shelf at scale, bypassing the 3–6-week autologous manufacturing timelines and patient-specific variability that constrain CAR-T programmes [ 64 ]. Their most important mechanistic property in this context is what they lack: CD3ε. Because TCEs activate T-cells exclusively through CD3ε engagement, a co-administered TCE will selectively recruit and redirect endogenous T-cells without triggering the infused NK product. This orthogonality permits genuine co-administration—TCE-driven T-cell cytotoxicity and NK-cell-mediated killing operating simultaneously across antigen-heterogeneous tumour populations—without the amplified cytokine release that complicates TCE plus CAR-T schedules. While dedicated CAR-NK plus TCE combination trials are only now entering early clinical development, the mechanistic logic is already sufficiently established to inform trial design [ 64 ].

6.4.3. Scheduling Architectures

Three deployment models are emerging from preclinical and early clinical experience, each suited to a distinct biological scenario. Their key features, clinical contexts, and practical considerations are summarised in Table 6 .

Selecting the right model is, at its core, a biomarker question—tumour antigen co-expression patterns, baseline TME immunophenotype, and real-time pharmacodynamic monitoring will together determine which approach is most appropriate for a given patient, linking these combination strategies directly to the biomarker frameworks discussed in Section 5.3 . What matters most, however, is the broader conceptual shift: TCEs are no longer best understood as standalone agents in competition with cellular therapies. They are, increasingly, tools that work alongside CAR-T and CAR-NK platforms—priming, redirecting, and rescuing them as the clinical situation evolves [ 64 , 70 , 82 ].

6.5. Advantages and Disadvantages of CD3-Based TCEs: A Balanced Appraisal

T-cell engagers have reached a point where honest appraisal is more useful than advocacy. Response rates are climbing, regulatory approvals are accumulating, and the engineering pipeline is genuinely impressive—but the limitations are real, and understanding them precisely is what separates a platform with lasting clinical impact from one that peaks early and plateaus.

6.5.1. What TCEs Do Well

The most fundamental advantage of CD3-based TCEs is mechanistic. By forcing direct CD3ε engagement independently of MHC class I presentation, they bypass one of the most pervasive immune evasion strategies in solid tumours—the downregulation of antigen presentation machinery that renders cancer cells invisible to conventional cytotoxic T-cell recognition [ 7 , 8 ]. This matters because HLA loss is not a rare event; it is a dominant resistance mechanism in lung, prostate, and colorectal cancers, and it is one that checkpoint inhibitors, which depend on endogenous T-cell recognition, cannot overcome. TCEs can. The resulting immunological synapse triggers serial killing—individual T-cells eliminating multiple tumour targets in sequence—which means that cytotoxic activity is sustained even when effector-to-target ratios are unfavourable, as they routinely are in solid tumour settings [ 7 , 8 ].

Practically, TCEs are far more accessible than cellular therapies. They require no lymphodepletion chemotherapy, no specialised transplant infrastructure, and no patient-specific manufacturing. Step-up dosing has reduced grade ≥ 3 cytokine release syndrome to below 1% in optimised regimens, enabling outpatient administration in community oncology settings that would never be considered for CAR-T programmes [ 72 ]. Critically, TCE-associated toxicities are generally reversible—treatment can be interrupted when adverse events emerge, held during intercurrent illness, and safely resumed after recovery. This flexibility is simply not available with cellular products, where the therapeutic effect, once initiated, cannot be paused [ 8 , 24 ]. For elderly patients, those with organ dysfunction from prior treatment, or those in settings without cellular therapy infrastructure, this reversibility is not a minor convenience—it is what makes treatment possible at all.

TCEs also occupy a uniquely productive mechanistic space alongside checkpoint inhibitors. Because both platforms act through the endogenous T-cell pool, they are genuinely synergistic rather than merely additive: checkpoint inhibition sustains the T-cell function that TCEs recruit, while TCE-driven immune activation upregulates PD-L1 in ways that make checkpoint blockade more effective. This bidirectional amplification has no equivalent in CAR-T combinations, where the infused cellular product operates largely independently of the endogenous immune landscape. Manufacturing innovation is progressively reducing cost. Knob-into-hole heterodimerisation and Pichia pastoris expression systems have driven yields to approximately 2.5 g/L, with cost modelling projecting a 40% reduction in cost of goods for weekly administered Fc-silenced TCEs compared with continuous-infusion formats [ 53 ]—positioning TCEs as genuinely scalable therapies rather than interventions available only to patients in well-resourced centres.

6.5.2. Where TCEs Fall Short

The most sobering limitation is durability. Despite objective response rates approaching 40% in selected populations, median progression-free survival remains modest—4.9 months for tarlatamab, with the majority of patients progressing within six to twelve months [ 9 ]. CAR-T cells can establish long-lived memory populations capable of occasional cure; TCEs cannot. The polyclonal activation that makes them broadly applicable simultaneously prevents the clonal expansion and immunological persistence that underpin durable remission [ 64 ]. Continuous or cyclical administration is therefore required to maintain target engagement, which accumulates toxicity, cost, and patient burden over time in ways that are only beginning to be characterised in longer follow-up studies.

Sustained TCE exposure also creates relentless selection pressure for antigen escape. In multiple myeloma, biallelic BCMA loss now accounts for approximately 30% of treatment failures, and soluble antigen shedding compounds the problem further—circulating BCMA acts as a pharmacological sink, sequestering therapeutic binding before it reaches tumour cells [ 65 , 93 ]. The very feature that makes TCEs effective—continuous immune pressure—thus undermines their own longevity in a proportion of patients that is not small.

Microenvironmental dependence is a ceiling that engineering has not yet raised. Tumours with minimal baseline CD3 + infiltration, dense desmoplastic stroma, or deep hypoxic niches remain largely refractory, because TCEs cannot manufacture T-cell activity from tissue that contains none [ 72 ]. CAR-T cells can be equipped with chemokine receptors—CCR2b, CXCR3—to actively traffic toward tumour deposits regardless of the local immune landscape; TCEs have no equivalent mechanism and remain dependent on conditions they cannot themselves create [ 64 ]. Finally, while ICANS rates are lower than with CAR-T therapy, the long-term neurological safety profile of continuous TCE administration—particularly for CNS-penetrant formats—is incompletely characterised, because early protocol-mandated discontinuation upon ICANS limits the follow-up data available [ 64 ]. Regulatory approval of novel multispecific constructs will require this gap to be prospectively addressed.

6.5.3. Where TCEs Fit

These strengths and limitations, taken together, define a clear clinical niche. TCEs are optimally deployed where rapid disease control is needed and manufacturing time is unavailable; where lymphodepletion is contraindicated; where cellular therapy infrastructure does not exist; and where the combination of checkpoint inhibition with a second active mechanism offers an incremental survival benefit over either alone. CAR-T cells retain their advantage for younger, fit patients seeking potential cure, for tumours with homogeneous antigen expression, and for settings where the upfront toxicity and cost are justified by the prospect of deep durable remission.

