The T cell antigen coupler (TAC) receptor is a novel synthetic receptor designed to harness the therapeutic potential of T cells in the absence of tonic signaling or receptor-associated toxicities. Studies have shown that TACs provide safe and durable antitumor immunity in multiple preclinical solid tumor models, supported by growing clinical evidence. TAC receptors function by targeting cancer-associated surface antigens while recapitulating natural T cell receptor (TCR) signaling, which includes TCR/CD3 recruitment and intracellular CD4 co-receptor activity. While other receptor designs exist that redirect TCR signaling toward cancer-associated antigens, TAC technology is unique in that antigen binding is physically separated from TCR/CD3 complex recruitment. Morey et al. [1] discovered that a single amino acid alteration in the TAC domain responsible for TCR recruitment in a Claudin 18.2-directed TAC receptor resulted in enhanced in vivo function. Analysis of the biophysical properties of these receptors revealed that TAC receptors with high TCR affinity were suboptimal compared to receptor architectures with lower TCR affinity and significantly faster off-rates. This work demonstrates that balancing TCR recruitment is critical when designing potent TAC-T cell receptors, a concept that may be more broadly applicable to other therapeutic approaches relying on TCR signaling.

Research Background and Main Research Content

In recent years, adoptive cell therapy for cancer has gained significant attention due to the clinical success of engineered chimeric antigen receptor (CAR) T-cell therapies. Briefly, autologous CAR-T therapy involves ex vivo engineering of a patient's T cells to express a synthetic receptor that recognizes tumor-specific antigens, enabling these engineered T cells to specifically target cancer cells. While current FDA-approved CAR-T therapies have achieved clinical success in treating hematological cancers such as B-cell non-Hodgkin lymphoma (NHL) or acute lymphoblastic leukemia (ALL), these approaches have failed to demonstrate efficacy against solid tumors. Furthermore, CAR-T therapies frequently induce severe side effects, including cytokine release syndrome and neurotoxicity, and exhibit significantly lower antigen sensitivity compared to natural T cell receptors (TCRs), rendering CAR-T therapies suboptimal for targeting cancer antigens expressed at low levels. To overcome these limitations, recent improvements in T cell engineering have refocused on methods where engineered receptors engage and utilize the endogenous TCR, rather than bypassing the TCR complex as in CAR-T therapy.

The TAC receptor (Figure 1A) comprises three functional domains: (1) a target-specific, interchangeable extracellular antigen recognition domain, such as a Claudin (CLDN)-18.2-specific nanobody (VHH); (2) a TCR-binding domain, typically an anti-CD3ε single-chain variable fragment (scFv) like UCHT1 (Figure 1A, B), which allows the TAC receptor to engage the endogenous TCR complex; and (3) a CD4 co-receptor domain that anchors the TAC receptor in the plasma membrane and provides co-receptor function. Preclinical testing of human T cells expressing TAC receptors has demonstrated potent efficacy against both antigen-positive liquid (BCMA+ or CD19+) and solid (HER2+ or CLDN18.2+) tumors in cellular and in vivo mouse models, without the side effects associated with CAR-T therapy. Based on these preclinical results, this technology has advanced to Phase 1/2 clinical trials, where autologous T cells engineered to express a CLDN18.2-targeting TAC receptor are being investigated for unresectable, locally advanced, or metastatic CLDN18.2+ solid tumors (TAC01-CLDN18.2).

Figure 1: T cell antigen coupler (TAC) receptor and UCHT1 scFv. (A) Mechanism of action of the TAC receptor. The TAC receptor (blue) recruits the endogenous T cell receptor (TCR; grey); the complex engages CD3e via the UCHT1 scFv domain. Antigen binding occurs separately via the antigen recruitment domain, e.g., a CLDN18.2-targeting single-domain nanobody (VHH). The TAC receptor is anchored in the plasma membrane via the CD4 co-receptor domain (intracellular, transmembrane, and extracellular flexible loop). (B) Murine (Ms) and humanized (Hu) UCHT1 scFvs were designed, comprising the UCHT1 light chain (VL; amino acids 1-108) and heavy chain (VH; amino acids 1-122) linked by flexible 4× or 3× GGGGS linkers, respectively. The mutated residue (Tyr54 (Y54T)) is located within the UCHT1 heavy chain.

