Combining T-Cell Engagers with a Tumor-Targeted IL-2 × IL-10 Immunocytokine: A Mechanistic Strategy to Address Toxicity, Potency, Objective Response, and Durability

Jacob C. McCright¹, Jeffery D. Smith¹, David Boclair¹, Pavel Khrimian¹, Christopher Yaen¹, John B. Mumm^2*

¹Deka Biosciences, Germantown, MD/USA

²1080  O'Brien Dr. Suite B, Menlo Park, CA 94025

---

Abstract

T-cell engagers (TCEs) have transformed the therapeutic landscape for hematological malignancies and are advancing rapidly into solid tumors, yet their clinical utility is constrained by four interrelated limitations: (1) cytokine release syndrome (CRS) and immune effector cell-associated neurotoxicity syndrome (ICANS), (2) suboptimal potency in cold tumors, (3) modest objective response rates outside select indications, and (4) limited durability driven by T-cell exhaustion. Recent preclinical and first-in-human data on DK2¹⁰ (EGFR), a tumor-anchored IL-2 × IL-10 immunocytokine, provides the mechanistic foundation for combining such a molecule with TCEs. IL-10 selectively abrogates IL-2- and IFN-γ-driven inflammatory cytokine cascades while sparing CD8⁺ effector activation, restores oxidative metabolism in tumor-infiltrating lymphocytes, and limits regulatory T-cell expansion. IL-2 delivered focally to the tumor synapse augments TCR-driven activation without requiring lymphodepletion. In a Phase 1 dose-escalation study, DK2¹⁰ (EGFR) showed dose-proportional IFN-γ induction without CRS-associated IL-6, TNF-α or IL-1β elevation at therapeutic doses. We argue that combining a tumor-anchored IL-2 × IL-10 immunocytokine with a TCE simultaneously addresses each of the four canonical TCE liabilities and represents a tractable, near-term clinical strategy.

---

Introduction

T-cell engagers (TCEs) are multispecific antibody constructs that physically link a tumor-associated antigen to a T-cell co-receptor, most commonly CD3, facilitating the formation of an immunological synapse between a cytotoxic T lymphocyte and a malignant cell irrespective of T-cell receptor (TCR) specificity^1,2. The clinical footprint of TCEs is anchored by blinatumomab (anti-CD19 × CD3) for B-cell acute lymphoblastic leukemia, and has expanded over the past five years to include teclistamab and elranatamab (BCMA × CD3) in multiple myeloma, glofitamab, mosunetuzumab and epcoritamab (CD20 × CD3) in B-cell non-Hodgkin lymphoma, and tarlatamab (DLL3 × CD3) in small-cell lung cancer (SCLC)^3–7. Despite this clinical momentum, four interrelated limitations constrain the patient pool and the depth of efficacy achievable with TCEs as monotherapy.

First, the very mechanism that underpins TCE potency, supraphysiological synapse formation between activated T cells and target-bearing cells, drives a well-defined pattern of toxicity dominated by cytokine release syndrome (CRS) and, in some indications, immune effector cell-associated neurotoxicity syndrome (ICANS). CRS occurs in 30–80% of patients across approved TCEs, with Grade ≥3 events in 10–20% and is the principal reason for the inpatient step-up dosing schemes that limit the outpatient feasibility of these therapies^8–10. Second, TCE-driven activation provides signal 1 (TCR ligation) without obligate co-stimulatory or cytokine support, which limits the depth and durability of T-cell effector function in the tumor microenvironment (TME). Third, objective response rates (ORRs) for TCE monotherapy in solid tumors have largely remained below the threshold required for monotherapy approvals. DLL3-directed tarlatamab in SCLC being a notable exception to this trend, reflecting the immunologically cold TME of most epithelial cancers, antigen heterogeneity, and frequent loss of MHC class I expression^11,12. Fourth, even in patients who respond, the durability of objective responses is typically constrained by the rapid onset of T-cell exhaustion under sustained CD3 engagement, characterized by upregulation of PD-1, LAG-3, TIGIT and TIM-3, metabolic dysfunction, and progressive loss of cytolytic capacity^13–15. Table 1 summarizes these four limitations, their clinical manifestations, and underlying mechanisms.

Strategies to mitigate these liabilities have included reduced-affinity CD3 binders, conditional or split TCEs, prophylactic tocilizumab, lymphodepleting chemotherapy, inclusion of costimulatory components, and concurrent checkpoint inhibition^16–18. None has yet fully decoupled TCE potency from CRS, restored durable cytotoxic T-cell function in cold tumors, or expanded the addressable solid-tumor population. We argue that a complementary strategy of pharmacologically supplying the cytokine signals that the synapse requires for sustained, productive cytolysis while simultaneously suppressing the inflammatory cascade that drives CRS addresses all four limitations through a single coordinated intervention. Recent preclinical and first-in-human data on DK2¹⁰ (EGFR), a tumor-anchored IL-2 × IL-10 immunocytokine developed by Deka Biosciences, provides the empirical foundation for this approach^19,20. We synthesize these data and outline the mechanistic rationale and translational considerations for a TCE plus tumor-anchored IL-2 × IL-10 combination.

Table 1: Principal clinical limitations of T-cell engager monotherapy. ORR, objective response rate; CRS, cytokine release syndrome; ICANS, immune effector cell-associated neurotoxicity syndrome; TCR, T-cell receptor; TME, tumor microenvironment; SCLC, small-cell lung cancer.