The more important point, however, is that this is no longer a binary choice. The combinatorial architectures described in Section 6.4 —priming, parallel delivery, sequential antigen-switching—reflect a growing clinical consensus that TCEs and cellular therapies are most powerful when deployed together, in the right order, guided by the right biomarkers. The question is no longer which platform to use. It is how to use them in sequence to give patients the best chance of a response that lasts.

6.6. TCEs in the Context of Standard of Care

TCEs perform best not where existing therapies work, but where they structurally fail. Across four tumour types with mature TCE evidence, a consistent pattern emerges: TCEs occupy spaces that chemotherapy, checkpoint inhibitors, and targeted agents cannot reach.

6.6.1. Small-Cell Lung Cancer

Extensive-stage SCLC exemplifies oncology’s promise and frustration. Platinum-etoposide produces high initial responses, but these are transient. Adding PD-L1 inhibitors extends median survival only modestly, to ~12–13 months, because SCLC is immunologically cold—driven by neuroendocrine plasticity that suppresses antigen presentation and renders checkpoint inhibition ineffective [ 9 ].

Tarlatamab addresses this gap directly. DLL3, broadly expressed on SCLC cells but absent from normal adult tissues, provides a tumour-restricted target independent of antigen presentation or pre-existing T-cell reactivity [ 66 ]. The DeLLphi-304 trial demonstrated improved overall survival versus chemotherapy in second-line disease (median 13.6 vs. 8.3 months; HR 0.60; 95% CI 0.47–0.77; p < 0.001), earning FDA approval in November 2025 [ 53 ]. This represents a mechanistically distinct intervention that circumvents immunological barriers which have limited the efficacy of previous approaches. Whether tarlatamab can move earlier is now being tested in DeLLphi-305 ( NCT05690945 ), combining TCE-driven cytotoxicity with checkpoint-sustained T-cell function.

6.6.2. Metastatic Castration-Resistant Prostate Cancer

The mCRPC landscape has expanded: AR pathway inhibitors, PARP inhibitors, and lutetium-177 PSMA radioligand therapy. Yet resistance to each is inevitable: secondary AR mutations, reversion mutations, and antigen heterogeneity or internalisation for LuPSMA [ 81 ]. Patients progressing through two or three lines have exhausted most mechanisms.

TCEs bring new mechanisms. JANX007’s conditional masking means PSMA-expressing cells with LuPSMA resistance remain targetable [ 36 ]. Xaluritamig’s STEAP1 target is expressed across mCRPC subtypes, including variants refractory to hormonal strategies [ 44 ]. The XALute phase III trial compares xaluritamig against cabazitaxel in post-taxane mCRPC [ 100 ]. If positive, TCEs become a genuine alternative to cytotoxic chemotherapy where quality of life matters.

6.6.3. Microsatellite Stable Colorectal Cancer

MSS CRC represents oncology’s most stubborn unmet need. Checkpoint inhibitors are inactive in MSS tumours (>85% of metastatic cases) due to low mutational burden, T-cell exclusion, and immunosuppressive myeloid landscapes. Standard care is FOLFOX or FOLFIRI with bevacizumab or cetuximab, with declining response rates and no immunotherapy option for most.

CEA-targeted TCEs engage this need directly, though cibisatamab’s 4–14% response rate reflects microenvironmental barrier depth [ 66 ]. The mechanistic understanding is clear: TGF-β-driven stromal exclusion walls T-cells away, and STING suppression prevents innate immune priming [ 73 , 82 ]. Combinations pairing CEA-TCB with TGF-β blockade, STING agonists, or vascular normalisation are now in evaluation [ 95 ]. MSS CRC may prove where TCE combination therapy demonstrates clearest value.

6.6.4. Uveal Melanoma

Uveal melanoma carries one of the lowest mutational burdens of any cancer, is universally unresponsive to checkpoint inhibition, and historically offered response rates below 5%. The biological reason—low neoantigen burden plus MHC class I restriction—is precisely the problem TCEs were designed to solve. Tebentafusp delivered 49% mortality reduction versus investigator’s choice [ 85 , 90 ]. This survival benefit occurred despite only 9% radiographic response, demonstrating T-cell engagement translates into benefit through immune-mediated control not captured by conventional criteria [ 85 , 90 ].

Where every preceding immunotherapy failed, a TCE succeeded—not because the immune landscape was favourable, but because the TCE created one. That proof-of-concept, now extending to SCLC with tarlatamab and testing in mCRPC with xaluritamig, transforms TCEs from biologics into a platform.

6.6.5. A Unifying Theme

Across these indications, the pattern is consistent. TCEs succeed where existing therapies have structurally failed—where checkpoint inhibitors cannot find T-cells, where cellular therapies cannot be delivered, where resistance has eliminated options. The challenge is deploying TCEs earlier and combining them with microenvironment-modulating strategies ( Section 5.2 ), biomarker frameworks ( Section 5.3 ), and cellular therapy platforms. The coming randomised trials will determine whether that ambition translates into survival outcomes patients have waited too long to see.

6.7. Limitations and Evidence Gaps

Several constraints temper definitive conclusions. Most solid tumour TCE data derive from single-arm phase I/II studies with median enrolment of 40–100 patients and follow-up under 12 months, limiting durability assessment [ 74 ]. Antigen density thresholds were predominantly derived retrospectively; prospective validation in registration trials remains incomplete [ 70 , 94 ]. The pharmacokinetics of masked and multispecific constructs are incompletely characterised, with optimal step-up dosing schedules unexamined in randomised comparisons. Combination strategy evidence, while mechanistically compelling, derives largely from single-arm expansion cohorts without placebo controls, introducing potential for selection bias [ 82 , 97 ]. Long-term neurotoxicity surveillance is constrained by early protocol-mandated discontinuation upon ICANS [ 24 ]. Health economic assessments rely on modelling (incremental cost-effectiveness ratios ≤ US$150,000/QALY) rather than mature phase III data [ 76 ].

6.8. Regulatory, Commercial, and Access Considerations

Regulatory frameworks have evolved pragmatically. Step-up dosing protocols constitute a cornerstone of cytokine release syndrome mitigation, established through FDA collaboration on tarlatamab [ 100 ]. Companion diagnostic development reflects flexibility: despite FDA approval of the VENTANA DLL3 (SP347) assay, tarlatamab’s label permits treatment regardless of baseline expression, acknowledging efficacy across antigen levels [ 63 , 98 ]. For orphan indications, evidentiary standards remain stringent. Tebentafusp’s approval required overall survival as primary endpoint—quality-of-life data were supplementary evidence—a precedent underscoring that survival advantage remains decisive even in ultra-rare diseases [ 85 , 90 ]. Manufacturing innovations are reducing cost of goods. Knob-into-hole engineering and Pichia pastoris expression systems have driven yields to 2.5 g/L [ 11 , 53 ]. Cost modelling predicts 40% reductions for weekly administered Fc-silenced TCEs versus continuous-infusion formats [ 11 , 53 ].