A unique feature of TAC technology is that TAC-mediated T cell activation can be coordinately regulated through both the antigen-binding domain (e.g., anti-CLDN18.2) and the engagement of the TCR complex via the TAC receptor's CD3e-binding UCHT1 scFv domain. This enables optimization of TAC receptors by fine-tuning TCR complex recruitment through the UCHT1 scFv domain without altering the affinity for the cell surface tumor antigen.

Figure 4: Binding kinetics analysis of UCHT1 scFv variants. (A) Binding of purified Ms- and Hu-UCHT1 scFv variants to CD3εδ analyzed by biolayer interferometry (BLI). (B-D) Binding affinity (KD; B), association rate (kon; C), and dissociation rate (koff; D) for each UCHT1 scFv variant are plotted for visual comparison.

Binding kinetics of purified soluble monomeric UCHT1 scFvs to recombinant biotinylated CD3εδ heterodimer purified from human HEK293 cells (ACRO Biosystems) were determined using biolayer interferometry (BLI). Comparison of murine and humanized UCHT1 scFv variants revealed distinct differences in binding and dissociation rates (Figure 4A). Affinity ranking was as follows: the wild-type prototype Ms-UCHT1 scFv exhibited the strongest affinity (0.25 ± 0.10 nM), followed by Ms-Y54T and Hu-WT (1.11 ± 0.29 nM and 1.81 ± 0.63 nM, respectively), and finally Hu-Y54T, which showed the weakest affinity (29.3 ± 4.0 nM). Association rate analysis (Figure C) revealed a trend toward slower association rates for both humanized variants compared to the murine variants. Dissociation rate (koff) analysis (Figure D) showed an inverse correlation with the observed KD, whereby lower affinity variants exhibited faster dissociation rates.

In summary, this study provides evidence for an affinity-mediated mechanism potentially responsible for enhanced in vivo efficacy of TAC receptors. When utilizing synthetic receptors, low to intermediate affinity binding appears preferable for T cell activation, including a study involving bispecific T cell engagers containing UCHT1 scFv variants, suggesting that optimal T cell activation may benefit from dynamic and low-affinity TCR recruitment rather than static signal integration. Therefore, this research may guide future rational design of improved TAC receptors and their clinical application, particularly by reducing TAC-TCR affinity. Endogenous TCR-mediated signaling is known to be both highly sensitive, capable of recognizing low antigen levels, and able to elicit potent T cell responses, properties stemming from low-affinity TCR-major histocompatibility complex (MHC) interactions. Consequently, the findings presented here may be more broadly applicable, where moderate TCR recruitment could promote dynamic and optimal TCR activity, offering new avenues for future cancer immunotherapy.

Tek Biotech (Tianjin) Co., Ltd. has established a comprehensive CAR-T/CAR-NK/TCR-T candidate sequence discovery platform. We are dedicated to providing high-quality targeted small molecule (Note: In this context, "small molecule" likely refers to antibody fragments like nanobodies or scFvs) antibody discovery services for scientists worldwide, as well as subsequent CAR/TCR design and construction, antibody humanization, antibody affinity maturation, cell killing assays, and animal model experiments. We also plan to develop TAC-T tumor-targeting therapeutic tools and related systems in the future, providing robust support for clients' research projects and the subsequent development of cancer therapeutics.

References

[1] Morey, T.M., Benatar, T., Xu, S.X. et al. Tuning TCR complex recruitment to the T cell antigen coupler (TAC) enhances TAC-T cell function. Sci Rep 15, 6769 (2025).