Limitation	Clinical manifestation	Underlying mechanism	Representative agents (Class)
Toxicity (CRS / ICANS)	Fever, hypotension, hypoxia; CRS in 30–80% of patients; Grade ≥3 in 10–20%; ICANS in subset	Massive systemic IL-6, IFN-γ, TNF-α release upon T-cell synapse formation; macrophage hyperactivation	Blinatumomab (CD19 × CD3); teclistamab (BCMA × CD3); tarlatamab (DLL3 × CD3)
Suboptimal potency	Variable depth of response; partial responses common; minimal residual disease persistence	T-cell engagers provide signal 1 (TCR) without obligate IL-2 co-stimulation in tumor microenvironment	Glofitamab (CD20 × CD3); epcoritamab (CD20 × CD3); mosunetuzumab (CD20 × CD3)
Limited objective response in solid tumors	ORR < 30% in most solid-tumor TCE programs outside select indications (e.g., DLL3+ SCLC)	Poor T-cell infiltration; immunosuppressive TME; antigen heterogeneity; loss of MHC-I	Tarlatamab (SCLC); cibisatamab/CEA-TCB (CEA × CD3, investigational)
T-cell exhaustion / loss of durability	Initial response followed by progression within months; rebound disease after treatment-free intervals	Sustained TCR engagement → PD-1/LAG-3/TIGIT upregulation; metabolic dysfunction; effector decay	Multiple TCE programs in heme malignancies and solid tumors

The mechanistic limits of TCE monotherapy

CRS and ICANS toxicity proportional to potency

TCE-induced CRS is the prototypical example of mechanism-driven toxicity in immuno-oncology. Synapse formation between an activated T cell and a target-bearing cell triggers IFN-γ release, which licenses macrophages and monocytes to produce IL-6, IL-1β and TNF-α; this in turn drives the constitutional symptoms (fever, hypotension, capillary leak) that define CRS^8,21. The clinical consequence is that potency and toxicity scale together. Step-up dosing, prophylactic corticosteroids, and tocilizumab attenuate but do not abolish the syndrome; they also blunt the immune activation that mediates anti-tumor activity. ICANS, occurring in a meaningful subset of CD19- and BCMA-directed TCE recipients, adds a separate neurotoxicity ceiling that further constrains dose escalation²².

Suboptimal cytotoxic function in cold tumors

Sustained T-cell cytolysis requires more than TCR ligation. IL-2—predominantly through engagement of the high-affinity heterotrimeric IL-2Rαβγ—drives clonal expansion, upregulation of cytotoxic granule machinery (granzyme B, perforin), and IFN-γ production²³. In the TME of most solid tumors, however, IL-2 is scarce, and the IL-2 that is produced is often outcompeted by Treg consumption. TCEs deliver signal 1 without restoring this missing cytokine context, which limits both the depth of the initial cytolytic burst and the proliferative response of recruited effectors. Solid-tumor TCE programs are over-represented among combinations with IL-2 muteins, lymphodepleting chemotherapy, or checkpoint inhibition for precisely this reason^16,18.

Objective response rates and the cold-tumor ceiling

Outside well-defined hematological niches and the DLL3+ SCLC indication, ORRs for TCE monotherapy in solid tumors have generally remained modest, with stable disease often the most frequent best response. Several biological features of the solid-tumor TME contribute to this; poor T-cell infiltration, antigen heterogeneity within the tumor, low surface antigen density on individual cells, MHC-I loss in subsets of tumors, and the immunosuppressive infiltrate of myeloid-derived suppressor cells, tumor-associated macrophages, and Tregs^11,12,24. None of these obstacles can be addressed by improvements to the TCE molecule itself.

Durability constrained by exhaustion

Continuous CD3 engagement the central design feature of TCE pharmacology, is also the proximate driver of T-cell exhaustion, the progressive loss of effector function under chronic antigenic stimulation. Exhausted T cells upregulate inhibitory receptors (PD-1, LAG-3, TIGIT, TIM-3), exhibit metabolic switching toward glycolysis with mitochondrial dysfunction, and produce diminished IFN-γ and granzyme B^13,15. In TCE-treated patients who experience an initial response, this exhaustion program is a principal driver of the relapse trajectory, and treatment-free intervals do not appear to reliably restore effector capacity once it has decayed.

Limitations of current combination approaches

Strategies to address these limitations have included reduced-affinity CD3 binders, co-stimulatory trispecific formats incorporating CD28 or 4-1BB as a second T-cell signal, and concurrent checkpoint inhibition^16–18. Although CD28-incorporating trispecifics provide a mechanistically sound approach to deepening T-cell activation, early clinical data reveal a corresponding safety liability: systemic or amplified CD28 agonism carries a well-documented risk of cytokine storm, as seen both with historical anti-CD28 superagonists and with tumor-targeted CD28 bispecifics combined with PD-1 blockade^16,17. None of the current combinatorial strategies has yet fully decoupled TCE potency from CRS, restored durable cytotoxic function in cold tumors, or expanded the addressable solid-tumor population.

DK2¹⁰ (EGFR): a tumor-anchored IL-2 × IL-10 immunocytokine

DK2¹⁰ (EGFR) is a tri-specific fusion protein in which wild-type human IL-2 is coupled to a high-affinity Epstein–Barr virus-derived IL-10 mutein via an EGFR-targeting single-chain variable fragment (scFv) scaffold. The EGFR-binding domain anchors the molecule on EGFR-expressing tumor cells, concentrating IL-2 and IL-10 signaling in the TME for several days following subcutaneous administration¹⁹. The molecule was rationally designed on the hypothesis that the inflammatory toxicity of IL-2 is driven by IL-2-induced IFN-γ and TNF-α, which IL-10 selectively suppresses without abolishing IL-2-mediated effector activation. DK2¹⁰ (EGFR) thus belongs to the broader class of immunocytokines and tumor-targeted cytokine-delivery agents — encompassing antibody–cytokine fusions, engineered IL-2 variants, and PD-1-targeted IL-2 — but is distinguished from prior single-cytokine designs by its dual IL-2 × IL-10 architecture, which couples an effector-promoting cytokine with an inflammation-restraining one in a single tumor-anchored molecule^30,31.