6.9. Outlook: Toward 2030

The coming decade will determine whether TCEs become foundational solid tumour therapies. Randomised trials comparing TCE plus tumour microenvironment modulators against monotherapy will read out by 2027–2028, clarifying incremental value [ 82 ]. The DeLLphi-305 trial (tarlatamab plus PD-L1 maintenance in first-line ES-SCLC) and XALute trial (xaluritamig versus cabazitaxel in post-taxane prostate cancer) represent critical efficacy benchmarks [ 44 , 55 , 101 ]. Earlier disease settings offer substantial opportunity. The first adjuvant TCE trial—masked anti-PSMA construct versus observation in high-risk post-prostatectomy prostate cancer—opens in 2026 [ 36 ]. Paediatric precision-oncology initiatives will evaluate B7-H3-directed constructs in neuroblastoma and Ewing sarcoma [ 60 , 62 ]. Convergence with cellular immunotherapy platforms—allogeneic CAR-T cells combined with TCEs—enters early clinical development [ 64 , 65 ]. By 2030, biomarker-stratified, sequentially activated regimens will likely become standard: TCEs primed by brief microenvironment modulation, dosed via step-up protocols, guided by real-time immune profiling. With tarlatamab established in second-line SCLC and xaluritamig phase III readouts imminent, TCEs are positioned to become standard-of-care options in biomarker-defined solid tumours. Realising these potential demands sustained cross-disciplinary collaboration—and unwavering commitment to equitable global access.

TCEs have crossed the threshold from proof-of-concept to platform therapy in solid tumours. The coming decade will determine the breadth, durability, and ultimate clinical role of that platform.

7. Conclusions

The therapeutic landscape for solid tumours is undergoing a profound transformation. Next-generation CD3-based T-cell engagers have matured from experimental haematology agents into sophisticated, biomarker-directed immunotherapies that actively engage with the tumour microenvironment. Conditional activation strategies—ranging from protease-masked prodrugs to hypoxia-gated formats—have already reshaped the safety and efficacy profiles of these agents, delivering objective response rates approaching 40–50% in molecularly selected populations such as DLL3-positive small-cell lung cancer, while maintaining grade ≥ 3 cytokine release syndrome rates below 3% through optimised dosing. Following the pathbreaking precedent of tebentafusp in uveal melanoma, these advances are now translating into tangible survival benefits in phase III settings, with traditional regulatory approval secured for tarlatamab and pivotal trials advancing for prostate-selective agents. What emerges from this trajectory is something more than incremental progress. Molecular engineering, biomarker-driven selection, and tumour microenvironment modulation have stopped behaving as parallel developments and started converging into a unified framework—one that is actively reshaping TCEs from narrowly applicable experimental agents into genuine platform therapies capable of reaching patients across multiple solid tumour types. Yet the biology of resistance remains formidable. Antigen escape, physical exclusion by desmoplastic stroma, and T-cell exhaustion continue to limit durability in early-phase studies. What is striking about the current moment is that none of these barriers remains without a rational engineering response. Antigen loss meets trispecific constructs, stromal exclusion meets conditional masking and vascular normalisation, and T-cell exhaustion meets IL-15 fusions and 4-1BB costimulation. For the first time, resistance biology is functioning less as an obstacle and more as a design specification for next-generation TCE platforms. Crucially, the field now possesses rational countermeasures for each axis: multi-target and trispecific constructs to outmanoeuvre antigen loss; innate immune priming through oncolytic viruses, STING agonists, or CD40-targeting adjuvants to remodel the extracellular matrix; and orthogonal payloads such as IL-15 super-agonists or checkpoint-inhibitory scFvs to rejuvenate exhausted effector populations. This modular engineering paradigm positions TCEs not as static drugs, but as adaptable platforms capable of iterative refinement across diverse oncologic contexts.

Looking toward the horizon, 2030 promises a fundamental realignment of the treatment paradigm. Phase III readouts for agents like xaluritamig, alongside emerging adjuvant and paediatric programmes, will determine whether TCEs anchor second-line therapy in biomarker-defined populations and extend into curative-intent settings. The vision—biomarker-stratified, sequentially activated regimens combining TCEs with short-course tumour microenvironment modulation and guided by real-time immune profiling—is fast becoming tangible reality. Delivering on this promise demands more than scientific ingenuity; it requires sustained cross-disciplinary collaboration among molecular engineers, translational immunologists, trialists, and regulatory scientists. Immediate priorities include prospective validation of composite biomarker algorithms, harmonisation of complementary diagnostics such as the VENTANA DLL3 assay, rigorous long-term safety surveillance, and—critically—ensuring equitable global access. For populations in low- and middle-income countries, where diagnostic infrastructure and cost remain prohibitive barriers, the challenge is as much about delivery as discovery. With continued innovation grounded in these collaborative principles, the coming generation of T-cell engagers holds genuine potential to deliver durable clinical benefit and, ultimately, transformative cancer control for solid tumours.

Disclaimer/Publisher’s Note: The statements, opinions and data contained in all publications are solely those of the individual author(s) and contributor(s) and not of MDPI and/or the editor(s). MDPI and/or the editor(s) disclaim responsibility for any injury to people or property resulting from any ideas, methods, instructions or products referred to in the content.

Author Contributions

Conceptualization, S.R.A.; methodology, S.R.A., M.Z. and H.H.; software, S.R.A. and A.M.A.; validation, A.O.A.M., A.A.G.A. and A.M.; formal analysis, S.R.A., M.Z., M.S.K., M.F.A. and H.H.; investigation, S.R.A. and A.M.A. and M.Z.; resources, S.R.A. and M.Z.; data curation, M.S.K., M.F.A., A.A.G.A., A.O.A.M. and A.M.; writing—original draft preparation, S.R.A.; writing—review and editing, M.Z., A.M.A., H.H. and A.M.; visualization, S.R.A., M.Z. and A.M.A.; supervision, S.R.A.; project administration, S.R.A. and M.Z.; funding acquisition, M.Z. All authors have read and agreed to the published version of the manuscript.

Data Availability Statement

The data generated during this study are included in this article.

Conflicts of Interest

The authors declare no conflicts of interest.