In primary human PBMC cultures, addition of IL-10 to IL-2 abolished IL-6, IL-1β and TNF-α elevation across donors while preserving IFN-γ induction characteristic of cytotoxic T-cell and NK-cell activation^{19, Fig 1}. In syngeneic murine models, daily systemic administration of IL-2 produced predictable vascular leak (Evans blue extravasation into lung), splenic Treg expansion and serum cytokine storm; coadministration of IL-10 fully reversed the vascular leak phenotype and prevented Treg accumulation while preserving GzmB⁺/Perforin⁺ NK-cell expansion and CD8⁺ effector memory T-cell accumulation^{19, Fig 2}. Genetic and pharmacological evaluation localized IL-2 toxicity to an IFN-γ- and TNF-α-dependent axis: IFN-γ-knockout, TNF-α-knockout, and anti-TNF-α-treated wild-type animals were all protected from IL-2 vascular leak ^{19, Fig 3}.

When wild-type IL-2 was structurally combined with an EBV-derived IL-10 mutein in the EGFR-anchored DK2¹⁰ scaffold, the molecule produced in-vitro tumor cytolysis comparable to that induced by equivalent doses of free IL-2 alone, whereas the unconjugated IL-2 × IL-10 combination produced more rapid and deeper cytolysis^{19, Fig 5G}. In a B16F10-hEGFR syngeneic model, DK2¹⁰ (EGFR) at 2 mg/kg outperformed the untargeted IL-2 + IL-10 combination, demonstrated CD8⁺, CD4⁺ and NK-cell-dependent activity, lost antitumor efficacy in IFN-γ-knockout hosts, indicating that IFN-γ is required to mediate its antitumor activity, and synergized with anti-PD-1 in the LL2-hEGFR model; in a PD-1-resistant patient-derived NSCLC xenograft, DK2¹⁰ (EGFR) monotherapy showed greater antitumor activity than anti-PD-1 monotherapy, although combination synergy was not tested in that model^{19, Fig 6}. In a 28-day GLP toxicology study in cynomolgus monkeys, DK2¹⁰ (EGFR) at doses up to 2.5 mg/kg subcutaneously thrice weekly produced expected wt-IL-2 pharmacodynamics of eosinophilia, IL-2Rα and IL-18 elevation, dose-proportional IFN-γ induction, however, with no clinically significant elevation of IL-6, TNF-α or IL-1β and no organ toxicity findings^{19, Fig 7}. Critically, the CTL:Treg ratio shifted toward effector dominance rather than the Treg-skewed expansion characteristic of recombinant IL-2.

First-in-human data from the DEKA-1 Phase 1 study (NCT05704985) corroborates the preclinical profile²⁰. Across dose levels of 2–16 mg administered subcutaneously thrice weekly, treatment-related adverse events of any grade were predominantly Grade 1–2 injection-site reactions (63%), fever (40%) and fatigue (31%); CRS occurred in 2.6% (a single Grade 2 event at 4 mg with elevated IL-6), with no Grade ≥3 CRS and no dose-limiting toxicities. IFN-γ induction was dose-proportional through 8 mg, reaching plasma concentrations up to ~200-fold over baseline (median 417 pg/mL), accompanied by IL-5-driven eosinophilia, IL-2Rα shedding, and T-cell and NK-cell expansion. Crucially, no concomitant elevation of IL-6, TNF-α or IL-1β was detected at therapeutic doses other than the single 4-mg event noted above. Of evaluable patients, approximately one-third had a best response of stable disease or mixed responses, with concomitant induction of IL-18, IL-18BP, soluble PD-L1, LAG-3 and TIGIT, indicating broad immune engagement^20,40. Together, these data establish the pharmacological profile of DK2¹⁰ (EGFR) and provide the empirical foundation for its combination with TCEs.

Figure 1: Mechanistic rationale for combining a T-cell engager (TCE) with the EGFR-anchored IL-2 × IL-10 immunocytokine DK2¹⁰ (EGFR). (A) TCE monotherapy forces a synapse between the tumor cell and a CD3⁺ T cell but provides no exogenous cytokine support; the resulting IFN-γ-driven macrophage activation produces IL-6, TNF-α and IL-1β that mediate cytokine release syndrome (CRS), while the absence of IL-2 in the TME limits potency, depth of response, and durability. (B) DK2¹⁰ (EGFR) is anchored on the EGFR⁺ tumor cell via its scFv arm, delivering wild-type IL-2 directly into the synapse to augment TCE-induced TCR signaling, while the EBV IL-10 mutein selectively suppresses IL-2- and IFN-γ-driven inflammatory cytokines, attenuating CRS without abolishing effector function. IL-10 additionally restores CD8+ T-cell metabolic fitness and reverses exhaustion-associated transcriptional programs. In both panels, tumor-associated macrophages are depicted as the proximate source of the IL-6, TNF-α and IL-1β that drive CRS: in (A) they are licensed by unopposed IFN-γ, whereas in (B) locally released IL-10 restrains their cytokine output while T- and NK-cell effector function is preserved.

Addressing CRS IL-10 as a selective inflammatory brake

The CRS-mitigating mechanism of IL-10 is selective rather than broadly immunosuppressive. IL-10 acts predominantly on monocytes and macrophages to suppress IRF1- and IRF8-mediated transcription of IL-6, TNF-α and IL-1β^{19, Figs 3,5}. Importantly, IL-10 does not block IFN-γ production by activated T cells or NK cells; rather, it interrupts the IFN-γ-to-macrophage axis that converts T-cell activation into systemic CRS. Because the EGFR-anchoring arm engages tumor cells rather than myeloid cells, DK2¹⁰ (EGFR) is not expected to deliver cytokine directly to macrophages; its influence on macrophage cytokine output is therefore indirect. Locally concentrated IL-10 diffusing from the anchored molecule acts on tumor-associated monocytes and macrophages to restrain IRF1/IRF8-dependent transcription of IL-6, TNF-α and IL-1β, while the preserved IFN-γ signal no longer translates into unchecked myeloid licensing. Because activated macrophages are the proximate source of the IL-6, TNF-α and IL-1β that mediate CRS, this restraint of the macrophage cytokine program is mechanistically positioned to limit toxicity; and because tumor-associated macrophages also help establish the immunosuppressive niche, the same modulation may additionally contribute to antitumor efficacy. This selectivity is the basis of the most therapeutically relevant property of DK2¹⁰ (EGFR): in the DEKA-1 study, IFN-γ rose 200-fold over baseline at therapeutic doses while IL-6, TNF-α and IL-1β remained essentially flat^20,40. Layered onto a TCE, this brake on the macrophage cytokine cascade is mechanistically positioned to attenuate the principal driver of TCE-induced CRS without compromising the effector function the TCE is designed to recruit.

Restoring potency IL-2 in the synapse

The pharmacological insight underpinning DK2¹⁰ (EGFR) that high-dose, locally retained IL-2 can be tolerated when paired with IL-10 is directly applicable to the TCE potency problem. EGFR is broadly expressed across epithelial tumors, and DK2¹⁰ (EGFR) accumulates and is retained on EGFR⁺ tumor cells for several days after dosing^19,25. A TCE that creates a forced synapse between a tumor cell and a T cell will, in a DK2¹⁰-treated TME, find that synapse already enriched for IL-2. IL-2 binding to IL-2Rαβγ on the synapsing T cell amplifies TCR-driven signal 1 with a strong cytokine signal 3, producing the depth of cytolytic activation that TCEs alone do not deliver in cold tumors^23,26. This is conceptually distinct from systemic IL-2 strategies (e.g., aldesleukin, NKTR-214) that flood the periphery: in a DK2¹⁰ + TCE combination, IL-2 is concentrated where the TCE is functioning.

Improving objective response broadening the addressable solid-tumor population

Two of the dominant constraints on TCE ORR in solid tumors, poor effector infiltration and Treg-mediated suppression, are directly addressed by the DK2¹⁰ mechanism. Preclinical data demonstrate that DK2¹⁰ (EGFR) drives CD8⁺ effector memory T-cell and NK-cell accumulation without the Treg expansion characteristic of unmodified IL-2^{19, Figs 2,7}. This shift in the CTL:Treg ratio toward effector dominance is a key point of differentiation from conventional high-dose IL-2 therapies (e.g., aldesleukin), whose preferential engagement of the high-affinity IL-2 receptor on CD25-high regulatory T cells drives Treg expansion and counteracts the intended effector response; the IL-2 × IL-10 architecture instead biases the balance toward cytotoxic effectors. Clinically, DEKA-1 reported expansion of ~40–350 unique peripheral T-cell clones beginning around Day 5, with changes in the peripheral repertoire correlating with the precision patient-selection assay^40,41, and absence of preferential Treg accumulation at therapeutic doses²⁰. Combined with a TCE that brings these expanded effectors into productive contact with tumor-associated antigen, the combination is mechanistically positioned to convert stable-disease responses into objective tumor regression. EGFR co-targeting with the scFv arm of DK2¹⁰ (EGFR) provides a degree of solid-tumor selectivity restricting the TCE potentiation to EGFR+ TME that may also support tolerability.

Extending durability IL-10, exhaustion, and metabolic fitness

IL-10 has a paradoxical relationship with anti-tumor immunity that has been progressively clarified over the past decade. While long classified as an immunosuppressive cytokine, work from multiple laboratories has established that IL-10 sustains the cytolytic capacity and metabolic fitness of tumor-infiltrating CD8⁺ T cells, suppresses exhaustion-associated transcriptional programs, and enhances the oxidative metabolism that distinguishes long-lived effector and memory populations from terminally exhausted cells^{27,28,37–39}. Mechanistically, this effect is cell-intrinsic: IL-10 engagement of the IL-10 receptor on CD8⁺ T cells signals through STAT3 to upregulate mitochondrial oxidative phosphorylation, restoring the metabolic capacity that terminally exhausted cells progressively lose. It is this metabolic reprogramming—rather than relief of an inhibitory checkpoint—that reinvigorates the proliferative and cytolytic function of exhausted effectors and links sustained IL-10 exposure to prolonged effector activity^37–39. In the DK2¹⁰ (EGFR) preclinical work, sorted CD8⁺ tumor-infiltrating lymphocytes from treated mice exhibited downregulation of exhaustion markers on nCounter analysis and elevated oxygen consumption rates compared with vehicle controls, with corresponding increases in tumor-reactive IFN-γ-producing cells on ELIspot^{19, Figs 6N–P}. In a TCE combination, this exhaustion-attenuating effect of IL-10 is mechanistically positioned to extend the duration of objective response by preventing the rapid functional decay that limits TCE durability in monotherapy use.

Table 2: Mechanistic rationale and supporting evidence for combining a T-cell engager (TCE) with DK210 (EGFR), an EGFR-anchored IL-2 × IL-10 immunocytokine. Evidence is drawn from preclinical studies in primary human PBMCs, syngeneic murine tumor models and cynomolgus monkey GLP toxicology, and from the DEKA-1 Phase 1 first-in-human dose-escalation study (NCT05704985).