References

1. Köhler G. Milstein C. Continuous cultures of fused cells secreting antibody of predefined specificity Nature 1975 256 495 497 10.1038/256495a0 1172191

2. Staerz U.D. Kanagawa O. Bevan M.J. Hybrid antibodies can target sites for attack by T cells Nature 1985 314 628 631 10.1038/314628a0 2859527

3. Baeuerle P.A. Reinhardt C. Bispecific T-cell engager (BiTE) antibodies for cancer therapy Cancer Res. 2009 69 4941 4944 10.1158/0008-5472.CAN-09-0547 19509221

4. Topp M.S. Gökbuget N. Stein A.S. Zugmaier G. O’BRien S. Bargou R.C. Dombret H. Fielding A.K. Heffner L. Larson R.A. Safety and activity of blinatumomab for adult patients with relapsed or refractory B-precursor acute lymphoblastic leukaemia Lancet Oncol. 2015 16 57 66 10.1016/S1470-2045(14)71170-2 25524800

5. U.S. Food and Drug Administration FDA Approves Blincyto (Blinatumomab) for Precursor B-Cell Acute Lymphoblastic Leukemia FDA Silver Spring, MD, USA 2014

6. Evitria. FDA-Approved Bispecific Antibodies: Comprehensive List & Insights Available online: https://www.evitria.com/journal/bispecific-antibodies/fda-approved-bispecific-antibodies (accessed on 1 March 2025)

7. Dustin M.L. Choudhuri K. Signaling and polarized communication across the T cell immunological synapse Annu. Rev. Cell Dev. Biol. 2016 32 303 325 10.1146/annurev-cellbio-100814-125330 27501450

8. Labrijn A.F. Janmaat M.L. Reichert J.M. Parren P.W.H. Bispecific antibodies: A mechanistic review of the pipeline Nat. Rev. Drug Discov. 2019 18 585 608 10.1038/s41573-019-0028-1 31175342

9. Ahn M.J. Cho B.C. Felip E. Korantzis I. Ohashi K. Majem M. Juan-Vidal O. Handzhiev S. Izumi H. Lee J.-S. Tarlatamab for patients with previously treated small-cell lung cancer N. Engl. J. Med. 2023 389 2063 2075 10.1056/NEJMoa2307980 37861218

10. Hernberg M. Pyrhönen S. Muhonen T. Regimens with or without interferon-α as treatment for metastatic melanoma and renal cell carcinoma: An overview of randomized trials J. Immunother. 1999 22 145 154 10.1097/00002371-199903000-00006 10093039

11. Segal N.H. Melero I. Moreno V. Steeghs N. Marabelle A. Rohrberg K. Rodriguez-Ruiz M.E. Eder J.P. Eng C. Manji G.A. CEA-CD3 bispecific antibody cibisatamab with or without atezolizumab in patients with CEA-positive solid tumours: Results of two multi-institutional Phase 1 trials Nat. Commun. 2024 15 4091 10.1038/s41467-024-48479-8 38750034 PMC11096172

12. Naing A. Mantia C. Morgensztern D. Kim T.Y. Li D. Kang Y.K. Marron T.U. Tripathi A. George S. Rini B.I. First-in-human study of SRF388, a first-in-class IL-27 targeting antibody, as monotherapy and in combination with pembrolizumab in patients with advanced solid tumors J. Clin. Oncol. 2022 40 2501 10.1200/JCO.2022.40.16_suppl.2501

13. McGranahan N. Swanton C. Clonal heterogeneity and tumor evolution Cell 2017 168 613 628 10.1016/j.cell.2017.01.018 28187284

14. Jain R.K. Normalizing tumor microenvironment to treat cancer: Bench to bedside to biomarkers J. Clin. Oncol. 2013 31 2205 2218 10.1200/JCO.2012.46.3653 23669226 PMC3731977

15. Thurber G.M. Schmidt M.M. Wittrup K.D. Antibody tumor penetration: Transport opposed by systemic and antigen-mediated clearance Adv. Drug Deliv. Rev. 2008 60 1421 1434 10.1016/j.addr.2008.04.012 18541331 PMC2820307

16. Marvel D. Gabrilovich D.I. Myeloid-derived suppressor cells in the tumor microenvironment: Expect the unexpected J. Clin. Investig. 2015 125 3356 3364 10.1172/JCI80005 26168215 PMC4588239

17. Facciabene A. Peng X. Hagemann I.S. Balint K. Barchetti A. Wang L.-P. Gimotty P.A. Gilks C.B. Lal P. Zhang L. Tumour hypoxia promotes tolerance and angiogenesis via CCL28 and Treg cells Nature 2011 475 226 230 10.1038/nature10169 21753853

18. Routy B. Le Chatelier E. Derosa L. Duong C.P.M. Alou M.T. Daillère R. Fluckiger A. Messaoudene M. Rauber C. Roberti M.P. Gut microbiome influences efficacy of PD-1-based immunotherapy against epithelial tumors Science 2018 359 91 97 10.1126/science.aan3706 29097494

19. Cattaruzza F. Nazeer A. To M. Hammond M. Koski C. Liu L.Y. Pete Yeung V. Rennerfeldt D.A. Henkensiefken A. Fox M. Precision-activated T-cell engagers targeting HER2 or EGFR and CD3 mitigate on-target, off-tumor toxicity for immunotherapy in solid tumors Nat. Cancer 2023 4 485 501 10.1038/s43018-023-00536-9 36997747 PMC10132983

20. Wu L. Seung E. Xu L. Rao E. Lord D.M. Wei R.R. Cortez-Retamozo V. Ospina B. Posternak V. Ulinski G. Trispecific antibodies enhance the therapeutic efficacy of tumor-directed T cells through T cell receptor co-stimulation Nat. Cancer 2020 1 86 98 10.1038/s43018-019-0004-z 35121834

21. Kloss C.C. Lee J. Zhang A. Chen F. Melenhorst J.J. Lacey S.F. Maus M.V. Fraietta J.A. Zhao Y. June C.H. Dominant-Negative TGF-β Receptor Enhances PSMA-Targeted Human CAR T Cell Proliferation And Augments Prostate Cancer Eradication Mol. Ther. 2018 26 1855 1866 10.1016/j.ymthe.2018.05.003 29807781 PMC6037129

22. Tanaka A. Sakaguchi S. Targeting Foxp3+ regulatory T cells in cancer immunotherapy Nat. Rev. Immunol. 2017 17 477 488 10.1038/nri.2017.38

23. Philip M. Schietinger A. CD8+ T cell differentiation and dysfunction in cancer Nat. Rev. Immunol. 2022 22 209 223 10.1038/s41577-021-00574-3 34253904 PMC9792152

24. Brahmer J.R. Lacchetti C. Schneider B.J. Atkins M.B. Brassil K.J. Caterino J.M. Chau I. Ernstoff M.S. Gardner J.M. Ginex P. Management of Immune-Related Adverse Events in Patients Treated With Immune Checkpoint Inhibitor Therapy: American Society of Clinical Oncology Clinical Practice Guideline J. Clin. Oncol. 2018 36 1714 1768 10.1200/JCO.2017.77.6385 29442540 PMC6481621

25. Demšar A. Geršak K. Gazić B. Blagus T. Goričar K. Jezernik G. Dolžan V. Grašič Kuhar C. Immune Infiltration, Effector T-Cell Enrichment, and Functional Context for Prediction of Pathologic Complete Response to Neoadjuvant Chemotherapy in Breast Cancer Int. J. Mol. Sci. 2026 27 2431 10.3390/ijms27052431 41828646 PMC12985902

26. Sergeeva O.A. Jin G. Sarkar M. Zeiger J. Dhindwal S. Chin S.S. Abulizi A. Lai Z. Brown C. DeMaria W.G. EVOLVE platform, a trispecific T cell engager with integrated CD2 costimulation, for the treatment of solid and hematologic tumors Proc. Natl. Acad. Sci. USA 2025 122 e2510829122 10.1073/pnas.2510829122 41252152 PMC12663936

27. Dewaele L. Fernandes R.A. Bispecific T-cell engagers for the recruitment of T cells in solid tumors: A literature review Immunother. Adv. 2025 5 ltae005 10.1093/immadv/ltae005 40083373 PMC11904783

28. Rodriguez-Otero P. Usmani S.Z. Cohen A.D. van de Donk N.W.C.J. Leleu X. Pérez-Larraya J.G. Manier S. Nooka A.K. Mateos M.V. Einsele H. International Myeloma Working Group immunotherapy committee consensus guidelines and recommendations for optimal use of T-cell-engaging bispecific antibodies in multiple myeloma Hematol. Oncol. 2024 42 e3078 10.1016/S1470-2045(24)00043-3 38697166

29. Spinazzola A. Iannantuono G.M. Gulley J.L. Giudice E. Filetti M. Sganga S. Bianco F.L. Floudas C.S. Daniele G. Current landscape of T-cell engagers in early-phase clinical development in solid cancers Front. Immunol. 2025 16 1665838 10.3389/fimmu.2025.1665838 41122165 PMC12536262

30. Liu M. Etherington M.S. Hanna A. Medina B.D. Vitiello G.A. Bowler T.G. Param N.J. Levin L. Rossi F. DeMatteo R.P. Oncogenic KIT modulates type I IFN–mediated antitumor immunity in GIST Cancer Immunol. Res. 2021 9 542 553 10.1158/2326-6066.cir-20-0692 33648985 PMC8102332

31. Frederickson R.M. Herzog R.W. An annis mirabilis for the molecular therapy journal family Mol. Ther. 2021 29 1 10.1016/j.ymthe.2020.12.011 33308439 PMC7791143

32. Desnoyers L.R. Vasiljeva O. Richardson J.H. Yang A. Menendez E.E.M. Liang T.W. Wong C. Bessette P.H. Kamath K. Moore S.J. Tumor-specific activation of an EGFR-targeting Probody enhances therapeutic index Sci. Transl. Med. 2013 5 207ra144 10.1126/scitranslmed.3006682 24132639

33. Topp M.S. Gökbuget N. Zugmaier G. Klappers P. Stelljes M. Neumann S. Viardot A. Marks R. Diedrich H. Faul C. Phase II trial of the anti-CD19 bispecific T cell-engager blinatumomab shows hematologic and molecular remissions in patients with relapsed or refractory B-precursor acute lymphoblastic leukemia J. Clin. Oncol. 2014 32 4134 4140 10.1200/JCO.2014.56.3247 25385737

34. Müller D. Karle A. Meissburger B. Höfig I. Stork R. Kontermann R.E. Improved pharmacokinetics of recombinant bispecific antibody molecules by fusion to human serum albumin J. Biol. Chem. 2007 282 12650 12660 10.1074/jbc.M700820200 17347147

35. Moore G.L. Bautista C. Pong E. Nguyen D.-H.T. Jacinto J. Eivazi A. Muchhal U.S. Karki S. Chu S.Y. Lazar G.A. A novel bispecific antibody format enables simultaneous bivalent and monovalent co-engagement of distinct target antigens mAbs 2011 3 546 557 10.4161/mabs.3.6.18123 22123055 PMC3242841

36. Janux Therapeutics JANX007 Phase 1 Data in Metastatic Castration-Resistant Prostate Cancer Press Release Janux Therapeutics San Diego, CA, USA 2025

37. Janux Therapeutics Program Update on Ongoing Phase 1 JANX008 Study Press Release Janux Therapeutics San Diego, CA, USA 2025

38. Vir Biotechnology First Patient Dosed in Phase 1 Trial of VIR-5525 Press Release Vir Biotechnology San Francisco, CA, USA 2025

39. Amgen Xaluritamig (AMG 509) Investigator’s Brochure Amgen Thousand Oaks, CA, USA 2024

40. Mountzios G. Sun L. Cho B.C. Demirci U. Baka S. Gümüş M. Lugini A. Zhu B. Yu Y. Korantzis I. Tarlatamab in small-cell lung cancer after platinum-based chemotherapy N. Engl. J. Med. 2025 393 349 361 10.1056/NEJMoa2502099 40454646

41. Kontermann R.E. Strategies for extended serum half-life of protein therapeutics Curr. Opin. Biotechnol. 2011 22 868 876 10.1016/j.copbio.2011.06.012 21862310

42. Kontermann R.E. Half-life extended biotherapeutics Expert Opin. Biol. Ther. 2016 16 903 915 10.1517/14712598.2016.1165661 26967759

43. Bar N. Martin T. Hofmeister C.C. Mateos M.-V. Hansson M. Paris L. Namburi S. Ribas P. Santoro A. Rodriguez-Otero P. Alnuctamab, a bivalent B-cell maturation antigen-targeting T cell engager for patients with relapsed or refractory multiple myeloma: Results from a phase 1, first-in-human study Leukemia 2026 40 481 490 10.1038/s41375-025-02841-x 41501501 PMC12960249

44. Kelly W.K. Danila D.C. Lin C.C. Lee J.-L. Matsubara N. Ward P.J. Armstrong A.J. Pook D. Kim M. Dorff T.B. Xaluritamig, a STEAP1 × CD3 XmAb 2+1 immune therapy for metastatic castration-resistant prostate cancer: Results from dose exploration in a first-in-human study Cancer Discov. 2024 14 76 89 10.1158/2159-8290.CD-23-0964 37861461 PMC10784743

45. Haber L. Olson K. Kelly M.P. Crawford A. DiLillo D.J. Tavaré R. Ullman E. Mao S. Canova L. Sineshchekova O. CD3 affinity tuning differentiates cytokine release Sci. Rep. 2021 11 14397 34257348 10.1038/s41598-021-93842-0 PMC8277787

46. National Library of Medicine Affinity-Tuned T-Cell Engager Trial. ClinicalTrials.gov Identifier NCT05987204 Available online: https://clinicaltrials.gov/study/NCT05987204 (accessed on 1 March 2026)

47. Wong H.C. Jeng E.K. Rhode P.R. The IL-15-based superagonist ALT-803 promotes the antigen-independent conversion of memory CD8+ T cells into innate-like effector cells with antitumor activity Oncoimmunology 2013 2 e26442 10.4161/onci.26442 24404427 PMC3881336

48. Schmiechen Z.C. Nanda H.A. Burrack A.L. Hickok G.H. Butler J.Z. Cruz-Hinojoza E. Maurice N.J. Geuenich M.J. Yu C. Tsai A.K. IL-15 complex enhances agonistic anti-CD40 + anti-PDL1 by correcting the T-bet to Tox ratio in CD8+ T cells infiltrating pancreatic ductal adenocarcinoma Cancer Immunol. Res. 2025 13 847 866 10.1158/2326-6066.CIR-24-0758 40072469 PMC12134750