TCE limitation	DK210-mediated mechanism	Supporting preclinical / clinical evidence
CRS	EBV-IL-10 mutein within DK210 suppresses IL-2- and IFN-γ-driven inflammatory cytokine release; ablates IL-6, TNF-α, IL-1β induction by acting on tumor-associated monocytes/macrophages, the proximate producers of these CRS cytokines	PBMC studies: pairwise reduction of IL-6, TNF-α with IL-2 + IL-10 vs IL-2 alone (p ≤ 0.001) [Ahn 2025, Fig 1]; NHP GLP study with no CRS-associated cytokine elevations [Ahn 2025, Fig 7]; DEKA-1 clinical: only 1/35 evaluable patients had Grade 2 CRS at 4 mg [Spira 2025]
Suboptimal potency	Wild-type IL-2 in DK210 delivers a sustained co-stimulatory signal at the tumor synapse, anchored to EGFR via the scFv scaffold	B16F10-hEGFR model: DK210 (EGFR) outperformed untargeted IL-2 + IL-10 combination at equivalent doses [Ahn 2025, Fig 6]; clinical IFN-γ induction up to 200-fold over baseline at 8 mg [Fu 2024]
VLS / vascular leak	IL-10 abolishes IL-2-driven vascular leak (Evans blue extravasation) in WT mice; effect is IFN-γ- and TNF-α-dependent	Quantitative Evans-blue assay in WT, IFN-γ-KO, TNF-α-KO mice and anti-TNF-α treated WT mice [Ahn 2025, Fig 3]
Treg expansion (immune-suppressive)	DK210 retains effector T cell and NK cell activation without preferential expansion of CD25+/Foxp3+ Tregs	Murine spleen flow cytometry showing GzmB+/Perforin+ NK and CD8+ Tem expansion without Treg accumulation [Ahn 2025, Fig 2]; NHP CTL:Treg ratio shifted toward effector dominance [Ahn 2025, Fig 7]
T-cell exhaustion	IL-10 attenuates exhaustion-associated transcriptional programs in tumor-infiltrating CD8+ T cells; preserves oxidative metabolism	Sorted TIL nCounter analysis showed downregulation of exhaustion markers; OCR measurements demonstrated metabolic fitness [Ahn 2025, Fig 6N–6P]
Durability	Repeated cycles of in-vitro tumor cytolysis sustained by DK210 even after drug removal; in-vivo combination with anti-PD-1 amplifies and extends response	5-cycle SK-BR-3 cytolysis assay [Ahn 2025, Fig 5]; LL2-hEGFR ] + anti-PD-1 model [Ahn 2025, Fig 6Q]; NSCLC PDX [Ahn 2025, Fig 6R]

Translational considerations and open questions

Several considerations deserve explicit attention as a TCE + tumor-anchored IL-2 × IL-10 combination moves toward clinical evaluation. First, sequencing and timing matter. In principle, DK2¹⁰ (EGFR) pretreatment could expand and prime the effector pool prior to TCE introduction, but a contemporaneous schedule that ensures both molecules are present in the TME during the cytolytic phase may be required to fully realize synergy. The thrice-weekly subcutaneous dosing of DK2¹⁰ (EGFR) in DEKA-1²⁰ is operationally compatible with most TCE schedules, including continuous infusion blinatumomab and step-up SC regimens for BCMA- and CD20-directed TCEs. As a practical starting point, the thrice-weekly subcutaneous schedule could be initiated as a short priming lead-in to expand and metabolically condition the effector pool, then continued contemporaneously through the TCE step-up and maintenance phases so that IL-2 and IL-10 are present in the tumor during the period of maximal CD3-driven cytolysis. Because DK2¹⁰ (EGFR) is administered subcutaneously and is not myelosuppressive, it does not require lymphodepletion and can be layered onto existing step-up regimens without adding cytopenia risk; sequencing studies comparing priming versus fully concurrent administration will be needed to define the optimal interval. Computational modeling—of immune-cell dynamics, of the hemodynamic forces and interstitial pressures that govern intratumoral drug distribution, and of multiphysics tumor-microenvironment normalization—offers a complementary, quantitative route to rational trial design and to optimizing the dose and schedule of both agents^34–36. Second, antigen-pair selection requires care. EGFR co-expression in the target indication allows DK2¹⁰ (EGFR) to anchor where the TCE engages a particularly attractive scenario in EGFR-expressing carcinomas (head and neck, colorectal, NSCLC, RCC, PDAC). The most readily addressable subgroups are likely tumors with high, relatively homogeneous EGFR expression—for example EGFR-amplified or EGFR-overexpressing head-and-neck, colorectal and squamous non-small-cell lung carcinomas—that also retain an inflamed or treatment-responsive immune profile, since both high antigen density and the presence of recruitable effector T cells should favor activity, whereas EGFR-low or immunologically excluded tumors are expected to respond poorly. EGFR is not tumor-restricted, however: it is constitutively expressed in normal epithelia, notably basal keratinocytes of the skin and the gastrointestinal mucosa, which underlies the dermatologic and GI on-target toxicities of systemic anti-EGFR agents. For DK2¹⁰ (EGFR) this risk is mitigated by tumor anchoring and local subcutaneous dosing, which favor accumulation in EGFR-high tumor tissue over sustained exposure of normal epithelium; nonetheless, the potential for on-target/off-tumor cytokine delivery to EGFR-expressing normal tissue should be monitored, particularly alongside a TCE that independently amplifies local T-cell activation. For TCEs targeting non-epithelial tumors or tumors without EGFR co-expression, alternative scFv arms in the Diakine^® platform (e.g., directed at distinct tumor antigens) would be required to maintain spatial co-localization²⁵.