49. Waldmann T.A. Dubois S. IL-15: From discovery to FDA approval J. Hematol. Oncol. 2025 18 64 10.1186/s13045-025-01664-8 39966991 PMC11837486

50. Von Kreudenstein T.S. Escobar-Carbrera E. Lario P.I. D’Angelo I. Brault K. Kelly J. Durocher Y. Baardsnes J. Woods R.J. Xie M.H. Improving biophysical properties of a bispecific antibody scaffold to aid developability: Quality by molecular design MAbs 2013 5 646 654 10.4161/mabs.25632 23924797 PMC3851217

51. Aromataris E. Lockwood C. Porritt K. Pilla B. Jordan Z. Systematic review methodology JBI Manual for Evidence Synthesis JBI Adelaide, Australia 2024

52. Jiao Y. Bai L. Wang J. Yan V.K.C. Chan E.W. Na R. Bishai D.M. Guan X. Li X. Analysis of drug consumption and expenditure of WHO essential medicines for cancer in 40 countries and regions between 2012 and 2022: A multinational drug use study Lancet Oncol. 2025 26 877 886 10.1016/S1470-2045(25)00159-7 40409320

53. Leighl N.B. Akamatsu H. Lim S.M. Cheng Y. Minchom A.R. Marmarelis M.E. Sanborn R.E. Yang J.C.H. Liu B. John T. Subcutaneous amivantamab vs intravenous amivantamab, both in combination with lazertinib, in refractory EGFR-mutated, advanced non-small cell lung cancer (NSCLC): Primary results, including overall survival (OS), from the global, phase 3, randomized controlled PALOMA-3 trial J. Clin. Oncol. 2024 42 LBA8505 10.1200/JCO.2024.42.17_suppl.LBA8505

54. Pruis M.A. Krebs M.G. Plummer R. De Vos F. Angevin E. Prenen H. Forster M.D. Clack G. Van der Aa A. Tjwa M. A Phase I Trial of the Dual MET Kinase/OCT-2 Inhibitor OMO-1 in Metastatic Solid Malignancies Including MET Exon 14 Mutated Lung Cancer Oncologist 2023 28 e1248 e1258 10.1093/oncolo/oyad146 37260332 PMC10712729

55. Page M.J. McKenzie J.E. Bossuyt P.M. Boutron I. Hoffmann T.C. Mulrow C.D. Shamseer L. Tetzlaff J.M. Akl E.A. Brennan S.E. The PRISMA 2020 statement: An updated guideline for reporting systematic reviews BMJ 2021 372 n71 10.1136/bmj.n71 33782057 PMC8005924

56. Molloy M.E. Aaron W.H. Barath M. Bush M.C. Callihan E.C. Carlin K. Cremin M. Evans T. Guerrero M.G. Hemmati G. HPN328, a trispecific T cell-activating protein construct targeting DLL3-expressing solid tumors Mol. Cancer Ther. 2024 23 1294 1304 10.1158/1535-7163.MCT-23-0524 38670552 PMC11372363

57. Meric-Bernstam F. Beeram M. Hamilton E. Oh D.-Y. Hanna D.L. Kang Y.-K. Elimova E. Chaves J. Goodwin R. Lee J. Zanidatamab, a novel bispecific antibody, for the treatment of locally advanced or metastatic HER2-expressing or HER2-amplified cancers: A phase 1, dose-escalation and expansion study Lancet Oncol. 2022 23 1558 1570 10.1016/S1470-2045(22)00621-0 36400106

58. Yu S. Zhang J. Yan Y. Yao X. Fang L. Xiong H. Liu Y. Chu Q. Zhou P. Wu K. A novel asymmetrical anti-HER2/CD3 bispecific antibody exhibits potent cytotoxicity for HER2-positive tumor cells J. Exp. Clin. Cancer Res. 2019 38 355 10.1186/s13046-019-1354-1 31412896 PMC6694677

59. Lee E.K. Yeku O. Winer I. Hamilton E. Richardson D. Zhang J. Konecny G. Anderson I. Wu X. Orr D. A phase I/II study of rinatabart sesutecan (Rina-S) in patients with advanced ovarian or endometrial cancer Ann. Oncol. 2024 35 S550 10.1016/j.annonc.2024.08.781

60. Vitanza N.A. Ronsley R. Choe M. Seidel K. Huang W. Rawlings-Rhea S.D. Beam M. Steinmetzer L. Wilson A.L. Brown C. Intracerebroventricular B7-H3-targeting CAR T cells for diffuse intrinsic pontine glioma: A phase 1 trial Nat. Med. 2025 31 861 868 10.1038/s41591-024-03451-3 39775044 PMC11922736

61. Goto H. CRS and ICANS in CAR T-cell therapy for large B-cell lymphoma J. Clin. Exp. Hematop. 2025 65 268 275 10.3960/jslrt.25072 41443883 PMC12884143

62. Maddalena G. van der Pol Y. Rivers Z. Dasari A. Rogers J.E. Guinney J. Kopetz S. Raghav K.P.S. Impact of KRAS codon-specific mutations on survival of patients with metastatic CRC (mCRC) treated with trifluridine-tipiracil (TAS) plus bevacizumab (Bev): A real-world analysis J. Clin. Oncol. 2024 42 201 10.1200/JCO.2024.42.3_suppl.201

63. Lee D.W. Santomasso B.D. Locatelli F. Ghobadi A. Turtle C.J. Brudno J.N. Maus M.V. Park J.H. Mead E. Pavletic S. ASTCT consensus grading for cytokine release syndrome and neurologic toxicity associated with immune effector cells Biol. Blood Marrow Transplant. 2019 25 625 638 10.1016/j.bbmt.2018.12.758 30592986 PMC12180426

64. Majzner R.G. Mackall C.L. Tumor antigen escape from CAR T-cell therapy Cancer Discov. 2018 8 1219 1226 10.1158/2159-8290.CD-18-0442 30135176

65. Cowan A.J. Pont M.J. Sather B.D. Turtle C.J. Till B.G. Libby E.N. Coffey D.G. Tuazon S.A. Wood B. Gooley T. γ-Secretase inhibitor in combination with BCMA chimeric antigen receptor T-cell immunotherapy for individuals with relapsed or refractory multiple myeloma: A phase 1, first-in-human trial Lancet Oncol. 2023 24 811 822 10.1016/S1470-2045(23)00246-2 37414012 PMC10783021

66. Paz-Ares L. Champiat S. Lai W.V. Izumi H. Govindan R. Boyer M. Hummel H.-D. Borghaei H. Johnson M.L. Steeghs N. Tarlatamab, a first-in-class DLL3-targeted bispecific T-cell engager, in recurrent small-cell lung cancer: An open-label, phase I study J. Clin. Oncol. 2023 41 2893 2903 10.1200/JCO.22.02823 36689692 PMC10414718

67. Joshi G.N. Yang M.T. Yoshioka J. Yanagawa F. 102 Multimodal spatial imaging and analysis in NSCLC tissue shows anticorrelation between IDO-1 and VISTA J. Immunother. Cancer 2024 12 A102 10.1136/jitc-2024-SITC2024.0102

68. Bose S. Ge L. Lee S.M. Van Tine B.A. Agulnik M. Weiss M.C. Lacuna K.P. Sender N. Ana S.S. Hsu F.J. A phase II trial with safety lead-in to evaluate the addition of sotigalimab, a CD40 agonistic monoclonal antibody, to standard-of-care doxorubicin for the treatment of advanced sarcoma J. Clin. Oncol. 2023 41 664 10.1200/JCO.2023.41.16_suppl.11565

69. Shen F. Kelly W.K. Pandit-Taskar N. McDevitt T. Smith R. Menard K. Packman K. Chaudhary R. Rupnow B. Yu M.K. Preclinical characterization of human Kallikrein 2 (hK2) as a novel target for the treatment of prostate cancer J. Clin. Oncol. 2024 42 202 10.1200/JCO.2024.42.4_suppl.202

70. Jiang S. Coombs A. Finch W. Thomas M. Morris T. Forster M. Tuthill M. 153 Turning the tumour-associated antigen survivin into a neoantigen-like cancer therapeutic vaccine J. Immunother. Cancer 2024 12 A153 10.1136/jitc-2024-SITC2024.0153

71. Curtius K. Abstract SY02-03: Stochastic modeling of clonal evolution in carcinogenesis Cancer Res. 2024 84 SY02-03 10.1158/1538-7445.AM2024-SY02-03

72. O’Hara M.H. O’Reilly E.M. Varadhachary G. Wolff R.A. Wainberg Z.A. Ko A.H. Fisher G. Rahma O. Lyman J.P. Cabanski C.R. CD40 agonistic monoclonal antibody APX005M (sotigalimab) and chemotherapy, with or without nivolumab, for the treatment of metastatic pancreatic adenocarcinoma: An open-label, multicentre, phase 1b study Lancet Oncol. 2021 22 118 131 10.1016/s1470-2045(20)30532-5 33387490

73. Dancsok A.R. Gao D. Lee A.F. Steigen S.E. Blay J.-Y. Thomas D.M. Maki R.G. Nielsen T.O. Demicco E.G. Tumor-associated macrophages and macrophage-related immune checkpoint expression in sarcomas OncoImmunology 2020 9 1747340 10.1080/2162402x.2020.1747340 PMC7153829 32313727

74. Mariathasan S. Turley S.J. Nickles D. Castiglioni A. Yuen K. Wang Y. Kadel E.E. III Koeppen H. Astarita J.L. Cubas R. TGFβ attenuates tumour response to PD-L1 blockade by contributing to exclusion of T cells Nature 2018 554 544 548 10.1038/nature25501 29443960 PMC6028240

75. Trachtman H. Fervenza F.C. Gipson D.S. Heering P. Jayne D.R. Peters H. Rota S. Remuzzi G. Rump L.C. Sellin L.K. A phase 1, single-dose study of fresolimumab, an anti-TGF-β antibody, in treatment-resistant primary focal segmental glomerulosclerosis Kidney Int. 2011 79 1236 1243 10.1038/ki.2011.33 21368745 PMC3257033

76. Loftus T.J. Rupji M. Jalal S.I. Dhippayom T. Bock A.M. Wagner C.B. Thavorn K. Chaiyakunapruk N. Cost-effectiveness of chimeric antigen receptor T-cell therapy for relapsed or refractory large B-cell lymphoma: A systematic review and meta-analysis J. Med. Econ. 2025 28 2216 2235 10.1080/13696998.2025.2600217 41405081

77. Formenti S.C. Lee P. Adams S. Goldberg J.D. Li X. Xie M.W. Ratikan J.A. Felix C. Hwang L. Faull K.F. Focal irradiation and systemic TGFβ blockade in metastatic breast cancer Clin. Cancer Res. 2018 24 2493 2504 10.1158/1078-0432.CCR-17-3322 29476019 PMC5999326

78. Teijeira A. Arribas B. Marre D. Karanikas V. Luri C. Cirella A. Olivera I. Cañamero M. Alvarez M. Ochoa M.C. Three-dimensional colon cancer organoids model the response to CEA-CD3 T-cell engagers Theranostics 2022 12 1373 1387 10.7150/thno.63359 35154495 PMC8771540

79. Giannella L. Delli Carpini G. Di Giuseppe J. Bogani G. Sopracordevole F. Clemente N. Giorda G. De Vincenzo R.P. Evangelista M.T. Gardella B. In Situ/Microinvasive Adenocarcinoma of the Uterine Cervix and HPV-Type Impact: Pathologic Features, Treatment Options, and Follow-Up Outcomes—Cervical Adenocarcinoma Study Group (CAS-Group) Cancers 2023 15 2876 10.3390/cancers15112876 37296839 PMC10252050

80. Sureshbabu S.K. Chaukar D. Chiplunkar S.V. Hypoxia regulates the differentiation and anti-tumor effector functions of γδT cells in oral cancer Clin. Exp. Immunol. 2020 201 40 57 10.1111/cei.13436 32255193 PMC7290089

81. Stuparu A.D. Capri J.R. Meyer C.A.L. Le T.M. Evans-Axelsson S.L. Current K. Lennox M. Mona C.E. Fendler W.P. Calais J. Mechanisms of Resistance to Prostate-Specific Membrane Antigen-Targeted Radioligand Therapy in a Mouse Model of Prostate Cancer J. Nucl. Med. 2021 62 989 995 10.2967/jnumed.120.256263 33277393 PMC8882874

82. Ball K. Dovedi S.J. Vajjah P. Phipps A. Strategies for clinical dose optimization of T cell-engaging therapies in oncology mAbs 2023 15 2181016 10.1080/19420862.2023.2181016 36823042 PMC9980545

83. Chan J.M. Quintanal-Villalonga Á. Gao V.R. Xie Y. Allaj V. Chaudhary O. Masilionis I. Egger J. Chow A. Walle T. Signatures of plasticity, metastasis, and immunosuppression in an atlas of human small cell lung cancer Cancer Cell 2021 39 1479 1496 10.1016/j.ccell.2021.09.008 34653364 PMC8628860

84. ClinicalTrials.gov. NCT05243797 : A Study to Compare Teclistamab Plus Lenalidomide and Dexamethasone Versus Lenalidomide and Dexamethasone in Participants with Newly Diagnosed Multiple Myeloma Available online: https://clinicaltrials.gov/ct2/show/NCT05243797 (accessed on 15 January 2026)

85. Nathan P. Hassel J.C. Rutkowski P. Baurain J.-F. Butler M.O. Schlaak M. Sullivan R.J. Ochsenreither S. Dummer R. Kirkwood J.M. Overall survival benefit with tebentafusp in metastatic uveal melanoma N. Engl. J. Med. 2021 385 1196 1206 10.1056/NEJMoa2103485 34551229