Third, Treg dynamics must be monitored. IL-10 has been shown preclinically and clinically to suppress Treg expansion³³. Similarly, the DK2¹⁰ (EGFR) data both preclinical and clinical demonstrate that the EGFR-anchored IL-2 × IL-10 format does not preferentially expand Foxp3⁺/CD25⁺ Tregs and instead shifts the CTL:Treg ratio in favor of effectors^19,20. Whether this preserves Treg homeostasis on a TCE background, where the synapse-driven activation may itself bias toward effector dominance, will require dedicated immune-monitoring in clinical combinations. Fourth, ICANS represents a separate hazard from CRS, and while the DK2¹⁰ mechanism plausibly addresses the underlying IL-6/TNF-α-driven blood–brain barrier dysfunction, the trial design should include neurological assessments adequate to detect attenuation of ICANS frequency and severity. Importantly, because ICANS is driven substantially by IL-6- and TNF-α-mediated endothelial activation and blood–brain-barrier disruption rather than by systemic CRS per se, IL-10-mediated suppression of these cytokines could in principle reduce ICANS even in the absence of—or out of proportion to—any reduction in CRS. This predicts a degree of neuroprotection that is mechanistically independent of CRS control, and the trial design should therefore assess ICANS and CRS endpoints separately rather than treating neurotoxicity solely as a downstream correlate of cytokine-release severity. Fifth, the durability hypothesis that IL-10-mediated exhaustion attenuation will extend the time to relapse on a TCE backbone is the most clinically meaningful but also the slowest endpoint to measure; trial designs should include adequately long follow-up and pre-specified exploratory analyses on TIL exhaustion phenotype, peripheral T-cell receptor repertoire diversity, and metabolic markers of T-cell fitness. In practical terms, a standardized monitoring panel might pair serial serum cytokines (IL-6, TNF-α, IL-1β and IFN-γ) and soluble immune markers (soluble PD-L1, IL-18, IL-18BP) for safety and pharmacodynamics with flow-cytometric quantification of the peripheral and, where feasible, intratumoral CTL:Treg ratio, TIL exhaustion-marker phenotyping (PD-1, LAG-3, TIGIT, TIM-3), T-cell-receptor repertoire sequencing, and ex-vivo measures of T-cell metabolic fitness such as mitochondrial oxygen-consumption rate.

Finally, an honest accounting of the limits of the available data: the published preclinical and clinical evidence summarized above is for DK2¹⁰ (EGFR) monotherapy or in combination with anti-PD-1, not yet for direct TCE combination. The mechanistic case is strong; the empirical case for the combination remains to be established and is the appropriate next clinical question. As discussed in Section 2.5, co-stimulatory CD28/4-1BB trispecific TCEs pursue similar goals but carry their own inflammatory liabilities^16,17, reinforcing the need for an approach that enhances T-cell function without compounding inflammatory risk.

Conclusions

T-cell engagers have established themselves as a transformative class in hematological oncology, but the four canonical limitations (1) CRS/ICANS toxicity, (2) suboptimal potency, (3) modest objective response rates in solid tumors, and (4) durability constrained by exhaustion remain barriers to broader clinical impact. Efforts to address these through co-stimulatory TCE formats have demonstrated mechanistic promise but have not yet resolved the underlying tension between enhanced T-cell activation and amplified inflammatory toxicity^16,17. Combining a TCE with a tumor-anchored IL-2 × IL-10 immunocytokine such as DK2¹⁰ (EGFR) addresses each of these limitations through a coordinated mechanistic intervention: IL-10 selectively brakes the macrophage cytokine cascade that drives CRS while sparing effector activation; IL-2 delivered into the tumor synapse augments TCE-induced TCR signaling to deepen response; the targeting strategy concentrates immune potentiation in the TME while limiting peripheral Treg expansion; and IL-10's effect on T-cell metabolic fitness has the potential to extend the durability of objective response. With Phase 1 monotherapy safety and pharmacodynamic data now available for DK2¹⁰ (EGFR) and a defined population of TCE programs in advanced solid tumors, the time is right to test this combination prospectively. We believe the available preclinical and first-in-human data justifies rapid clinical translation. Critically, the elements required for near-term translation are already in hand: DK2¹⁰ (EGFR) has completed Phase 1 dose escalation in the DEKA-1 study with an established subcutaneous schedule, a defined dose range, and a first-in-human profile showing dose-proportional IFN-γ induction without dose-limiting CRS²⁰. Combined with the clinical and regulatory maturity of approved TCE backbones, these data support proceeding directly to a biomarker-embedded Phase 1b combination study rather than awaiting additional monotherapy results. The optimal biologic dose of DK2¹⁰ (EGFR) is estimated to be 6 – 8 mgs/dose from this study. These doses achieve a sustained 5 – 10 ng/mL in systemic circulation of 24 – 48 hrs., consistent with dose levels likely necessary for both suppression of CRS and activation of T cells.

Declarations

Funding

This work received no specific external funding.

Conflicts of interest

Each author should declare relevant conflicts. Example: Author 1 declares no conflicts of interest. Author 2 reports consulting fees from [Company]. Author 3 holds equity in [Company] and has served on the advisory board of [Company].] Author 6 declares no conflicts of interest and does not hold equity in the company.

Ethics and consent statement

This article is a Mini Review and does not contain any new studies with human participants or animals performed by the authors. All cited human and animal data are reproduced from published sources [19,20] in which the originating authors confirmed appropriate institutional ethics review and informed consent.

Data availability statement

Data sharing is not applicable to this Mini Review as no new datasets were generated. The clinical and preclinical data discussed herein are available in the original publications cited [19,20] and at ClinicalTrials.gov (NCT05704985).