86. U.S. Food and Drug Administration Multi-Discipline Review: Kimmtrak (Tebentafusp-tebn). BLA 761228 FDA Silver Spring, MD, USA 2022

87. U.S. Food and Drug Administration FDA Approves Tebentafusp-tebn for Unresectable or Metastatic Uveal Melanoma FDA Silver Spring, MD, USA 2022

88. ClinicalTrials.gov. NCT04471727 : Spatial Multi-Omics Analysis of T-Cell Engager Tumor Microenvironment Remodeling Available online: https://clinicaltrials.gov/study/NCT04471727 (accessed on 1 February 2026)

89. Wang Y. Sun W. Wang B. Evaluating the efficacy and safety of tebentafusp in metastatic uveal melanoma: A 2025 update systematic review and meta-analysis Front. Oncol. 2025 15 1667282 10.3389/fonc.2025.1667282 41170451 PMC12568024

90. Howlett S. Carter T.J. Shaw H.M. Nathan P.D. Tebentafusp: A first-in-class treatment for metastatic uveal melanoma Ther. Adv. Med. Oncol. 2023 15 17588359231160140 10.1177/17588359231160140 36970111 PMC10031621

91. Marcoux N. Gettinger S.N. O’Kane G. Arbour K.C. Neal J.W. Husain H. Evans T.L. Brahmer J.R. Muzikansky A. Bonomi P.D. EGFR-mutant adenocarcinomas that transform to small-cell lung cancer and other neuroendocrine carcinomas: Clinical outcomes J. Clin. Oncol. 2019 37 278 285 10.1200/JCO.18.01585 30550363 PMC7001776

92. Cooper A.J. Herzberg B. Parma M. Wang K. Alder L. Weber U.M. Patil T. Joshi U. Puri S. Saqi A. Multi-institution analysis of tarlatamab for the treatment of EGFR-mutant transformed small cell lung cancer J. Clin. Oncol. 2024 42 8505 10.1200/PO-25-00538 PMC12670137 41166675

93. Arenas E.J. Martínez-Sabadell A. Rius Ruiz I. Román Alonso M. Escorihuela M. Luque A. Fajardo C.A. Gros A. Klein C. Arribas J. Acquired cancer cell resistance to T cell bispecific antibodies and CAR T targeting HER2 through JAK2 down-modulation Nat. Commun. 2021 12 1237 10.1038/s41467-021-21445-4 33623012 PMC7902842

94. Shin D.S. Zaretsky J.M. Escuin-Ordinas H. Garcia-Diaz A. Hu-Lieskovan S. Kalbasi A. Grasso C.S. Hugo W. Sandoval S. Torrejon D.Y. Primary resistance to PD-1 blockade mediated by JAK1/2 mutations Cancer Discov. 2017 7 188 201 10.1158/2159-8290.CD-16-1223 27903500 PMC5296316

95. Letouzé E. Moreau P. Munshi N. Samur M.K. Minvielle S. Touzeau C. Mechanisms of resistance to bispecific T-cell engagers in multiple myeloma and their clinical implications Blood Adv. 2024 8 2952 2959 10.1182/bloodadvances.2023012354 38513088 PMC11302375

96. Zhang Y. Feng Y. Zhou S. Gao S. Xiong B. Gao X. Song Y. Liu L. Wang C. Yang Y. Establishment of a model of LPS-induced inflammatory injury in human aortic endothelial cells Biomed. Pharmacother. 2024 174 116576 10.1016/j.biopha.2024.116576 38593707

97. Philipp N. Kazerani M. Nicholls A. Vick B. Wulf J. Straub T. Scheurer M. Muth A. Hänel G. Nixdorf D. T-cell exhaustion induced by continuous bispecific molecule exposure is ameliorated by treatment-free intervals Blood 2022 140 1104 1118 10.1182/blood.2022015956 35878001 PMC10652962

98. Ding J. Yeong C. Advances in DLL3-targeted therapies for small cell lung cancer: Challenges, opportunities, and future directions Front. Oncol. 2024 14 1504139 10.3389/fonc.2024.1504139 39703856 PMC11655346

99. Majzner R. Ramakrishna S. Mochizuki A. Patel S. Chinnasamy H. Yeom K. Schultz L. Richards R. Campen C. Reschke A. EPCT-14. GD2 CAR T-CELLS MEDIATE CLINICAL ACTIVITY AND MANAGEABLE TOXICITY IN CHILDREN AND YOUNG ADULTS WITH H3K27M-MUTATED DIPG AND SPINAL CORD DMG Neuro Oncol. 2021 23 i49 i50 10.1093/neuonc/noab090.200

100. Nolan-Stevaux O. Smith J. Double-conditional T-cell engagers: Precision activation in the tumor microenvironment J. Exp. Med. 2025 223 e20251652 10.1084/jem.20251652 41474966

101. Graf K.C. Davis J.A. Cendagorta A. Granger K. Gaffney K.J. Green K. Hess B.T. Hashmi H. ‘Fast but not so furious’: A condensed step-up dosing schedule of teclistamab for relapsed/refractory multiple myeloma eJHaem 2024 5 793 797 10.1002/jha2.906 39157607 PMC11327708

102. Khandare J. Ramesh A. Aland G. Patil P. Deshpande T. Khutale G. Jayant S. Joshi A. Joshi K.C. Chugh B. Effect of circulating tumor cells in clinically stable patients on the conundrum of recurrence with cellular residual disease J. Clin. Oncol. 2024 42 e15040 10.1200/JCO.2024.42.16_suppl.e15040

103. Qiu Y. Han Y. Guo B. A Biomarker-Based Dose-Schedule Optimization Design for Immunotherapy Trials Stat. Med. 2026 45 e70357 10.1002/sim.70357 41572569 PMC12828111

104. National Institute for Health and Care Excellence Final Appraisal Document: Tebentafusp for Treating Unresectable or Metastatic Uveal Melanoma (TA925) NICE London, UK 2023

105. Seo H. Chen J. González-Avalos E. Samaniego-Castruita D. Das A. Wang Y.H. López-Moyado I.F. Georges R.O. Zhang W. Onodera A. TOX and TOX2 transcription factors cooperate with NR4A transcription factors to impose CD8+ T cell exhaustion Proc. Natl. Acad. Sci. USA 2019 116 12410 12415 10.1073/pnas.1905675116 31152140 PMC6589758

106. Lee H. Durante M. Skerget S. Vishwamitra D. Benaoudia S. Ahn S. Poorebrahim M. Barakat E. Jung D. Leblay N. Impact of soluble BCMA and non–T-cell factors on refractoriness to BCMA-targeting T-cell engagers in multiple myeloma Blood 2024 144 2637 2651 10.1182/blood.2024026212 39321344 PMC11738017

107. Ros J. Balconi F. Baraibar I. Gonzalez N.S. Salva F. Tabernero J. Elez E. Advances in immune checkpoint inhibitor combination strategies for microsatellite stable colorectal cancer Front. Oncol. 2023 13 1112276 10.3389/fonc.2023.1112276 36816981 PMC9932591