Author contributions

Authors 1 and 6 conceptualized the review and drafted the manuscript. Author 2 contributed to literature synthesis and figure preparation. Author 3-5 critically revised the manuscript for important intellectual content. All authors read and approved of the final manuscript.

References

1. Goebeler ME, Bargou RC. T cell-engaging therapies BiTEs and beyond. Nat Rev Clin Oncol. 2020; 17(7): 418–34. doi: 10.1038/s41571-020-0347-5.
2. Klinger M, Benjamin J, Kischel R, et al. Harnessing T cells to fight cancer with BiTE® antibody constructs past developments and future directions. Immunol Rev. 2016; 270(1): 193–208. doi: 10.1111/imr.12393.
3. Kantarjian H, Stein A, Gökbuget N, et al. Blinatumomab versus chemotherapy for advanced acute lymphoblastic leukemia. N Engl J Med. 2017; 376(9): 836–47. doi: 10.1056/NEJMoa1609783.
4. Moreau P, Garfall AL, van de Donk NWCJ, et al. Teclistamab in relapsed or refractory multiple myeloma. N Engl J Med. 2022; 387(6): 495–505. doi: 10.1056/NEJMoa2203478.
5. Bannerji R, Arnason JE, Advani RH, et al. Odronextamab, a human CD20 × CD3 bispecific antibody, in patients with relapsed/refractory B-cell non-Hodgkin lymphoma. Lancet Haematol. 2022; 9(5): e327–e339. doi: 10.1016/S2352-3026(22)00072-2.
6. Dickinson MJ, Carlo-Stella C, Morschhauser F, et al. Glofitamab for relapsed or refractory diffuse large B-cell lymphoma. N Engl J Med. 2022; 387(24): 2220–31. doi: 10.1056/NEJMoa2206913.
7. Ahn MJ, Cho BC, Felip E, et al. Tarlatamab for patients with previously treated small-cell lung cancer. N Engl J Med. 2023; 389(22): 2063–75. doi: 10.1056/NEJMoa2307980.
8. Shimabukuro-Vornhagen A, Gödel P, Subklewe M, et al. Cytokine release syndrome. J Immunother Cancer. 2018; 6(1): 56. doi: 10.1186/s40425-018-0343-9.
9. Lee DW, Santomasso BD, Locke FL, et al. ASTCT consensus grading for cytokine release syndrome and neurologic toxicity. Biol Blood Marrow Transplant. 2019; 25(4): 625–38. doi: 10.1016/j.bbmt.2018.12.758.
10. Crombie JL, Graff T, Falchi L, et al. Consensus recommendations on the management of toxicity associated with CD3 × CD20 bispecific antibody therapy. Blood. 2024; 143(16): 1565–75. doi: 10.1182/blood.2023022432.
11. Middelburg J, Kemper K, Engelberts P, et al. Overcoming challenges for CD3-bispecific antibody therapy in solid tumors. Cancers (Basel). 2021; 13(2): 287. doi: 10.3390/cancers13020287.
12. Tapia-Galisteo A, Compte M, Álvarez-Vallina L, et al. When three is not a crowd: trispecific antibodies for enhanced cancer immunotherapy. Theranostics. 2023; 13(3): 1028–41. doi: 10.7150/thno.81494.
13. Wherry EJ, Kurachi M. Molecular and cellular insights into T cell exhaustion. Nat Rev Immunol. 2015; 15(8): 486–99. doi: 10.1038/nri3862.
14. Beltra JC, Manne S, Abdel-Hakeem MS, et al. Developmental relationships of four exhausted CD8+ T cell subsets reveals underlying transcriptional and epigenetic landscape control mechanisms. Immunity. 2020; 52(5): 825–841.e8. doi: 10.1016/j.immuni.2020.04.014.
15. Philip M, Schietinger A. CD8+ T cell differentiation and dysfunction in cancer. Nat Rev Immunol. 2022; 22(4): 209–23. doi: 10.1038/s41577-021-00574-3.
16. Stein A, Franklin JL, Chia VM, et al. Benefit-risk assessment of blinatumomab in the treatment of relapsed/refractory B-cell precursor acute lymphoblastic leukemia. Drug Saf. 2019; 42(5): 587–601. doi: 10.1007/s40264-018-0760-1.
17. Trinklein ND, Pham D, Schellenberger U, et al. Efficient tumor killing and minimal cytokine release with novel T-cell agonist bispecific antibodies. MAbs. 2019; 11(4): 639–52. doi: 10.1080/19420862.2019.1574521.
18. Schoenfeld AJ, Lee SM, Doger de Speville B, et al. Lifileucel, an autologous tumor-infiltrating lymphocyte monotherapy, in patients with advanced non-small cell lung cancer. Cancer Discov. 2024; 14(8): 1389–402. doi: 10.1158/2159-8290.CD-23-1334.
19. Ahn JJ, Dudics S, Langan DP, et al. Coupling IL-2 with IL-10 to mitigate toxicity and enhance antitumor immunity. Cell Rep Med. 2025; 6(8): 102257. doi: 10.1016/j.xcrm.2025.102257.
20. Spira AI, Tolcher AW, Carneiro BA, et al. Phase 1 study of DK210 (EGFR), a tumor-targeted IL-2 × IL-10 dual immunocytokine, in advanced cancer patients: dose escalation, immune activation, and safety results. J Clin Oncol. 2025; 43(16 Suppl): 2589. doi: 10.1200/JCO.2025.43.16_suppl.2589.
21. Teachey DT, Lacey SF, Shaw PA, et al. Identification of predictive biomarkers for cytokine release syndrome after chimeric antigen receptor T-cell therapy for acute lymphoblastic leukemia. Cancer Discov. 2016; 6(6): 664–79. doi: 10.1158/2159-8290.CD-16-0040.
22. Santomasso BD, Park JH, Salloum D, et al. Clinical and biological correlates of neurotoxicity associated with CAR T-cell therapy in patients with B-cell acute lymphoblastic leukemia. Cancer Discov. 2018; 8(8): 958–71. doi: 10.1158/2159-8290.CD-17-1319.
23. Boyman O, Sprent J. The role of interleukin-2 during homeostasis and activation of the immune system. Nat Rev Immunol. 2012; 12(3): 180–90. doi: 10.1038/nri3156.
24. Garrido F, Aptsiauri N, Doorduijn EM, et al. The urgent need to recover MHC class I in cancers for effective immunotherapy. Curr Opin Immunol. 2016; 39: 44–51. doi: 10.1016/j.coi.2015.12.007.
25. Deka Biosciences. Diakine platform mechanism of action DK210 (EGFR). Available from: https://www.dekabiosciences.com/moa/ [Accessed 2026].
26. Liao W, Lin JX, Leonard WJ. Interleukin-2 at the crossroads of effector responses, tolerance, and immunotherapy. Immunity. 2013; 38(1): 13–25. doi: 10.1016/j.immuni.2013.01.004.
27. Mumm JB, Emmerich J, Zhang X, et al. IL-10 elicits IFN-γ-dependent tumor immune surveillance. Cancer Cell. 2011; 20(6): 781–96. doi: 10.1016/j.ccr.2011.11.003.
28. Naing A, Wong DJ, Infante JR, et al. Pegilodecakin combined with pembrolizumab or nivolumab for patients with advanced solid tumors. Lancet Oncol. 2019; 20(11): 1544–55. doi: 10.1016/S1470-2045(19)30514-5.
29. Hashimoto M, Araki K, Cardenas MA, et al. PD-1 combination therapy with IL-2 modifies CD8+ T cell exhaustion program. Nature. 2022; 610(7930): 173–81. doi: 10.1038/s41586-022-05257-0.
30. Charych DH, Hoch U, Langowski JL, et al. NKTR-214, an engineered cytokine with biased IL2 receptor binding, increased tumor exposure, and marked efficacy in mouse tumor models. Clin Cancer Res. 2016; 22(3): 680–90. doi: 10.1158/1078-0432.CCR-15-1631.
31. Diab A, Tannir NM, Bentebibel SE, et al. Bempegaldesleukin (NKTR-214) plus nivolumab in patients with advanced solid tumors: phase I dose-escalation study of safety, efficacy, and immune activation (PIVOT-02). Cancer Discov. 2020; 10(8): 1158–73. doi: 10.1158/2159-8290.CD-19-1510.
32. U.S. National Library of Medicine. ClinicalTrials.gov. Phase 1 study of DK210 (EGFR) in patients with advanced solid cancer (DEKA-1). Identifier NCT05704985. Available from: https://clinicaltrials.gov/study/NCT05704985.
33. Chan IH, Wu V, Bilardello M, et al. PEG-rIL-10 treatment decreases FoxP3+ Tregs despite upregulation of intratumoral IDO. Oncoimmunology. 2016; 5(7): e1197458. doi: 10.1080/2162402X.2016.1197458.
34. Mirala F, Moradi Kashkooli F, Soltani M. Microvascular and stromal normalization combined with nano-immunotherapy: a synergistic strategy to overcome drug delivery barriers in solid tumors. Microvascular Research. 2026; 166: 104955. doi: 10.1016/j.mvr.2026.104955.
35. Razavi Z, Soltani M, Souri M, et al. CRISPR-driven biosensors: a new frontier in rapid and accurate disease detection. Crit Rev Anal Chem. 2024. doi: 10.1080/10408347.2024.2400267.
36. Mirala F, Soltani M. A mechanical reprogramming: synergistic tumor microenvironment normalization improves nano-immunotherapy delivery in prostate cancer via multiphysics modeling. Drug Deliv Transl Res. 2026. doi: 10.1007/s13346-026-02051-7.
37. Guo Y, Xie YQ, Gao M, et al. Metabolic reprogramming of terminally exhausted CD8+ T cells by IL-10 enhances anti-tumor immunity. Nat Immunol. 2021; 22(6): 746–756. doi: 10.1038/s41590-021-00940-2.
38. Hanna BS, Llaó-Cid L, Iskar M, et al. Interleukin-10 receptor signaling promotes the maintenance of a PD-1int TCF-1+ CD8+ T cell population that sustains anti-tumor immunity. Immunity. 2021; 54(12): 2825–2841.e10. doi: 10.1016/j.immuni.2021.11.004.
39. Zhao Y, Chen J, Andreatta M, et al. IL-10-expressing CAR T cells resist dysfunction and mediate durable clearance of solid tumors and metastases. Nat Biotechnol. 2024; 42(11): 1693–1704. doi: 10.1038/s41587-023-02060-8.
40. Fu S, Naqash AR, Ayanambakkam A, et al. Safety and activity of Diakine DK210 (EGFR), a next-generation tumor-targeted IL-2 × IL-10 dual immunocytokine, in patients with advanced cancer: initial results of the phase 1 first-in-human trial. J Clin Oncol. 2024;42(16 Suppl): 2584. doi: 10.1200/JCO.2024.42.16_suppl.2584.
41. Naqash AR, Ayanambakkam A, Spira A, et al. Immune biomarkers from a phase 1 first-in-human trial treating advanced cancer patients with DK210 (EGFR). Presented at: 37th Annual Meeting of the Society for Immunotherapy of Cancer (SITC); November 6–10, 2024; Houston, TX. Abstract 43.